chore: bump version to 0.1.3

feat: add record spread syntax { ...base, field: val }
Adds spread operator for records, allowing concise record updates: let p2 = { ...p, x: 5.0 } Changes across the full pipeline: - Lexer: new DotDotDot (...) token - AST: optional spread field on Record variant - Parser: detect ... at start of record expression - Typechecker: merge spread record fields with explicit overrides - Interpreter: evaluate spread, overlay explicit fields - JS backend: emit native JS spread syntax - C backend: copy spread into temp, assign overrides - Formatter, linter, LSP, symbol table: propagate spread Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-02-18 23:06:10 -05:00 · 2026-02-18 23:05:27 -05:00 · 2026-02-18 23:05:12 -05:00 · 2026-02-18 23:05:03 -05:00 · 2026-02-18 22:01:41 -05:00 · 2026-02-18 21:16:44 -05:00
119 changed files with 20031 additions and 5323 deletions
--- a/.gitignore
+++ b/.gitignore
@@ -1,6 +1,14 @@
 /target
 /result
 # Claude Code project instructions
 CLAUDE.md
 # Build output
 _site/
 docs/*.html
 docs/*.css
 # Test binaries
 hello
 test_rc
--- a/CLAUDE.md
+++ b/CLAUDE.md
@@ -0,0 +1,177 @@
 # Lux Project Notes
 ## Development Environment
 This is a **Nix environment**. Tools like `cargo`, `rustc`, `clippy`, etc. are not available in the base shell.
 To run Rust/Cargo commands, use one of:
 ```bash
 nix develop --command cargo test
 nix develop --command cargo build
 nix develop --command cargo clippy
 nix develop --command cargo fmt
 ```
 Or enter the development shell first:
 ```bash
 nix develop
 # then run commands normally
 cargo test
 ```
 The `lux` binary can be run directly if already built:
 ```bash
 ./target/debug/lux test
 ./target/release/lux <file.lux>
 ```
 For additional tools not in the dev shell:
 ```bash
 nix-shell -p <program>
 ```
 ## Development Workflow
 When making changes:
 1. **Always run tests**: `cargo check && cargo test` - fix all errors and warnings
 2. **Lint the Lux code**: `./target/release/lux lint` - fix warnings
 3. **Check Lux code**: `./target/release/lux check` - type check + lint in one pass
 4. **Format Lux code**: `./target/release/lux fmt` - auto-format all .lux files
 5. **Write tests**: Add tests to cover new code
 6. **Document features**: Provide documentation and tutorials for new features/frameworks
 7. **Fix language limitations**: If you encounter parser/type system limitations, fix them (without regressions on guarantees or speed)
 8. **Git commits**: Always use `--no-gpg-sign` flag
 ### Post-work checklist (run after each committable change)
 **MANDATORY: Run the full validation script after every committable change:**
 ```bash
 ./scripts/validate.sh
 ```
 This script runs ALL of the following checks and will fail if any regress:
 1. `cargo check` — no Rust compilation errors
 2. `cargo test` — all Rust tests pass (currently 387)
 3. `cargo build --release` — release binary builds
 4. `lux test` on every package (path, frontmatter, xml, rss, markdown) — all 286 package tests pass
 5. `lux check` on every package — type checking + lint passes
 If `validate.sh` is not available or you need to run manually:
 ```bash
 nix develop --command cargo check       # No Rust errors
 nix develop --command cargo test        # All Rust tests pass
 nix develop --command cargo build --release  # Build release binary
 cd ../packages/path && ../../lang/target/release/lux test        # Package tests
 cd ../packages/frontmatter && ../../lang/target/release/lux test
 cd ../packages/xml && ../../lang/target/release/lux test
 cd ../packages/rss && ../../lang/target/release/lux test
 cd ../packages/markdown && ../../lang/target/release/lux test
 ```
 **Do NOT commit if any check fails.** Fix the issue first.
 ### Commit after every piece of work
 **After completing each logical unit of work, commit immediately.** This is NOT optional — every fix, feature, or change MUST be committed right away. Do not let changes accumulate uncommitted across multiple features. Each commit should be a single logical change (one feature, one bugfix, etc.). Use `--no-gpg-sign` flag for all commits.
 **Commit workflow:**
 1. Make the change
 2. Run `./scripts/validate.sh` (all 13 checks must pass)
 3. `git add` the relevant files
 4. `git commit --no-gpg-sign -m "type: description"` (use conventional commits: fix/feat/chore/docs)
 5. Move on to the next task
 **Never skip committing.** If you fixed a bug, commit it. If you added a feature, commit it. If you updated docs, commit it. Do not batch unrelated changes into one commit.
 **IMPORTANT: Always verify Lux code you write:**
 - Run with interpreter: `./target/release/lux file.lux`
 - Compile to binary: `./target/release/lux compile file.lux`
 - Both must work before claiming code is functional
 - The C backend has limited effect support (Console, File only - no HttpServer, Http, etc.)
 ## CLI Commands & Aliases
 | Command | Alias | Description |
 |---------|-------|-------------|
 | `lux fmt` | `lux f` | Format .lux files |
 | `lux test` | `lux t` | Run test suite |
 | `lux check` | `lux k` | Type check + lint |
 | `lux lint` | `lux l` | Lint only (with `--explain` for detailed help) |
 | `lux serve` | `lux s` | Static file server |
 | `lux compile` | `lux c` | Compile to binary |
 ## Documenting Lux Language Errors
 When working on any major task that involves writing Lux code, **document every language error, limitation, or surprising behavior** you encounter. This log is optimized for LLM consumption so future sessions can avoid repeating mistakes.
 **File:** Maintain an `ISSUES.md` in the relevant project directory (e.g., `~/src/blu-site/ISSUES.md`).
 **Format for each entry:**
 ```markdown
 ## Issue N: <Short descriptive title>
 **Category**: Parser limitation | Type checker gap | Missing feature | Runtime error | Documentation gap
 **Severity**: High | Medium | Low
 **Status**: Open | **Fixed** (commit hash or version)
 <1-2 sentence description of the problem>
 **Reproduction:**
 ```lux
 // Minimal code that triggers the issue
 ```
 **Error message:** `<exact error text>`
 **Workaround:** <how to accomplish the goal despite the limitation>
 **Fix:** <if fixed, what was changed and where>
 ```
 **Rules:**
 - Add new issues as you encounter them during any task
 - When a previously documented issue gets fixed, update its status to **Fixed** and note the commit/version
 - Remove entries that are no longer relevant (e.g., the feature was redesigned entirely)
 - Keep the summary table at the bottom of ISSUES.md in sync with the entries
 - Do NOT duplicate issues already documented -- check existing entries first
 ## Code Quality
 - Fix all compiler warnings before committing
 - Ensure all tests pass (currently 387 tests)
 - Add new tests when adding features
 - Keep examples and documentation in sync
 ## Lux Language Notes
 ### Top-level expressions
 Bare `run` expressions are not allowed at top-level. You must wrap them in a `let` binding:
 ```lux
 // WRONG: parse error
 run main() with {}
 // CORRECT
 let output = run main() with {}
 ```
 ### String methods
 Lux uses module-qualified function calls, not method syntax on primitives:
 ```lux
 // WRONG: not valid syntax
 path.endsWith(".html")
 // CORRECT
 String.endsWith(path, ".html")
 ```
 ### Available String functions
 Key string functions (all in `String.` namespace):
 - `String.length(s)` - get length
 - `String.startsWith(s, prefix)` - check prefix
 - `String.endsWith(s, suffix)` - check suffix
 - `String.split(s, delimiter)` - split into list
 - `String.join(list, delimiter)` - join list
 - `String.substring(s, start, end)` - extract substring
 - `String.indexOf(s, needle)` - find position (returns Option)
 - `String.replace(s, old, new)` - replace occurrences
 - `String.trim(s)` - trim whitespace
 - `String.toLower(s)` / `String.toUpper(s)` - case conversion
--- a/Cargo.lock
+++ b/Cargo.lock
@@ -135,16 +135,6 @@ dependencies = [
 "libc",
 ]
 [[package]]
 name = "core-foundation"
 version = "0.10.1"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "b2a6cd9ae233e7f62ba4e9353e81a88df7fc8a5987b8d445b4d90c879bd156f6"
 dependencies = [
 "core-foundation-sys",
 "libc",
 ]
 [[package]]
 name = "core-foundation-sys"
 version = "0.8.7"
@@ -235,7 +225,7 @@ source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "39cab71617ae0d63f51a36d69f866391735b51691dbda63cf6f96d042b63efeb"
 dependencies = [
 "libc",
- "windows-sys 0.61.2",
+ "windows-sys 0.59.0",
 ]
 [[package]]
@@ -297,21 +287,6 @@ version = "0.1.5"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "d9c4f5dac5e15c24eb999c26181a6ca40b39fe946cbe4c263c7209467bc83af2"
 [[package]]
 name = "foreign-types"
 version = "0.3.2"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "f6f339eb8adc052cd2ca78910fda869aefa38d22d5cb648e6485e4d3fc06f3b1"
 dependencies = [
 "foreign-types-shared",
 ]
 [[package]]
 name = "foreign-types-shared"
 version = "0.1.1"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "00b0228411908ca8685dba7fc2cdd70ec9990a6e753e89b6ac91a84c40fbaf4b"
 [[package]]
 name = "form_urlencoded"
 version = "1.2.2"
@@ -552,16 +527,17 @@ dependencies = [
 ]
 [[package]]
-name = "hyper-tls"
+name = "hyper-rustls"
-version = "0.5.0"
+version = "0.24.2"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "d6183ddfa99b85da61a140bea0efc93fdf56ceaa041b37d553518030827f9905"
+checksum = "ec3efd23720e2049821a693cbc7e65ea87c72f1c58ff2f9522ff332b1491e590"
 dependencies = [
- "bytes",
+ "futures-util",
 "http",
 "hyper",
- "native-tls",
+ "rustls",
 "tokio",
- "tokio-native-tls",
+ "tokio-rustls",
 ]
 [[package]]
@@ -794,7 +770,7 @@ dependencies = [
 [[package]]
 name = "lux"
-version = "0.1.0"
+version = "0.1.2"
 dependencies = [
 "lsp-server",
 "lsp-types",
@@ -843,23 +819,6 @@ dependencies = [
 "windows-sys 0.61.2",
 ]
 [[package]]
 name = "native-tls"
 version = "0.2.16"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "9d5d26952a508f321b4d3d2e80e78fc2603eaefcdf0c30783867f19586518bdc"
 dependencies = [
 "libc",
 "log",
 "openssl",
 "openssl-probe",
 "openssl-sys",
 "schannel",
 "security-framework",
 "security-framework-sys",
 "tempfile",
 ]
 [[package]]
 name = "nibble_vec"
 version = "0.1.0"
@@ -905,50 +864,6 @@ version = "1.21.3"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "42f5e15c9953c5e4ccceeb2e7382a716482c34515315f7b03532b8b4e8393d2d"
 [[package]]
 name = "openssl"
 version = "0.10.75"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "08838db121398ad17ab8531ce9de97b244589089e290a384c900cb9ff7434328"
 dependencies = [
 "bitflags 2.10.0",
 "cfg-if",
 "foreign-types",
 "libc",
 "once_cell",
 "openssl-macros",
 "openssl-sys",
 ]
 [[package]]
 name = "openssl-macros"
 version = "0.1.1"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "a948666b637a0f465e8564c73e89d4dde00d72d4d473cc972f390fc3dcee7d9c"
 dependencies = [
 "proc-macro2",
 "quote",
 "syn",
 ]
 [[package]]
 name = "openssl-probe"
 version = "0.2.1"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "7c87def4c32ab89d880effc9e097653c8da5d6ef28e6b539d313baaacfbafcbe"
 [[package]]
 name = "openssl-sys"
 version = "0.9.111"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "82cab2d520aa75e3c58898289429321eb788c3106963d0dc886ec7a5f4adc321"
 dependencies = [
 "cc",
 "libc",
 "pkg-config",
 "vcpkg",
 ]
 [[package]]
 name = "parking_lot"
 version = "0.12.5"
@@ -1203,15 +1118,15 @@ dependencies = [
 "http",
 "http-body",
 "hyper",
- "hyper-tls",
+ "hyper-rustls",
 "ipnet",
 "js-sys",
 "log",
 "mime",
 "native-tls",
 "once_cell",
 "percent-encoding",
 "pin-project-lite",
 "rustls",
 "rustls-pemfile",
 "serde",
 "serde_json",
@@ -1219,15 +1134,30 @@ dependencies = [
 "sync_wrapper",
 "system-configuration",
 "tokio",
- "tokio-native-tls",
+ "tokio-rustls",
 "tower-service",
 "url",
 "wasm-bindgen",
 "wasm-bindgen-futures",
 "web-sys",
 "webpki-roots",
 "winreg",
 ]
 [[package]]
 name = "ring"
 version = "0.17.14"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "a4689e6c2294d81e88dc6261c768b63bc4fcdb852be6d1352498b114f61383b7"
 dependencies = [
 "cc",
 "cfg-if",
 "getrandom 0.2.17",
 "libc",
 "untrusted",
 "windows-sys 0.52.0",
 ]
 [[package]]
 name = "rusqlite"
 version = "0.31.0"
@@ -1252,7 +1182,19 @@ dependencies = [
 "errno",
 "libc",
 "linux-raw-sys",
- "windows-sys 0.61.2",
+ "windows-sys 0.59.0",
 ]
 [[package]]
 name = "rustls"
 version = "0.21.12"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "3f56a14d1f48b391359b22f731fd4bd7e43c97f3c50eee276f3aa09c94784d3e"
 dependencies = [
 "log",
 "ring",
 "rustls-webpki",
 "sct",
 ]
 [[package]]
@@ -1264,6 +1206,16 @@ dependencies = [
 "base64 0.21.7",
 ]
 [[package]]
 name = "rustls-webpki"
 version = "0.101.7"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "8b6275d1ee7a1cd780b64aca7726599a1dbc893b1e64144529e55c3c2f745765"
 dependencies = [
 "ring",
 "untrusted",
 ]
 [[package]]
 name = "rustversion"
 version = "1.0.22"
@@ -1298,15 +1250,6 @@ version = "1.0.23"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "9774ba4a74de5f7b1c1451ed6cd5285a32eddb5cccb8cc655a4e50009e06477f"
 [[package]]
 name = "schannel"
 version = "0.1.28"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "891d81b926048e76efe18581bf793546b4c0eaf8448d72be8de2bbee5fd166e1"
 dependencies = [
 "windows-sys 0.61.2",
 ]
 [[package]]
 name = "scopeguard"
 version = "1.2.0"
@@ -1314,26 +1257,13 @@ source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "94143f37725109f92c262ed2cf5e59bce7498c01bcc1502d7b9afe439a4e9f49"
 [[package]]
-name = "security-framework"
+name = "sct"
-version = "3.6.0"
+version = "0.7.1"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "d17b898a6d6948c3a8ee4372c17cb384f90d2e6e912ef00895b14fd7ab54ec38"
+checksum = "da046153aa2352493d6cb7da4b6e5c0c057d8a1d0a9aa8560baffdd945acd414"
 dependencies = [
- "bitflags 2.10.0",
+ "ring",
- "core-foundation 0.10.1",
+ "untrusted",
 "core-foundation-sys",
 "libc",
 "security-framework-sys",
 ]
 [[package]]
 name = "security-framework-sys"
 version = "2.16.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "321c8673b092a9a42605034a9879d73cb79101ed5fd117bc9a597b89b4e9e61a"
 dependencies = [
 "core-foundation-sys",
 "libc",
 ]
 [[package]]
@@ -1521,7 +1451,7 @@ source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "ba3a3adc5c275d719af8cb4272ea1c4a6d668a777f37e115f6d11ddbc1c8e0e7"
 dependencies = [
 "bitflags 1.3.2",
- "core-foundation 0.9.4",
+ "core-foundation",
 "system-configuration-sys",
 ]
@@ -1545,7 +1475,7 @@ dependencies = [
 "getrandom 0.4.1",
 "once_cell",
 "rustix",
- "windows-sys 0.61.2",
+ "windows-sys 0.59.0",
 ]
 [[package]]
@@ -1619,16 +1549,6 @@ dependencies = [
 "windows-sys 0.61.2",
 ]
 [[package]]
 name = "tokio-native-tls"
 version = "0.3.1"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "bbae76ab933c85776efabc971569dd6119c580d8f5d448769dec1764bf796ef2"
 dependencies = [
 "native-tls",
 "tokio",
 ]
 [[package]]
 name = "tokio-postgres"
 version = "0.7.16"
@@ -1655,6 +1575,16 @@ dependencies = [
 "whoami",
 ]
 [[package]]
 name = "tokio-rustls"
 version = "0.24.1"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "c28327cf380ac148141087fbfb9de9d7bd4e84ab5d2c28fbc911d753de8a7081"
 dependencies = [
 "rustls",
 "tokio",
 ]
 [[package]]
 name = "tokio-util"
 version = "0.7.18"
@@ -1750,6 +1680,12 @@ version = "0.2.6"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "ebc1c04c71510c7f702b52b7c350734c9ff1295c464a03335b00bb84fc54f853"
 [[package]]
 name = "untrusted"
 version = "0.9.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "8ecb6da28b8a351d773b68d5825ac39017e680750f980f3a1a85cd8dd28a47c1"
 [[package]]
 name = "url"
 version = "2.5.8"
@@ -1941,6 +1877,12 @@ dependencies = [
 "wasm-bindgen",
 ]
 [[package]]
 name = "webpki-roots"
 version = "0.25.4"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "5f20c57d8d7db6d3b86154206ae5d8fba62dd39573114de97c2cb0578251f8e1"
 [[package]]
 name = "whoami"
 version = "2.1.1"
--- a/Cargo.toml
+++ b/Cargo.toml
@@ -1,6 +1,6 @@
 [package]
 name = "lux"
-version = "0.1.0"
+version = "0.1.3"
 edition = "2021"
 description = "A functional programming language with first-class effects, schema evolution, and behavioral types"
 license = "MIT"
@@ -13,7 +13,7 @@ lsp-types = "0.94"
 serde = { version = "1", features = ["derive"] }
 serde_json = "1"
 rand = "0.8"
-reqwest = { version = "0.11", features = ["blocking", "json"] }
+reqwest = { version = "0.11", default-features = false, features = ["blocking", "json", "rustls-tls"] }
 tiny_http = "0.12"
 rusqlite = { version = "0.31", features = ["bundled"] }
 postgres = "0.19"
--- a/PACKAGES.md
+++ b/PACKAGES.md
@@ -0,0 +1,367 @@
 # Lux Package Ecosystem Plan
 ## Current State
 ### Stdlib (built-in)
 | Module | Coverage |
 |--------|----------|
 | String | Comprehensive (split, join, trim, indexOf, replace, etc.) |
 | List | Good (map, filter, fold, head, tail, concat, range, find, any, all, take, drop) |
 | Option | Basic (map, flatMap, getOrElse, isSome, isNone) |
 | Result | Basic (map, flatMap, getOrElse, isOk, isErr) |
 | Math | Basic (abs, min, max, sqrt, pow, floor, ceil, round) |
 | Json | Comprehensive (parse, stringify, get, typed extractors, constructors) |
 | File | Good (read, write, append, exists, delete, readDir, isDir, mkdir) |
 | Console | Good (print, read, readLine, readInt) |
 | Process | Good (exec, execStatus, env, args, exit, cwd) |
 | Http | Basic (get, post, put, delete, setHeader) |
 | HttpServer | Basic (listen, accept, respond) |
 | Time | Minimal (now, sleep) |
 | Random | Basic (int, float, bool) |
 | Sql | Good (SQLite: open, query, execute, transactions) |
 | Postgres | Good (connect, query, execute, transactions) |
 | Schema | Niche (versioned data migration) |
 | Test | Good (assert, assertEqual, assertTrue) |
 | Concurrent | Experimental (spawn, await, yield, cancel) |
 | Channel | Experimental (create, send, receive) |
 ### Registry (pkgs.lux) - 3 packages
 | Package | Version | Notes |
 |---------|---------|-------|
 | json | 1.0.0 | Wraps stdlib Json with convenience functions (getPath, getString, etc.) |
 | http-client | 0.1.0 | Wraps stdlib Http with JSON helpers, URL encoding |
 | testing | 0.1.0 | Wraps stdlib Test with describe/it structure |
 ---
 ## Gap Analysis
 ### What's Missing vs Other Languages
 Compared to ecosystems like Rust/cargo, Go, Python, Elm, Gleam:
 | Category | Gap | Impact | Notes |
 |----------|-----|--------|-------|
 | **Collections** | No HashMap, Set, Queue, Stack | Critical | List-of-pairs with O(n) lookup is the only option |
 | **Sorting** | No List.sort or List.sortBy | High | Must implement insertion sort manually |
 | **Date/Time** | Only `Time.now()` (epoch ms), no parsing/formatting | High | blu-site does string-based date formatting manually |
 | **Markdown** | No markdown parser | High | blu-site has 300+ lines of hand-rolled markdown |
 | **XML/RSS** | No XML generation | High | Can't generate RSS feeds or sitemaps |
 | **Regex** | No pattern matching on strings | High | Character-by-character scanning required |
 | **Path** | No file path utilities | Medium | basename/dirname manually reimplemented |
 | **YAML/TOML** | No config file parsing (beyond JSON) | Medium | Frontmatter parsing is manual |
 | **Template** | No string templating | Medium | HTML built via raw string concatenation |
 | **URL** | No URL parsing/encoding | Medium | http-client has basic urlEncode but no parser |
 | **Crypto** | No hashing (SHA256, etc.) | Medium | Can't do checksums, content hashing |
 | **Base64** | No encoding/decoding | Low | Needed for data URIs, some auth |
 | **CSV** | No CSV parsing | Low | Common data format |
 | **UUID** | No UUID generation | Low | Useful for IDs |
 | **Logging** | No structured logging | Low | Just Console.print |
 | **CLI** | No argument parsing library | Low | Manual arg handling |
 ### What Should Be Stdlib vs Package
 **Should be stdlib additions** (too fundamental to be packages):
 - HashMap / Map type (requires runtime support)
 - List.sort / List.sortBy (fundamental operation)
 - Better Time module (date parsing, formatting)
 - Regex (needs runtime/C support for performance)
 - Path module (cross-platform file path handling)
 **Should be packages** (application-level, opinionated, composable):
 - markdown
 - xml
 - rss/atom
 - frontmatter
 - template
 - csv
 - crypto
 - ssg (static site generator framework)
 ---
 ## Priority Package Plans
 Ordered by what unblocks blu-site fixes first, then general ecosystem value.
 ---
 ### Package 1: `markdown` (Priority: HIGHEST)
 **Why:** The 300-line markdown parser in blu-site's main.lux is general-purpose code that belongs in a reusable package. It's also the most complex part of blu-site and has known bugs (e.g., `### ` inside list items renders literally).
 **Scope:**
 ```
 markdown/
  lux.toml
  lib.lux          # Public API: parse, parseInline
  src/
    inline.lux     # Inline parsing (bold, italic, links, images, code)
    block.lux      # Block parsing (headings, lists, code blocks, blockquotes, hr)
    types.lux      # AST types (optional - could emit HTML directly)
 ```
 **Public API:**
 ```lux
 // Convert markdown string to HTML string
 pub fn toHtml(markdown: String): String
 // Convert inline markdown only (no blocks)
 pub fn inlineToHtml(text: String): String
 // Escape HTML entities
 pub fn escapeHtml(s: String): String
 ```
 **Improvements over current blu-site code:**
 - Fix heading-inside-list-item rendering (`- ### Title` should work)
 - Support nested lists (currently flat only)
 - Support reference-style links `[text][ref]`
 - Handle edge cases (empty lines in code blocks, nested blockquotes)
 - Proper HTML entity escaping in more contexts
 **Depends on:** Nothing (pure string processing)
 **Estimated size:** ~400-500 lines of Lux
 ---
 ### Package 2: `xml` (Priority: HIGH)
 **Why:** Needed for RSS/Atom feed generation, sitemap.xml, and robots.txt generation. General-purpose XML builder that doesn't try to parse XML (which would need regex), just emits it.
 **Scope:**
 ```
 xml/
  lux.toml
  lib.lux          # Public API: element, document, serialize
 ```
 **Public API:**
 ```lux
 type XmlNode =
    | Element(String, List<XmlAttr>, List<XmlNode>)
    | Text(String)
    | CData(String)
    | Comment(String)
    | Declaration(String, String)  // version, encoding
 type XmlAttr =
    | Attr(String, String)
 // Build an XML element
 pub fn element(tag: String, attrs: List<XmlAttr>, children: List<XmlNode>): XmlNode
 // Build a text node (auto-escapes)
 pub fn text(content: String): XmlNode
 // Build a CDATA section
 pub fn cdata(content: String): XmlNode
 // Serialize XML tree to string
 pub fn serialize(node: XmlNode): String
 // Serialize with XML declaration header
 pub fn document(version: String, encoding: String, root: XmlNode): String
 // Convenience: self-closing element
 pub fn selfClosing(tag: String, attrs: List<XmlAttr>): XmlNode
 ```
 **Depends on:** Nothing
 **Estimated size:** ~150-200 lines
 ---
 ### Package 3: `rss` (Priority: HIGH)
 **Why:** Directly needed for blu-site's #6 priority fix (add RSS feed). Builds on `xml` package.
 **Scope:**
 ```
 rss/
  lux.toml         # depends on xml
  lib.lux          # Public API: feed, item, toXml, toAtom
 ```
 **Public API:**
 ```lux
 type FeedInfo =
    | FeedInfo(String, String, String, String, String)
    // title, link, description, language, lastBuildDate
 type FeedItem =
    | FeedItem(String, String, String, String, String, String)
    // title, link, description, pubDate, guid, categories (comma-separated)
 // Generate RSS 2.0 XML string
 pub fn toRss(info: FeedInfo, items: List<FeedItem>): String
 // Generate Atom 1.0 XML string
 pub fn toAtom(info: FeedInfo, items: List<FeedItem>): String
 ```
 **Depends on:** `xml`
 **Estimated size:** ~100-150 lines
 ---
 ### Package 4: `frontmatter` (Priority: HIGH)
 **Why:** blu-site has ~50 lines of fragile frontmatter parsing. This is a common need for any content-driven Lux project. The current parser uses `String.indexOf(line, ": ")` which breaks on values containing `: `.
 **Scope:**
 ```
 frontmatter/
  lux.toml
  lib.lux          # Public API: parse
 ```
 **Public API:**
 ```lux
 type FrontmatterResult =
    | FrontmatterResult(List<(String, String)>, String)
    // key-value pairs, remaining body
 // Parse frontmatter from a string (--- delimited YAML-like header)
 pub fn parse(content: String): FrontmatterResult
 // Get a value by key from parsed frontmatter
 pub fn get(pairs: List<(String, String)>, key: String): Option<String>
 // Get a value or default
 pub fn getOrDefault(pairs: List<(String, String)>, key: String, default: String): String
 // Parse a space-separated tag string into a list
 pub fn parseTags(tagString: String): List<String>
 ```
 **Improvements over current blu-site code:**
 - Handle values with `: ` in them (only split on first `: `)
 - Handle multi-line values (indented continuation)
 - Handle quoted values with embedded newlines
 - Strip quotes from values consistently
 **Depends on:** Nothing
 **Estimated size:** ~100-150 lines
 ---
 ### Package 5: `path` (Priority: MEDIUM)
 **Why:** blu-site manually implements `basename` and `dirname`. Any file-processing Lux program needs these. Tiny but universally useful.
 **Scope:**
 ```
 path/
  lux.toml
  lib.lux
 ```
 **Public API:**
 ```lux
 // Get filename from path: "/foo/bar.txt" -> "bar.txt"
 pub fn basename(p: String): String
 // Get directory from path: "/foo/bar.txt" -> "/foo"
 pub fn dirname(p: String): String
 // Get file extension: "file.txt" -> "txt", "file" -> ""
 pub fn extension(p: String): String
 // Remove file extension: "file.txt" -> "file"
 pub fn stem(p: String): String
 // Join path segments: join("foo", "bar") -> "foo/bar"
 pub fn join(a: String, b: String): String
 // Normalize path: "foo//bar/../baz" -> "foo/baz"
 pub fn normalize(p: String): String
 // Check if path is absolute
 pub fn isAbsolute(p: String): Bool
 ```
 **Depends on:** Nothing
 **Estimated size:** ~80-120 lines
 ---
 ### Package 6: `sitemap` (Priority: MEDIUM)
 **Why:** Directly needed for blu-site's #9 priority fix. Simple package that generates sitemap.xml.
 **Scope:**
 ```
 sitemap/
  lux.toml         # depends on xml
  lib.lux
 ```
 **Public API:**
 ```lux
 type SitemapEntry =
    | SitemapEntry(String, String, String, String)
    // url, lastmod (ISO date), changefreq, priority
 // Generate sitemap.xml string
 pub fn generate(entries: List<SitemapEntry>): String
 // Generate a simple robots.txt pointing to the sitemap
 pub fn robotsTxt(sitemapUrl: String): String
 ```
 **Depends on:** `xml`
 **Estimated size:** ~50-70 lines
 ---
 ### Package 7: `ssg` (Priority: LOW - future)
 **Why:** Once markdown, frontmatter, rss, sitemap, and path packages exist, the remaining logic in blu-site's main.lux is generic SSG framework code: read content dirs, parse posts, sort by date, generate section indexes, generate tag pages, copy static assets. This could be extracted into a framework package that other Lux users could use to build their own static sites.
 **This should wait** until the foundation packages above are stable and battle-tested through blu-site usage.
 ---
 ## Non-Package Stdlib Improvements Needed
 These gaps are too fundamental to be packages and should be added to the Lux language itself:
 ### HashMap (Critical)
 Every package above that needs key-value lookups (frontmatter, xml attributes, etc.) is working around the lack of HashMap with `List<(String, String)>`. This is O(n) per lookup and makes code verbose. A stdlib `Map` module would transform the ecosystem.
 ### List.sort / List.sortBy (High)
 blu-site implements insertion sort manually. Every content-driven app needs sorting. This should be a stdlib function.
 ### Time.format / Time.parse (High)
 blu-site manually parses "2025-01-15" by substring extraction and maps month numbers to names. A proper date/time library (even just ISO 8601 parsing and basic formatting) would help every package above.
 ---
 ## Implementation Order
 ```
 Phase 1 (unblock blu-site fixes):
  1. markdown     - extract from blu-site, fix bugs, publish
  2. frontmatter  - extract from blu-site, improve robustness
  3. path         - tiny, universally useful
  4. xml          - needed by rss and sitemap
 Phase 2 (complete blu-site features):
  5. rss          - depends on xml
  6. sitemap      - depends on xml
 Phase 3 (ecosystem growth):
  7. template     - string templating (mustache-like)
  8. csv          - data processing
  9. cli          - argument parsing
  10. ssg         - framework extraction from blu-site
 ```
 Each package should be developed in its own directory under `~/src/`, published to the git.qrty.ink registry, and tested by integrating it into blu-site.
--- a/README.md
+++ b/README.md
@@ -2,15 +2,22 @@
 A functional programming language with first-class effects, schema evolution, and behavioral types.
-## Vision
+## Philosophy
-Most programming languages treat three critical concerns as afterthoughts:
+**Make the important things visible.**
-1. **Effects** — What can this code do? (Hidden, untraceable, untestable)
+Most languages hide what matters most: what code can do (effects), how data changes over time (schema evolution), and what guarantees functions provide (behavioral properties). Lux makes all three first-class, compiler-checked language features.
 2. **Data Evolution** — Types change, data persists. (Manual migrations, runtime failures)
 3. **Behavioral Properties** — Is this idempotent? Does it terminate? (Comments and hope)
-Lux makes these first-class language features. The compiler knows what your code does, how your data evolves, and what properties your functions guarantee.
+| Principle | What it means |
 |-----------|--------------|
 | **Explicit over implicit** | Effects in types — see what code does |
 | **Composition over configuration** | No DI frameworks — effects compose naturally |
 | **Safety without ceremony** | Type inference + explicit signatures where they matter |
 | **Practical over academic** | Familiar syntax, ML semantics, no monads |
 | **One right way** | Opinionated formatter, integrated tooling, built-in test framework |
 | **Tools are the language** | `lux fmt/lint/check/test/compile` — one binary, not seven tools |
 See [docs/PHILOSOPHY.md](./docs/PHILOSOPHY.md) for the full philosophy with language comparisons and design rationale.
 ## Core Principles
@@ -144,6 +151,7 @@ fn main(): Unit with {Console} =
 - String, List, Option, Result, Math, JSON modules
 - Console, File, Http, Random, Time, Process effects
 - SQL effect (SQLite with transactions)
 - PostgreSQL effect (connection pooling ready)
 - DOM effect (40+ browser operations)
 See:
--- a/build.rs
+++ b/build.rs
@@ -0,0 +1,38 @@
 use std::path::PathBuf;
 fn main() {
    // Capture the absolute C compiler path at build time so the binary is self-contained.
    // This is critical for Nix builds where cc/gcc live in /nix/store paths.
    let cc_path = std::env::var("CC").ok()
        .filter(|s| !s.is_empty())
        .and_then(|s| resolve_absolute(&s))
        .or_else(|| find_in_path("cc"))
        .or_else(|| find_in_path("gcc"))
        .or_else(|| find_in_path("clang"))
        .unwrap_or_default();
    println!("cargo:rustc-env=LUX_CC_PATH={}", cc_path);
    println!("cargo:rerun-if-env-changed=CC");
    println!("cargo:rerun-if-env-changed=PATH");
 }
 /// Resolve a command name to its absolute path by searching PATH.
 fn find_in_path(cmd: &str) -> Option<String> {
    let path_var = std::env::var("PATH").ok()?;
    for dir in path_var.split(':') {
        let candidate = PathBuf::from(dir).join(cmd);
        if candidate.is_file() {
            return Some(candidate.to_string_lossy().into_owned());
        }
    }
    None
 }
 /// If the path is already absolute and exists, return it. Otherwise search PATH.
 fn resolve_absolute(cmd: &str) -> Option<String> {
    let p = PathBuf::from(cmd);
    if p.is_absolute() && p.is_file() {
        return Some(cmd.to_string());
    }
    find_in_path(cmd)
 }
--- a/docs/COMPILER_OPTIMIZATIONS.md
+++ b/docs/COMPILER_OPTIMIZATIONS.md
@@ -0,0 +1,400 @@
 # Compiler Optimizations from Behavioral Types
 This document describes optimization opportunities enabled by Lux's behavioral type system. When functions are annotated with properties like `is pure`, `is total`, `is idempotent`, `is deterministic`, or `is commutative`, the compiler gains knowledge that enables aggressive optimizations.
 ## Overview
 | Property | Key Optimizations |
 |----------|-------------------|
 | `is pure` | Memoization, CSE, dead code elimination, auto-parallelization |
 | `is total` | No exception handling, aggressive inlining, loop unrolling |
 | `is deterministic` | Result caching, test reproducibility, parallel execution |
 | `is idempotent` | Duplicate call elimination, retry optimization |
 | `is commutative` | Argument reordering, parallel reduction, algebraic simplification |
 ## Pure Function Optimizations
 When a function is marked `is pure`:
 ### 1. Memoization (Automatic Caching)
 ```lux
 fn fib(n: Int): Int is pure =
    if n <= 1 then n else fib(n - 1) + fib(n - 2)
 ```
 **Optimization**: The compiler can automatically memoize results. Since `fib` is pure, `fib(10)` will always return the same value, so we can cache it.
 **Implementation approach**:
 - Maintain a hash map of argument → result mappings
 - Before computing, check if result exists
 - Store results after computation
 - Use LRU eviction for memory management
 **Impact**: Reduces exponential recursive calls to linear time.
 ### 2. Common Subexpression Elimination (CSE)
 ```lux
 fn compute(x: Int): Int is pure =
    expensive(x) + expensive(x)  // Same call twice
 ```
 **Optimization**: The compiler recognizes both calls are identical and computes `expensive(x)` only once.
 **Transformed to**:
 ```lux
 fn compute(x: Int): Int is pure =
    let temp = expensive(x)
    temp + temp
 ```
 **Impact**: Eliminates redundant computation.
 ### 3. Dead Code Elimination
 ```lux
 fn example(): Int is pure = {
    let unused = expensiveComputation()  // Result not used
    42
 }
 ```
 **Optimization**: Since `expensiveComputation` is pure (no side effects), and its result is unused, the entire call can be eliminated.
 **Impact**: Removes unnecessary work.
 ### 4. Auto-Parallelization
 ```lux
 fn processAll(items: List<Item>): List<Result> is pure =
    List.map(items, processItem)  // processItem is pure
 ```
 **Optimization**: Since `processItem` is pure, each invocation is independent. The compiler can automatically parallelize the map operation.
 **Implementation approach**:
 - Detect pure functions in map/filter/fold operations
 - Split work across available cores
 - Merge results (order-preserving for map)
 **Impact**: Linear speedup with core count for CPU-bound operations.
 ### 5. Speculative Execution
 ```lux
 fn decide(cond: Bool, a: Int, b: Int): Int is pure =
    if cond then computeA(a) else computeB(b)
 ```
 **Optimization**: Both branches can be computed in parallel before the condition is known, since neither has side effects.
 **Impact**: Reduced latency when condition evaluation is slow.
 ## Total Function Optimizations
 When a function is marked `is total`:
 ### 1. Exception Handling Elimination
 ```lux
 fn safeCompute(x: Int): Int is total =
    complexCalculation(x)
 ```
 **Optimization**: No try/catch blocks needed around calls to `safeCompute`. The compiler knows it will never throw or fail.
 **Generated code difference**:
 ```c
 // Without is total - needs error checking
 Result result = safeCompute(x);
 if (result.is_error) { handle_error(); }
 // With is total - direct call
 int result = safeCompute(x);
 ```
 **Impact**: Reduced code size, better branch prediction.
 ### 2. Aggressive Inlining
 ```lux
 fn square(x: Int): Int is total = x * x
 fn sumOfSquares(a: Int, b: Int): Int is total =
    square(a) + square(b)
 ```
 **Optimization**: Total functions are safe to inline aggressively because:
 - They won't change control flow unexpectedly
 - They won't introduce exception handling complexity
 - Their termination is guaranteed
 **Impact**: Eliminates function call overhead, enables further optimizations.
 ### 3. Loop Unrolling
 ```lux
 fn sumList(xs: List<Int>): Int is total =
    List.fold(xs, 0, fn(acc: Int, x: Int): Int is total => acc + x)
 ```
 **Optimization**: When the list size is known at compile time and the fold function is total, the loop can be fully unrolled.
 **Impact**: Eliminates loop overhead, enables vectorization.
 ### 4. Termination Assumptions
 ```lux
 fn processRecursive(data: Tree): Result is total =
    match data {
        Leaf(v) => Result.single(v),
        Node(left, right) => {
            let l = processRecursive(left)
            let r = processRecursive(right)
            Result.merge(l, r)
        }
    }
 ```
 **Optimization**: The compiler can assume this recursion terminates, allowing optimizations like:
 - Converting recursion to iteration
 - Allocating fixed stack space
 - Tail call optimization
 **Impact**: Stack safety, predictable memory usage.
 ## Deterministic Function Optimizations
 When a function is marked `is deterministic`:
 ### 1. Compile-Time Evaluation
 ```lux
 fn hashConstant(s: String): Int is deterministic = computeHash(s)
 let key = hashConstant("api_key")  // Constant input
 ```
 **Optimization**: Since the input is a compile-time constant and the function is deterministic, the result can be computed at compile time.
 **Transformed to**:
 ```lux
 let key = 7823491  // Pre-computed
 ```
 **Impact**: Zero runtime cost for constant computations.
 ### 2. Result Caching Across Runs
 ```lux
 fn parseConfig(path: String): Config is deterministic with {File} =
    Json.parse(File.read(path))
 ```
 **Optimization**: Results can be cached persistently. If the file hasn't changed, the cached result is valid.
 **Implementation approach**:
 - Hash inputs (including file contents)
 - Store results in persistent cache
 - Validate cache on next run
 **Impact**: Faster startup times, reduced I/O.
 ### 3. Reproducible Parallel Execution
 ```lux
 fn renderImages(images: List<Image>): List<Bitmap> is deterministic =
    List.map(images, render)
 ```
 **Optimization**: Deterministic parallel execution guarantees same results regardless of scheduling order. This enables:
 - Work stealing without synchronization concerns
 - Speculative execution without rollback complexity
 - Distributed computation across machines
 **Impact**: Easier parallelization, simpler distributed systems.
 ## Idempotent Function Optimizations
 When a function is marked `is idempotent`:
 ### 1. Duplicate Call Elimination
 ```lux
 fn setFlag(config: Config, flag: Bool): Config is idempotent =
    { ...config, enabled: flag }
 fn configure(c: Config): Config is idempotent =
    c |> setFlag(true) |> setFlag(true) |> setFlag(true)
 ```
 **Optimization**: Multiple consecutive calls with the same arguments can be collapsed to one.
 **Transformed to**:
 ```lux
 fn configure(c: Config): Config is idempotent =
    setFlag(c, true)
 ```
 **Impact**: Eliminates redundant operations.
 ### 2. Retry Optimization
 ```lux
 fn sendRequest(data: Request): Response is idempotent with {Http} =
    Http.put("/api/resource", data)
 fn reliableSend(data: Request): Response with {Http} =
    retry(3, fn(): Response => sendRequest(data))
 ```
 **Optimization**: The retry mechanism knows the operation is safe to retry without side effects accumulating.
 **Implementation approach**:
 - No need for transaction logs
 - No need for "already processed" checks
 - Simple retry loop
 **Impact**: Simpler error recovery, reduced complexity.
 ### 3. Convergent Computation
 ```lux
 fn normalize(value: Float): Float is idempotent =
    clamp(round(value, 2), 0.0, 1.0)
 ```
 **Optimization**: In iterative algorithms, the compiler can detect when a value has converged (applying the function no longer changes it).
 ```lux
 // Can terminate early when values stop changing
 fn iterateUntilStable(values: List<Float>): List<Float> =
    let normalized = List.map(values, normalize)
    if normalized == values then values
    else iterateUntilStable(normalized)
 ```
 **Impact**: Early termination of iterative algorithms.
 ## Commutative Function Optimizations
 When a function is marked `is commutative`:
 ### 1. Argument Reordering
 ```lux
 fn multiply(a: Int, b: Int): Int is commutative = a * b
 // In a computation
 multiply(expensiveA(), cheapB())
 ```
 **Optimization**: Evaluate the cheaper argument first to enable short-circuit optimizations or better register allocation.
 **Impact**: Improved instruction scheduling.
 ### 2. Parallel Reduction
 ```lux
 fn add(a: Int, b: Int): Int is commutative = a + b
 fn sum(xs: List<Int>): Int =
    List.fold(xs, 0, add)
 ```
 **Optimization**: Since `add` is commutative (and associative), the fold can be parallelized:
 ```
 [1, 2, 3, 4, 5, 6, 7, 8]
    ↓ parallel reduce
 [(1+2), (3+4), (5+6), (7+8)]
    ↓ parallel reduce
 [(3+7), (11+15)]
    ↓ parallel reduce
 [36]
 ```
 **Impact**: O(log n) parallel reduction instead of O(n) sequential.
 ### 3. Algebraic Simplification
 ```lux
 fn add(a: Int, b: Int): Int is commutative = a + b
 // Expression: add(x, add(y, z))
 ```
 **Optimization**: Commutative operations can be reordered for simplification:
 - `add(x, 0)` → `x`
 - `add(add(x, 1), add(y, 1))` → `add(add(x, y), 2)`
 **Impact**: Constant folding, strength reduction.
 ## Combined Property Optimizations
 Properties can be combined for even more powerful optimizations:
 ### Pure + Deterministic + Total
 ```lux
 fn computeKey(data: String): Int
    is pure
    is deterministic
    is total = {
    // Hash computation
    List.fold(String.chars(data), 0, fn(acc: Int, c: Char): Int =>
        acc * 31 + Char.code(c))
 }
 ```
 **Enabled optimizations**:
 - Compile-time evaluation for constants
 - Automatic memoization at runtime
 - Parallel execution in batch operations
 - No exception handling needed
 - Safe to inline anywhere
 ### Idempotent + Commutative
 ```lux
 fn setUnionItem<T>(set: Set<T>, item: T): Set<T>
    is idempotent
    is commutative = {
    Set.add(set, item)
 }
 ```
 **Enabled optimizations**:
 - Parallel set building (order doesn't matter)
 - Duplicate insertions are free (idempotent)
 - Reorder insertions for cache locality
 ## Implementation Status
 | Optimization | Status |
 |--------------|--------|
 | Pure: CSE | Planned |
 | Pure: Dead code elimination | Partial (basic) |
 | Pure: Auto-parallelization | Planned |
 | Total: Exception elimination | Planned |
 | Total: Aggressive inlining | Partial |
 | Deterministic: Compile-time eval | Planned |
 | Idempotent: Duplicate elimination | Planned |
 | Commutative: Parallel reduction | Planned |
 ## Adding New Optimizations
 When implementing new optimizations based on behavioral types:
 1. **Verify the property is correct**: The optimization is only valid if the property holds
 2. **Consider combinations**: Multiple properties together enable more optimizations
 3. **Measure impact**: Profile before and after to ensure benefit
 4. **Handle `assume`**: Functions using `assume` bypass verification but still enable optimizations (risk is on the programmer)
 ## Future Work
 1. **Inter-procedural analysis**: Track properties across function boundaries
 2. **Automatic property inference**: Derive properties when not explicitly stated
 3. **Profile-guided optimization**: Use runtime data to decide when to apply optimizations
 4. **LLVM integration**: Pass behavioral hints to LLVM for backend optimizations
--- a/docs/COMPREHENSIVE_ROADMAP.md
+++ b/docs/COMPREHENSIVE_ROADMAP.md
--- a/docs/PHILOSOPHY.md
+++ b/docs/PHILOSOPHY.md
@@ -0,0 +1,449 @@
 # The Lux Philosophy
 ## In One Sentence
 **Make the important things visible.**
 ## The Three Pillars
 Most programming languages hide the things that matter most in production:
 1. **What can this code do?** — Side effects are invisible in function signatures
 2. **How does data change over time?** — Schema evolution is a deployment problem, not a language one
 3. **What guarantees does this code provide?** — Properties like idempotency live in comments and hope
 Lux makes all three first-class, compiler-checked language features.
 ---
 ## Core Principles
 ### 1. Explicit Over Implicit
 Every function signature tells you what it does:
 ```lux
 fn processOrder(order: Order): Receipt with {Database, Email, Logger}
 ```
 You don't need to read the body, trace call chains, or check documentation. The signature *is* the documentation. Code review becomes: "should this function really send emails?"
 **What this means in practice:**
 - Effects are declared in types, not hidden behind interfaces
 - No dependency injection frameworks — just swap handlers
 - No mocking libraries — test with different effect implementations
 - No "spooky action at a distance" — if a function can fail, its type says so
 **How this compares:**
 | Language | Side effects | Lux equivalent |
 |----------|-------------|----------------|
 | JavaScript | Anything, anywhere, silently | `with {Console, Http, File}` |
 | Python | Implicit, discovered by reading code | Effect declarations in signature |
 | Java | Checked exceptions (partial), DI frameworks | Effects + handlers |
 | Go | Return error values (partial) | `with {Fail}` or `Result` |
 | Rust | `unsafe` blocks, `Result`/`Option` | Effects for I/O, Result for values |
 | Haskell | Monad transformers (explicit but heavy) | Effects (explicit and lightweight) |
 | Koka | Algebraic effects (similar) | Same family, more familiar syntax |
 ### 2. Composition Over Configuration
 Things combine naturally without glue code:
 ```lux
 // Multiple effects compose by listing them
 fn sync(id: UserId): User with {Database, Http, Logger} = ...
 // Handlers compose by providing them
 run sync(id) with {
    Database = postgres(conn),
    Http = realHttp,
    Logger = consoleLogger
 }
 ```
 No monad transformers. No middleware stacks. No factory factories. Effects are sets; they union naturally.
 **What this means in practice:**
 - Functions compose with `|>` (pipes)
 - Effects compose by set union
 - Types compose via generics and ADTs
 - Tests compose by handler substitution
 ### 3. Safety Without Ceremony
 The type system catches errors at compile time, but doesn't make you fight it:
 ```lux
 // Type inference keeps code clean
 let x = 42                    // Int, inferred
 let names = ["Alice", "Bob"]  // List<String>, inferred
 // But function signatures are always explicit
 fn greet(name: String): String = "Hello, {name}"
 ```
 **The balance:**
 - Function signatures: always annotated (documentation + API contract)
 - Local bindings: inferred (reduces noise in implementation)
 - Effects: declared or inferred (explicit at boundaries, lightweight inside)
 - Behavioral properties: opt-in (`is pure`, `is total` — add when valuable)
 ### 4. Practical Over Academic
 Lux borrows from the best of programming language research, but wraps it in familiar syntax:
 ```lux
 // This is algebraic effects. But it reads like normal code.
 fn main(): Unit with {Console} = {
    Console.print("What's your name?")
    let name = Console.readLine()
    Console.print("Hello, {name}!")
 }
 ```
 Compare with Haskell's equivalent:
 ```haskell
 main :: IO ()
 main = do
    putStrLn "What's your name?"
    name <- getLine
    putStrLn ("Hello, " ++ name ++ "!")
 ```
 Both are explicit about effects. Lux chooses syntax that reads like imperative code while maintaining the same guarantees.
 **What this means in practice:**
 - ML-family semantics, C-family appearance
 - No monads to learn (effects replace them)
 - No category theory prerequisites
 - The learning curve is: functions → types → effects (days, not months)
 ### 5. One Right Way
 Like Go and Python, Lux favors having one obvious way to do things:
 - **One formatter** (`lux fmt`) — opinionated, not configurable, ends all style debates
 - **One test framework** (built-in `Test` effect) — no framework shopping
 - **One way to handle effects** — declare, handle, compose
 - **One package manager** (`lux pkg`) — integrated, not bolted on
 This is a deliberate rejection of the JavaScript/Ruby approach where every project assembles its own stack from dozens of competing libraries.
 ### 6. Tools Are Part of the Language
 The compiler, linter, formatter, LSP, package manager, and test runner are one thing, not seven:
 ```bash
 lux fmt        # Format
 lux lint       # Lint (with --explain for education)
 lux check      # Type check + lint
 lux test       # Run tests
 lux compile    # Build a binary
 lux serve      # Serve files
 lux --lsp      # Editor integration
 ```
 This follows Go's philosophy: a language is its toolchain. The formatter knows the AST. The linter knows the type system. The LSP knows the effects. They're not afterthoughts.
 ---
 ## Design Decisions and Their Reasons
 ### Why algebraic effects instead of monads?
 Monads are powerful but have poor ergonomics for composition. Combining `IO`, `State`, and `Error` in Haskell requires monad transformers — a notoriously difficult concept. Effects compose naturally:
 ```lux
 // Just list the effects you need. No transformers.
 fn app(): Unit with {Console, File, Http, Time} = ...
 ```
 ### Why not just `async/await`?
 `async/await` solves one effect (concurrency). Effects solve all of them: I/O, state, randomness, failure, concurrency, logging, databases. One mechanism, universally applicable.
 ### Why require function type annotations?
 Three reasons:
 1. **Documentation**: Every function signature is self-documenting
 2. **Error messages**: Inference failures produce confusing errors; annotations localize them
 3. **API stability**: Changing a function body shouldn't silently change its type
 ### Why an opinionated formatter?
 Style debates waste engineering time. `gofmt` proved that an opinionated, non-configurable formatter eliminates an entire category of bikeshedding. `lux fmt` does the same.
 ### Why immutable by default?
 Mutable state is the root of most concurrency bugs and many logic bugs. Immutability makes code easier to reason about. When you need state, the `State` effect makes it explicit and trackable.
 ### Why behavioral types?
 Properties like "this function is idempotent" or "this function always terminates" are critical for correctness but typically live in comments. Making them part of the type system means:
 - The compiler can verify them (or generate property tests)
 - Callers can require them (`where F is idempotent`)
 - They serve as machine-readable documentation
 ---
 ## Comparison with Popular Languages
 ### JavaScript / TypeScript (SO #1 / #6 by usage)
 | Aspect | JavaScript/TypeScript | Lux |
 |--------|----------------------|-----|
 | **Type system** | Optional/gradual (TS) | Required, Hindley-Milner |
 | **Side effects** | Anywhere, implicit | Declared in types |
 | **Testing** | Mock libraries (Jest, etc.) | Swap effect handlers |
 | **Formatting** | Prettier (configurable) | `lux fmt` (opinionated) |
 | **Package management** | npm (massive ecosystem) | `lux pkg` (small ecosystem) |
 | **Paradigm** | Multi-paradigm | Functional-first |
 | **Null safety** | Optional chaining (partial) | `Option<T>`, no null |
 | **Error handling** | try/catch (unchecked) | `Result<T, E>` + `Fail` effect |
 | **Shared** | Familiar syntax, first-class functions, closures, string interpolation |
 **What Lux learns from JS/TS:** Familiar syntax matters. String interpolation, arrow functions, and readable code lower the barrier to entry.
 **What Lux rejects:** Implicit `any`, unchecked exceptions, the "pick your own adventure" toolchain.
 ### Python (SO #4 by usage, #1 most desired)
 | Aspect | Python | Lux |
 |--------|--------|-----|
 | **Type system** | Optional (type hints) | Required, static |
 | **Side effects** | Implicit | Explicit |
 | **Performance** | Slow (interpreted) | Faster (compiled to C) |
 | **Syntax** | Whitespace-significant | Braces/keywords |
 | **Immutability** | Mutable by default | Immutable by default |
 | **Tooling** | Fragmented (black, ruff, mypy, pytest...) | Unified (`lux` binary) |
 | **Shared** | Clean syntax philosophy, "one way to do it", readability focus |
 **What Lux learns from Python:** Readability counts. The Zen of Python's emphasis on one obvious way to do things resonates with Lux's design.
 **What Lux rejects:** Dynamic typing, mutable-by-default, fragmented tooling.
 ### Rust (SO #1 most admired)
 | Aspect | Rust | Lux |
 |--------|------|-----|
 | **Memory** | Ownership/borrowing (manual) | Reference counting (automatic) |
 | **Type system** | Traits, generics, lifetimes | ADTs, effects, generics |
 | **Side effects** | Implicit (except `unsafe`) | Explicit (effect system) |
 | **Error handling** | `Result<T, E>` + `?` | `Result<T, E>` + `Fail` effect |
 | **Performance** | Zero-cost, systems-level | Good, not systems-level |
 | **Learning curve** | Steep (ownership) | Moderate (effects) |
 | **Pattern matching** | Excellent, exhaustive | Excellent, exhaustive |
 | **Shared** | ADTs, pattern matching, `Option`/`Result`, no null, immutable by default, strong type system |
 **What Lux learns from Rust:** ADTs with exhaustive matching, `Option`/`Result` instead of null/exceptions, excellent error messages, integrated tooling (cargo model).
 **What Lux rejects:** Ownership complexity (Lux uses GC/RC instead), lifetimes, `unsafe`.
 ### Go (SO #13 by usage, #11 most admired)
 | Aspect | Go | Lux |
 |--------|-----|-----|
 | **Type system** | Structural, simple | HM inference, ADTs |
 | **Side effects** | Implicit | Explicit |
 | **Error handling** | Multiple returns (`val, err`) | `Result<T, E>` + effects |
 | **Formatting** | `gofmt` (opinionated) | `lux fmt` (opinionated) |
 | **Tooling** | All-in-one (`go` binary) | All-in-one (`lux` binary) |
 | **Concurrency** | Goroutines + channels | `Concurrent` + `Channel` effects |
 | **Generics** | Added late, limited | First-class from day one |
 | **Shared** | Opinionated formatter, unified tooling, practical philosophy |
 **What Lux learns from Go:** Unified toolchain, opinionated formatting, simplicity as a feature, fast compilation.
 **What Lux rejects:** Verbose error handling (`if err != nil`), no ADTs, no generics (historically), nil.
 ### Java / C# (SO #7 / #8 by usage)
 | Aspect | Java/C# | Lux |
 |--------|---------|-----|
 | **Paradigm** | OOP-first | FP-first |
 | **Effects** | DI frameworks (Spring, etc.) | Language-level effects |
 | **Testing** | Mock frameworks (Mockito, etc.) | Handler swapping |
 | **Null safety** | Nullable (Java), nullable ref types (C#) | `Option<T>` |
 | **Boilerplate** | High (getters, setters, factories) | Low (records, inference) |
 | **Shared** | Static typing, generics, pattern matching (recent), established ecosystems |
 **What Lux learns from Java/C#:** Enterprise needs (database effects, HTTP, serialization) matter. Testability is a first-class concern.
 **What Lux rejects:** OOP ceremony, DI frameworks, null, boilerplate.
 ### Haskell / OCaml / Elm (FP family)
 | Aspect | Haskell | Elm | Lux |
 |--------|---------|-----|-----|
 | **Effects** | Monads + transformers | Cmd/Sub (Elm Architecture) | Algebraic effects |
 | **Learning curve** | Steep | Moderate | Moderate |
 | **Error messages** | Improving | Excellent | Good (aspiring to Elm-quality) |
 | **Practical focus** | Academic-leaning | Web-focused | General-purpose |
 | **Syntax** | Unique | Unique | Familiar (C-family feel) |
 | **Shared** | Immutability, ADTs, pattern matching, type inference, no null |
 **What Lux learns from Haskell:** Effects must be explicit. Types must be powerful. Purity matters.
 **What Lux learns from Elm:** Error messages should teach. Tooling should be integrated. Simplicity beats power.
 **What Lux rejects (from Haskell):** Monad transformers, academic syntax, steep learning curve.
 ### Gleam / Elixir (SO #2 / #3 most admired, 2025)
 | Aspect | Gleam | Elixir | Lux |
 |--------|-------|--------|-----|
 | **Type system** | Static, HM | Dynamic | Static, HM |
 | **Effects** | No special tracking | Implicit | First-class |
 | **Concurrency** | BEAM (built-in) | BEAM (built-in) | Effect-based |
 | **Error handling** | `Result` | Pattern matching on tuples | `Result` + `Fail` effect |
 | **Shared** | Friendly errors, pipe operator, functional style, immutability |
 **What Lux learns from Gleam:** Friendly developer experience, clear error messages, and pragmatic FP resonate with developers.
 ---
 ## Tooling Philosophy Audit
 ### Does the linter follow the philosophy?
 **Yes, strongly.** The linter embodies "make the important things visible":
 - `could-be-pure`: Nudges users toward declaring purity — making guarantees visible
 - `could-be-total`: Same for termination
 - `unnecessary-effect-decl`: Keeps effect signatures honest — don't claim effects you don't use
 - `unused-variable/import/function`: Keeps code focused — everything visible should be meaningful
 - `single-arm-match` / `manual-map-option`: Teaches idiomatic patterns
 The category system (correctness > suspicious > idiom > style > pedantic) reflects the philosophy of being practical, not academic: real bugs are errors, style preferences are opt-in.
 ### Does the formatter follow the philosophy?
 **Yes, with one gap.** The formatter is opinionated and non-configurable, matching the "one right way" principle. It enforces consistent style across all Lux code.
 **Gap:** `max_width` and `trailing_commas` are declared in `FormatConfig` but never used. This is harmless but inconsistent — either remove the unused config or implement line wrapping.
 ### Does the type checker follow the philosophy?
 **Yes.** The type checker embodies every core principle:
 - Effects are tracked and verified in function types
 - Behavioral properties are checked where possible
 - Error messages include context and suggestions
 - Type inference reduces ceremony while maintaining safety
 ---
 ## What Could Be Improved
 ### High-value additions (improve experience significantly, low verbosity cost)
 1. **Pipe-friendly standard library**
   - Currently: `List.map(myList, fn(x: Int): Int => x * 2)`
   - Better: Allow `myList |> List.map(fn(x: Int): Int => x * 2)`
   - Many languages (Elixir, F#, Gleam) make the pipe operator the primary composition tool. If the first argument of stdlib functions is always the data, pipes become natural. This is a **library convention**, not a language change.
   - **LLM impact:** Pipe chains are easier for LLMs to generate and read — linear data flow with no nesting.
   - **Human impact:** Reduces cognitive load. Reading left-to-right matches how humans think about data transformation.
 2. **Exhaustive `match` warnings for non-enum types**
   - The linter warns about `wildcard-on-small-enum`, but could also warn when a match on `Option` or `Result` uses a wildcard instead of handling both cases explicitly.
   - **Both audiences:** Prevents subtle bugs where new variants are silently caught by `_`.
 3. **Error message improvements toward Elm quality**
   - Current errors show the right information but could be more conversational and suggest fixes more consistently.
   - Example improvement: When a function is called with wrong argument count, show the expected signature and highlight which argument is wrong.
   - **LLM impact:** Structured error messages with clear "expected X, got Y" patterns are easier for LLMs to parse and fix.
   - **Human impact:** Friendly errors reduce frustration, especially for beginners.
 4. **`let ... else` for fallible destructuring**
   - Rust's `let ... else` pattern handles the "unwrap or bail" case elegantly:
     ```lux
     let Some(value) = maybeValue else return defaultValue
     ```
   - Currently requires a full `match` expression for this common pattern.
   - **Both audiences:** Reduces boilerplate for the most common Option/Result handling pattern.
 5. **Trait/typeclass system for overloading**
   - Currently `toString`, `==`, and similar operations are built-in. A trait system would let users define their own:
     ```lux
     trait Show<T> { fn show(value: T): String }
     impl Show<User> { fn show(u: User): String = "User({u.name})" }
     ```
   - **Note:** This exists partially. Expanding it would enable more generic programming without losing explicitness.
   - **LLM impact:** Traits provide clear, greppable contracts. LLMs can generate trait impls from examples.
 ### Medium-value additions (good improvements, some verbosity cost)
 6. **Named arguments or builder pattern for records**
   - When functions take many parameters, the linter already warns at 5+. Named arguments or record-punning would help:
     ```lux
     fn createUser({ name, email, age }: UserConfig): User = ...
     createUser({ name: "Alice", email: "alice@ex.com", age: 30 })
     ```
   - **Trade-off:** Adds syntax, but the linter already pushes users toward records for many params.
 7. **Async/concurrent effect sugar**
   - The `Concurrent` effect exists but could benefit from syntactic sugar:
     ```lux
     let (a, b) = concurrent {
         fetch("/api/users"),
         fetch("/api/posts")
     }
     ```
   - **Trade-off:** Adds syntax, but concurrent code is important enough to warrant it.
 8. **Module-level documentation with `///` doc comments**
   - The `missing-doc-comment` lint exists, but the doc generation system could be enhanced with richer doc comments that include examples, parameter descriptions, and effect documentation.
   - **LLM impact:** Structured documentation is the single highest-value feature for LLM code understanding.
 ### Lower-value or risky additions (consider carefully)
 9. **Type inference for function return types**
   - Would reduce ceremony: `fn double(x: Int) = x * 2` instead of `fn double(x: Int): Int = x * 2`
   - **Risk:** Violates the "function signatures are documentation" principle. A body change could silently change the API. Current approach is the right trade-off.
 10. **Operator overloading**
    - Tempting for numeric types, but quickly leads to the C++ problem where `+` could mean anything.
    - **Risk:** Violates "make the important things visible" — you can't tell what `a + b` does.
    - **Better:** Keep operators for built-in numeric types. Use named functions for everything else.
 11. **Macros**
    - Powerful but drastically complicate tooling, error messages, and readability.
    - **Risk:** Rust's macro system is powerful but produces some of the worst error messages in the language.
    - **Better:** Solve specific problems with language features (effects, generics) rather than a general metaprogramming escape hatch.
 ---
 ## The LLM Perspective
 Lux has several properties that make it unusually well-suited for LLM-assisted programming:
 1. **Effect signatures are machine-readable contracts.** An LLM reading `fn f(): T with {Database, Logger}` knows exactly what capabilities to provide when generating handler code.
 2. **Behavioral properties are verifiable assertions.** `is pure`, `is idempotent` give LLMs clear constraints to check their own output against.
 3. **The opinionated formatter eliminates style ambiguity.** LLMs don't need to guess indentation, brace style, or naming conventions — `lux fmt` handles it.
 4. **Exhaustive pattern matching forces completeness.** LLMs that generate `match` expressions are reminded by the compiler when they miss cases.
 5. **Small, consistent standard library.** `List.map`, `String.split`, `Option.map` — uniform `Module.function` convention is easy to learn from few examples.
 6. **Effect-based testing needs no framework knowledge.** An LLM doesn't need to know Jest, pytest, or JUnit — just swap handlers.
 **What would help LLMs more:**
 - Structured error output (JSON mode) for programmatic error fixing
 - Example-rich documentation that LLMs can learn patterns from
 - A canonical set of "Lux patterns" (like Go's proverbs) that encode best practices in memorable form
 ---
 ## Summary
 Lux's philosophy can be compressed to five words: **Make the important things visible.**
 This manifests as:
 - **Effects in types** — see what code does
 - **Properties in types** — see what code guarantees
 - **Versions in types** — see how data evolves
 - **One tool for everything** — see how to build
 - **One format for all** — see consistent style
 The language is in the sweet spot between Haskell's rigor and Python's practicality, with Go's tooling philosophy and Elm's developer experience aspirations. It doesn't try to be everything — it tries to make the things that matter most in real software visible, composable, and verifiable.
--- a/docs/ROADMAP.md
+++ b/docs/ROADMAP.md
@@ -53,6 +53,7 @@
 | SQL effect (query, execute) | P1 | 2 weeks | ✅ Complete |
 | Transaction effect | P2 | 1 week | ✅ Complete |
 | Connection pooling | P2 | 1 week | ❌ Missing |
 | PostgreSQL support | P1 | 2 weeks | ✅ Complete |
 ### Phase 1.3: Web Server Framework
@@ -207,8 +208,11 @@
 |------|----------|--------|--------|
 | Package manager (lux pkg) | P1 | 3 weeks | ✅ Complete |
 | Module loader integration | P1 | 1 week | ✅ Complete |
-| Package registry | P2 | 2 weeks | ✅ Complete (server + CLI commands) |
+| Package registry server | P2 | 2 weeks | ✅ Complete |
-| Dependency resolution | P2 | 2 weeks | ❌ Missing |
+| Registry CLI (search, publish) | P2 | 1 week | ✅ Complete |
 | Lock file generation | P1 | 1 week | ✅ Complete |
 | Version constraint parsing | P1 | 1 week | ✅ Complete |
 | Transitive dependency resolution | P2 | 2 weeks | ⚠️ Basic (direct deps only) |
 **Package Manager Features:**
 - `lux pkg init` - Initialize project with lux.toml
@@ -300,6 +304,8 @@
 - ✅ Random effect (int, float, range, bool)
 - ✅ Time effect (now, sleep)
 - ✅ Test effect (assert, assertEqual, assertTrue, assertFalse)
 - ✅ SQL effect (SQLite with transactions)
 - ✅ Postgres effect (PostgreSQL connections)
 **Module System:**
 - ✅ Imports, exports, aliases
@@ -319,7 +325,7 @@
 - ✅ C backend (functions, closures, pattern matching, lists)
 - ✅ JS backend (full language support, browser & Node.js)
 - ✅ REPL with history
- ✅ Basic LSP server
+- ✅ LSP server (diagnostics, hover, completions, go-to-definition, references, symbols)
 - ✅ Formatter
 - ✅ Watch mode
 - ✅ Debugger (basic)
--- a/docs/SQL_DESIGN_ANALYSIS.md
+++ b/docs/SQL_DESIGN_ANALYSIS.md
@@ -0,0 +1,330 @@
 # SQL in Lux: Built-in Effect vs Package
 ## Executive Summary
 This document analyzes whether SQL database access should be a built-in language feature (as it currently is) or a separate package. After comparing approaches across 12+ languages, the recommendation is:
 **Keep SQL as a built-in effect, but refactor the implementation to be more modular.**
 ## Current Implementation
 Lux currently implements SQL as a built-in effect:
 ```lux
 fn main(): Unit with {Console, Sql} = {
    let db = Sql.openMemory()
    Sql.execute(db, "CREATE TABLE users (...)")
    let users = Sql.query(db, "SELECT * FROM users")
    Sql.close(db)
 }
 ```
 The implementation uses rusqlite (SQLite) compiled directly into the Lux binary.
 ## How Other Languages Handle Database Access
 ### Languages with Built-in Database Support
 | Language | Approach | Notes |
 |----------|----------|-------|
 | **Python** | `sqlite3` in stdlib | Most languages have SQLite in stdlib |
 | **Ruby** | `sqlite3` gem + AR are common | ActiveRecord is de facto standard |
 | **Go** | `database/sql` interface in stdlib | Drivers are packages |
 | **Elixir** | Ecto as separate package | But universally used |
 | **PHP** | PDO in core | Multiple backends |
 ### Languages with Package-Only Database Support
 | Language | Approach | Notes |
 |----------|----------|-------|
 | **Rust** | rusqlite, diesel, sqlx packages | No stdlib database |
 | **Node.js** | pg, mysql2, better-sqlite3 | Packages only |
 | **Haskell** | postgresql-simple, persistent | Packages only |
 | **OCaml** | caqti, postgresql-ocaml | Packages only |
 ### Analysis of Each Approach
 #### Go's Model: Interface in Stdlib + Driver Packages
 ```go
 import (
    "database/sql"
    _ "github.com/lib/pq"  // PostgreSQL driver
 )
 db, _ := sql.Open("postgres", "...")
 rows, _ := db.Query("SELECT * FROM users")
 ```
 **Pros:**
 - Standard interface for all databases
 - Type-safe at compile time
 - Drivers are swappable
 **Cons:**
 - Requires understanding interfaces
 - Need external packages for actual database
 #### Python's Model: SQLite in Stdlib
 ```python
 import sqlite3
 conn = sqlite3.connect('example.db')
 c = conn.cursor()
 c.execute('SELECT * FROM users')
 ```
 **Pros:**
 - Zero dependencies for getting started
 - Great for learning/prototyping
 - Always available
 **Cons:**
 - Other databases need packages
 - stdlib vs package API differences
 #### Rust's Model: Everything is Packages
 ```rust
 use rusqlite::{Connection, Result};
 fn main() -> Result<()> {
    let conn = Connection::open("test.db")?;
    conn.execute("CREATE TABLE users (...)", [])?;
    Ok(())
 }
 ```
 **Pros:**
 - Minimal core language
 - Best-in-class implementations
 - Clear ownership
 **Cons:**
 - Cargo.toml management
 - Version conflicts possible
 - Learning curve for package ecosystem
 #### Elixir's Model: Strong Package Ecosystem
 ```elixir
 # Ecto is technically a package but universally used
 Repo.all(from u in User, where: u.age > 18)
 ```
 **Pros:**
 - Best API emerges naturally
 - Core team can focus on language
 - Community ownership
 **Cons:**
 - Package can become outdated
 - Multiple competing solutions
 ## Arguments For Built-in SQL
 ### 1. Effect System Integration
 The most compelling argument: **SQL fits naturally into Lux's effect system.**
 ```lux
 // The effect signature documents database access
 fn fetchUser(id: Int): User with {Sql} = { ... }
 // Handlers enable testing without mocks
 handler testDatabase(): Sql { ... }
 ```
 This is harder to achieve with packages - they'd need to integrate deeply with the effect system.
 ### 2. Zero-Dependency Getting Started
 New users can immediately:
 - Follow tutorials that use databases
 - Build real applications
 - Learn effects with practical examples
 ```bash
 lux run database_example.lux
 # Just works - no package installation
 ```
 ### 3. Guaranteed API Stability
 Built-in effects have stable, documented APIs. Package APIs can change between versions.
 ### 4. Teaching Functional Effects
 SQL is an excellent teaching example for effects:
 - Clear side effects (I/O to database)
 - Handler swapping for testing
 - Transaction scoping
 ### 5. Practical Utility
 90%+ of real applications need database access. Making it trivial benefits most users.
 ## Arguments For SQL as Package
 ### 1. Smaller Binary Size
 rusqlite adds significant binary size (~2-3MB). Package-based approach lets users opt-in.
 ### 2. Database Backend Choice
 Currently locked to SQLite. A package ecosystem could offer:
 - `lux-sqlite`
 - `lux-postgres`
 - `lux-mysql`
 - `lux-mongodb`
 ### 3. Faster Core Language Evolution
 Core team focuses on language; community builds integrations.
 ### 4. Better Specialization
 Dedicated package maintainers might build better database tooling than core team.
 ### 5. Multiple Competing Implementations
 Competition drives quality. The best SQL package wins adoption.
 ## Comparison Matrix
 | Factor | Built-in | Package |
 |--------|----------|---------|
 | Effect integration | Excellent | Needs design work |
 | Learning curve | Low | Medium |
 | Binary size | Larger | User controls |
 | Database options | Limited | Unlimited |
 | API stability | Guaranteed | Version-dependent |
 | Getting started | Instant | Requires install |
 | Testing story | Built-in handlers | Package-specific |
 | Maintenance burden | Core team | Community |
 ## Recommendation
 ### Keep SQL as Built-in Effect, With Changes
 **Rationale:**
 1. **Effect system is Lux's differentiator** - SQL showcases it perfectly
 2. **Practicality matters** - 90% of apps need databases
 3. **Teaching value** - SQL is ideal for learning effects
 4. **Handler testing** - Built-in integration enables powerful testing
 ### Proposed Architecture
 ```
 Core Lux
 ├── Sql effect (interface only)
 │   ├── open/close
 │   ├── execute/query
 │   └── transaction operations
 │
 └── Default SQLite handler (built-in)
    └── Uses rusqlite
 Future packages (optional)
 ├── lux-postgres  -- PostgreSQL handler
 ├── lux-mysql     -- MySQL handler
 └── lux-redis     -- Redis (key-value, not Sql)
 ```
 ### Specific Changes to Consider
 1. **Make SQLite compilation optional**
   ```toml
   # Cargo.toml
   [features]
   default = ["sqlite"]
   sqlite = ["rusqlite"]
   ```
 2. **Define stable Sql effect interface**
   ```lux
   effect Sql {
       fn open(path: String): SqlConn
       fn close(conn: SqlConn): Unit
       fn execute(conn: SqlConn, sql: String): Int
       fn query(conn: SqlConn, sql: String): List<SqlRow>
       // ...
   }
   ```
 3. **Allow package handlers to implement Sql**
   ```lux
   // In lux-postgres package
   handler postgresHandler(connStr: String): Sql { ... }
   // Usage
   run myApp() with {
       Sql -> postgresHandler("postgres://...")
   }
   ```
 4. **Add connection pooling to core**
   Important for production, should be standard.
 ## Comparison to Similar Decisions
 ### Console Effect
 Console is built-in. Nobody questions this because:
 - Universally needed
 - Simple interface
 - Hard to get wrong
 SQL is similar but more complex.
 ### HTTP Effect
 HTTP client is built-in in Lux. This was the right call because:
 - Most apps need HTTP
 - Complex to implement well
 - Effect system integration important
 SQL follows same reasoning.
 ### File Effect
 File I/O is built-in. Same rationale applies.
 ## What Other Effect-System Languages Do
 | Language | Database | Built-in? |
 |----------|----------|-----------|
 | **Koka** | No database support | N/A |
 | **Eff** | No database support | N/A |
 | **Frank** | No database support | N/A |
 | **Unison** | Abilities + packages | Both |
 Lux is pioneering practical effects. Built-in SQL makes sense.
 ## Conclusion
 SQL should remain a built-in effect in Lux because:
 1. It demonstrates the power of effects for real-world use
 2. It enables the handler-based testing story
 3. It removes friction for most applications
 4. It serves as a teaching example for effects
 However, the implementation should evolve to:
 - Support multiple database backends via handlers
 - Make SQLite optional for minimal binaries
 - Provide connection pooling
 - Add parameterized query support
 This hybrid approach gives users the best of both worlds: immediate productivity with built-in SQLite, and flexibility through package-provided handlers for other databases.
 ---
 ## Future Work
 1. **Parameterized queries** - Critical for SQL injection prevention
 2. **Connection pooling** - Required for production servers
 3. **PostgreSQL handler** - Most requested database
 4. **Migration support** - Schema evolution tooling
 5. **Type-safe queries** - Compile-time SQL checking (ambitious)
--- a/docs/WEBSITE_PLAN.md
+++ b/docs/WEBSITE_PLAN.md
--- a/docs/guide/05-effects.md
+++ b/docs/guide/05-effects.md
@@ -53,6 +53,10 @@ Lux provides several built-in effects:
 | `Process` | `exec`, `env`, `args`, `cwd`, `exit` | System processes |
 | `Http` | `get`, `post`, `put`, `delete` | HTTP client |
 | `HttpServer` | `listen`, `accept`, `respond`, `stop` | HTTP server |
 | `Sql` | `open`, `openMemory`, `close`, `execute`, `query`, `queryOne`, `beginTx`, `commit`, `rollback` | SQLite database |
 | `Postgres` | `connect`, `close`, `execute`, `query`, `queryOne` | PostgreSQL database |
 | `Concurrent` | `spawn`, `await`, `yield`, `sleep`, `cancel`, `isRunning`, `taskCount` | Concurrent tasks |
 | `Channel` | `create`, `send`, `receive`, `tryReceive`, `close` | Inter-task communication |
 | `Test` | `assert`, `assertEqual`, `assertTrue`, `assertFalse` | Testing |
 Example usage:
--- a/docs/guide/09-stdlib.md
+++ b/docs/guide/09-stdlib.md
@@ -320,6 +320,114 @@ fn example(): Int with {Fail} = {
 }
 ```
 ### Sql (SQLite)
 ```lux
 fn example(): Unit with {Sql, Console} = {
    let conn = Sql.open("mydb.sqlite")     // Open database file
    // Or: let conn = Sql.openMemory()     // In-memory database
    // Execute statements (returns row count)
    Sql.execute(conn, "CREATE TABLE users (id INTEGER, name TEXT)")
    Sql.execute(conn, "INSERT INTO users VALUES (1, 'Alice')")
    // Query returns list of rows
    let rows = Sql.query(conn, "SELECT * FROM users")
    // Query for single row
    let user = Sql.queryOne(conn, "SELECT * FROM users WHERE id = 1")
    // Transactions
    Sql.beginTx(conn)
    Sql.execute(conn, "UPDATE users SET name = 'Bob' WHERE id = 1")
    Sql.commit(conn)   // Or: Sql.rollback(conn)
    Sql.close(conn)
 }
 ```
 ### Postgres (PostgreSQL)
 ```lux
 fn example(): Unit with {Postgres, Console} = {
    let conn = Postgres.connect("postgres://user:pass@localhost/mydb")
    // Execute statements
    Postgres.execute(conn, "INSERT INTO users (name) VALUES ('Alice')")
    // Query returns list of rows
    let rows = Postgres.query(conn, "SELECT * FROM users")
    // Query for single row
    let user = Postgres.queryOne(conn, "SELECT * FROM users WHERE id = 1")
    Postgres.close(conn)
 }
 ```
 ### Concurrent (Parallel Tasks)
 ```lux
 fn example(): Unit with {Concurrent, Console} = {
    // Spawn concurrent tasks
    let task1 = Concurrent.spawn(fn(): Int => expensiveComputation(1))
    let task2 = Concurrent.spawn(fn(): Int => expensiveComputation(2))
    // Do other work while tasks run
    Console.print("Tasks spawned, doing other work...")
    // Wait for tasks to complete
    let result1 = Concurrent.await(task1)
    let result2 = Concurrent.await(task2)
    Console.print("Results: " + toString(result1) + ", " + toString(result2))
    // Check task status
    if Concurrent.isRunning(task1) then
        Concurrent.cancel(task1)
    // Non-blocking sleep
    Concurrent.sleep(100)  // 100ms
    // Yield to allow other tasks to run
    Concurrent.yield()
    // Get active task count
    let count = Concurrent.taskCount()
 }
 ```
 ### Channel (Inter-Task Communication)
 ```lux
 fn example(): Unit with {Concurrent, Channel, Console} = {
    // Create a channel for communication
    let ch = Channel.create()
    // Spawn producer task
    let producer = Concurrent.spawn(fn(): Unit => {
        Channel.send(ch, 1)
        Channel.send(ch, 2)
        Channel.send(ch, 3)
        Channel.close(ch)
    })
    // Consumer receives values
    match Channel.receive(ch) {
        Some(value) => Console.print("Received: " + toString(value)),
        None => Console.print("Channel closed")
    }
    // Non-blocking receive
    match Channel.tryReceive(ch) {
        Some(value) => Console.print("Got: " + toString(value)),
        None => Console.print("No value available")
    }
    Concurrent.await(producer)
 }
 ```
 ### Test
 Native testing framework:
@@ -360,6 +468,10 @@ fn main(): Unit with {Console} = {
 | Random | int, float, bool |
 | State | get, put |
 | Fail | fail |
 | Sql | open, openMemory, close, execute, query, queryOne, beginTx, commit, rollback |
 | Postgres | connect, close, execute, query, queryOne |
 | Concurrent | spawn, await, yield, sleep, cancel, isRunning, taskCount |
 | Channel | create, send, receive, tryReceive, close |
 | Test | assert, assertEqual, assertTrue, assertFalse |
 ## Next
--- a/docs/guide/12-behavioral-types.md
+++ b/docs/guide/12-behavioral-types.md
@@ -0,0 +1,449 @@
 # Chapter 12: Behavioral Types
 Lux's behavioral types let you make **compile-time guarantees** about function behavior. Unlike comments or documentation, these are actually verified by the compiler.
 ## Why Behavioral Types Matter
 Consider these real-world scenarios:
 1. **Payment processing**: You retry a failed charge. If the function isn't idempotent, you might charge the customer twice.
 2. **Caching**: You cache a computation. If the function isn't deterministic, you'll serve stale/wrong results.
 3. **Parallelization**: You run tasks in parallel. If they aren't pure, you'll have race conditions.
 4. **Infinite loops**: A function never returns. If it was supposed to be total, you have a bug.
 **Behavioral types catch these bugs at compile time.**
 ## The Five Properties
 ### 1. Pure (`is pure`)
 A pure function has **no side effects**. It only depends on its inputs.
 ```lux
 // GOOD: No effects, just computation
 fn add(a: Int, b: Int): Int is pure = a + b
 fn double(x: Int): Int is pure = x * 2
 fn greet(name: String): String is pure = "Hello, " + name
 // ERROR: Pure function cannot have effects
 fn impure(x: Int): Int is pure with {Console} =
    Console.print("x = " + toString(x))  // Compiler error!
    x
 ```
 **What the compiler checks:**
 - Function must have an empty effect set
 - No calls to effectful operations
 **When to use `is pure`:**
 - Mathematical functions
 - Data transformations
 - Any function that should be cacheable
 **Compiler optimizations enabled:**
 - Memoization (cache results)
 - Common subexpression elimination
 - Parallel execution
 - Dead code elimination (if result unused)
 ### 2. Total (`is total`)
 A total function **always terminates** and **never fails**. It produces a value for every valid input.
 ```lux
 // GOOD: Always terminates (structural recursion)
 fn factorial(n: Int): Int is total =
    if n <= 1 then 1 else n * factorial(n - 1)
 // GOOD: Non-recursive is always total
 fn max(a: Int, b: Int): Int is total =
    if a > b then a else b
 // GOOD: List operations that terminate
 fn length<T>(list: List<T>): Int is total =
    match list {
        [] => 0,
        [_, ...rest] => 1 + length(rest)  // Structurally decreasing
    }
 // ERROR: Uses Fail effect
 fn divide(a: Int, b: Int): Int is total with {Fail} =
    if b == 0 then Fail.fail("division by zero")  // Compiler error!
    else a / b
 // ERROR: May not terminate (not structurally decreasing)
 fn collatz(n: Int): Int is total =
    if n == 1 then 1
    else if n % 2 == 0 then collatz(n / 2)
    else collatz(3 * n + 1)  // Not structurally smaller!
 ```
 **What the compiler checks:**
 - No `Fail` effect used
 - Recursive calls must have at least one structurally decreasing argument
 **When to use `is total`:**
 - Core business logic that must never crash
 - Mathematical functions
 - Data structure operations
 **Compiler optimizations enabled:**
 - No exception handling overhead
 - Aggressive inlining
 - Removal of termination checks
 ### 3. Deterministic (`is deterministic`)
 A deterministic function produces the **same output for the same input**, every time.
 ```lux
 // GOOD: Same input = same output
 fn hash(s: String): Int is deterministic =
    List.fold(String.chars(s), 0, fn(acc: Int, c: String): Int => acc * 31 + charCode(c))
 fn formatDate(year: Int, month: Int, day: Int): String is deterministic =
    toString(year) + "-" + padZero(month) + "-" + padZero(day)
 // ERROR: Random is non-deterministic
 fn generateId(): String is deterministic with {Random} =
    "id-" + toString(Random.int(0, 1000000))  // Compiler error!
 // ERROR: Time is non-deterministic
 fn timestamp(): Int is deterministic with {Time} =
    Time.now()  // Compiler error!
 ```
 **What the compiler checks:**
 - No `Random` effect
 - No `Time` effect
 **When to use `is deterministic`:**
 - Hashing functions
 - Serialization/formatting
 - Test helpers
 **Compiler optimizations enabled:**
 - Result caching
 - Parallel execution with consistent results
 - Test reproducibility
 ### 4. Idempotent (`is idempotent`)
 An idempotent function satisfies: `f(f(x)) == f(x)`. Applying it multiple times has the same effect as applying it once.
 ```lux
 // GOOD: Pattern 1 - Constants
 fn alwaysZero(x: Int): Int is idempotent = 0
 // GOOD: Pattern 2 - Identity
 fn identity<T>(x: T): T is idempotent = x
 // GOOD: Pattern 3 - Projection
 fn getName(person: Person): String is idempotent = person.name
 // GOOD: Pattern 4 - Clamping
 fn clampPositive(x: Int): Int is idempotent =
    if x < 0 then 0 else x
 // GOOD: Pattern 5 - Absolute value
 fn abs(x: Int): Int is idempotent =
    if x < 0 then 0 - x else x
 // ERROR: Not idempotent (increment changes value each time)
 fn increment(x: Int): Int is idempotent = x + 1  // f(f(1)) = 3, not 2
 // If you're certain a function is idempotent but the compiler can't verify:
 fn normalize(s: String): String assume is idempotent =
    String.toLower(String.trim(s))
 ```
 **What the compiler checks:**
 - Pattern recognition: constants, identity, projections, clamping, abs
 **When to use `is idempotent`:**
 - Setting configuration
 - Database upserts
 - API PUT/DELETE operations (REST semantics)
 - Retry-safe operations
 **Real-world example - safe retries:**
 ```lux
 // Payment processing with safe retries
 fn chargeCard(amount: Int, cardId: String): Receipt
    is idempotent
    with {Payment, Logger} = {
    Logger.log("Charging card " + cardId)
    Payment.charge(amount, cardId)
 }
 // Safe to retry because chargeCard is idempotent
 fn processWithRetry(amount: Int, cardId: String): Receipt with {Payment, Logger, Fail} = {
    let result = retry(3, fn(): Receipt => chargeCard(amount, cardId))
    match result {
        Ok(receipt) => receipt,
        Err(e) => Fail.fail("Payment failed after 3 attempts: " + e)
    }
 }
 ```
 ### 5. Commutative (`is commutative`)
 A commutative function satisfies: `f(a, b) == f(b, a)`. The order of arguments doesn't matter.
 ```lux
 // GOOD: Addition is commutative
 fn add(a: Int, b: Int): Int is commutative = a + b
 // GOOD: Multiplication is commutative
 fn multiply(a: Int, b: Int): Int is commutative = a * b
 // GOOD: Min/max are commutative
 fn minimum(a: Int, b: Int): Int is commutative =
    if a < b then a else b
 // ERROR: Subtraction is not commutative (3 - 2 != 2 - 3)
 fn subtract(a: Int, b: Int): Int is commutative = a - b  // Compiler error!
 // ERROR: Wrong number of parameters
 fn triple(a: Int, b: Int, c: Int): Int is commutative = a + b + c  // Must have exactly 2
 ```
 **What the compiler checks:**
 - Must have exactly 2 parameters
 - Body must be a commutative operation (+, *, min, max, ==, !=, &&, ||)
 **When to use `is commutative`:**
 - Mathematical operations
 - Set operations (union, intersection)
 - Merging/combining functions
 **Compiler optimizations enabled:**
 - Argument reordering for efficiency
 - Parallel reduction
 - Algebraic simplifications
 ## Combining Properties
 Properties can be combined for stronger guarantees:
 ```lux
 // Pure + deterministic + total = perfect for caching
 fn computeHash(data: String): Int
    is pure
    is deterministic
    is total = {
    List.fold(String.chars(data), 0, fn(acc: Int, c: String): Int =>
        acc * 31 + charCode(c)
    )
 }
 // Pure + idempotent = safe transformation
 fn normalizeEmail(email: String): String
    is pure
    is idempotent = {
    String.toLower(String.trim(email))
 }
 // Commutative + pure = parallel reduction friendly
 fn merge(a: Record, b: Record): Record
    is pure
    is commutative = {
    { ...a, ...b }  // Last wins, but both contribute
 }
 ```
 ## Property Constraints in Where Clauses
 You can require function arguments to have certain properties:
 ```lux
 // Higher-order function that requires a pure function
 fn map<T, U>(list: List<T>, f: fn(T): U is pure): List<U> is pure =
    match list {
        [] => [],
        [x, ...rest] => [f(x), ...map(rest, f)]
    }
 // Only accepts idempotent functions - safe to retry
 fn retry<T>(times: Int, action: fn(): T is idempotent): Result<T, String> = {
    if times <= 0 then Err("No attempts left")
    else {
        match tryCall(action) {
            Ok(result) => Ok(result),
            Err(e) => retry(times - 1, action)  // Safe because action is idempotent
        }
    }
 }
 // Only accepts deterministic functions - safe to cache
 fn memoize<K, V>(f: fn(K): V is deterministic): fn(K): V = {
    let cache = HashMap.new()
    fn(key: K): V => {
        match cache.get(key) {
            Some(v) => v,
            None => {
                let v = f(key)
                cache.set(key, v)
                v
            }
        }
    }
 }
 // Usage:
 let cachedHash = memoize(computeHash)  // OK: computeHash is deterministic
 let badCache = memoize(generateRandom)  // ERROR: generateRandom is not deterministic
 ```
 ## The `assume` Escape Hatch
 Sometimes you know a function has a property but the compiler can't verify it. Use `assume`:
 ```lux
 // Compiler can't verify this is idempotent, but we know it is
 fn setUserStatus(userId: String, status: String): Unit
    assume is idempotent
    with {Database} = {
    Database.execute("UPDATE users SET status = ? WHERE id = ?", [status, userId])
 }
 // Use assume sparingly - it bypasses compiler checks!
 // If you're wrong, you may have subtle bugs.
 ```
 **Warning**: `assume` tells the compiler to trust you. If you're wrong, the optimization or guarantee may be invalid.
 ## Compiler Optimizations
 When the compiler knows behavioral properties, it can optimize aggressively:
 | Property | Optimizations |
 |----------|---------------|
 | `is pure` | Memoization, CSE, dead code elimination, parallelization |
 | `is total` | No exception handling, aggressive inlining |
 | `is deterministic` | Result caching, parallel execution |
 | `is idempotent` | Retry optimization, duplicate call elimination |
 | `is commutative` | Argument reordering, parallel reduction |
 ### Example: Automatic Memoization
 ```lux
 fn expensiveComputation(n: Int): Int
    is pure
    is deterministic
    is total = {
    // Complex calculation...
    fib(n)
 }
 // The compiler may automatically cache results because:
 // - pure: no side effects, so caching is safe
 // - deterministic: same input = same output
 // - total: will always return a value
 ```
 ### Example: Safe Parallelization
 ```lux
 fn processItems(items: List<Item>): List<Result>
    is pure = {
    List.map(items, processItem)
 }
 // If processItem is pure, the compiler can parallelize this automatically
 ```
 ## Practical Examples
 ### Example 1: Financial Calculations
 ```lux
 // Interest calculation - pure, deterministic, total
 fn calculateInterest(principal: Int, rate: Float, years: Int): Float
    is pure
    is deterministic
    is total = {
    let r = rate / 100.0
    Float.fromInt(principal) * Math.pow(1.0 + r, Float.fromInt(years))
 }
 // Transaction validation - pure, total
 fn validateTransaction(tx: Transaction): Result<Transaction, String>
    is pure
    is total = {
    if tx.amount <= 0 then Err("Amount must be positive")
    else if tx.from == tx.to then Err("Cannot transfer to self")
    else Ok(tx)
 }
 ```
 ### Example 2: Data Processing Pipeline
 ```lux
 // Each step is pure and deterministic
 fn cleanData(raw: String): String is pure is deterministic =
    raw |> String.trim |> String.toLower
 fn parseRecord(line: String): Result<Record, String> is pure is deterministic =
    match String.split(line, ",") {
        [name, age, email] => Ok({ name, age: parseInt(age), email }),
        _ => Err("Invalid format")
    }
 fn validateRecord(record: Record): Bool is pure is deterministic is total =
    String.length(record.name) > 0 && record.age > 0
 // Pipeline can be parallelized because all functions are pure + deterministic
 fn processFile(contents: String): List<Record> is pure is deterministic = {
    contents
    |> String.lines
    |> List.map(cleanData)
    |> List.map(parseRecord)
    |> List.filterMap(fn(r: Result<Record, String>): Option<Record> =>
        match r { Ok(v) => Some(v), Err(_) => None })
    |> List.filter(validateRecord)
 }
 ```
 ### Example 3: Idempotent API Handlers
 ```lux
 // PUT /users/:id - idempotent by REST semantics
 fn handlePutUser(id: String, data: UserData): Response
    is idempotent
    with {Database, Logger} = {
    Logger.log("PUT /users/" + id)
    Database.upsert("users", id, data)
    Response.ok({ id, ...data })
 }
 // DELETE /users/:id - idempotent by REST semantics
 fn handleDeleteUser(id: String): Response
    is idempotent
    with {Database, Logger} = {
    Logger.log("DELETE /users/" + id)
    Database.delete("users", id)  // Safe to call multiple times
    Response.noContent()
 }
 ```
 ## Summary
 | Property | Meaning | Compiler Checks | Use Case |
 |----------|---------|-----------------|----------|
 | `is pure` | No effects | Empty effect set | Caching, parallelization |
 | `is total` | Always terminates | No Fail, structural recursion | Core logic |
 | `is deterministic` | Same in = same out | No Random/Time | Caching, testing |
 | `is idempotent` | f(f(x)) = f(x) | Pattern recognition | Retries, APIs |
 | `is commutative` | f(a,b) = f(b,a) | 2 params, commutative op | Math, merging |
 ## What's Next?
 - [Chapter 13: Schema Evolution](./13-schema-evolution.md) - Version your data types
 - [Tutorials](../tutorials/README.md) - Practical projects
--- a/docs/guide/13-schema-evolution.md
+++ b/docs/guide/13-schema-evolution.md
@@ -0,0 +1,573 @@
 # Chapter 13: Schema Evolution
 Data structures change over time. Fields get added, removed, or renamed. Types get split or merged. Without careful handling, these changes break systems—old data can't be read, services fail, migrations corrupt data.
 Lux's **schema evolution** system makes these changes safe and automatic.
 ## The Problem
 Consider a real scenario:
 ```lux
 // Version 1: Simple user
 type User {
    name: String
 }
 // Later, you need email addresses
 type User {
    name: String,
    email: String   // Breaking change! Old data doesn't have this.
 }
 ```
 In most languages, this breaks everything. Existing users in your database don't have email addresses. Deserializing old data fails. Services crash.
 Lux solves this with **versioned types** and **automatic migrations**.
 ## Versioned Types
 Add a version annotation to any type:
 ```lux
 // Version 1: Original definition
 type User @v1 {
    name: String
 }
 // Version 2: Added email field
 type User @v2 {
    name: String,
    email: String,
    // How to migrate from v1
    from @v1 = { name: old.name, email: "unknown@example.com" }
 }
 // Version 3: Split name into first/last
 type User @v3 {
    firstName: String,
    lastName: String,
    email: String,
    // How to migrate from v2
    from @v2 = {
        firstName: String.split(old.name, " ") |> List.head |> Option.getOrElse(""),
        lastName: String.split(old.name, " ") |> List.tail |> List.head |> Option.getOrElse(""),
        email: old.email
    }
 }
 ```
 The `@latest` alias always refers to the most recent version:
 ```lux
 type User @latest {
    firstName: String,
    lastName: String,
    email: String,
    from @v2 = { ... }
 }
 // These are equivalent:
 fn createUser(first: String, last: String, email: String): User@latest = ...
 fn createUser(first: String, last: String, email: String): User@v3 = ...
 ```
 ## Migration Syntax
 ### Basic Migration
 ```lux
 type Config @v2 {
    theme: String,
    fontSize: Int,
    // 'old' refers to the v1 value
    from @v1 = {
        theme: old.theme,
        fontSize: 14  // New field with default
    }
 }
 ```
 ### Computed Fields
 ```lux
 type Order @v2 {
    items: List<Item>,
    total: Int,
    itemCount: Int,  // New computed field
    from @v1 = {
        items: old.items,
        total: old.total,
        itemCount: List.length(old.items)
    }
 }
 ```
 ### Removing Fields
 When removing fields, simply don't include them in the new version:
 ```lux
 type Settings @v1 {
    theme: String,
    legacyMode: Bool,  // To be removed
    volume: Int
 }
 type Settings @v2 {
    theme: String,
    volume: Int,
    // legacyMode is dropped - just don't migrate it
    from @v1 = {
        theme: old.theme,
        volume: old.volume
    }
 }
 ```
 ### Renaming Fields
 ```lux
 type Product @v1 {
    name: String,
    cost: Int  // Old field name
 }
 type Product @v2 {
    name: String,
    price: Int,  // Renamed from 'cost'
    from @v1 = {
        name: old.name,
        price: old.cost  // Map old field to new name
    }
 }
 ```
 ### Complex Transformations
 ```lux
 type Address @v1 {
    fullAddress: String  // "123 Main St, New York, NY 10001"
 }
 type Address @v2 {
    street: String,
    city: String,
    state: String,
    zip: String,
    from @v1 = {
        let parts = String.split(old.fullAddress, ", ")
        {
            street: List.get(parts, 0) |> Option.getOrElse(""),
            city: List.get(parts, 1) |> Option.getOrElse(""),
            state: List.get(parts, 2)
                |> Option.map(fn(s: String): String => String.split(s, " ") |> List.head |> Option.getOrElse(""))
                |> Option.getOrElse(""),
            zip: List.get(parts, 2)
                |> Option.map(fn(s: String): String => String.split(s, " ") |> List.last |> Option.getOrElse(""))
                |> Option.getOrElse("")
        }
    }
 }
 ```
 ## Working with Versioned Values
 The `Schema` module provides runtime operations for versioned values:
 ### Creating Versioned Values
 ```lux
 // Create a value tagged with a specific version
 let userV1 = Schema.versioned("User", 1, { name: "Alice" })
 let userV2 = Schema.versioned("User", 2, { name: "Alice", email: "alice@example.com" })
 ```
 ### Checking Versions
 ```lux
 let user = Schema.versioned("User", 1, { name: "Alice" })
 let version = Schema.getVersion(user)  // Returns 1
 // Version-aware logic
 if version < 2 then
    Console.print("Legacy user format")
 else
    Console.print("Modern user format")
 ```
 ### Migrating Values
 ```lux
 // Migrate to a specific version
 let userV1 = Schema.versioned("User", 1, { name: "Alice" })
 let userV2 = Schema.migrate(userV1, 2)  // Uses declared migration
 let version = Schema.getVersion(userV2)  // Now 2
 // Chain migrations (v1 -> v2 -> v3)
 let userV3 = Schema.migrate(userV1, 3)  // Applies v1->v2, then v2->v3
 ```
 ## Auto-Generated Migrations
 For simple changes, Lux can **automatically generate** migrations:
 ```lux
 type Profile @v1 {
    name: String
 }
 // Adding a field with a default? Migration is auto-generated
 type Profile @v2 {
    name: String,
    bio: String = ""  // Default value provided
 }
 // The compiler generates this for you:
 // from @v1 = { name: old.name, bio: "" }
 ```
 Auto-migration works for:
 - Adding fields with default values
 - Keeping existing fields unchanged
 You must write explicit migrations for:
 - Field renaming
 - Field removal (to confirm intent)
 - Type changes
 - Computed/derived fields
 ## Practical Examples
 ### Example 1: API Response Versioning
 ```lux
 type ApiResponse @v1 {
    status: String,
    data: String
 }
 type ApiResponse @v2 {
    status: String,
    data: String,
    meta: { timestamp: Int, version: String },
    from @v1 = {
        status: old.status,
        data: old.data,
        meta: { timestamp: 0, version: "legacy" }
    }
 }
 // Version-aware API client
 fn handleResponse(raw: ApiResponse@v1): ApiResponse@v2 = {
    Schema.migrate(Schema.versioned("ApiResponse", 1, raw), 2)
 }
 ```
 ### Example 2: Database Record Evolution
 ```lux
 // Original schema
 type Customer @v1 {
    name: String,
    address: String
 }
 // Split address into components
 type Customer @v2 {
    name: String,
    street: String,
    city: String,
    country: String,
    from @v1 = {
        let parts = String.split(old.address, ", ")
        {
            name: old.name,
            street: List.get(parts, 0) |> Option.getOrElse(old.address),
            city: List.get(parts, 1) |> Option.getOrElse("Unknown"),
            country: List.get(parts, 2) |> Option.getOrElse("Unknown")
        }
    }
 }
 // Load and migrate on read
 fn loadCustomer(id: String): Customer@v2 with {Database} = {
    let record = Database.query("SELECT * FROM customers WHERE id = ?", [id])
    let version = record.schema_version  // Stored version
    if version == 1 then
        let v1 = Schema.versioned("Customer", 1, {
            name: record.name,
            address: record.address
        })
        Schema.migrate(v1, 2)
    else
        { name: record.name, street: record.street, city: record.city, country: record.country }
 }
 ```
 ### Example 3: Configuration Files
 ```lux
 type AppConfig @v1 {
    debug: Bool,
    port: Int
 }
 type AppConfig @v2 {
    debug: Bool,
    port: Int,
    logLevel: String,  // New in v2
    from @v1 = {
        debug: old.debug,
        port: old.port,
        logLevel: if old.debug then "debug" else "info"
    }
 }
 type AppConfig @v3 {
    environment: String,  // Replaces debug flag
    port: Int,
    logLevel: String,
    from @v2 = {
        environment: if old.debug then "development" else "production",
        port: old.port,
        logLevel: old.logLevel
    }
 }
 // Load config with automatic migration
 fn loadConfig(path: String): AppConfig@v3 with {File} = {
    let json = File.read(path)
    let parsed = Json.parse(json)
    let version = Json.getInt(parsed, "version") |> Option.getOrElse(1)
    match version {
        1 => {
            let v1 = Schema.versioned("AppConfig", 1, {
                debug: Json.getBool(parsed, "debug") |> Option.getOrElse(false),
                port: Json.getInt(parsed, "port") |> Option.getOrElse(8080)
            })
            Schema.migrate(v1, 3)
        },
        2 => {
            let v2 = Schema.versioned("AppConfig", 2, {
                debug: Json.getBool(parsed, "debug") |> Option.getOrElse(false),
                port: Json.getInt(parsed, "port") |> Option.getOrElse(8080),
                logLevel: Json.getString(parsed, "logLevel") |> Option.getOrElse("info")
            })
            Schema.migrate(v2, 3)
        },
        _ => {
            // Already v3
            {
                environment: Json.getString(parsed, "environment") |> Option.getOrElse("production"),
                port: Json.getInt(parsed, "port") |> Option.getOrElse(8080),
                logLevel: Json.getString(parsed, "logLevel") |> Option.getOrElse("info")
            }
        }
    }
 }
 ```
 ### Example 4: Event Sourcing
 ```lux
 // Event types evolve over time
 type UserCreated @v1 {
    userId: String,
    name: String,
    timestamp: Int
 }
 type UserCreated @v2 {
    userId: String,
    name: String,
    email: String,
    createdAt: Int,  // Renamed from timestamp
    from @v1 = {
        userId: old.userId,
        name: old.name,
        email: "",  // Not captured in v1
        createdAt: old.timestamp
    }
 }
 // Process events regardless of version
 fn processEvent(event: UserCreated@v1 | UserCreated@v2): Unit with {Console} = {
    let normalized = Schema.migrate(event, 2)  // Always work with v2
    Console.print("User created: " + normalized.name + " at " + toString(normalized.createdAt))
 }
 ```
 ## Compile-Time Safety
 The compiler catches schema evolution errors:
 ```lux
 type User @v2 {
    name: String,
    email: String
    // ERROR: Migration references non-existent field
    from @v1 = { name: old.username, email: old.email }
    //                    ^^^^^^^^ 'username' does not exist in User@v1
 }
 ```
 ```lux
 type User @v2 {
    name: String,
    email: String
    // ERROR: Migration missing required field
    from @v1 = { name: old.name }
    //                           ^ Missing 'email' field
 }
 ```
 ```lux
 type User @v2 {
    name: String,
    age: Int
    // ERROR: Type mismatch in migration
    from @v1 = { name: old.name, age: old.birthYear }
    //                                ^^^^^^^^^^^^^ Expected Int, found String
 }
 ```
 ## Compatibility Checking
 Lux tracks compatibility between versions:
 | Change Type | Backward Compatible | Forward Compatible |
 |-------------|--------------------|--------------------|
 | Add optional field (with default) | Yes | Yes |
 | Add required field | No | Yes (with migration) |
 | Remove field | Yes (with migration) | No |
 | Rename field | No | No (need migration) |
 | Change field type | No | No (need migration) |
 The compiler warns about breaking changes:
 ```lux
 type User @v1 {
    name: String,
    email: String
 }
 type User @v2 {
    name: String
    // Warning: Removing 'email' is a breaking change
    // Existing v2 consumers expect this field
 }
 ```
 ## Best Practices
 ### 1. Always Version Production Types
 ```lux
 // Good: Versioned from the start
 type Order @v1 {
    id: String,
    items: List<Item>,
    total: Int
 }
 // Bad: Unversioned type is hard to evolve
 type Order {
    id: String,
    items: List<Item>,
    total: Int
 }
 ```
 ### 2. Keep Migrations Simple
 ```lux
 // Good: Simple, direct mapping
 from @v1 = {
    name: old.name,
    email: old.email |> Option.getOrElse("")
 }
 // Avoid: Complex logic in migrations
 from @v1 = {
    name: old.name,
    email: {
        // Don't put complex business logic here
        let domain = inferDomainFromName(old.name)
        let local = String.toLower(String.replace(old.name, " ", "."))
        local + "@" + domain
    }
 }
 ```
 ### 3. Test Migrations
 ```lux
 fn testUserMigration(): Unit with {Test} = {
    let v1User = Schema.versioned("User", 1, { name: "Alice" })
    let v2User = Schema.migrate(v1User, 2)
    Test.assertEqual(v2User.name, "Alice")
    Test.assertEqual(v2User.email, "unknown@example.com")
 }
 ```
 ### 4. Document Breaking Changes
 ```lux
 type User @v3 {
    // BREAKING: 'name' split into firstName/lastName
    // Migration: name.split(" ")[0] -> firstName, name.split(" ")[1] -> lastName
    firstName: String,
    lastName: String,
    email: String,
    from @v2 = { ... }
 }
 ```
 ## Schema Module Reference
 | Function | Description |
 |----------|-------------|
 | `Schema.versioned(typeName, version, value)` | Create a versioned value |
 | `Schema.getVersion(value)` | Get the version of a value |
 | `Schema.migrate(value, targetVersion)` | Migrate to a target version |
 | `Schema.isCompatible(v1, v2)` | Check if versions are compatible |
 ## Summary
 Schema evolution in Lux provides:
 - **Versioned types** with `@v1`, `@v2`, `@latest` annotations
 - **Explicit migrations** with `from @vN = { ... }` syntax
 - **Automatic migrations** for simple field additions with defaults
 - **Runtime operations** via the `Schema` module
 - **Compile-time safety** catching migration errors early
 - **Migration chaining** for multi-step upgrades
 This system ensures your data can evolve safely over time, without breaking existing code or losing information.
 ## What's Next?
 - [Tutorials](../tutorials/README.md) - Build real projects
 - [Standard Library Reference](../stdlib/README.md) - Complete API docs
--- a/examples/behavioral.lux
+++ b/examples/behavioral.lux
@@ -1,36 +1,19 @@
-// Demonstrating behavioral properties in Lux
+fn add(a: Int, b: Int): Int is pure = a + b
 // Behavioral properties are compile-time guarantees about function behavior
 //
 // Expected output:
 // add(5, 3) = 8
 // factorial(5) = 120
 // multiply(7, 6) = 42
 // abs(-5) = 5
-// A pure function - no side effects, same input always gives same output
+fn factorial(n: Int): Int is deterministic = if n <= 1 then 1 else n * factorial(n - 1)
 fn add(a: Int, b: Int): Int is pure =
    a + b
-// A deterministic function - same input always gives same output
+fn multiply(a: Int, b: Int): Int is commutative = a * b
 fn factorial(n: Int): Int is deterministic =
    if n <= 1 then 1
    else n * factorial(n - 1)
-// A commutative function - order of arguments doesn't matter
+fn abs(x: Int): Int is idempotent = if x < 0 then 0 - x else x
 fn multiply(a: Int, b: Int): Int is commutative =
    a * b
 // An idempotent function - absolute value
 fn abs(x: Int): Int is idempotent =
    if x < 0 then 0 - x else x
 // Test the functions
 let sumResult = add(5, 3)
 let factResult = factorial(5)
 let productResult = multiply(7, 6)
 let absResult = abs(0 - 5)
 // Print results
 fn printResults(): Unit with {Console} = {
    Console.print("add(5, 3) = " + toString(sumResult))
    Console.print("factorial(5) = " + toString(factResult))
--- a/examples/behavioral_types.lux
+++ b/examples/behavioral_types.lux
@@ -1,82 +1,42 @@
 // Behavioral Types Demo
 // Demonstrates compile-time verification of function properties
 // ============================================================
 // PART 1: Pure Functions
 // ============================================================
 // Pure functions have no side effects
 fn add(a: Int, b: Int): Int is pure = a + b
 fn subtract(a: Int, b: Int): Int is pure = a - b
 // ============================================================
 // PART 2: Commutative Functions
 // ============================================================
 // Commutative functions: f(a, b) = f(b, a)
 fn multiply(a: Int, b: Int): Int is commutative = a * b
 fn sum(a: Int, b: Int): Int is commutative = a + b
-// ============================================================
+fn abs(x: Int): Int is idempotent = if x < 0 then 0 - x else x
 // PART 3: Idempotent Functions
 // ============================================================
 // Idempotent functions: f(f(x)) = f(x)
 fn abs(x: Int): Int is idempotent =
    if x < 0 then 0 - x else x
 fn identity(x: Int): Int is idempotent = x
-// ============================================================
+fn factorial(n: Int): Int is deterministic = if n <= 1 then 1 else n * factorial(n - 1)
 // PART 4: Deterministic Functions
 // ============================================================
-// Deterministic functions always produce the same output for the same input
+fn fib(n: Int): Int is deterministic = if n <= 1 then n else fib(n - 1) + fib(n - 2)
 fn factorial(n: Int): Int is deterministic =
    if n <= 1 then 1 else n * factorial(n - 1)
-fn fib(n: Int): Int is deterministic =
+fn sumTo(n: Int): Int is total = if n <= 0 then 0 else n + sumTo(n - 1)
    if n <= 1 then n else fib(n - 1) + fib(n - 2)
-// ============================================================
+fn power(base: Int, exp: Int): Int is total = if exp <= 0 then 1 else base * power(base, exp - 1)
 // PART 5: Total Functions
 // ============================================================
 // Total functions are defined for all inputs (no infinite loops, no exceptions)
 fn sumTo(n: Int): Int is total =
    if n <= 0 then 0 else n + sumTo(n - 1)
 fn power(base: Int, exp: Int): Int is total =
    if exp <= 0 then 1 else base * power(base, exp - 1)
 // ============================================================
 // RESULTS
 // ============================================================
 fn main(): Unit with {Console} = {
    Console.print("=== Behavioral Types Demo ===")
    Console.print("")
    Console.print("Part 1: Pure functions")
    Console.print("  add(5, 3) = " + toString(add(5, 3)))
    Console.print("  subtract(10, 4) = " + toString(subtract(10, 4)))
    Console.print("")
    Console.print("Part 2: Commutative functions")
    Console.print("  multiply(7, 6) = " + toString(multiply(7, 6)))
    Console.print("  sum(10, 20) = " + toString(sum(10, 20)))
    Console.print("")
    Console.print("Part 3: Idempotent functions")
    Console.print("  abs(-42) = " + toString(abs(0 - 42)))
    Console.print("  identity(100) = " + toString(identity(100)))
    Console.print("")
    Console.print("Part 4: Deterministic functions")
    Console.print("  factorial(5) = " + toString(factorial(5)))
    Console.print("  fib(10) = " + toString(fib(10)))
    Console.print("")
    Console.print("Part 5: Total functions")
    Console.print("  sumTo(10) = " + toString(sumTo(10)))
    Console.print("  power(2, 8) = " + toString(power(2, 8)))
--- a/examples/builtin_effects.lux
+++ b/examples/builtin_effects.lux
@@ -1,31 +1,7 @@
-// Demonstrating built-in effects in Lux
+fn safeDivide(a: Int, b: Int): Int with {Fail} = if b == 0 then Fail.fail("Division by zero") else a / b
 //
 // Lux provides several built-in effects:
 // - Console: print and read from terminal
 // - Fail: early termination with error
 // - State: get/put mutable state (requires runtime initialization)
 // - Reader: read-only environment access (requires runtime initialization)
 //
 // This example demonstrates Console and Fail effects.
 //
 // Expected output:
 // Starting computation...
 // Step 1: validating input
 // Step 2: processing
 // Result: 42
 // Done!
-// A function that can fail
+fn validatePositive(n: Int): Int with {Fail} = if n < 0 then Fail.fail("Negative number not allowed") else n
 fn safeDivide(a: Int, b: Int): Int with {Fail} =
    if b == 0 then Fail.fail("Division by zero")
    else a / b
 // A function that validates input
 fn validatePositive(n: Int): Int with {Fail} =
    if n < 0 then Fail.fail("Negative number not allowed")
    else n
 // A computation that uses multiple effects
 fn compute(input: Int): Int with {Console, Fail} = {
    Console.print("Starting computation...")
    Console.print("Step 1: validating input")
@@ -36,7 +12,6 @@ fn compute(input: Int): Int with {Console, Fail} = {
    result
 }
 // Main function
 fn main(): Unit with {Console} = {
    let result = run compute(21) with {}
    Console.print("Done!")
--- a/examples/counter.lux
+++ b/examples/counter.lux
@@ -1,14 +1,3 @@
 // Counter Example - A simple interactive counter using TEA pattern
 //
 // This example demonstrates:
 // - Model-View-Update architecture (TEA)
 // - Html DSL for describing UI (inline version)
 // - Message-based state updates
 // ============================================================================
 // Html Types (subset of stdlib/html)
 // ============================================================================
 type Html<M> =
    | Element(String, List<Attr<M>>, List<Html<M>>)
    | Text(String)
@@ -19,130 +8,96 @@ type Attr<M> =
    | Id(String)
    | OnClick(M)
-// Html builder helpers
+fn div<M>(attrs: List<Attr<M>>, children: List<Html<M>>): Html<M> = Element("div", attrs, children)
 fn div<M>(attrs: List<Attr<M>>, children: List<Html<M>>): Html<M> =
    Element("div", attrs, children)
-fn span<M>(attrs: List<Attr<M>>, children: List<Html<M>>): Html<M> =
+fn span<M>(attrs: List<Attr<M>>, children: List<Html<M>>): Html<M> = Element("span", attrs, children)
    Element("span", attrs, children)
-fn h1<M>(attrs: List<Attr<M>>, children: List<Html<M>>): Html<M> =
+fn h1<M>(attrs: List<Attr<M>>, children: List<Html<M>>): Html<M> = Element("h1", attrs, children)
    Element("h1", attrs, children)
-fn button<M>(attrs: List<Attr<M>>, children: List<Html<M>>): Html<M> =
+fn button<M>(attrs: List<Attr<M>>, children: List<Html<M>>): Html<M> = Element("button", attrs, children)
    Element("button", attrs, children)
-fn text<M>(content: String): Html<M> =
+fn text<M>(content: String): Html<M> = Text(content)
    Text(content)
-fn class<M>(name: String): Attr<M> =
+fn class<M>(name: String): Attr<M> = Class(name)
    Class(name)
-fn onClick<M>(msg: M): Attr<M> =
+fn onClick<M>(msg: M): Attr<M> = OnClick(msg)
    OnClick(msg)
 // ============================================================================
 // Model - The application state (using ADT wrapper)
 // ============================================================================
 type Model =
    | Counter(Int)
 fn getCount(model: Model): Int =
    match model {
-        Counter(n) => n
+    Counter(n) => n,
-    }
+}
 fn init(): Model = Counter(0)
 // ============================================================================
 // Messages - Events that can occur
 // ============================================================================
 type Msg =
    | Increment
    | Decrement
    | Reset
 // ============================================================================
 // Update - State transitions
 // ============================================================================
 fn update(model: Model, msg: Msg): Model =
    match msg {
-        Increment => Counter(getCount(model) + 1),
+    Increment => Counter(getCount(model) + 1),
-        Decrement => Counter(getCount(model) - 1),
+    Decrement => Counter(getCount(model) - 1),
-        Reset => Counter(0)
+    Reset => Counter(0),
-    }
+}
 // ============================================================================
 // View - Render the UI
 // ============================================================================
 fn viewCounter(count: Int): Html<Msg> = {
    let countText = text(toString(count))
    let countSpan = span([class("count")], [countText])
    let displayDiv = div([class("counter-display")], [countSpan])
    let minusBtn = button([onClick(Decrement), class("btn")], [text("-")])
    let resetBtn = button([onClick(Reset), class("btn btn-reset")], [text("Reset")])
    let plusBtn = button([onClick(Increment), class("btn")], [text("+")])
    let buttonsDiv = div([class("counter-buttons")], [minusBtn, resetBtn, plusBtn])
    let title = h1([], [text("Counter")])
    div([class("counter-app")], [title, displayDiv, buttonsDiv])
 }
 fn view(model: Model): Html<Msg> = viewCounter(getCount(model))
 // ============================================================================
 // Debug: Print Html structure
 // ============================================================================
 fn showAttr(attr: Attr<Msg>): String =
    match attr {
-        Class(s) => "class=\"" + s + "\"",
+    Class(s) => "class=\"" + s + "\"",
-        Id(s) => "id=\"" + s + "\"",
+    Id(s) => "id=\"" + s + "\"",
-        OnClick(msg) => match msg {
+    OnClick(msg) => match msg {
-            Increment => "onclick=\"Increment\"",
+    Increment => "onclick=\"Increment\"",
-            Decrement => "onclick=\"Decrement\"",
+    Decrement => "onclick=\"Decrement\"",
-            Reset => "onclick=\"Reset\""
+    Reset => "onclick=\"Reset\"",
-        }
+},
-    }
+}
 fn showAttrs(attrs: List<Attr<Msg>>): String =
    match List.head(attrs) {
-        None => "",
+    None => "",
-        Some(a) => match List.tail(attrs) {
+    Some(a) => match List.tail(attrs) {
-            None => showAttr(a),
+    None => showAttr(a),
-            Some(rest) => showAttr(a) + " " + showAttrs(rest)
+    Some(rest) => showAttr(a) + " " + showAttrs(rest),
-        }
+},
-    }
+}
 fn showChildren(children: List<Html<Msg>>, indent: Int): String =
    match List.head(children) {
-        None => "",
+    None => "",
-        Some(c) => match List.tail(children) {
+    Some(c) => match List.tail(children) {
-            None => showHtml(c, indent),
+    None => showHtml(c, indent),
-            Some(rest) => showHtml(c, indent) + showChildren(rest, indent)
+    Some(rest) => showHtml(c, indent) + showChildren(rest, indent),
-        }
+},
-    }
+}
 fn showHtml(html: Html<Msg>, indent: Int): String =
    match html {
-        Empty => "",
+    Empty => "",
-        Text(s) => s,
+    Text(s) => s,
-        Element(tag, attrs, children) => {
+    Element(tag, attrs, children) => {
-            let attrStr = showAttrs(attrs)
+    let attrStr = showAttrs(attrs)
-            let attrPart = if String.length(attrStr) > 0 then " " + attrStr else ""
+    let attrPart = if String.length(attrStr) > 0 then " " + attrStr else ""
-            let childStr = showChildren(children, indent + 2)
+    let childStr = showChildren(children, indent + 2)
-            "<" + tag + attrPart + ">" + childStr + "</" + tag + ">"
+    "<" + tag + attrPart + ">" + childStr + "</" + tag + ">"
-        }
+},
-    }
+}
 // ============================================================================
 // Entry point
 // ============================================================================
 fn main(): Unit with {Console} = {
    let model = init()
@@ -150,24 +105,19 @@ fn main(): Unit with {Console} = {
    Console.print("")
    Console.print("Initial count: " + toString(getCount(model)))
    Console.print("")
    let m1 = update(model, Increment)
    Console.print("After Increment: " + toString(getCount(m1)))
    let m2 = update(m1, Increment)
    Console.print("After Increment: " + toString(getCount(m2)))
    let m3 = update(m2, Increment)
    Console.print("After Increment: " + toString(getCount(m3)))
    let m4 = update(m3, Decrement)
    Console.print("After Decrement: " + toString(getCount(m4)))
    let m5 = update(m4, Reset)
    Console.print("After Reset: " + toString(getCount(m5)))
    Console.print("")
    Console.print("=== View (HTML Structure) ===")
    Console.print(showHtml(view(m2), 0))
 }
 let output = run main() with {}
--- a/examples/datatypes.lux
+++ b/examples/datatypes.lux
@@ -1,57 +1,37 @@
 // Demonstrating algebraic data types and pattern matching
 //
 // Expected output:
 // Tree sum: 8
 // Tree depth: 3
 // Safe divide 10/2: Result: 5
 // Safe divide 10/0: Division by zero!
 // Define a binary tree
 type Tree =
    | Leaf(Int)
    | Node(Tree, Tree)
 // Sum all values in a tree
 fn sumTree(tree: Tree): Int =
    match tree {
-        Leaf(n) => n,
+    Leaf(n) => n,
-        Node(left, right) => sumTree(left) + sumTree(right)
+    Node(left, right) => sumTree(left) + sumTree(right),
-    }
+}
 // Find the depth of a tree
 fn depth(tree: Tree): Int =
    match tree {
-        Leaf(_) => 1,
+    Leaf(_) => 1,
-        Node(left, right) => {
+    Node(left, right) => {
-            let leftDepth = depth(left)
+    let leftDepth = depth(left)
-            let rightDepth = depth(right)
+    let rightDepth = depth(right)
-            1 + (if leftDepth > rightDepth then leftDepth else rightDepth)
+    1 + if leftDepth > rightDepth then leftDepth else rightDepth
-        }
+},
-    }
+}
 // Example tree:
 //       Node
 //      /    \
 //    Node   Leaf(5)
 //   /    \
 // Leaf(1) Leaf(2)
 let myTree = Node(Node(Leaf(1), Leaf(2)), Leaf(5))
 let treeSum = sumTree(myTree)
 let treeDepth = depth(myTree)
-// Option type example
+fn safeDivide(a: Int, b: Int): Option<Int> = if b == 0 then None else Some(a / b)
 fn safeDivide(a: Int, b: Int): Option<Int> =
    if b == 0 then None
    else Some(a / b)
 fn showResult(result: Option<Int>): String =
    match result {
-        None => "Division by zero!",
+    None => "Division by zero!",
-        Some(n) => "Result: " + toString(n)
+    Some(n) => "Result: " + toString(n),
-    }
+}
 // Print results
 fn printResults(): Unit with {Console} = {
    Console.print("Tree sum: " + toString(treeSum))
    Console.print("Tree depth: " + toString(treeDepth))
--- a/examples/effects.lux
+++ b/examples/effects.lux
@@ -1,17 +1,8 @@
 // Demonstrating algebraic effects in Lux
 //
 // Expected output:
 // [info] Processing data...
 // [debug] Result computed
 // Final result: 42
 // Define a custom logging effect
 effect Logger {
    fn log(level: String, msg: String): Unit
    fn getLevel(): String
 }
 // A function that uses the Logger effect
 fn processData(data: Int): Int with {Logger} = {
    Logger.log("info", "Processing data...")
    let result = data * 2
@@ -19,17 +10,15 @@ fn processData(data: Int): Int with {Logger} = {
    result
 }
 // A handler that prints logs to console
 handler consoleLogger: Logger {
    fn log(level, msg) = Console.print("[" + level + "] " + msg)
    fn getLevel() = "debug"
 }
 // Run and print
 fn main(): Unit with {Console} = {
    let result = run processData(21) with {
-        Logger = consoleLogger
+    Logger = consoleLogger,
-    }
+}
    Console.print("Final result: " + toString(result))
 }
--- a/examples/factorial.lux
+++ b/examples/factorial.lux
@@ -1,16 +1,7 @@
-// Factorial function demonstrating recursion
+fn factorial(n: Int): Int = if n <= 1 then 1 else n * factorial(n - 1)
 //
 // Expected output: 10! = 3628800
 fn factorial(n: Int): Int =
    if n <= 1 then 1
    else n * factorial(n - 1)
 // Calculate factorial of 10
 let result = factorial(10)
-// Print result using Console effect
+fn showResult(): Unit with {Console} = Console.print("10! = " + toString(result))
 fn showResult(): Unit with {Console} =
    Console.print("10! = " + toString(result))
 let output = run showResult() with {}
--- a/examples/file_io.lux
+++ b/examples/file_io.lux
@@ -1,9 +1,6 @@
 // File I/O example - demonstrates the File effect
 //
 // This script reads a file, counts lines/words, and writes a report
 fn countLines(content: String): Int = {
-    let lines = String.split(content, "\n")
+    let lines = String.split(content, "
 ")
    List.length(lines)
 }
@@ -14,35 +11,28 @@ fn countWords(content: String): Int = {
 fn analyzeFile(path: String): Unit with {File, Console} = {
    Console.print("Analyzing file: " + path)
    if File.exists(path) then {
-        let content = File.read(path)
+    let content = File.read(path)
-        let lines = countLines(content)
+    let lines = countLines(content)
-        let words = countWords(content)
+    let words = countWords(content)
-        let chars = String.length(content)
+    let chars = String.length(content)
-
+    Console.print("  Lines: " + toString(lines))
-        Console.print("  Lines: " + toString(lines))
+    Console.print("  Words: " + toString(words))
-        Console.print("  Words: " + toString(words))
+    Console.print("  Chars: " + toString(chars))
-        Console.print("  Chars: " + toString(chars))
+} else {
-    } else {
+    Console.print("  Error: File not found!")
-        Console.print("  Error: File not found!")
+}
    }
 }
 fn main(): Unit with {File, Console} = {
    Console.print("=== Lux File Analyzer ===")
    Console.print("")
    // Analyze this file itself
    analyzeFile("examples/file_io.lux")
    Console.print("")
    // Analyze hello.lux
    analyzeFile("examples/hello.lux")
    Console.print("")
-
+    let report = "File analysis complete.
-    // Write a report
+Analyzed 2 files."
    let report = "File analysis complete.\nAnalyzed 2 files."
    File.write("/tmp/lux_report.txt", report)
    Console.print("Report written to /tmp/lux_report.txt")
 }
--- a/examples/functional.lux
+++ b/examples/functional.lux
@@ -1,55 +1,39 @@
 // Demonstrating functional programming features
 //
 // Expected output:
 // apply(double, 21) = 42
 // compose(addOne, double)(5) = 11
 // pipe: 5 |> double |> addOne |> square = 121
 // curried add5(10) = 15
 // partial times3(7) = 21
 // record transform = 30
 // Higher-order functions
 fn apply(f: fn(Int): Int, x: Int): Int = f(x)
-fn compose(f: fn(Int): Int, g: fn(Int): Int): fn(Int): Int =
+fn compose(f: fn(Int): Int, g: fn(Int): Int): fn(Int): Int = fn(x: Int): Int => f(g(x))
    fn(x: Int): Int => f(g(x))
 // Basic functions
 fn double(x: Int): Int = x * 2
 fn addOne(x: Int): Int = x + 1
 fn square(x: Int): Int = x * x
 // Using apply
 let result1 = apply(double, 21)
 // Using compose
 let doubleAndAddOne = compose(addOne, double)
 let result2 = doubleAndAddOne(5)
-// Using the pipe operator
+let result3 = square(addOne(double(5)))
 let result3 = 5 |> double |> addOne |> square
-// Currying example
+fn add(a: Int): fn(Int): Int = fn(b: Int): Int => a + b
 fn add(a: Int): fn(Int): Int =
    fn(b: Int): Int => a + b
 let add5 = add(5)
 let result4 = add5(10)
 // Partial application simulation
 fn multiply(a: Int, b: Int): Int = a * b
 let times3 = fn(x: Int): Int => multiply(3, x)
 let result5 = times3(7)
-// Working with records
+let transform = fn(record: { x: Int, y: Int }): Int => record.x + record.y
 let transform = fn(record: { x: Int, y: Int }): Int =>
    record.x + record.y
 let point = { x: 10, y: 20 }
 let recordSum = transform(point)
 // Print all results
 fn printResults(): Unit with {Console} = {
    Console.print("apply(double, 21) = " + toString(result1))
    Console.print("compose(addOne, double)(5) = " + toString(result2))
--- a/examples/generics.lux
+++ b/examples/generics.lux
@@ -1,54 +1,43 @@
 // Demonstrating generic type parameters in Lux
 //
 // Expected output:
 // identity(42) = 42
 // identity("hello") = hello
 // first(MkPair(1, "one")) = 1
 // second(MkPair(1, "one")) = one
 // map(Some(21), double) = Some(42)
 // Generic identity function
 fn identity<T>(x: T): T = x
 // Generic pair type
 type Pair<A, B> =
    | MkPair(A, B)
 fn first<A, B>(p: Pair<A, B>): A =
    match p {
-        MkPair(a, _) => a
+    MkPair(a, _) => a,
-    }
+}
 fn second<A, B>(p: Pair<A, B>): B =
    match p {
-        MkPair(_, b) => b
+    MkPair(_, b) => b,
-    }
+}
 // Generic map function for Option
 fn mapOption<T, U>(opt: Option<T>, f: fn(T): U): Option<U> =
    match opt {
-        None => None,
+    None => None,
-        Some(x) => Some(f(x))
+    Some(x) => Some(f(x)),
-    }
+}
 // Helper function for testing
 fn double(x: Int): Int = x * 2
 // Test usage
 let id_int = identity(42)
 let id_str = identity("hello")
 let pair = MkPair(1, "one")
 let fst = first(pair)
 let snd = second(pair)
 let doubled = mapOption(Some(21), double)
 fn showOption(opt: Option<Int>): String =
    match opt {
-        None => "None",
+    None => "None",
-        Some(x) => "Some(" + toString(x) + ")"
+    Some(x) => "Some(" + toString(x) + ")",
-    }
+}
 fn printResults(): Unit with {Console} = {
    Console.print("identity(42) = " + toString(id_int))
--- a/examples/handlers.lux
+++ b/examples/handlers.lux
@@ -1,21 +1,8 @@
 // Demonstrating resumable effect handlers in Lux
 //
 // Handlers can use `resume(value)` to return a value to the effect call site
 // and continue the computation. This enables powerful control flow patterns.
 //
 // Expected output:
 // [INFO] Starting computation
 // [DEBUG] Intermediate result: 10
 // [INFO] Computation complete
 // Final result: 20
 // Define a custom logging effect
 effect Logger {
    fn log(level: String, msg: String): Unit
    fn getLogLevel(): String
 }
 // A function that uses the Logger effect
 fn compute(): Int with {Logger} = {
    Logger.log("INFO", "Starting computation")
    let x = 10
@@ -25,20 +12,19 @@ fn compute(): Int with {Logger} = {
    result
 }
 // A handler that prints logs with brackets and resumes with Unit
 handler prettyLogger: Logger {
-    fn log(level, msg) = {
+    fn log(level, msg) = 
-        Console.print("[" + level + "] " + msg)
+        {
-        resume(())
+    Console.print("[" + level + "] " + msg)
-    }
+    resume(())
 }
    fn getLogLevel() = resume("DEBUG")
 }
 // Main function
 fn main(): Unit with {Console} = {
    let result = run compute() with {
-        Logger = prettyLogger
+    Logger = prettyLogger,
-    }
+}
    Console.print("Final result: " + toString(result))
 }
--- a/examples/hello.lux
+++ b/examples/hello.lux
@@ -1,10 +1,3 @@
-// Hello World in Lux
+fn greet(): Unit with {Console} = Console.print("Hello, World!")
 // Demonstrates basic effect usage
 //
 // Expected output: Hello, World!
 fn greet(): Unit with {Console} =
    Console.print("Hello, World!")
 // Run the greeting with the Console effect
 let output = run greet() with {}
--- a/examples/http.lux
+++ b/examples/http.lux
@@ -1,91 +1,72 @@
 // HTTP example - demonstrates the Http effect
 //
 // This script makes HTTP requests and parses JSON responses
 fn main(): Unit with {Console, Http} = {
    Console.print("=== Lux HTTP Example ===")
    Console.print("")
    // Make a GET request to a public API
    Console.print("Fetching data from httpbin.org...")
    Console.print("")
    match Http.get("https://httpbin.org/get") {
-        Ok(response) => {
+    Ok(response) => {
-            Console.print("GET request successful!")
+    Console.print("GET request successful!")
-            Console.print("  Status: " + toString(response.status))
+    Console.print("  Status: " + toString(response.status))
-            Console.print("  Body length: " + toString(String.length(response.body)) + " bytes")
+    Console.print("  Body length: " + toString(String.length(response.body)) + " bytes")
-            Console.print("")
+    Console.print("")
-
+    match Json.parse(response.body) {
-            // Parse the JSON response
+    Ok(json) => {
-            match Json.parse(response.body) {
+    Console.print("Parsed JSON response:")
-                Ok(json) => {
+    match Json.get(json, "origin") {
-                    Console.print("Parsed JSON response:")
+    Some(origin) => match Json.asString(origin) {
-                    match Json.get(json, "origin") {
+    Some(ip) => Console.print("  Your IP: " + ip),
-                        Some(origin) => match Json.asString(origin) {
+    None => Console.print("  origin: (not a string)"),
-                            Some(ip) => Console.print("  Your IP: " + ip),
+},
-                            None => Console.print("  origin: (not a string)")
+    None => Console.print("  origin: (not found)"),
-                        },
+}
-                        None => Console.print("  origin: (not found)")
+    match Json.get(json, "url") {
-                    }
+    Some(url) => match Json.asString(url) {
-                    match Json.get(json, "url") {
+    Some(u) => Console.print("  URL: " + u),
-                        Some(url) => match Json.asString(url) {
+    None => Console.print("  url: (not a string)"),
-                            Some(u) => Console.print("  URL: " + u),
+},
-                            None => Console.print("  url: (not a string)")
+    None => Console.print("  url: (not found)"),
-                        },
+}
-                        None => Console.print("  url: (not found)")
+},
-                    }
+    Err(e) => Console.print("JSON parse error: " + e),
-                },
+}
-                Err(e) => Console.print("JSON parse error: " + e)
+},
-            }
+    Err(e) => Console.print("GET request failed: " + e),
-        },
+}
        Err(e) => Console.print("GET request failed: " + e)
    }
    Console.print("")
    Console.print("--- POST Request ---")
    Console.print("")
    // Make a POST request with JSON body
    let requestBody = Json.object([("message", Json.string("Hello from Lux!")), ("version", Json.int(1))])
    Console.print("Sending POST with JSON body...")
    Console.print("  Body: " + Json.stringify(requestBody))
    Console.print("")
    match Http.postJson("https://httpbin.org/post", requestBody) {
-        Ok(response) => {
+    Ok(response) => {
-            Console.print("POST request successful!")
+    Console.print("POST request successful!")
-            Console.print("  Status: " + toString(response.status))
+    Console.print("  Status: " + toString(response.status))
-
+    match Json.parse(response.body) {
-            // Parse and extract what we sent
+    Ok(json) => match Json.get(json, "json") {
-            match Json.parse(response.body) {
+    Some(sentJson) => {
-                Ok(json) => match Json.get(json, "json") {
+    Console.print("  Server received:")
-                    Some(sentJson) => {
+    Console.print("    " + Json.stringify(sentJson))
-                        Console.print("  Server received:")
+},
-                        Console.print("    " + Json.stringify(sentJson))
+    None => Console.print("  (no json field in response)"),
-                    },
+},
-                    None => Console.print("  (no json field in response)")
+    Err(e) => Console.print("JSON parse error: " + e),
-                },
+}
-                Err(e) => Console.print("JSON parse error: " + e)
+},
-            }
+    Err(e) => Console.print("POST request failed: " + e),
-        },
+}
        Err(e) => Console.print("POST request failed: " + e)
    }
    Console.print("")
    Console.print("--- Headers ---")
    Console.print("")
    // Show response headers
    match Http.get("https://httpbin.org/headers") {
-        Ok(response) => {
+    Ok(response) => {
-            Console.print("Response headers (first 5):")
+    Console.print("Response headers (first 5):")
-            let count = 0
+    let count = 0
-            // Note: Can't easily iterate with effects in callbacks, so just show count
+    Console.print("  Total headers: " + toString(List.length(response.headers)))
-            Console.print("  Total headers: " + toString(List.length(response.headers)))
+},
-        },
+    Err(e) => Console.print("Request failed: " + e),
-        Err(e) => Console.print("Request failed: " + e)
+}
    }
 }
 let result = run main() with {}
--- a/examples/http_api.lux
+++ b/examples/http_api.lux
@@ -1,85 +1,48 @@
-// HTTP API Example
+fn httpOk(body: String): { status: Int, body: String } = { status: 200, body: body }
 //
 // A complete REST API demonstrating:
 // - Route matching with path parameters
 // - Response builders
 // - JSON construction
 //
 // Run with: lux examples/http_api.lux
 // Test with:
 //   curl http://localhost:8080/
 //   curl http://localhost:8080/users
 //   curl http://localhost:8080/users/42
-// ============================================================
+fn httpCreated(body: String): { status: Int, body: String } = { status: 201, body: body }
 // Response Helpers
 // ============================================================
-fn httpOk(body: String): { status: Int, body: String } =
+fn httpNotFound(body: String): { status: Int, body: String } = { status: 404, body: body }
    { status: 200, body: body }
-fn httpCreated(body: String): { status: Int, body: String } =
+fn httpBadRequest(body: String): { status: Int, body: String } = { status: 400, body: body }
    { status: 201, body: body }
-fn httpNotFound(body: String): { status: Int, body: String } =
+fn jsonEscape(s: String): String = String.replace(String.replace(s, "\\", "\\\\"), "\"", "\\\"")
    { status: 404, body: body }
-fn httpBadRequest(body: String): { status: Int, body: String } =
+fn jsonStr(key: String, value: String): String = "\"" + jsonEscape(key) + "\":\"" + jsonEscape(value) + "\""
    { status: 400, body: body }
-// ============================================================
+fn jsonNum(key: String, value: Int): String = "\"" + jsonEscape(key) + "\":" + toString(value)
 // JSON Helpers
 // ============================================================
-fn jsonEscape(s: String): String =
+fn jsonObj(content: String): String = toString(" + content + ")
    String.replace(String.replace(s, "\\", "\\\\"), "\"", "\\\"")
-fn jsonStr(key: String, value: String): String =
+fn jsonArr(content: String): String = "[" + content + "]"
    "\"" + jsonEscape(key) + "\":\"" + jsonEscape(value) + "\""
-fn jsonNum(key: String, value: Int): String =
+fn jsonError(message: String): String = jsonObj(jsonStr("error", message))
    "\"" + jsonEscape(key) + "\":" + toString(value)
 fn jsonObj(content: String): String =
    "{" + content + "}"
 fn jsonArr(content: String): String =
    "[" + content + "]"
 fn jsonError(message: String): String =
    jsonObj(jsonStr("error", message))
 // ============================================================
 // Path Matching
 // ============================================================
 fn pathMatches(path: String, pattern: String): Bool = {
    let pathParts = String.split(path, "/")
    let patternParts = String.split(pattern, "/")
-    if List.length(pathParts) != List.length(patternParts) then false
+    if List.length(pathParts) != List.length(patternParts) then false else matchParts(pathParts, patternParts)
    else matchParts(pathParts, patternParts)
 }
 fn matchParts(pathParts: List<String>, patternParts: List<String>): Bool = {
-    if List.length(pathParts) == 0 then true
+    if List.length(pathParts) == 0 then true else {
-    else {
+    match List.head(pathParts) {
-        match List.head(pathParts) {
+    None => true,
-            None => true,
+    Some(pathPart) => {
-            Some(pathPart) => {
+    match List.head(patternParts) {
-                match List.head(patternParts) {
+    None => true,
-                    None => true,
+    Some(patternPart) => {
-                    Some(patternPart) => {
+    let isMatch = if String.startsWith(patternPart, ":") then true else pathPart == patternPart
-                        let isMatch = if String.startsWith(patternPart, ":") then true else pathPart == patternPart
+    if isMatch then {
-                        if isMatch then {
+    let restPath = Option.getOrElse(List.tail(pathParts), [])
-                            let restPath = Option.getOrElse(List.tail(pathParts), [])
+    let restPattern = Option.getOrElse(List.tail(patternParts), [])
-                            let restPattern = Option.getOrElse(List.tail(patternParts), [])
+    matchParts(restPath, restPattern)
-                            matchParts(restPath, restPattern)
+} else false
-                        } else false
+},
-                    }
+}
-                }
+},
-            }
+}
-        }
+}
    }
 }
 fn getPathSegment(path: String, index: Int): Option<String> = {
@@ -87,15 +50,9 @@ fn getPathSegment(path: String, index: Int): Option<String> = {
    List.get(parts, index + 1)
 }
-// ============================================================
+fn indexHandler(): { status: Int, body: String } = httpOk(jsonObj(jsonStr("message", "Welcome to Lux HTTP API")))
 // Handlers
 // ============================================================
-fn indexHandler(): { status: Int, body: String } =
+fn healthHandler(): { status: Int, body: String } = httpOk(jsonObj(jsonStr("status", "healthy")))
    httpOk(jsonObj(jsonStr("message", "Welcome to Lux HTTP API")))
 fn healthHandler(): { status: Int, body: String } =
    httpOk(jsonObj(jsonStr("status", "healthy")))
 fn listUsersHandler(): { status: Int, body: String } = {
    let user1 = jsonObj(jsonNum("id", 1) + "," + jsonStr("name", "Alice"))
@@ -105,12 +62,12 @@ fn listUsersHandler(): { status: Int, body: String } = {
 fn getUserHandler(path: String): { status: Int, body: String } = {
    match getPathSegment(path, 1) {
-        Some(id) => {
+    Some(id) => {
-            let body = jsonObj(jsonStr("id", id) + "," + jsonStr("name", "User " + id))
+    let body = jsonObj(jsonStr("id", id) + "," + jsonStr("name", "User " + id))
-            httpOk(body)
+    httpOk(body)
-        },
+},
-        None => httpNotFound(jsonError("User not found"))
+    None => httpNotFound(jsonError("User not found")),
-    }
+}
 }
 fn createUserHandler(body: String): { status: Int, body: String } = {
@@ -118,34 +75,21 @@ fn createUserHandler(body: String): { status: Int, body: String } = {
    httpCreated(newUser)
 }
 // ============================================================
 // Router
 // ============================================================
 fn router(method: String, path: String, body: String): { status: Int, body: String } = {
-    if method == "GET" && path == "/" then indexHandler()
+    if method == "GET" && path == "/" then indexHandler() else if method == "GET" && path == "/health" then healthHandler() else if method == "GET" && path == "/users" then listUsersHandler() else if method == "GET" && pathMatches(path, "/users/:id") then getUserHandler(path) else if method == "POST" && path == "/users" then createUserHandler(body) else httpNotFound(jsonError("Not found: " + path))
    else if method == "GET" && path == "/health" then healthHandler()
    else if method == "GET" && path == "/users" then listUsersHandler()
    else if method == "GET" && pathMatches(path, "/users/:id") then getUserHandler(path)
    else if method == "POST" && path == "/users" then createUserHandler(body)
    else httpNotFound(jsonError("Not found: " + path))
 }
 // ============================================================
 // Server
 // ============================================================
 fn serveLoop(remaining: Int): Unit with {Console, HttpServer} = {
    if remaining <= 0 then {
-        Console.print("Max requests reached, stopping server.")
+    Console.print("Max requests reached, stopping server.")
-        HttpServer.stop()
+    HttpServer.stop()
-    } else {
+} else {
-        let req = HttpServer.accept()
+    let req = HttpServer.accept()
-        Console.print(req.method + " " + req.path)
+    Console.print(req.method + " " + req.path)
-        let resp = router(req.method, req.path, req.body)
+    let resp = router(req.method, req.path, req.body)
-        HttpServer.respond(resp.status, resp.body)
+    HttpServer.respond(resp.status, resp.body)
-        serveLoop(remaining - 1)
+    serveLoop(remaining - 1)
-    }
+}
 }
 fn main(): Unit with {Console, HttpServer} = {
--- a/examples/http_router.lux
+++ b/examples/http_router.lux
@@ -1,24 +1,4 @@
-// HTTP Router Example
+fn indexHandler(): { status: Int, body: String } = httpOk("Welcome to Lux HTTP Framework!")
 //
 // Demonstrates the HTTP helper library with:
 // - Path pattern matching
 // - Response builders
 // - JSON helpers
 //
 // Run with: lux examples/http_router.lux
 // Test with:
 //   curl http://localhost:8080/
 //   curl http://localhost:8080/users
 //   curl http://localhost:8080/users/42
 import stdlib/http
 // ============================================================
 // Route Handlers
 // ============================================================
 fn indexHandler(): { status: Int, body: String } =
    httpOk("Welcome to Lux HTTP Framework!")
 fn listUsersHandler(): { status: Int, body: String } = {
    let user1 = jsonObject(jsonJoin([jsonNumber("id", 1), jsonString("name", "Alice")]))
@@ -29,44 +9,31 @@ fn listUsersHandler(): { status: Int, body: String } = {
 fn getUserHandler(path: String): { status: Int, body: String } = {
    match getPathSegment(path, 1) {
-        Some(id) => {
+    Some(id) => {
-            let body = jsonObject(jsonJoin([jsonString("id", id), jsonString("name", "User " + id)]))
+    let body = jsonObject(jsonJoin([jsonString("id", id), jsonString("name", "User " + id)]))
-            httpOk(body)
+    httpOk(body)
-        },
+},
-        None => httpNotFound(jsonErrorMsg("User ID required"))
+    None => httpNotFound(jsonErrorMsg("User ID required")),
-    }
+}
 }
-fn healthHandler(): { status: Int, body: String } =
+fn healthHandler(): { status: Int, body: String } = httpOk(jsonObject(jsonString("status", "healthy")))
    httpOk(jsonObject(jsonString("status", "healthy")))
 // ============================================================
 // Router
 // ============================================================
 fn router(method: String, path: String, body: String): { status: Int, body: String } = {
-    if method == "GET" && path == "/" then indexHandler()
+    if method == "GET" && path == "/" then indexHandler() else if method == "GET" && path == "/health" then healthHandler() else if method == "GET" && path == "/users" then listUsersHandler() else if method == "GET" && pathMatches(path, "/users/:id") then getUserHandler(path) else httpNotFound(jsonErrorMsg("Not found: " + path))
    else if method == "GET" && path == "/health" then healthHandler()
    else if method == "GET" && path == "/users" then listUsersHandler()
    else if method == "GET" && pathMatches(path, "/users/:id") then getUserHandler(path)
    else httpNotFound(jsonErrorMsg("Not found: " + path))
 }
 // ============================================================
 // Server
 // ============================================================
 fn serveLoop(remaining: Int): Unit with {Console, HttpServer} = {
    if remaining <= 0 then {
-        Console.print("Max requests reached, stopping server.")
+    Console.print("Max requests reached, stopping server.")
-        HttpServer.stop()
+    HttpServer.stop()
-    } else {
+} else {
-        let req = HttpServer.accept()
+    let req = HttpServer.accept()
-        Console.print(req.method + " " + req.path)
+    Console.print(req.method + " " + req.path)
-        let resp = router(req.method, req.path, req.body)
+    let resp = router(req.method, req.path, req.body)
-        HttpServer.respond(resp.status, resp.body)
+    HttpServer.respond(resp.status, resp.body)
-        serveLoop(remaining - 1)
+    serveLoop(remaining - 1)
-    }
+}
 }
 fn main(): Unit with {Console, HttpServer} = {
--- a/examples/jit_test.lux
+++ b/examples/jit_test.lux
@@ -1,13 +1,6 @@
-// Test file for JIT compilation
+fn fib(n: Int): Int = if n <= 1 then n else fib(n - 1) + fib(n - 2)
 // This uses only features the JIT supports: integers, arithmetic, conditionals, functions
-fn fib(n: Int): Int =
+fn factorial(n: Int): Int = if n <= 1 then 1 else n * factorial(n - 1)
    if n <= 1 then n
    else fib(n - 1) + fib(n - 2)
 fn factorial(n: Int): Int =
    if n <= 1 then 1
    else n * factorial(n - 1)
 fn main(): Unit with {Console} = {
    let fibResult = fib(30)
--- a/examples/json.lux
+++ b/examples/json.lux
@@ -1,107 +1,79 @@
 // JSON example - demonstrates JSON parsing and manipulation
 //
 // This script parses JSON, extracts values, and builds new JSON structures
 fn main(): Unit with {Console, File} = {
    Console.print("=== Lux JSON Example ===")
    Console.print("")
    // First, build some JSON programmatically
    Console.print("=== Building JSON ===")
    Console.print("")
    let name = Json.string("Alice")
    let age = Json.int(30)
    let active = Json.bool(true)
    let scores = Json.array([Json.int(95), Json.int(87), Json.int(92)])
    let person = Json.object([("name", name), ("age", age), ("active", active), ("scores", scores)])
    Console.print("Built JSON:")
    let pretty = Json.prettyPrint(person)
    Console.print(pretty)
    Console.print("")
    // Stringify to a compact string
    let jsonStr = Json.stringify(person)
    Console.print("Compact: " + jsonStr)
    Console.print("")
    // Write to file and read back to test parsing
    File.write("/tmp/test.json", jsonStr)
    Console.print("Written to /tmp/test.json")
    Console.print("")
    // Read and parse from file
    Console.print("=== Parsing JSON ===")
    Console.print("")
    let content = File.read("/tmp/test.json")
    Console.print("Read from file: " + content)
    Console.print("")
    match Json.parse(content) {
-        Ok(json) => {
+    Ok(json) => {
-            Console.print("Parse succeeded!")
+    Console.print("Parse succeeded!")
-            Console.print("")
+    Console.print("")
-
+    Console.print("Extracting fields:")
-            // Get string field
+    match Json.get(json, "name") {
-            Console.print("Extracting fields:")
+    Some(nameJson) => match Json.asString(nameJson) {
-            match Json.get(json, "name") {
+    Some(n) => Console.print("  name: " + n),
-                Some(nameJson) => match Json.asString(nameJson) {
+    None => Console.print("  name: (not a string)"),
-                    Some(n) => Console.print("  name: " + n),
+},
-                    None => Console.print("  name: (not a string)")
+    None => Console.print("  name: (not found)"),
-                },
+}
-                None => Console.print("  name: (not found)")
+    match Json.get(json, "age") {
-            }
+    Some(ageJson) => match Json.asInt(ageJson) {
-
+    Some(a) => Console.print("  age: " + toString(a)),
-            // Get int field
+    None => Console.print("  age: (not an int)"),
-            match Json.get(json, "age") {
+},
-                Some(ageJson) => match Json.asInt(ageJson) {
+    None => Console.print("  age: (not found)"),
-                    Some(a) => Console.print("  age: " + toString(a)),
+}
-                    None => Console.print("  age: (not an int)")
+    match Json.get(json, "active") {
-                },
+    Some(activeJson) => match Json.asBool(activeJson) {
-                None => Console.print("  age: (not found)")
+    Some(a) => Console.print("  active: " + toString(a)),
-            }
+    None => Console.print("  active: (not a bool)"),
-
+},
-            // Get bool field
+    None => Console.print("  active: (not found)"),
-            match Json.get(json, "active") {
+}
-                Some(activeJson) => match Json.asBool(activeJson) {
+    match Json.get(json, "scores") {
-                    Some(a) => Console.print("  active: " + toString(a)),
+    Some(scoresJson) => match Json.asArray(scoresJson) {
-                    None => Console.print("  active: (not a bool)")
+    Some(arr) => {
-                },
+    Console.print("  scores: " + toString(List.length(arr)) + " items")
-                None => Console.print("  active: (not found)")
+    match Json.getIndex(scoresJson, 0) {
-            }
+    Some(firstJson) => match Json.asInt(firstJson) {
-
+    Some(first) => Console.print("    first score: " + toString(first)),
-            // Get array field
+    None => Console.print("    first score: (not an int)"),
-            match Json.get(json, "scores") {
+},
-                Some(scoresJson) => match Json.asArray(scoresJson) {
+    None => Console.print("    (no first element)"),
-                    Some(arr) => {
+}
-                        Console.print("  scores: " + toString(List.length(arr)) + " items")
+},
-                        // Get first score
+    None => Console.print("  scores: (not an array)"),
-                        match Json.getIndex(scoresJson, 0) {
+},
-                            Some(firstJson) => match Json.asInt(firstJson) {
+    None => Console.print("  scores: (not found)"),
-                                Some(first) => Console.print("    first score: " + toString(first)),
+}
-                                None => Console.print("    first score: (not an int)")
+    Console.print("")
-                            },
+    Console.print("Object keys:")
-                            None => Console.print("    (no first element)")
+    match Json.keys(json) {
-                        }
+    Some(ks) => Console.print("  " + String.join(ks, ", ")),
-                    },
+    None => Console.print("  (not an object)"),
-                    None => Console.print("  scores: (not an array)")
+}
-                },
+},
-                None => Console.print("  scores: (not found)")
+    Err(e) => Console.print("Parse error: " + e),
-            }
+}
            Console.print("")
            // Get the keys
            Console.print("Object keys:")
            match Json.keys(json) {
                Some(ks) => Console.print("  " + String.join(ks, ", ")),
                None => Console.print("  (not an object)")
            }
        },
        Err(e) => Console.print("Parse error: " + e)
    }
    Console.print("")
    Console.print("=== JSON Null Check ===")
    let nullVal = Json.null()
--- a/examples/modules/main.lux
+++ b/examples/modules/main.lux
@@ -1,17 +1,9 @@
 // Main program that imports modules
 import examples/modules/math_utils
 import examples/modules/string_utils
 fn main(): Unit with {Console} = {
    Console.print("=== Testing Module Imports ===")
    // Use math_utils
    Console.print("square(5) = " + toString(math_utils.square(5)))
    Console.print("cube(3) = " + toString(math_utils.cube(3)))
    Console.print("factorial(6) = " + toString(math_utils.factorial(6)))
    Console.print("sumRange(1, 10) = " + toString(math_utils.sumRange(1, 10)))
    // Use string_utils
    Console.print(string_utils.greet("World"))
    Console.print(string_utils.exclaim("Modules work"))
    Console.print("repeat(\"ab\", 3) = " + string_utils.repeat("ab", 3))
--- a/examples/modules/main_selective.lux
+++ b/examples/modules/main_selective.lux
@@ -1,15 +1,7 @@
 // Test selective imports
 import examples/modules/math_utils.{square, factorial}
 import examples/modules/string_utils as str
 fn main(): Unit with {Console} = {
    Console.print("=== Selective & Aliased Imports ===")
    // Direct imports (no module prefix)
    Console.print("square(7) = " + toString(square(7)))
    Console.print("factorial(5) = " + toString(factorial(5)))
    // Aliased import
    Console.print(str.greet("Lux"))
    Console.print(str.exclaim("Aliased imports work"))
 }
--- a/examples/modules/main_wildcard.lux
+++ b/examples/modules/main_wildcard.lux
@@ -1,10 +1,5 @@
 // Test wildcard imports
 import examples/modules/math_utils.*
 fn main(): Unit with {Console} = {
    Console.print("=== Wildcard Imports ===")
    // All functions available directly
    Console.print("square(4) = " + toString(square(4)))
    Console.print("cube(4) = " + toString(cube(4)))
    Console.print("factorial(4) = " + toString(factorial(4)))
--- a/examples/modules/math_utils.lux
+++ b/examples/modules/math_utils.lux
@@ -1,14 +1,7 @@
-// Math utilities module
+fn square(n: Int): Int = n * n
 // Exports: square, cube, factorial
-pub fn square(n: Int): Int = n * n
+fn cube(n: Int): Int = n * n * n
-pub fn cube(n: Int): Int = n * n * n
+fn factorial(n: Int): Int = if n <= 1 then 1 else n * factorial(n - 1)
-pub fn factorial(n: Int): Int =
+fn sumRange(start: Int, end: Int): Int = if start > end then 0 else start + sumRange(start + 1, end)
    if n <= 1 then 1
    else n * factorial(n - 1)
 pub fn sumRange(start: Int, end: Int): Int =
    if start > end then 0
    else start + sumRange(start + 1, end)
--- a/examples/modules/string_utils.lux
+++ b/examples/modules/string_utils.lux
@@ -1,11 +1,5 @@
-// String utilities module
+fn repeat(s: String, n: Int): String = if n <= 0 then "" else s + repeat(s, n - 1)
 // Exports: repeat, exclaim
-pub fn repeat(s: String, n: Int): String =
+fn exclaim(s: String): String = s + "!"
    if n <= 0 then ""
    else s + repeat(s, n - 1)
-pub fn exclaim(s: String): String = s + "!"
+fn greet(name: String): String = "Hello, " + name + "!"
 pub fn greet(name: String): String =
    "Hello, " + name + "!"
--- a/examples/modules/use_stdlib.lux
+++ b/examples/modules/use_stdlib.lux
@@ -1,17 +1,9 @@
 // Example using the standard library
 import std/prelude.*
 import std/option as opt
 fn main(): Unit with {Console} = {
    Console.print("=== Using Standard Library ===")
    // Prelude functions
    Console.print("identity(42) = " + toString(identity(42)))
    Console.print("not(true) = " + toString(not(true)))
    Console.print("and(true, false) = " + toString(and(true, false)))
    Console.print("or(true, false) = " + toString(or(true, false)))
    // Option utilities
    let x = opt.some(10)
    let y = opt.none()
    Console.print("isSome(Some(10)) = " + toString(opt.isSome(x)))
--- a/examples/pipelines.lux
+++ b/examples/pipelines.lux
@@ -1,47 +1,31 @@
 // Demonstrating the pipe operator and functional data processing
 //
 // Expected output:
 // 5 |> double |> addTen |> square = 400
 // Pipeline result2 = 42
 // process(1) = 144
 // process(2) = 196
 // process(3) = 256
 // clamped = 0
 // composed = 121
 // Basic transformations
 fn double(x: Int): Int = x * 2
 fn addTen(x: Int): Int = x + 10
 fn square(x: Int): Int = x * x
 fn negate(x: Int): Int = -x
-// Using the pipe operator for data transformation
+let result1 = square(addTen(double(5)))
 let result1 = 5 |> double |> addTen |> square
-// Chaining multiple operations
+let result2 = addTen(double(addTen(double(3))))
 let result2 = 3 |> double |> addTen |> double |> addTen
-// More complex pipelines
+fn process(n: Int): Int = square(addTen(double(n)))
 fn process(n: Int): Int =
    n |> double |> addTen |> square
 // Multiple values through same pipeline
 let a = process(1)
 let b = process(2)
 let c = process(3)
-// Conditional in pipeline
+fn clampPositive(x: Int): Int = if x < 0 then 0 else x
 fn clampPositive(x: Int): Int =
    if x < 0 then 0 else x
-let clamped = -5 |> double |> clampPositive
+let clamped = clampPositive(double(-5))
 // Function composition using pipe
 fn increment(x: Int): Int = x + 1
-let composed = 5 |> double |> increment |> square
+let composed = square(increment(double(5)))
 // Print results
 fn printResults(): Unit with {Console} = {
    Console.print("5 |> double |> addTen |> square = " + toString(result1))
    Console.print("Pipeline result2 = " + toString(result2))
--- a/examples/postgres_demo.lux
+++ b/examples/postgres_demo.lux
@@ -1,72 +1,42 @@
-// PostgreSQL Database Example
+fn jsonStr(key: String, value: String): String = "\"" + key + "\":\"" + value + "\""
 //
 // Demonstrates the Postgres effect for database operations.
 //
 // Prerequisites:
 //   - PostgreSQL server running locally
 //   - Database 'testdb' created
 //   - User 'testuser' with password 'testpass'
 //
 // To set up:
 //   createdb testdb
 //   psql testdb -c "CREATE TABLE users (id SERIAL PRIMARY KEY, name TEXT, email TEXT);"
 //
 // Run with: lux examples/postgres_demo.lux
-// ============================================================
+fn jsonNum(key: String, value: Int): String = "\"" + key + "\":" + toString(value)
 // Helper Functions
 // ============================================================
-fn jsonStr(key: String, value: String): String =
+fn jsonObj(content: String): String = toString(" + content + ")
    "\"" + key + "\":\"" + value + "\""
 fn jsonNum(key: String, value: Int): String =
    "\"" + key + "\":" + toString(value)
 fn jsonObj(content: String): String =
    "{" + content + "}"
 // ============================================================
 // Database Operations
 // ============================================================
 // Insert a user
 fn insertUser(connId: Int, name: String, email: String): Int with {Console, Postgres} = {
    let sql = "INSERT INTO users (name, email) VALUES ('" + name + "', '" + email + "') RETURNING id"
    Console.print("Inserting user: " + name)
    match Postgres.queryOne(connId, sql) {
-        Some(row) => {
+    Some(row) => {
-            Console.print("  Inserted with ID: " + toString(row.id))
+    Console.print("  Inserted with ID: " + toString(row.id))
-            row.id
+    row.id
-        },
+},
-        None => {
+    None => {
-            Console.print("  Insert failed")
+    Console.print("  Insert failed")
-            -1
+    -1
-        }
+},
-    }
+}
 }
 // Get all users
 fn getUsers(connId: Int): Unit with {Console, Postgres} = {
    Console.print("Fetching all users...")
    let rows = Postgres.query(connId, "SELECT id, name, email FROM users ORDER BY id")
    Console.print("  Found " + toString(List.length(rows)) + " users:")
-    List.forEach(rows, fn(row: { id: Int, name: String, email: String }): Unit with {Console} => {
+    List.forEach(rows, fn(row: { id: Int, name: String, email: String }): Unit => {
-        Console.print("    - " + toString(row.id) + ": " + row.name + " <" + row.email + ">")
+    Console.print("    - " + toString(row.id) + ": " + row.name + " <" + row.email + ">")
-    })
+})
 }
 // Get user by ID
 fn getUserById(connId: Int, id: Int): Unit with {Console, Postgres} = {
    let sql = "SELECT id, name, email FROM users WHERE id = " + toString(id)
    Console.print("Looking up user " + toString(id) + "...")
    match Postgres.queryOne(connId, sql) {
-        Some(row) => Console.print("  Found: " + row.name + " <" + row.email + ">"),
+    Some(row) => Console.print("  Found: " + row.name + " <" + row.email + ">"),
-        None => Console.print("  User not found")
+    None => Console.print("  User not found"),
-    }
+}
 }
 // Update user email
 fn updateUserEmail(connId: Int, id: Int, newEmail: String): Unit with {Console, Postgres} = {
    let sql = "UPDATE users SET email = '" + newEmail + "' WHERE id = " + toString(id)
    Console.print("Updating user " + toString(id) + " email to " + newEmail)
@@ -74,7 +44,6 @@ fn updateUserEmail(connId: Int, id: Int, newEmail: String): Unit with {Console,
    Console.print("  Rows affected: " + toString(affected))
 }
 // Delete user
 fn deleteUser(connId: Int, id: Int): Unit with {Console, Postgres} = {
    let sql = "DELETE FROM users WHERE id = " + toString(id)
    Console.print("Deleting user " + toString(id))
@@ -82,104 +51,63 @@ fn deleteUser(connId: Int, id: Int): Unit with {Console, Postgres} = {
    Console.print("  Rows affected: " + toString(affected))
 }
 // ============================================================
 // Transaction Example
 // ============================================================
 fn transactionDemo(connId: Int): Unit with {Console, Postgres} = {
    Console.print("")
    Console.print("=== Transaction Demo ===")
    // Start transaction
    Console.print("Beginning transaction...")
    Postgres.beginTx(connId)
    // Make some changes
    insertUser(connId, "TxUser1", "tx1@example.com")
    insertUser(connId, "TxUser2", "tx2@example.com")
    // Show users before commit
    Console.print("Users before commit:")
    getUsers(connId)
    // Commit the transaction
    Console.print("Committing transaction...")
    Postgres.commit(connId)
    Console.print("Transaction committed!")
 }
 // ============================================================
 // Main
 // ============================================================
 fn main(): Unit with {Console, Postgres} = {
    Console.print("========================================")
    Console.print("  PostgreSQL Demo")
    Console.print("========================================")
    Console.print("")
    // Connect to database
    Console.print("Connecting to PostgreSQL...")
    let connStr = "host=localhost user=testuser password=testpass dbname=testdb"
    let connId = Postgres.connect(connStr)
    Console.print("Connected! Connection ID: " + toString(connId))
    Console.print("")
    // Create table if not exists
    Console.print("Creating users table...")
    Postgres.execute(connId, "CREATE TABLE IF NOT EXISTS users (id SERIAL PRIMARY KEY, name TEXT NOT NULL, email TEXT NOT NULL)")
    Console.print("")
    // Clear table for demo
    Console.print("Clearing existing data...")
    Postgres.execute(connId, "DELETE FROM users")
    Console.print("")
    // Insert some users
    Console.print("=== Inserting Users ===")
    let id1 = insertUser(connId, "Alice", "alice@example.com")
    let id2 = insertUser(connId, "Bob", "bob@example.com")
    let id3 = insertUser(connId, "Charlie", "charlie@example.com")
    Console.print("")
    // Query all users
    Console.print("=== All Users ===")
    getUsers(connId)
    Console.print("")
    // Query single user
    Console.print("=== Single User Lookup ===")
    getUserById(connId, id2)
    Console.print("")
    // Update user
    Console.print("=== Update User ===")
    updateUserEmail(connId, id2, "bob.new@example.com")
    getUserById(connId, id2)
    Console.print("")
    // Delete user
    Console.print("=== Delete User ===")
    deleteUser(connId, id3)
    getUsers(connId)
    Console.print("")
    // Transaction demo
    transactionDemo(connId)
    Console.print("")
    // Final state
    Console.print("=== Final State ===")
    getUsers(connId)
    Console.print("")
    // Close connection
    Console.print("Closing connection...")
    Postgres.close(connId)
    Console.print("Done!")
 }
 // Note: This will fail if PostgreSQL is not running
 // To test the syntax only, you can comment out the last line
 let output = run main() with {}
--- a/examples/property_testing.lux
+++ b/examples/property_testing.lux
@@ -1,18 +1,6 @@
 // Property-Based Testing Example
 //
 // This example demonstrates property-based testing in Lux,
 // where we verify properties hold for randomly generated inputs.
 //
 // Run with: lux examples/property_testing.lux
 // ============================================================
 // Generator Functions (using Random effect)
 // ============================================================
 let CHARS = "abcdefghijklmnopqrstuvwxyz"
-fn genInt(min: Int, max: Int): Int with {Random} =
+fn genInt(min: Int, max: Int): Int with {Random} = Random.int(min, max)
    Random.int(min, max)
 fn genIntList(min: Int, max: Int, maxLen: Int): List<Int> with {Random} = {
    let len = Random.int(0, maxLen)
@@ -20,10 +8,7 @@ fn genIntList(min: Int, max: Int, maxLen: Int): List<Int> with {Random} = {
 }
 fn genIntListHelper(min: Int, max: Int, len: Int): List<Int> with {Random} = {
-    if len <= 0 then
+    if len <= 0 then [] else List.concat([Random.int(min, max)], genIntListHelper(min, max, len - 1))
        []
    else
        List.concat([Random.int(min, max)], genIntListHelper(min, max, len - 1))
 }
 fn genChar(): String with {Random} = {
@@ -37,195 +22,147 @@ fn genString(maxLen: Int): String with {Random} = {
 }
 fn genStringHelper(len: Int): String with {Random} = {
-    if len <= 0 then
+    if len <= 0 then "" else genChar() + genStringHelper(len - 1)
        ""
    else
        genChar() + genStringHelper(len - 1)
 }
 // ============================================================
 // Test Runner State
 // ============================================================
 fn printResult(name: String, passed: Bool, count: Int): Unit with {Console} = {
-    if passed then
+    if passed then Console.print("  PASS " + name + " (" + toString(count) + " tests)") else Console.print("  FAIL " + name)
        Console.print("  PASS " + name + " (" + toString(count) + " tests)")
    else
        Console.print("  FAIL " + name)
 }
 // ============================================================
 // Property Tests
 // ============================================================
 // Test: List reverse is involutive
 fn testReverseInvolutive(n: Int, count: Int): Bool with {Console, Random} = {
    if n <= 0 then {
-        printResult("reverse(reverse(xs)) == xs", true, count)
+    printResult("reverse(reverse(xs)) == xs", true, count)
-        true
+    true
-    } else {
+} else {
-        let xs = genIntList(0, 100, 20)
+    let xs = genIntList(0, 100, 20)
-        if List.reverse(List.reverse(xs)) == xs then
+    if List.reverse(List.reverse(xs)) == xs then testReverseInvolutive(n - 1, count) else {
-            testReverseInvolutive(n - 1, count)
+    printResult("reverse(reverse(xs)) == xs", false, count - n + 1)
-        else {
+    false
-            printResult("reverse(reverse(xs)) == xs", false, count - n + 1)
+}
-            false
+}
        }
    }
 }
 // Test: List reverse preserves length
 fn testReverseLength(n: Int, count: Int): Bool with {Console, Random} = {
    if n <= 0 then {
-        printResult("length(reverse(xs)) == length(xs)", true, count)
+    printResult("length(reverse(xs)) == length(xs)", true, count)
-        true
+    true
-    } else {
+} else {
-        let xs = genIntList(0, 100, 20)
+    let xs = genIntList(0, 100, 20)
-        if List.length(List.reverse(xs)) == List.length(xs) then
+    if List.length(List.reverse(xs)) == List.length(xs) then testReverseLength(n - 1, count) else {
-            testReverseLength(n - 1, count)
+    printResult("length(reverse(xs)) == length(xs)", false, count - n + 1)
-        else {
+    false
-            printResult("length(reverse(xs)) == length(xs)", false, count - n + 1)
+}
-            false
+}
        }
    }
 }
 // Test: List map preserves length
 fn testMapLength(n: Int, count: Int): Bool with {Console, Random} = {
    if n <= 0 then {
-        printResult("length(map(xs, f)) == length(xs)", true, count)
+    printResult("length(map(xs, f)) == length(xs)", true, count)
-        true
+    true
-    } else {
+} else {
-        let xs = genIntList(0, 100, 20)
+    let xs = genIntList(0, 100, 20)
-        if List.length(List.map(xs, fn(x) => x * 2)) == List.length(xs) then
+    if List.length(List.map(xs, fn(x: _) => x * 2)) == List.length(xs) then testMapLength(n - 1, count) else {
-            testMapLength(n - 1, count)
+    printResult("length(map(xs, f)) == length(xs)", false, count - n + 1)
-        else {
+    false
-            printResult("length(map(xs, f)) == length(xs)", false, count - n + 1)
+}
-            false
+}
        }
    }
 }
 // Test: List concat length is sum
 fn testConcatLength(n: Int, count: Int): Bool with {Console, Random} = {
    if n <= 0 then {
-        printResult("length(xs ++ ys) == length(xs) + length(ys)", true, count)
+    printResult("length(xs ++ ys) == length(xs) + length(ys)", true, count)
-        true
+    true
-    } else {
+} else {
-        let xs = genIntList(0, 50, 10)
+    let xs = genIntList(0, 50, 10)
-        let ys = genIntList(0, 50, 10)
+    let ys = genIntList(0, 50, 10)
-        if List.length(List.concat(xs, ys)) == List.length(xs) + List.length(ys) then
+    if List.length(List.concat(xs, ys)) == List.length(xs) + List.length(ys) then testConcatLength(n - 1, count) else {
-            testConcatLength(n - 1, count)
+    printResult("length(xs ++ ys) == length(xs) + length(ys)", false, count - n + 1)
-        else {
+    false
-            printResult("length(xs ++ ys) == length(xs) + length(ys)", false, count - n + 1)
+}
-            false
+}
        }
    }
 }
 // Test: Addition is commutative
 fn testAddCommutative(n: Int, count: Int): Bool with {Console, Random} = {
    if n <= 0 then {
-        printResult("a + b == b + a", true, count)
+    printResult("a + b == b + a", true, count)
-        true
+    true
-    } else {
+} else {
-        let a = genInt(-1000, 1000)
+    let a = genInt(-1000, 1000)
-        let b = genInt(-1000, 1000)
+    let b = genInt(-1000, 1000)
-        if a + b == b + a then
+    if a + b == b + a then testAddCommutative(n - 1, count) else {
-            testAddCommutative(n - 1, count)
+    printResult("a + b == b + a", false, count - n + 1)
-        else {
+    false
-            printResult("a + b == b + a", false, count - n + 1)
+}
-            false
+}
        }
    }
 }
 // Test: Multiplication is associative
 fn testMulAssociative(n: Int, count: Int): Bool with {Console, Random} = {
    if n <= 0 then {
-        printResult("(a * b) * c == a * (b * c)", true, count)
+    printResult("(a * b) * c == a * (b * c)", true, count)
-        true
+    true
-    } else {
+} else {
-        let a = genInt(-100, 100)
+    let a = genInt(-100, 100)
-        let b = genInt(-100, 100)
+    let b = genInt(-100, 100)
-        let c = genInt(-100, 100)
+    let c = genInt(-100, 100)
-        if (a * b) * c == a * (b * c) then
+    if a * b * c == a * b * c then testMulAssociative(n - 1, count) else {
-            testMulAssociative(n - 1, count)
+    printResult("(a * b) * c == a * (b * c)", false, count - n + 1)
-        else {
+    false
-            printResult("(a * b) * c == a * (b * c)", false, count - n + 1)
+}
-            false
+}
        }
    }
 }
 // Test: String concat length is sum
 fn testStringConcatLength(n: Int, count: Int): Bool with {Console, Random} = {
    if n <= 0 then {
-        printResult("length(s1 + s2) == length(s1) + length(s2)", true, count)
+    printResult("length(s1 + s2) == length(s1) + length(s2)", true, count)
-        true
+    true
-    } else {
+} else {
-        let s1 = genString(10)
+    let s1 = genString(10)
-        let s2 = genString(10)
+    let s2 = genString(10)
-        if String.length(s1 + s2) == String.length(s1) + String.length(s2) then
+    if String.length(s1 + s2) == String.length(s1) + String.length(s2) then testStringConcatLength(n - 1, count) else {
-            testStringConcatLength(n - 1, count)
+    printResult("length(s1 + s2) == length(s1) + length(s2)", false, count - n + 1)
-        else {
+    false
-            printResult("length(s1 + s2) == length(s1) + length(s2)", false, count - n + 1)
+}
-            false
+}
        }
    }
 }
 // Test: Zero is identity for addition
 fn testAddIdentity(n: Int, count: Int): Bool with {Console, Random} = {
    if n <= 0 then {
-        printResult("x + 0 == x && 0 + x == x", true, count)
+    printResult("x + 0 == x && 0 + x == x", true, count)
-        true
+    true
-    } else {
+} else {
-        let x = genInt(-10000, 10000)
+    let x = genInt(-10000, 10000)
-        if x + 0 == x && 0 + x == x then
+    if x + 0 == x && 0 + x == x then testAddIdentity(n - 1, count) else {
-            testAddIdentity(n - 1, count)
+    printResult("x + 0 == x && 0 + x == x", false, count - n + 1)
-        else {
+    false
-            printResult("x + 0 == x && 0 + x == x", false, count - n + 1)
+}
-            false
+}
        }
    }
 }
 // Test: Filter reduces or maintains length
 fn testFilterLength(n: Int, count: Int): Bool with {Console, Random} = {
    if n <= 0 then {
-        printResult("length(filter(xs, p)) <= length(xs)", true, count)
+    printResult("length(filter(xs, p)) <= length(xs)", true, count)
-        true
+    true
-    } else {
+} else {
-        let xs = genIntList(0, 100, 20)
+    let xs = genIntList(0, 100, 20)
-        if List.length(List.filter(xs, fn(x) => x > 50)) <= List.length(xs) then
+    if List.length(List.filter(xs, fn(x: _) => x > 50)) <= List.length(xs) then testFilterLength(n - 1, count) else {
-            testFilterLength(n - 1, count)
+    printResult("length(filter(xs, p)) <= length(xs)", false, count - n + 1)
-        else {
+    false
-            printResult("length(filter(xs, p)) <= length(xs)", false, count - n + 1)
+}
-            false
+}
        }
    }
 }
 // Test: Empty list is identity for concat
 fn testConcatIdentity(n: Int, count: Int): Bool with {Console, Random} = {
    if n <= 0 then {
-        printResult("concat(xs, []) == xs && concat([], xs) == xs", true, count)
+    printResult("concat(xs, []) == xs && concat([], xs) == xs", true, count)
-        true
+    true
-    } else {
+} else {
-        let xs = genIntList(0, 100, 10)
+    let xs = genIntList(0, 100, 10)
-        if List.concat(xs, []) == xs && List.concat([], xs) == xs then
+    if List.concat(xs, []) == xs && List.concat([], xs) == xs then testConcatIdentity(n - 1, count) else {
-            testConcatIdentity(n - 1, count)
+    printResult("concat(xs, []) == xs && concat([], xs) == xs", false, count - n + 1)
-        else {
+    false
-            printResult("concat(xs, []) == xs && concat([], xs) == xs", false, count - n + 1)
+}
-            false
+}
        }
    }
 }
 // ============================================================
 // Main
 // ============================================================
 fn main(): Unit with {Console, Random} = {
    Console.print("========================================")
@@ -234,7 +171,6 @@ fn main(): Unit with {Console, Random} = {
    Console.print("")
    Console.print("Running 100 iterations per property...")
    Console.print("")
    testReverseInvolutive(100, 100)
    testReverseLength(100, 100)
    testMapLength(100, 100)
@@ -245,7 +181,6 @@ fn main(): Unit with {Console, Random} = {
    testAddIdentity(100, 100)
    testFilterLength(100, 100)
    testConcatIdentity(100, 100)
    Console.print("")
    Console.print("========================================")
    Console.print("  All property tests completed!")
--- a/examples/random.lux
+++ b/examples/random.lux
@@ -1,39 +1,22 @@
 // Demonstrating Random and Time effects in Lux
 //
 // Expected output (values will vary):
 // Rolling dice...
 // Die 1: <random 1-6>
 // Die 2: <random 1-6>
 // Die 3: <random 1-6>
 // Coin flip: <true/false>
 // Random float: <0.0-1.0>
 // Current time: <timestamp>
 // Roll a single die (1-6)
 fn rollDie(): Int with {Random} = Random.int(1, 6)
 // Roll multiple dice and print results
 fn rollDice(count: Int): Unit with {Random, Console} = {
    if count > 0 then {
-        let value = rollDie()
+    let value = rollDie()
-        Console.print("Die " + toString(4 - count) + ": " + toString(value))
+    Console.print("Die " + toString(4 - count) + ": " + toString(value))
-        rollDice(count - 1)
+    rollDice(count - 1)
-    } else {
+} else {
-        ()
+    ()
-    }
+}
 }
 // Main function demonstrating random effects
 fn main(): Unit with {Random, Console, Time} = {
    Console.print("Rolling dice...")
    rollDice(3)
    let coin = Random.bool()
    Console.print("Coin flip: " + toString(coin))
    let f = Random.float()
    Console.print("Random float: " + toString(f))
    let now = Time.now()
    Console.print("Current time: " + toString(now))
 }
--- a/examples/schema_evolution.lux
+++ b/examples/schema_evolution.lux
@@ -1,67 +1,41 @@
-// Schema Evolution Demo
+type User = {
 // Demonstrates version tracking and automatic migrations
 // ============================================================
 // PART 1: Type-Declared Migrations
 // ============================================================
 // Define a versioned type with a migration from v1 to v2
 type User @v2 {
    name: String,
    email: String,
    // Migration from v1: add default email
    from @v1 = { name: old.name, email: "unknown@example.com" }
 }
 // Create a v1 user
 let v1_user = Schema.versioned("User", 1, { name: "Alice" })
 let v1_version = Schema.getVersion(v1_user)  // 1
-// Migrate to v2 - uses the declared migration automatically
+let v1_version = Schema.getVersion(v1_user)
 let v2_user = Schema.migrate(v1_user, 2)
 let v2_version = Schema.getVersion(v2_user)  // 2
-// ============================================================
+let v2_version = Schema.getVersion(v2_user)
 // PART 2: Runtime Schema Operations (separate type)
 // ============================================================
 // Create versioned values for a different type (no migration)
 let config1 = Schema.versioned("Config", 1, "debug")
 let config2 = Schema.versioned("Config", 2, "release")
-// Check versions
+let c1 = Schema.getVersion(config1)
-let c1 = Schema.getVersion(config1)  // 1
+
-let c2 = Schema.getVersion(config2)  // 2
+let c2 = Schema.getVersion(config2)
 // Migrate config (auto-migration since no explicit migration defined)
 let upgradedConfig = Schema.migrate(config1, 2)
 let upgradedConfigVersion = Schema.getVersion(upgradedConfig)  // 2
-// ============================================================
+let upgradedConfigVersion = Schema.getVersion(upgradedConfig)
 // PART 2: Practical Example - API Versioning
 // ============================================================
-// Simulate different API response versions
+fn createResponseV1(data: String): { version: Int, payload: String } = { version: 1, payload: data }
 fn createResponseV1(data: String): { version: Int, payload: String } =
    { version: 1, payload: data }
-fn createResponseV2(data: String, timestamp: Int): { version: Int, payload: String, meta: { ts: Int } } =
+fn createResponseV2(data: String, timestamp: Int): { version: Int, payload: String, meta: { ts: Int } } = { version: 2, payload: data, meta: { ts: timestamp } }
    { version: 2, payload: data, meta: { ts: timestamp } }
-// Version-aware processing
+fn getPayload(response: { version: Int, payload: String }): String = response.payload
 fn getPayload(response: { version: Int, payload: String }): String =
    response.payload
 let resp1 = createResponseV1("Hello")
 let resp2 = createResponseV2("World", 1234567890)
 let payload1 = getPayload(resp1)
 let payload2 = resp2.payload
-// ============================================================
+let payload2 = resp2.payload
 // RESULTS
 // ============================================================
 fn main(): Unit with {Console} = {
    Console.print("=== Schema Evolution Demo ===")
--- a/examples/shell.lux
+++ b/examples/shell.lux
@@ -1,58 +1,43 @@
 // Shell/Process example - demonstrates the Process effect
 //
 // This script runs shell commands and uses environment variables
 fn main(): Unit with {Process, Console} = {
    Console.print("=== Lux Shell Example ===")
    Console.print("")
    // Get current working directory
    let cwd = Process.cwd()
    Console.print("Current directory: " + cwd)
    Console.print("")
    // Get environment variables
    Console.print("Environment variables:")
    match Process.env("USER") {
-        Some(user) => Console.print("  USER: " + user),
+    Some(user) => Console.print("  USER: " + user),
-        None => Console.print("  USER: (not set)")
+    None => Console.print("  USER: (not set)"),
-    }
+}
    match Process.env("HOME") {
-        Some(home) => Console.print("  HOME: " + home),
+    Some(home) => Console.print("  HOME: " + home),
-        None => Console.print("  HOME: (not set)")
+    None => Console.print("  HOME: (not set)"),
-    }
+}
    match Process.env("SHELL") {
-        Some(shell) => Console.print("  SHELL: " + shell),
+    Some(shell) => Console.print("  SHELL: " + shell),
-        None => Console.print("  SHELL: (not set)")
+    None => Console.print("  SHELL: (not set)"),
-    }
+}
    Console.print("")
    // Run shell commands
    Console.print("Running shell commands:")
    let date = Process.exec("date")
    Console.print("  date: " + String.trim(date))
    let kernel = Process.exec("uname -r")
    Console.print("  kernel: " + String.trim(kernel))
    let files = Process.exec("ls examples/*.lux | wc -l")
    Console.print("  .lux files in examples/: " + String.trim(files))
    Console.print("")
    // Command line arguments
    Console.print("Command line arguments:")
    let args = Process.args()
    let argCount = List.length(args)
    if argCount == 0 then {
-        Console.print("  (no arguments)")
+    Console.print("  (no arguments)")
-    } else {
+} else {
-        Console.print("  Count: " + toString(argCount))
+    Console.print("  Count: " + toString(argCount))
-        match List.head(args) {
+    match List.head(args) {
-            Some(first) => Console.print("  First: " + first),
+    Some(first) => Console.print("  First: " + first),
-            None => Console.print("  First: (empty)")
+    None => Console.print("  First: (empty)"),
-        }
+}
-    }
+}
 }
 let result = run main() with {}
--- a/examples/showcase/README.md
+++ b/examples/showcase/README.md
@@ -0,0 +1,107 @@
 # Task Manager Showcase
 This example demonstrates Lux's three killer features in a practical, real-world context.
 ## Running the Example
 ```bash
 lux run examples/showcase/task_manager.lux
 ```
 ## Features Demonstrated
 ### 1. Algebraic Effects
 Every function signature shows exactly what side effects it can perform:
 ```lux
 fn createTask(title: String, priority: String): Task@latest
    with {TaskStore, Random} = { ... }
 ```
 - `TaskStore` - database operations
 - `Random` - random number generation
 - No hidden I/O or surprise calls
 ### 2. Behavioral Types
 Compile-time guarantees about function behavior:
 ```lux
 fn formatTask(task: Task@latest): String
    is pure           // No side effects
    is deterministic  // Same input = same output
    is total          // Always terminates
 ```
 ```lux
 fn completeTask(id: String): Option<Task@latest>
    is idempotent     // Safe to retry
    with {TaskStore}
 ```
 ### 3. Schema Evolution
 Versioned types with automatic migration:
 ```lux
 type Task @v2 {
    id: String,
    title: String,
    done: Bool,
    priority: String,  // New in v2
    from @v1 = { ...old, priority: "medium" }
 }
 ```
 ### 4. Handler Swapping (Testing)
 Test without mocks by swapping effect handlers:
 ```lux
 // Production
 run processOrders() with {
    TaskStore = PostgresTaskStore,
    Logger = CloudLogger
 }
 // Testing
 run processOrders() with {
    TaskStore = InMemoryTaskStore,
    Logger = SilentLogger
 }
 ```
 ## Why This Matters
 | Traditional Languages | Lux |
 |----------------------|-----|
 | Side effects are implicit | Effects in type signatures |
 | Runtime crashes | Compile-time verification |
 | Complex mocking frameworks | Simple handler swapping |
 | Manual migration code | Automatic schema evolution |
 | Hope for retry safety | Verified idempotency |
 ## File Structure
 ```
 showcase/
 ├── README.md           # This file
 └── task_manager.lux    # Main example with all features
 ```
 ## Key Sections in the Code
 1. **Versioned Data Types** - `Task @v1`, `@v2`, `@v3` with migrations
 2. **Pure Functions** - `is pure`, `is total`, `is deterministic`, `is idempotent`
 3. **Effects** - `effect TaskStore` and `effect Logger`
 4. **Effect Handlers** - `InMemoryTaskStore`, `ConsoleLogger`
 5. **Testing** - `runTestScenario()` with swapped handlers
 6. **Migration Demo** - `demonstrateMigration()`
 ## Next Steps
 - Read the [Behavioral Types Guide](../../docs/guide/12-behavioral-types.md)
 - Read the [Schema Evolution Guide](../../docs/guide/13-schema-evolution.md)
 - Explore [more examples](../)
--- a/examples/showcase/ask_pattern.lux
+++ b/examples/showcase/ask_pattern.lux
@@ -1,15 +1,3 @@
 // The "Ask" Pattern - Resumable Effects
 //
 // Unlike exceptions which unwind the stack, effect handlers can
 // RESUME with a value. This enables "ask the environment" patterns.
 //
 // Expected output:
 // Need config: api_url
 // Got: https://api.example.com
 // Need config: timeout
 // Got: 30
 // Configured with url=https://api.example.com, timeout=30
 effect Config {
    fn get(key: String): String
 }
@@ -25,14 +13,13 @@ fn configure(): String with {Config, Console} = {
 }
 handler envConfig: Config {
-    fn get(key) =
+    fn get(key) = if key == "api_url" then resume("https://api.example.com") else if key == "timeout" then resume("30") else resume("unknown")
        if key == "api_url" then resume("https://api.example.com")
        else if key == "timeout" then resume("30")
        else resume("unknown")
 }
 fn main(): Unit with {Console} = {
-    let result = run configure() with { Config = envConfig }
+    let result = run configure() with {
    Config = envConfig,
 }
    Console.print(result)
 }
--- a/examples/showcase/custom_logging.lux
+++ b/examples/showcase/custom_logging.lux
@@ -1,15 +1,3 @@
 // Custom Logging with Effects
 //
 // This demonstrates how effects let you abstract side effects.
 // The same code can be run with different logging implementations.
 //
 // Expected output:
 // [INFO] Starting computation
 // [DEBUG] x = 10
 // [INFO] Processing
 // [DEBUG] result = 20
 // Final: 20
 effect Log {
    fn info(msg: String): Unit
    fn debug(msg: String): Unit
@@ -26,18 +14,22 @@ fn computation(): Int with {Log} = {
 }
 handler consoleLogger: Log {
-    fn info(msg) = {
+    fn info(msg) = 
-        Console.print("[INFO] " + msg)
+        {
-        resume(())
+    Console.print("[INFO] " + msg)
-    }
+    resume(())
-    fn debug(msg) = {
+}
-        Console.print("[DEBUG] " + msg)
+    fn debug(msg) = 
-        resume(())
+        {
-    }
+    Console.print("[DEBUG] " + msg)
    resume(())
 }
 }
 fn main(): Unit with {Console} = {
-    let result = run computation() with { Log = consoleLogger }
+    let result = run computation() with {
    Log = consoleLogger,
 }
    Console.print("Final: " + toString(result))
 }
--- a/examples/showcase/early_return.lux
+++ b/examples/showcase/early_return.lux
@@ -1,37 +1,18 @@
 // Early Return with Fail Effect
 //
 // The Fail effect provides clean early termination.
 // Functions declare their failure modes in the type signature.
 //
 // Expected output:
 // Parsing "42"...
 // Result: 42
 // Parsing "100"...
 // Result: 100
 // Dividing 100 by 4...
 // Result: 25
 fn parsePositive(s: String): Int with {Fail, Console} = {
    Console.print("Parsing \"" + s + "\"...")
-    if s == "42" then 42
+    if s == "42" then 42 else if s == "100" then 100 else Fail.fail("Invalid number: " + s)
    else if s == "100" then 100
    else Fail.fail("Invalid number: " + s)
 }
 fn safeDivide(a: Int, b: Int): Int with {Fail, Console} = {
    Console.print("Dividing " + toString(a) + " by " + toString(b) + "...")
-    if b == 0 then Fail.fail("Division by zero")
+    if b == 0 then Fail.fail("Division by zero") else a / b
    else a / b
 }
 fn main(): Unit with {Console} = {
    // These succeed
    let n1 = run parsePositive("42") with {}
    Console.print("Result: " + toString(n1))
    let n2 = run parsePositive("100") with {}
    Console.print("Result: " + toString(n2))
    let n3 = run safeDivide(100, 4) with {}
    Console.print("Result: " + toString(n3))
 }
--- a/examples/showcase/effect_composition.lux
+++ b/examples/showcase/effect_composition.lux
@@ -1,16 +1,3 @@
 // Effect Composition - Combine multiple effects cleanly
 //
 // Unlike monad transformers (which have ordering issues),
 // effects can be freely combined without boilerplate.
 // Each handler handles its own effect, ignoring others.
 //
 // Expected output:
 // [LOG] Starting computation
 // Generated: 7
 // [LOG] Processing value
 // [LOG] Done
 // Result: 14
 effect Log {
    fn log(msg: String): Unit
 }
@@ -30,8 +17,8 @@ handler consoleLog: Log {
 fn main(): Unit with {Console} = {
    let result = run computation() with {
-        Log = consoleLog
+    Log = consoleLog,
-    }
+}
    Console.print("Generated: " + toString(result / 2))
    Console.print("Result: " + toString(result))
 }
--- a/examples/showcase/higher_order.lux
+++ b/examples/showcase/higher_order.lux
@@ -1,38 +1,19 @@
 // Higher-Order Functions and Closures
 //
 // Functions are first-class values in Lux.
 // Closures capture their environment.
 //
 // Expected output:
 // Square of 5: 25
 // Cube of 3: 27
 // Add 10 to 5: 15
 // Add 10 to 20: 30
 // Composed: 15625 (cube(square(5)) = cube(25) = 15625)
 fn apply(f: fn(Int): Int, x: Int): Int = f(x)
-fn compose(f: fn(Int): Int, g: fn(Int): Int): fn(Int): Int =
+fn compose(f: fn(Int): Int, g: fn(Int): Int): fn(Int): Int = fn(x: Int): Int => f(g(x))
    fn(x: Int): Int => f(g(x))
 fn square(n: Int): Int = n * n
 fn cube(n: Int): Int = n * n * n
-fn makeAdder(n: Int): fn(Int): Int =
+fn makeAdder(n: Int): fn(Int): Int = fn(x: Int): Int => x + n
    fn(x: Int): Int => x + n
 fn main(): Unit with {Console} = {
    // Apply functions
    Console.print("Square of 5: " + toString(apply(square, 5)))
    Console.print("Cube of 3: " + toString(apply(cube, 3)))
    // Closures
    let add10 = makeAdder(10)
    Console.print("Add 10 to 5: " + toString(add10(5)))
    Console.print("Add 10 to 20: " + toString(add10(20)))
    // Function composition
    let squareThenCube = compose(cube, square)
    Console.print("Composed: " + toString(squareThenCube(5)))
 }
--- a/examples/showcase/pattern_matching.lux
+++ b/examples/showcase/pattern_matching.lux
@@ -1,16 +1,3 @@
 // Algebraic Data Types and Pattern Matching
 //
 // Lux has powerful ADTs with exhaustive pattern matching.
 // The type system ensures all cases are handled.
 //
 // Expected output:
 // Evaluating: (2 + 3)
 // Result: 5
 // Evaluating: ((1 + 2) * (3 + 4))
 // Result: 21
 // Evaluating: (10 - (2 * 3))
 // Result: 4
 type Expr =
    | Num(Int)
    | Add(Expr, Expr)
@@ -19,19 +6,19 @@ type Expr =
 fn eval(e: Expr): Int =
    match e {
-        Num(n) => n,
+    Num(n) => n,
-        Add(a, b) => eval(a) + eval(b),
+    Add(a, b) => eval(a) + eval(b),
-        Sub(a, b) => eval(a) - eval(b),
+    Sub(a, b) => eval(a) - eval(b),
-        Mul(a, b) => eval(a) * eval(b)
+    Mul(a, b) => eval(a) * eval(b),
-    }
+}
 fn showExpr(e: Expr): String =
    match e {
-        Num(n) => toString(n),
+    Num(n) => toString(n),
-        Add(a, b) => "(" + showExpr(a) + " + " + showExpr(b) + ")",
+    Add(a, b) => "(" + showExpr(a) + " + " + showExpr(b) + ")",
-        Sub(a, b) => "(" + showExpr(a) + " - " + showExpr(b) + ")",
+    Sub(a, b) => "(" + showExpr(a) + " - " + showExpr(b) + ")",
-        Mul(a, b) => "(" + showExpr(a) + " * " + showExpr(b) + ")"
+    Mul(a, b) => "(" + showExpr(a) + " * " + showExpr(b) + ")",
-    }
+}
 fn evalAndPrint(e: Expr): Unit with {Console} = {
    Console.print("Evaluating: " + showExpr(e))
@@ -39,15 +26,10 @@ fn evalAndPrint(e: Expr): Unit with {Console} = {
 }
 fn main(): Unit with {Console} = {
    // (2 + 3)
    let e1 = Add(Num(2), Num(3))
    evalAndPrint(e1)
    // ((1 + 2) * (3 + 4))
    let e2 = Mul(Add(Num(1), Num(2)), Add(Num(3), Num(4)))
    evalAndPrint(e2)
    // (10 - (2 * 3))
    let e3 = Sub(Num(10), Mul(Num(2), Num(3)))
    evalAndPrint(e3)
 }
--- a/examples/showcase/task_manager.lux
+++ b/examples/showcase/task_manager.lux
@@ -0,0 +1,419 @@
 // =============================================================================
 // Task Manager API - A Showcase of Lux's Unique Features
 // =============================================================================
 //
 // This example demonstrates Lux's three killer features:
 //
 // 1. ALGEBRAIC EFFECTS - Every side effect is explicit in function signatures
 //    - No hidden I/O, no surprise database calls
 //    - Testing is trivial: just swap handlers
 //
 // 2. BEHAVIORAL TYPES - Compile-time guarantees about function behavior
 //    - `is pure` - no side effects, safe to cache
 //    - `is total` - always terminates, never fails
 //    - `is idempotent` - safe to retry without side effects
 //    - `is deterministic` - same input = same output
 //
 // 3. SCHEMA EVOLUTION - Versioned types with automatic migration
 //    - Data structures evolve safely over time
 //    - Old data automatically upgrades
 //
 // To run: lux run examples/showcase/task_manager.lux
 // =============================================================================
 // =============================================================================
 // PART 1: VERSIONED DATA TYPES (Schema Evolution)
 // =============================================================================
 // Task v1: Our original data model (simple)
 type Task @v1 {
    id: String,
    title: String,
    done: Bool
 }
 // Task v2: Added priority field
 // The `from @v1` clause defines how to migrate old data automatically
 type Task @v2 {
    id: String,
    title: String,
    done: Bool,
    priority: String,  // New field: "low", "medium", "high"
    // Migration: old tasks get "medium" priority by default
    from @v1 = {
        id: old.id,
        title: old.title,
        done: old.done,
        priority: "medium"
    }
 }
 // Task v3: Added due date and tags
 // Migrations chain automatically: v1 → v2 → v3
 type Task @v3 {
    id: String,
    title: String,
    done: Bool,
    priority: String,
    dueDate: Option<Int>,   // Unix timestamp, optional
    tags: List<String>,      // New: categorization
    from @v2 = {
        id: old.id,
        title: old.title,
        done: old.done,
        priority: old.priority,
        dueDate: None,  // No due date for migrated tasks
        tags: []        // Empty tags for migrated tasks
    }
 }
 // Use @latest to always refer to the newest version
 type TaskList = List<Task@latest>
 // =============================================================================
 // PART 2: PURE FUNCTIONS WITH BEHAVIORAL TYPES
 // =============================================================================
 // Pure function: no side effects, safe to cache, parallelize, eliminate if unused
 // The compiler verifies `is pure` - if you try to call an effect, it errors.
 fn formatTask(task: Task@latest): String
    is pure
    is deterministic
    is total = {
    let status = if task.done then "[x]" else "[ ]"
    let priority = match task.priority {
        "high" => "!!",
        "medium" => "!",
        _ => ""
    }
    status + " " + priority + task.title
 }
 // Idempotent function: f(f(x)) = f(x)
 // Safe to apply multiple times without changing the result
 // Critical for retry logic - the compiler verifies this property
 fn normalizeTitle(title: String): String
    is pure
    is idempotent = {
    title
    |> String.trim
    |> String.toLower
 }
 // Total function: always terminates, never throws
 // No Fail effect allowed, recursion must be structurally decreasing
 fn countCompleted(tasks: TaskList): Int
    is pure
    is total = {
    match tasks {
        [] => 0,
        [task, ...rest] =>
            (if task.done then 1 else 0) + countCompleted(rest)
    }
 }
 // Commutative function: f(a, b) = f(b, a)
 // Enables parallel reduction and argument reordering optimizations
 fn maxPriority(a: String, b: String): String
    is pure
    is commutative = {
    let priorityValue = fn(p: String): Int =>
        match p {
            "high" => 3,
            "medium" => 2,
            "low" => 1,
            _ => 0
        }
    if priorityValue(a) > priorityValue(b) then a else b
 }
 // Filter tasks by criteria - pure, can be cached and parallelized
 fn filterByPriority(tasks: TaskList, priority: String): TaskList
    is pure
    is deterministic = {
    List.filter(tasks, fn(t: Task@latest): Bool => t.priority == priority)
 }
 fn filterPending(tasks: TaskList): TaskList
    is pure
    is deterministic = {
    List.filter(tasks, fn(t: Task@latest): Bool => !t.done)
 }
 fn filterCompleted(tasks: TaskList): TaskList
    is pure
    is deterministic = {
    List.filter(tasks, fn(t: Task@latest): Bool => t.done)
 }
 // =============================================================================
 // PART 3: EFFECTS - EXPLICIT SIDE EFFECTS
 // =============================================================================
 // Custom effect for task storage
 // This declares WHAT operations are available, not HOW they work
 effect TaskStore {
    fn save(task: Task@latest): Result<Task@latest, String>
    fn getById(id: String): Option<Task@latest>
    fn getAll(): TaskList
    fn delete(id: String): Bool
 }
 // Service functions declare their effects in the type signature
 // Anyone reading the signature knows exactly what side effects can occur
 // Create a new task - requires TaskStore and Random effects
 fn createTask(title: String, priority: String): Task@latest
    with {TaskStore, Random} = {
    let id = "task_" + toString(Random.int(10000, 99999))
    let task = {
        id: id,
        title: normalizeTitle(title),  // Uses our idempotent normalizer
        done: false,
        priority: priority,
        dueDate: None,
        tags: []
    }
    match TaskStore.save(task) {
        Ok(saved) => saved,
        Err(_) => task  // Return unsaved if storage fails
    }
 }
 // Complete a task - idempotent, safe to retry
 // If the network fails mid-request, retry is safe
 fn completeTask(id: String): Option<Task@latest>
    is idempotent  // Compiler verifies this is safe to retry
    with {TaskStore} = {
    match TaskStore.getById(id) {
        None => None,
        Some(task) => {
            // Setting done = true is idempotent: already done? stays done
            let updated = { ...task, done: true }
            match TaskStore.save(updated) {
                Ok(saved) => Some(saved),
                Err(_) => None
            }
        }
    }
 }
 // Get task summary - logging effect, but computation is pure
 fn getTaskSummary(): { total: Int, completed: Int, pending: Int, highPriority: Int }
    with {TaskStore, Logger} = {
    let tasks = TaskStore.getAll()
    Logger.log("Fetched " + toString(List.length(tasks)) + " tasks")
    // These computations are pure - could be parallelized
    let completed = countCompleted(tasks)
    let pending = List.length(tasks) - completed
    let highPriority = List.length(filterByPriority(tasks, "high"))
    { total: List.length(tasks), completed: completed, pending: pending, highPriority: highPriority }
 }
 // =============================================================================
 // PART 4: EFFECT HANDLERS - SWAP IMPLEMENTATIONS
 // =============================================================================
 // In-memory handler for testing
 // This handler stores tasks in a mutable list - perfect for unit tests
 handler InMemoryTaskStore: TaskStore {
    let tasks: List<Task@latest> = []
    fn save(task: Task@latest): Result<Task@latest, String> = {
        // Remove existing task with same ID (if any), then add new
        tasks = List.filter(tasks, fn(t: Task@latest): Bool => t.id != task.id)
        tasks = List.concat(tasks, [task])
        Ok(task)
    }
    fn getById(id: String): Option<Task@latest> = {
        List.find(tasks, fn(t: Task@latest): Bool => t.id == id)
    }
    fn getAll(): TaskList = tasks
    fn delete(id: String): Bool = {
        let before = List.length(tasks)
        tasks = List.filter(tasks, fn(t: Task@latest): Bool => t.id != task.id)
        List.length(tasks) < before
    }
 }
 // Logging handler - wraps another handler with logging
 handler LoggingTaskStore(inner: TaskStore): TaskStore with {Logger} {
    fn save(task: Task@latest): Result<Task@latest, String> = {
        Logger.log("Saving task: " + task.id)
        inner.save(task)
    }
    fn getById(id: String): Option<Task@latest> = {
        Logger.log("Getting task: " + id)
        inner.getById(id)
    }
    fn getAll(): TaskList = {
        Logger.log("Getting all tasks")
        inner.getAll()
    }
    fn delete(id: String): Bool = {
        Logger.log("Deleting task: " + id)
        inner.delete(id)
    }
 }
 // Simple logger effect and handler
 effect Logger {
    fn log(message: String): Unit
 }
 handler ConsoleLogger: Logger with {Console} {
    fn log(message: String): Unit = {
        Console.print("[LOG] " + message)
    }
 }
 handler SilentLogger: Logger {
    fn log(message: String): Unit = {
        // Do nothing - useful for tests
    }
 }
 // =============================================================================
 // PART 5: TESTING - SWAP HANDLERS, NO MOCKS NEEDED
 // =============================================================================
 // Test helper: creates a controlled environment
 fn runTestScenario(): Unit with {Console} = {
    Console.print("=== Running Test Scenario ===")
    Console.print("")
    // Use in-memory storage and silent logging for tests
    // No database, no file I/O, no network - pure in-memory testing
    let result = run {
        // Create some tasks
        let task1 = createTask("Write documentation", "high")
        let task2 = createTask("Fix bug #123", "medium")
        let task3 = createTask("Review PR", "low")
        // Complete one task
        completeTask(task1.id)
        // Get summary
        getTaskSummary()
    } with {
        TaskStore = InMemoryTaskStore,
        Logger = SilentLogger,
        Random = {
            // Deterministic "random" for tests
            let counter = 0
            fn int(min: Int, max: Int): Int = {
                counter = counter + 1
                min + (counter * 12345) % (max - min)
            }
        }
    }
    Console.print("Test Results:")
    Console.print("  Total tasks: " + toString(result.total))
    Console.print("  Completed: " + toString(result.completed))
    Console.print("  Pending: " + toString(result.pending))
    Console.print("  High priority: " + toString(result.highPriority))
    Console.print("")
    // Verify results
    if result.total == 3 &&
       result.completed == 1 &&
       result.pending == 2 &&
       result.highPriority == 1 {
        Console.print("All tests passed!")
    } else {
        Console.print("Test failed!")
    }
 }
 // =============================================================================
 // PART 6: SCHEMA MIGRATION DEMO
 // =============================================================================
 fn demonstrateMigration(): Unit with {Console} = {
    Console.print("=== Schema Evolution Demo ===")
    Console.print("")
    // Simulate loading a v1 task (from old database/API)
    let oldTask = Schema.versioned("Task", 1, {
        id: "legacy_001",
        title: "Old task from v1",
        done: false
    })
    Console.print("Loaded v1 task:")
    Console.print("  Version: " + toString(Schema.getVersion(oldTask)))
    Console.print("")
    // Migrate to latest version automatically
    let migratedTask = Schema.migrate(oldTask, 3)
    Console.print("After migration to v3:")
    Console.print("  Version: " + toString(Schema.getVersion(migratedTask)))
    Console.print("  Has priority: " + migratedTask.priority)  // Added by v2 migration
    Console.print("  Has tags: " + toString(List.length(migratedTask.tags)) + " tags")  // Added by v3
    Console.print("")
    Console.print("Old data seamlessly upgraded!")
 }
 // =============================================================================
 // PART 7: MAIN - PUTTING IT ALL TOGETHER
 // =============================================================================
 fn main(): Unit with {Console} = {
    Console.print("╔═══════════════════════════════════════════════════════════╗")
    Console.print("║         Lux Task Manager - Feature Showcase               ║")
    Console.print("╚═══════════════════════════════════════════════════════════╝")
    Console.print("")
    // Demonstrate pure functions
    Console.print("--- Pure Functions (Behavioral Types) ---")
    let sampleTask = {
        id: "demo",
        title: "Learn Lux",
        done: false,
        priority: "high",
        dueDate: None,
        tags: ["learning", "programming"]
    }
    Console.print("Formatted task: " + formatTask(sampleTask))
    Console.print("Normalized title: " + normalizeTitle("  HELLO WORLD  "))
    Console.print("")
    // Demonstrate schema evolution
    demonstrateMigration()
    Console.print("")
    // Run tests with swapped handlers
    runTestScenario()
    Console.print("")
    Console.print("╔═══════════════════════════════════════════════════════════╗")
    Console.print("║  Key Takeaways:                                           ║")
    Console.print("║                                                           ║")
    Console.print("║  1. Effects in signatures = no hidden side effects        ║")
    Console.print("║  2. Behavioral types = compile-time guarantees            ║")
    Console.print("║  3. Handler swapping = easy testing without mocks         ║")
    Console.print("║  4. Schema evolution = safe data migrations               ║")
    Console.print("╚═══════════════════════════════════════════════════════════╝")
 }
 // Run the showcase
 let _ = run main() with {}
--- a/examples/standard/factorial.lux
+++ b/examples/standard/factorial.lux
@@ -1,14 +1,6 @@
-// Factorial - compute n!
+fn factorial(n: Int): Int = if n <= 1 then 1 else n * factorial(n - 1)
-// Recursive version
+fn factorialTail(n: Int, acc: Int): Int = if n <= 1 then acc else factorialTail(n - 1, n * acc)
 fn factorial(n: Int): Int =
    if n <= 1 then 1
    else n * factorial(n - 1)
 // Tail-recursive version (optimized)
 fn factorialTail(n: Int, acc: Int): Int =
    if n <= 1 then acc
    else factorialTail(n - 1, n * acc)
 fn factorial2(n: Int): Int = factorialTail(n, 1)
--- a/examples/standard/fizzbuzz.lux
+++ b/examples/standard/fizzbuzz.lux
@@ -1,22 +1,11 @@
-// FizzBuzz - print numbers 1-100, but:
+fn fizzbuzz(n: Int): String = if n % 15 == 0 then "FizzBuzz" else if n % 3 == 0 then "Fizz" else if n % 5 == 0 then "Buzz" else toString(n)
 // - multiples of 3: print "Fizz"
 // - multiples of 5: print "Buzz"
 // - multiples of both: print "FizzBuzz"
 fn fizzbuzz(n: Int): String =
    if n % 15 == 0 then "FizzBuzz"
    else if n % 3 == 0 then "Fizz"
    else if n % 5 == 0 then "Buzz"
    else toString(n)
 fn printFizzbuzz(i: Int, max: Int): Unit with {Console} =
-    if i > max then ()
+    if i > max then () else {
-    else {
+    Console.print(fizzbuzz(i))
-        Console.print(fizzbuzz(i))
+    printFizzbuzz(i + 1, max)
-        printFizzbuzz(i + 1, max)
+}
    }
-fn main(): Unit with {Console} =
+fn main(): Unit with {Console} = printFizzbuzz(1, 100)
    printFizzbuzz(1, 100)
 let output = run main() with {}
--- a/examples/standard/guessing_game.lux
+++ b/examples/standard/guessing_game.lux
@@ -1,42 +1,17 @@
-// Number guessing game - demonstrates Random and Console effects
+fn checkGuess(guess: Int, secret: Int): String = if guess == secret then "Correct" else if guess < secret then "Too low" else "Too high"
 //
 // Expected output:
 // Welcome to the Guessing Game!
 // Target number: 42
 // Simulating guesses...
 // Guess 50: Too high!
 // Guess 25: Too low!
 // Guess 37: Too low!
 // Guess 43: Too high!
 // Guess 40: Too low!
 // Guess 41: Too low!
 // Guess 42: Correct!
 // Found in 7 attempts!
 // Game logic - check a guess against the secret
 fn checkGuess(guess: Int, secret: Int): String =
    if guess == secret then "Correct"
    else if guess < secret then "Too low"
    else "Too high"
 // Binary search simulation to find the number
 fn binarySearch(low: Int, high: Int, secret: Int, attempts: Int): Int with {Console} = {
-    let mid = (low + high) / 2
+    let mid = low + high / 2
    let result = checkGuess(mid, secret)
    Console.print("Guess " + toString(mid) + ": " + result + "!")
-
+    if result == "Correct" then attempts else if result == "Too low" then binarySearch(mid + 1, high, secret, attempts + 1) else binarySearch(low, mid - 1, secret, attempts + 1)
    if result == "Correct" then attempts
    else if result == "Too low" then binarySearch(mid + 1, high, secret, attempts + 1)
    else binarySearch(low, mid - 1, secret, attempts + 1)
 }
 fn main(): Unit with {Console} = {
    Console.print("Welcome to the Guessing Game!")
    // Use a fixed "secret" for reproducible output
    let secret = 42
    Console.print("Target number: " + toString(secret))
    Console.print("Simulating guesses...")
    let attempts = binarySearch(1, 100, secret, 1)
    Console.print("Found in " + toString(attempts) + " attempts!")
 }
--- a/examples/standard/hello_world.lux
+++ b/examples/standard/hello_world.lux
@@ -1,7 +1,3 @@
-// The classic first program
+fn main(): Unit with {Console} = Console.print("Hello, World!")
 // Expected output: Hello, World!
 fn main(): Unit with {Console} =
    Console.print("Hello, World!")
 let output = run main() with {}
--- a/examples/standard/primes.lux
+++ b/examples/standard/primes.lux
@@ -1,25 +1,14 @@
-// Prime number utilities
+fn isPrime(n: Int): Bool = if n < 2 then false else isPrimeHelper(n, 2)
-fn isPrime(n: Int): Bool =
+fn isPrimeHelper(n: Int, i: Int): Bool = if i * i > n then true else if n % i == 0 then false else isPrimeHelper(n, i + 1)
    if n < 2 then false
    else isPrimeHelper(n, 2)
-fn isPrimeHelper(n: Int, i: Int): Bool =
+fn findPrimes(count: Int): Unit with {Console} = findPrimesHelper(2, count)
    if i * i > n then true
    else if n % i == 0 then false
    else isPrimeHelper(n, i + 1)
 // Find first n primes
 fn findPrimes(count: Int): Unit with {Console} =
    findPrimesHelper(2, count)
 fn findPrimesHelper(current: Int, remaining: Int): Unit with {Console} =
-    if remaining <= 0 then ()
+    if remaining <= 0 then () else if isPrime(current) then {
-    else if isPrime(current) then {
+    Console.print(toString(current))
-        Console.print(toString(current))
+    findPrimesHelper(current + 1, remaining - 1)
-        findPrimesHelper(current + 1, remaining - 1)
+} else findPrimesHelper(current + 1, remaining)
    }
    else findPrimesHelper(current + 1, remaining)
 fn main(): Unit with {Console} = {
    Console.print("First 20 prime numbers:")
--- a/examples/standard/stdlib_demo.lux
+++ b/examples/standard/stdlib_demo.lux
@@ -1,6 +1,3 @@
 // Standard Library Demo
 // Demonstrates the built-in modules: List, String, Option, Math
 fn main(): Unit with {Console} = {
    Console.print("=== List Operations ===")
    let nums = [1, 2, 3, 4, 5]
@@ -11,7 +8,6 @@ fn main(): Unit with {Console} = {
    Console.print("Length: " + toString(List.length(nums)))
    Console.print("Reversed: " + toString(List.reverse(nums)))
    Console.print("Range 1-5: " + toString(List.range(1, 6)))
    Console.print("")
    Console.print("=== String Operations ===")
    let text = "  Hello, World!  "
@@ -22,7 +18,6 @@ fn main(): Unit with {Console} = {
    Console.print("Contains 'World': " + toString(String.contains(text, "World")))
    Console.print("Split by comma: " + toString(String.split("a,b,c", ",")))
    Console.print("Join with dash: " + String.join(["x", "y", "z"], "-"))
    Console.print("")
    Console.print("=== Option Operations ===")
    let some_val = Some(42)
@@ -31,7 +26,6 @@ fn main(): Unit with {Console} = {
    Console.print("None mapped: " + toString(Option.map(none_val, fn(x: Int): Int => x * 2)))
    Console.print("Some(42) getOrElse(0): " + toString(Option.getOrElse(some_val, 0)))
    Console.print("None getOrElse(0): " + toString(Option.getOrElse(none_val, 0)))
    Console.print("")
    Console.print("=== Math Operations ===")
    Console.print("abs(-5): " + toString(Math.abs(-5)))
--- a/examples/statemachine.lux
+++ b/examples/statemachine.lux
@@ -1,13 +1,3 @@
 // State machine example using algebraic data types
 // Demonstrates pattern matching for state transitions
 //
 // Expected output:
 // Initial light: red
 // After transition: green
 // After two transitions: yellow
 // Door: Closed -> Open -> Closed -> Locked
 // Traffic light state machine
 type TrafficLight =
    | Red
    | Yellow
@@ -15,26 +5,25 @@ type TrafficLight =
 fn nextLight(light: TrafficLight): TrafficLight =
    match light {
-        Red => Green,
+    Red => Green,
-        Green => Yellow,
+    Green => Yellow,
-        Yellow => Red
+    Yellow => Red,
-    }
+}
 fn canGo(light: TrafficLight): Bool =
    match light {
-        Green => true,
+    Green => true,
-        Yellow => false,
+    Yellow => false,
-        Red => false
+    Red => false,
-    }
+}
 fn lightColor(light: TrafficLight): String =
    match light {
-        Red => "red",
+    Red => "red",
-        Yellow => "yellow",
+    Yellow => "yellow",
-        Green => "green"
+    Green => "green",
-    }
+}
 // Door state machine
 type DoorState =
    | Open
    | Closed
@@ -48,31 +37,34 @@ type DoorAction =
 fn applyAction(state: DoorState, action: DoorAction): DoorState =
    match (state, action) {
-        (Closed, OpenDoor) => Open,
+    (Closed, OpenDoor) => Open,
-        (Open, CloseDoor) => Closed,
+    (Open, CloseDoor) => Closed,
-        (Closed, LockDoor) => Locked,
+    (Closed, LockDoor) => Locked,
-        (Locked, UnlockDoor) => Closed,
+    (Locked, UnlockDoor) => Closed,
-        _ => state
+    _ => state,
-    }
+}
 fn doorStateName(state: DoorState): String =
    match state {
-        Open => "Open",
+    Open => "Open",
-        Closed => "Closed",
+    Closed => "Closed",
-        Locked => "Locked"
+    Locked => "Locked",
-    }
+}
 // Test the state machines
 let light1 = Red
 let light2 = nextLight(light1)
 let light3 = nextLight(light2)
 let door1 = Closed
 let door2 = applyAction(door1, OpenDoor)
 let door3 = applyAction(door2, CloseDoor)
 let door4 = applyAction(door3, LockDoor)
 // Print results
 fn printResults(): Unit with {Console} = {
    Console.print("Initial light: " + lightColor(light1))
    Console.print("After transition: " + lightColor(light2))
--- a/examples/stress_rc.lux
+++ b/examples/stress_rc.lux
@@ -1,8 +1,4 @@
 // Stress test for RC system with large lists
 // Tests FBIP optimization with single-owner chains
 fn processChain(n: Int): Int = {
    // Single owner chain - FBIP should reuse lists
    let nums = List.range(1, n)
    let doubled = List.map(nums, fn(x: Int): Int => x * 2)
    let filtered = List.filter(doubled, fn(x: Int): Bool => x > n)
@@ -12,13 +8,10 @@ fn processChain(n: Int): Int = {
 fn main(): Unit = {
    Console.print("=== RC Stress Test ===")
    // Run multiple iterations of list operations
    let result1 = processChain(100)
    let result2 = processChain(200)
    let result3 = processChain(500)
    let result4 = processChain(1000)
    Console.print("Completed 4 chains")
    Console.print("Sizes: 100, 200, 500, 1000")
 }
--- a/examples/stress_shared_rc.lux
+++ b/examples/stress_shared_rc.lux
@@ -1,12 +1,7 @@
 // Stress test for RC system WITH shared references
 // Forces rc>1 path by keeping aliases
 fn processWithAlias(n: Int): Int = {
    let nums = List.range(1, n)
-    let alias = nums  // This increments rc, forcing copy path
+    let alias = nums
-    let _len = List.length(alias)  // Use the alias
+    let _len = List.length(alias)
    // Now nums has rc>1, so map must allocate new
    let doubled = List.map(nums, fn(x: Int): Int => x * 2)
    let filtered = List.filter(doubled, fn(x: Int): Bool => x > n)
    let reversed = List.reverse(filtered)
@@ -15,12 +10,9 @@ fn processWithAlias(n: Int): Int = {
 fn main(): Unit = {
    Console.print("=== RC Stress Test (Shared Refs) ===")
    // Run multiple iterations with shared references
    let result1 = processWithAlias(100)
    let result2 = processWithAlias(200)
    let result3 = processWithAlias(500)
    let result4 = processWithAlias(1000)
    Console.print("Completed 4 chains with shared refs")
 }
--- a/examples/tailcall.lux
+++ b/examples/tailcall.lux
@@ -1,45 +1,25 @@
-// Demonstrating tail call optimization (TCO) in Lux
+fn factorialTCO(n: Int, acc: Int): Int = if n <= 1 then acc else factorialTCO(n - 1, n * acc)
 // TCO allows recursive functions to run in constant stack space
 //
 // Expected output:
 // factorial(20) = 2432902008176640000
 // fib(30) = 832040
 // sumTo(1000) = 500500
 // countdown(10000) completed
 // Factorial with accumulator - tail recursive
 fn factorialTCO(n: Int, acc: Int): Int =
    if n <= 1 then acc
    else factorialTCO(n - 1, n * acc)
 fn factorial(n: Int): Int = factorialTCO(n, 1)
-// Fibonacci with accumulator - tail recursive
+fn fibTCO(n: Int, a: Int, b: Int): Int = if n <= 0 then a else fibTCO(n - 1, b, a + b)
 fn fibTCO(n: Int, a: Int, b: Int): Int =
    if n <= 0 then a
    else fibTCO(n - 1, b, a + b)
 fn fib(n: Int): Int = fibTCO(n, 0, 1)
-// Count down - simple tail recursion
+fn countdown(n: Int): Int = if n <= 0 then 0 else countdown(n - 1)
 fn countdown(n: Int): Int =
    if n <= 0 then 0
    else countdown(n - 1)
-// Sum with accumulator - tail recursive
+fn sumToTCO(n: Int, acc: Int): Int = if n <= 0 then acc else sumToTCO(n - 1, acc + n)
 fn sumToTCO(n: Int, acc: Int): Int =
    if n <= 0 then acc
    else sumToTCO(n - 1, acc + n)
 fn sumTo(n: Int): Int = sumToTCO(n, 0)
 // Test the functions
 let fact20 = factorial(20)
 let fib30 = fib(30)
 let sum1000 = sumTo(1000)
 let countResult = countdown(10000)
 // Print results
 fn printResults(): Unit with {Console} = {
    Console.print("factorial(20) = " + toString(fact20))
    Console.print("fib(30) = " + toString(fib30))
--- a/examples/test_fbip_allocs.lux
+++ b/examples/test_fbip_allocs.lux
@@ -1,17 +1,8 @@
 // This test shows FBIP optimization by comparing allocation counts
 // With FBIP (rc=1): lists are reused in-place
 // Without FBIP (rc>1): new lists are allocated
 fn main(): Unit = {
    Console.print("=== FBIP Allocation Test ===")
    // Case 1: Single owner (FBIP active) - should reuse list
    let a = List.range(1, 100)
    let b = List.map(a, fn(x: Int): Int => x * 2)
    let c = List.filter(b, fn(x: Int): Bool => x > 50)
    let d = List.reverse(c)
    Console.print("Single owner chain done")
    // The allocation count will show FBIP is working
    // if allocations are low relative to operations performed
 }
--- a/examples/test_fbip_clean.lux
+++ b/examples/test_fbip_clean.lux
@@ -1,5 +1,4 @@
 fn main(): Unit = {
    // Test FBIP without string operations
    let nums = [1, 2, 3, 4, 5]
    let doubled = List.map(nums, fn(x: Int): Int => x * 2)
    let filtered = List.filter(doubled, fn(x: Int): Bool => x > 4)
--- a/examples/test_lists.lux
+++ b/examples/test_lists.lux
@@ -1,6 +1,3 @@
 // List Operations Test Suite
 // Run with: lux test examples/test_lists.lux
 fn test_list_length(): Unit with {Test} = {
    Test.assertEqual(0, List.length([]))
    Test.assertEqual(1, List.length([1]))
--- a/examples/test_math.lux
+++ b/examples/test_math.lux
@@ -1,6 +1,3 @@
 // Math Test Suite
 // Run with: lux test examples/test_math.lux
 fn test_addition(): Unit with {Test} = {
    Test.assertEqual(4, 2 + 2)
    Test.assertEqual(0, 0 + 0)
--- a/examples/test_ownership_transfer.lux
+++ b/examples/test_ownership_transfer.lux
@@ -1,21 +1,10 @@
 // Test demonstrating ownership transfer with aliases
 // The ownership transfer optimization ensures FBIP still works
 // even when variables are aliased, because ownership is transferred
 // rather than reference count being incremented.
 fn main(): Unit = {
    Console.print("=== Ownership Transfer Test ===")
    let a = List.range(1, 100)
    // Ownership transfers from 'a' to 'alias', keeping rc=1
    let alias = a
    let len1 = List.length(alias)
    // Since ownership transferred, 'a' still has rc=1
    // FBIP can still optimize map/filter/reverse
    let b = List.map(a, fn(x: Int): Int => x * 2)
    let c = List.filter(b, fn(x: Int): Bool => x > 50)
    let d = List.reverse(c)
    Console.print("Ownership transfer chain done")
 }
--- a/examples/test_rc_comparison.lux
+++ b/examples/test_rc_comparison.lux
@@ -1,17 +1,13 @@
 fn main(): Unit = {
    Console.print("=== Allocation Comparison ===")
    // FBIP path (rc=1): list is reused
    Console.print("Test 1: FBIP path")
    let a1 = List.range(1, 50)
    let b1 = List.map(a1, fn(x: Int): Int => x * 2)
    let c1 = List.reverse(b1)
    Console.print("FBIP done")
    // To show non-FBIP, we need concat which doesn't have FBIP
    Console.print("Test 2: Non-FBIP path (concat)")
    let x = List.range(1, 25)
    let y = List.range(26, 50)
-    let z = List.concat(x, y)  // concat always allocates new
+    let z = List.concat(x, y)
    Console.print("Concat done")
 }
--- a/examples/traits.lux
+++ b/examples/traits.lux
@@ -1,21 +1,11 @@
 // Demonstrating type classes (traits) in Lux
 //
 // Expected output:
 // RGB color: rgb(255,128,0)
 // Red color: red
 // Green color: green
 // Define a simple Printable trait
 trait Printable {
    fn format(value: Int): String
 }
 // Implement Printable
 impl Printable for Int {
    fn format(value: Int): String = "Number: " + toString(value)
 }
 // A Color type with pattern matching
 type Color =
    | Red
    | Green
@@ -24,18 +14,18 @@ type Color =
 fn colorName(c: Color): String =
    match c {
-        Red => "red",
+    Red => "red",
-        Green => "green",
+    Green => "green",
-        Blue => "blue",
+    Blue => "blue",
-        RGB(r, g, b) => "rgb(" + toString(r) + "," + toString(g) + "," + toString(b) + ")"
+    RGB(r, g, b) => "rgb(" + toString(r) + "," + toString(g) + "," + toString(b) + ")",
-    }
+}
 // Test
 let myColor = RGB(255, 128, 0)
 let redColor = Red
 let greenColor = Green
 // Print results
 fn printResults(): Unit with {Console} = {
    Console.print("RGB color: " + colorName(myColor))
    Console.print("Red color: " + colorName(redColor))
--- a/examples/versioning.lux
+++ b/examples/versioning.lux
@@ -1,15 +1,3 @@
 // Demonstrating Schema Evolution in Lux
 //
 // Lux provides versioned types to help manage data evolution over time.
 // The Schema module provides functions for creating and migrating versioned values.
 //
 // Expected output:
 // Created user v1: Alice (age unknown)
 // User version: 1
 // Migrated to v2: Alice (age unknown)
 // User version after migration: 2
 // Create a versioned User value at v1
 fn createUserV1(name: String): Unit with {Console} = {
    let user = Schema.versioned("User", 1, { name: name })
    let version = Schema.getVersion(user)
@@ -17,7 +5,6 @@ fn createUserV1(name: String): Unit with {Console} = {
    Console.print("User version: " + toString(version))
 }
 // Migrate a user to v2
 fn migrateUserToV2(name: String): Unit with {Console} = {
    let userV1 = Schema.versioned("User", 1, { name: name })
    let userV2 = Schema.migrate(userV1, 2)
@@ -26,7 +13,6 @@ fn migrateUserToV2(name: String): Unit with {Console} = {
    Console.print("User version after migration: " + toString(newVersion))
 }
 // Main
 fn main(): Unit with {Console} = {
    createUserV1("Alice")
    migrateUserToV2("Alice")
--- a/examples/web/counter.lux
+++ b/examples/web/counter.lux
@@ -1,62 +1,38 @@
-// Simple Counter for Browser
+type Model =
-// Compile with: lux compile examples/web/counter.lux --target js -o examples/web/counter.js
+    | Counter(Int)
-// ============================================================================
+fn getCount(m: Model): Int =
-// Model
+    match m {
-// ============================================================================
+    Counter(n) => n,
-
+}
 type Model = | Counter(Int)
 fn getCount(m: Model): Int = match m { Counter(n) => n }
 fn init(): Model = Counter(0)
-// ============================================================================
+type Msg =
-// Messages
+    | Increment
-// ============================================================================
+    | Decrement
-
+    | Reset
 type Msg = | Increment | Decrement | Reset
 // ============================================================================
 // Update
 // ============================================================================
 fn update(model: Model, msg: Msg): Model =
    match msg {
-        Increment => Counter(getCount(model) + 1),
+    Increment => Counter(getCount(model) + 1),
-        Decrement => Counter(getCount(model) - 1),
+    Decrement => Counter(getCount(model) - 1),
-        Reset => Counter(0)
+    Reset => Counter(0),
-    }
+}
 // ============================================================================
 // View - Returns HTML string for simplicity
 // ============================================================================
 fn view(model: Model): String = {
    let count = getCount(model)
-    "<div class=\"counter\">" +
+    "<div class=\"counter\">" + "<h1>Lux Counter</h1>" + "<div class=\"display\">" + toString(count) + "</div>" + "<div class=\"buttons\">" + "<button onclick=\"dispatch('Decrement')\">-</button>" + "<button onclick=\"dispatch('Reset')\">Reset</button>" + "<button onclick=\"dispatch('Increment')\">+</button>" + "</div>" + "</div>"
        "<h1>Lux Counter</h1>" +
        "<div class=\"display\">" + toString(count) + "</div>" +
        "<div class=\"buttons\">" +
            "<button onclick=\"dispatch('Decrement')\">-</button>" +
            "<button onclick=\"dispatch('Reset')\">Reset</button>" +
            "<button onclick=\"dispatch('Increment')\">+</button>" +
        "</div>" +
    "</div>"
 }
 // ============================================================================
 // Export for browser runtime
 // ============================================================================
 fn luxInit(): Model = init()
 fn luxUpdate(model: Model, msgName: String): Model =
    match msgName {
-        "Increment" => update(model, Increment),
+    "Increment" => update(model, Increment),
-        "Decrement" => update(model, Decrement),
+    "Decrement" => update(model, Decrement),
-        "Reset" => update(model, Reset),
+    "Reset" => update(model, Reset),
-        _ => model
+    _ => model,
-    }
+}
 fn luxView(model: Model): String = view(model)
--- a/flake.nix
+++ b/flake.nix
@@ -14,6 +14,7 @@
        pkgs = import nixpkgs { inherit system overlays; };
        rustToolchain = pkgs.rust-bin.stable.latest.default.override {
          extensions = [ "rust-src" "rust-analyzer" ];
          targets = [ "x86_64-unknown-linux-musl" ];
        };
      in
      {
@@ -22,8 +23,8 @@
            rustToolchain
            cargo-watch
            cargo-edit
-            pkg-config
+            # Static builds
-            openssl
+            pkgsStatic.stdenv.cc
            # Benchmark tools
            hyperfine
            poop
@@ -43,7 +44,7 @@
            printf "\n"
            printf "  \033[1;35m╦  ╦ ╦╦ ╦\033[0m\n"
            printf "  \033[1;35m║  ║ ║╔╣\033[0m\n"
-            printf "  \033[1;35m╩═╝╚═╝╩ ╩\033[0m  v0.1.0\n"
+            printf "  \033[1;35m╩═╝╚═╝╩ ╩\033[0m  v0.1.3\n"
            printf "\n"
            printf "  Functional language with first-class effects\n"
            printf "\n"
@@ -61,18 +62,47 @@
        packages.default = pkgs.rustPlatform.buildRustPackage {
          pname = "lux";
-          version = "0.1.0";
+          version = "0.1.3";
          src = ./.;
          cargoLock.lockFile = ./Cargo.lock;
          nativeBuildInputs = [ pkgs.pkg-config ];
          buildInputs = [ pkgs.openssl ];
          doCheck = false;
        };
-        # Benchmark scripts
+        packages.static = let
          muslPkgs = import nixpkgs {
            inherit system;
            crossSystem = {
              config = "x86_64-unknown-linux-musl";
              isStatic = true;
            };
          };
        in muslPkgs.rustPlatform.buildRustPackage {
          pname = "lux";
          version = "0.1.3";
          src = ./.;
          cargoLock.lockFile = ./Cargo.lock;
          CARGO_BUILD_TARGET = "x86_64-unknown-linux-musl";
          CARGO_BUILD_RUSTFLAGS = "-C target-feature=+crt-static";
          doCheck = false;
          postInstall = ''
            $STRIP $out/bin/lux 2>/dev/null || true
          '';
        };
        apps = {
          # Release automation
          release = {
            type = "app";
            program = toString (pkgs.writeShellScript "lux-release" ''
              exec ${self}/scripts/release.sh "$@"
            '');
          };
          # Benchmark scripts
          # Run hyperfine benchmark comparison
          bench = {
            type = "app";
--- a/projects/lux-compiler/lexer.lux
+++ b/projects/lux-compiler/lexer.lux
@@ -0,0 +1,512 @@
 // Lux Lexer — Self-hosted lexer for the Lux language
 //
 // This is the first component of the Lux-in-Lux compiler.
 // It tokenizes Lux source code into a list of tokens.
 //
 // Design:
 //   - Recursive descent character scanning
 //   - Immutable state (ParseState tracks chars + position)
 //   - Pattern matching for all token types
 // === Token types ===
 type TokenKind =
    // Literals
    | TkInt(Int)
    | TkFloat(String)
    | TkString(String)
    | TkChar(Char)
    | TkBool(Bool)
    // Identifiers
    | TkIdent(String)
    // Keywords
    | TkFn | TkLet | TkIf | TkThen | TkElse | TkMatch
    | TkWith | TkEffect | TkHandler | TkRun | TkResume
    | TkType | TkImport | TkPub | TkAs | TkFrom
    | TkTrait | TkImpl | TkFor
    // Behavioral
    | TkIs | TkPure | TkTotal | TkIdempotent
    | TkDeterministic | TkCommutative
    | TkWhere | TkAssume
    // Operators
    | TkPlus | TkMinus | TkStar | TkSlash | TkPercent
    | TkEq | TkEqEq | TkNe | TkLt | TkLe | TkGt | TkGe
    | TkAnd | TkOr | TkNot
    | TkPipe | TkPipeGt | TkArrow | TkThinArrow
    | TkDot | TkColon | TkColonColon | TkComma | TkSemi | TkAt
    // Delimiters
    | TkLParen | TkRParen | TkLBrace | TkRBrace
    | TkLBracket | TkRBracket
    // Special
    | TkUnderscore | TkNewline | TkEof
    // Doc comment
    | TkDocComment(String)
 type Token =
    | Token(TokenKind, Int, Int)  // kind, start, end
 type LexState =
    | LexState(List<Char>, Int)  // chars, position
 type LexResult =
    | LexOk(Token, LexState)
    | LexErr(String, Int)
 // === Character utilities ===
 fn peek(state: LexState): Option<Char> =
    match state {
        LexState(chars, pos) => List.get(chars, pos)
    }
 fn peekAt(state: LexState, offset: Int): Option<Char> =
    match state {
        LexState(chars, pos) => List.get(chars, pos + offset)
    }
 fn advance(state: LexState): LexState =
    match state {
        LexState(chars, pos) => LexState(chars, pos + 1)
    }
 fn position(state: LexState): Int =
    match state { LexState(_, pos) => pos }
 fn isDigit(c: Char): Bool =
    c == '0' || c == '1' || c == '2' || c == '3' || c == '4' ||
    c == '5' || c == '6' || c == '7' || c == '8' || c == '9'
 fn isAlpha(c: Char): Bool =
    (c >= 'a' && c <= 'z') || (c >= 'A' && c <= 'Z') || c == '_'
 fn isAlphaNumeric(c: Char): Bool =
    isAlpha(c) || isDigit(c)
 fn isWhitespace(c: Char): Bool =
    c == ' ' || c == '\t' || c == '\r'
 // === Core lexing ===
 fn skipLineComment(state: LexState): LexState =
    match peek(state) {
        None => state,
        Some(c) =>
            if c == '\n' then state
            else skipLineComment(advance(state))
    }
 fn skipWhitespaceAndComments(state: LexState): LexState =
    match peek(state) {
        None => state,
        Some(c) =>
            if isWhitespace(c) then
                skipWhitespaceAndComments(advance(state))
            else if c == '/' then
                match peekAt(state, 1) {
                    Some('/') =>
                        // Check for doc comment (///)
                        match peekAt(state, 2) {
                            Some('/') => state,  // Don't skip doc comments
                            _ => skipWhitespaceAndComments(skipLineComment(advance(advance(state))))
                        },
                    _ => state
                }
            else state
    }
 // Collect identifier characters
 fn collectIdent(state: LexState, acc: List<Char>): (List<Char>, LexState) =
    match peek(state) {
        None => (acc, state),
        Some(c) =>
            if isAlphaNumeric(c) then
                collectIdent(advance(state), List.concat(acc, [c]))
            else (acc, state)
    }
 // Collect number characters (digits only)
 fn collectDigits(state: LexState, acc: List<Char>): (List<Char>, LexState) =
    match peek(state) {
        None => (acc, state),
        Some(c) =>
            if isDigit(c) then
                collectDigits(advance(state), List.concat(acc, [c]))
            else (acc, state)
    }
 // Convert list of digit chars to int
 fn charsToInt(chars: List<Char>): Int =
    List.fold(chars, 0, fn(acc, c) => acc * 10 + charToDigit(c))
 fn charToDigit(c: Char): Int =
    if c == '0' then 0
    else if c == '1' then 1
    else if c == '2' then 2
    else if c == '3' then 3
    else if c == '4' then 4
    else if c == '5' then 5
    else if c == '6' then 6
    else if c == '7' then 7
    else if c == '8' then 8
    else 9
 // Map identifier string to keyword token or ident
 fn identToToken(name: String): TokenKind =
    if name == "fn" then TkFn
    else if name == "let" then TkLet
    else if name == "if" then TkIf
    else if name == "then" then TkThen
    else if name == "else" then TkElse
    else if name == "match" then TkMatch
    else if name == "with" then TkWith
    else if name == "effect" then TkEffect
    else if name == "handler" then TkHandler
    else if name == "run" then TkRun
    else if name == "resume" then TkResume
    else if name == "type" then TkType
    else if name == "true" then TkBool(true)
    else if name == "false" then TkBool(false)
    else if name == "import" then TkImport
    else if name == "pub" then TkPub
    else if name == "as" then TkAs
    else if name == "from" then TkFrom
    else if name == "trait" then TkTrait
    else if name == "impl" then TkImpl
    else if name == "for" then TkFor
    else if name == "is" then TkIs
    else if name == "pure" then TkPure
    else if name == "total" then TkTotal
    else if name == "idempotent" then TkIdempotent
    else if name == "deterministic" then TkDeterministic
    else if name == "commutative" then TkCommutative
    else if name == "where" then TkWhere
    else if name == "assume" then TkAssume
    else TkIdent(name)
 // Lex a string literal (after opening quote consumed)
 fn lexStringBody(state: LexState, acc: List<Char>): (List<Char>, LexState) =
    match peek(state) {
        None => (acc, state),
        Some(c) =>
            if c == '"' then (acc, advance(state))
            else if c == '\\' then
                match peekAt(state, 1) {
                    Some('n') => lexStringBody(advance(advance(state)), List.concat(acc, ['\n'])),
                    Some('t') => lexStringBody(advance(advance(state)), List.concat(acc, ['\t'])),
                    Some('\\') => lexStringBody(advance(advance(state)), List.concat(acc, ['\\'])),
                    Some('"') => lexStringBody(advance(advance(state)), List.concat(acc, ['"'])),
                    _ => lexStringBody(advance(state), List.concat(acc, [c]))
                }
            else lexStringBody(advance(state), List.concat(acc, [c]))
    }
 // Lex a char literal (after opening quote consumed)
 fn lexCharLiteral(state: LexState): LexResult =
    let start = position(state) - 1;
    match peek(state) {
        None => LexErr("Unexpected end of input in char literal", start),
        Some(c) =>
            if c == '\\' then
                match peekAt(state, 1) {
                    Some('n') =>
                        match peekAt(state, 2) {
                            Some('\'') => LexOk(Token(TkChar('\n'), start, position(state) + 3), advance(advance(advance(state)))),
                            _ => LexErr("Expected closing quote", position(state))
                        },
                    Some('t') =>
                        match peekAt(state, 2) {
                            Some('\'') => LexOk(Token(TkChar('\t'), start, position(state) + 3), advance(advance(advance(state)))),
                            _ => LexErr("Expected closing quote", position(state))
                        },
                    Some('\\') =>
                        match peekAt(state, 2) {
                            Some('\'') => LexOk(Token(TkChar('\\'), start, position(state) + 3), advance(advance(advance(state)))),
                            _ => LexErr("Expected closing quote", position(state))
                        },
                    _ => LexErr("Unknown escape sequence", position(state))
                }
            else
                match peekAt(state, 1) {
                    Some('\'') => LexOk(Token(TkChar(c), start, position(state) + 2), advance(advance(state))),
                    _ => LexErr("Expected closing quote", position(state))
                }
    }
 // Collect doc comment text (after /// consumed)
 fn collectDocComment(state: LexState, acc: List<Char>): (List<Char>, LexState) =
    match peek(state) {
        None => (acc, state),
        Some(c) =>
            if c == '\n' then (acc, state)
            else collectDocComment(advance(state), List.concat(acc, [c]))
    }
 // Lex a single token
 fn lexToken(state: LexState): LexResult =
    let state = skipWhitespaceAndComments(state);
    let start = position(state);
    match peek(state) {
        None => LexOk(Token(TkEof, start, start), state),
        Some(c) =>
            if c == '\n' then
                LexOk(Token(TkNewline, start, start + 1), advance(state))
            // Numbers
            else if isDigit(c) then
                let result = collectDigits(state, []);
                match result {
                    (digits, nextState) =>
                        // Check for float
                        match peek(nextState) {
                            Some('.') =>
                                match peekAt(nextState, 1) {
                                    Some(d) =>
                                        if isDigit(d) then
                                            let fracResult = collectDigits(advance(nextState), []);
                                            match fracResult {
                                                (fracDigits, finalState) =>
                                                    let intPart = String.join(List.map(digits, fn(ch) => String.fromChar(ch)), "");
                                                    let fracPart = String.join(List.map(fracDigits, fn(ch) => String.fromChar(ch)), "");
                                                    LexOk(Token(TkFloat(intPart + "." + fracPart), start, position(finalState)), finalState)
                                            }
                                        else
                                            LexOk(Token(TkInt(charsToInt(digits)), start, position(nextState)), nextState),
                                    None =>
                                        LexOk(Token(TkInt(charsToInt(digits)), start, position(nextState)), nextState)
                                },
                            _ => LexOk(Token(TkInt(charsToInt(digits)), start, position(nextState)), nextState)
                        }
                }
            // Identifiers and keywords
            else if isAlpha(c) then
                let result = collectIdent(state, []);
                match result {
                    (chars, nextState) =>
                        let name = String.join(List.map(chars, fn(ch) => String.fromChar(ch)), "");
                        LexOk(Token(identToToken(name), start, position(nextState)), nextState)
                }
            // String literals
            else if c == '"' then
                let result = lexStringBody(advance(state), []);
                match result {
                    (chars, nextState) =>
                        let str = String.join(List.map(chars, fn(ch) => String.fromChar(ch)), "");
                        LexOk(Token(TkString(str), start, position(nextState)), nextState)
                }
            // Char literals
            else if c == '\'' then
                lexCharLiteral(advance(state))
            // Doc comments (///)
            else if c == '/' then
                match peekAt(state, 1) {
                    Some('/') =>
                        match peekAt(state, 2) {
                            Some('/') =>
                                // Skip the "/// " prefix
                                let docState = advance(advance(advance(state)));
                                let docState = match peek(docState) {
                                    Some(' ') => advance(docState),
                                    _ => docState
                                };
                                let result = collectDocComment(docState, []);
                                match result {
                                    (chars, nextState) =>
                                        let text = String.join(List.map(chars, fn(ch) => String.fromChar(ch)), "");
                                        LexOk(Token(TkDocComment(text), start, position(nextState)), nextState)
                                },
                            _ => LexOk(Token(TkSlash, start, start + 1), advance(state))
                        },
                    _ => LexOk(Token(TkSlash, start, start + 1), advance(state))
                }
            // Two-character operators
            else if c == '=' then
                match peekAt(state, 1) {
                    Some('=') => LexOk(Token(TkEqEq, start, start + 2), advance(advance(state))),
                    Some('>') => LexOk(Token(TkArrow, start, start + 2), advance(advance(state))),
                    _ => LexOk(Token(TkEq, start, start + 1), advance(state))
                }
            else if c == '!' then
                match peekAt(state, 1) {
                    Some('=') => LexOk(Token(TkNe, start, start + 2), advance(advance(state))),
                    _ => LexOk(Token(TkNot, start, start + 1), advance(state))
                }
            else if c == '<' then
                match peekAt(state, 1) {
                    Some('=') => LexOk(Token(TkLe, start, start + 2), advance(advance(state))),
                    _ => LexOk(Token(TkLt, start, start + 1), advance(state))
                }
            else if c == '>' then
                match peekAt(state, 1) {
                    Some('=') => LexOk(Token(TkGe, start, start + 2), advance(advance(state))),
                    _ => LexOk(Token(TkGt, start, start + 1), advance(state))
                }
            else if c == '&' then
                match peekAt(state, 1) {
                    Some('&') => LexOk(Token(TkAnd, start, start + 2), advance(advance(state))),
                    _ => LexErr("Expected '&&'", start)
                }
            else if c == '|' then
                match peekAt(state, 1) {
                    Some('|') => LexOk(Token(TkOr, start, start + 2), advance(advance(state))),
                    Some('>') => LexOk(Token(TkPipeGt, start, start + 2), advance(advance(state))),
                    _ => LexOk(Token(TkPipe, start, start + 1), advance(state))
                }
            else if c == '-' then
                match peekAt(state, 1) {
                    Some('>') => LexOk(Token(TkThinArrow, start, start + 2), advance(advance(state))),
                    _ => LexOk(Token(TkMinus, start, start + 1), advance(state))
                }
            else if c == ':' then
                match peekAt(state, 1) {
                    Some(':') => LexOk(Token(TkColonColon, start, start + 2), advance(advance(state))),
                    _ => LexOk(Token(TkColon, start, start + 1), advance(state))
                }
            // Single-character tokens
            else if c == '+' then LexOk(Token(TkPlus, start, start + 1), advance(state))
            else if c == '*' then LexOk(Token(TkStar, start, start + 1), advance(state))
            else if c == '%' then LexOk(Token(TkPercent, start, start + 1), advance(state))
            else if c == '.' then LexOk(Token(TkDot, start, start + 1), advance(state))
            else if c == ',' then LexOk(Token(TkComma, start, start + 1), advance(state))
            else if c == ';' then LexOk(Token(TkSemi, start, start + 1), advance(state))
            else if c == '@' then LexOk(Token(TkAt, start, start + 1), advance(state))
            else if c == '(' then LexOk(Token(TkLParen, start, start + 1), advance(state))
            else if c == ')' then LexOk(Token(TkRParen, start, start + 1), advance(state))
            else if c == '{' then LexOk(Token(TkLBrace, start, start + 1), advance(state))
            else if c == '}' then LexOk(Token(TkRBrace, start, start + 1), advance(state))
            else if c == '[' then LexOk(Token(TkLBracket, start, start + 1), advance(state))
            else if c == ']' then LexOk(Token(TkRBracket, start, start + 1), advance(state))
            else if c == '_' then
                // Check if it's just underscore or start of ident
                match peekAt(state, 1) {
                    Some(next) =>
                        if isAlphaNumeric(next) then
                            let result = collectIdent(state, []);
                            match result {
                                (chars, nextState) =>
                                    let name = String.join(List.map(chars, fn(ch) => String.fromChar(ch)), "");
                                    LexOk(Token(TkIdent(name), start, position(nextState)), nextState)
                            }
                        else LexOk(Token(TkUnderscore, start, start + 1), advance(state)),
                    None => LexOk(Token(TkUnderscore, start, start + 1), advance(state))
                }
            else LexErr("Unexpected character: " + String.fromChar(c), start)
    }
 // Lex all tokens from source
 fn lexAll(state: LexState, acc: List<Token>): List<Token> =
    match lexToken(state) {
        LexErr(msg, pos) =>
            // On error, skip the character and continue
            List.concat(acc, [Token(TkEof, pos, pos)]),
        LexOk(token, nextState) =>
            match token {
                Token(TkEof, _, _) => List.concat(acc, [token]),
                Token(TkNewline, _, _) =>
                    // Skip consecutive newlines
                    lexAll(nextState, List.concat(acc, [token])),
                _ => lexAll(nextState, List.concat(acc, [token]))
            }
    }
 // Public API: tokenize a source string
 fn tokenize(source: String): List<Token> =
    let chars = String.chars(source);
    let state = LexState(chars, 0);
    lexAll(state, [])
 // === Token display ===
 fn tokenKindToString(kind: TokenKind): String =
    match kind {
        TkInt(n) => "Int(" + toString(n) + ")",
        TkFloat(s) => "Float(" + s + ")",
        TkString(s) => "String(\"" + s + "\")",
        TkChar(c) => "Char('" + String.fromChar(c) + "')",
        TkBool(b) => if b then "true" else "false",
        TkIdent(name) => "Ident(" + name + ")",
        TkFn => "fn", TkLet => "let", TkIf => "if",
        TkThen => "then", TkElse => "else", TkMatch => "match",
        TkWith => "with", TkEffect => "effect", TkHandler => "handler",
        TkRun => "run", TkResume => "resume", TkType => "type",
        TkImport => "import", TkPub => "pub", TkAs => "as",
        TkFrom => "from", TkTrait => "trait", TkImpl => "impl", TkFor => "for",
        TkIs => "is", TkPure => "pure", TkTotal => "total",
        TkIdempotent => "idempotent", TkDeterministic => "deterministic",
        TkCommutative => "commutative", TkWhere => "where", TkAssume => "assume",
        TkPlus => "+", TkMinus => "-", TkStar => "*", TkSlash => "/",
        TkPercent => "%", TkEq => "=", TkEqEq => "==", TkNe => "!=",
        TkLt => "<", TkLe => "<=", TkGt => ">", TkGe => ">=",
        TkAnd => "&&", TkOr => "||", TkNot => "!",
        TkPipe => "|", TkPipeGt => "|>",
        TkArrow => "=>", TkThinArrow => "->",
        TkDot => ".", TkColon => ":", TkColonColon => "::",
        TkComma => ",", TkSemi => ";", TkAt => "@",
        TkLParen => "(", TkRParen => ")", TkLBrace => "{", TkRBrace => "}",
        TkLBracket => "[", TkRBracket => "]",
        TkUnderscore => "_", TkNewline => "\\n", TkEof => "EOF",
        TkDocComment(text) => "DocComment(\"" + text + "\")",
        _ => "?"
    }
 fn tokenToString(token: Token): String =
    match token {
        Token(kind, start, end) =>
            tokenKindToString(kind) + " [" + toString(start) + ".." + toString(end) + "]"
    }
 // === Tests ===
 fn printTokens(tokens: List<Token>): Unit with {Console} =
    match List.head(tokens) {
        None => Console.print(""),
        Some(t) => {
            Console.print("  " + tokenToString(t));
            match List.tail(tokens) {
                Some(rest) => printTokens(rest),
                None => Console.print("")
            }
        }
    }
 fn testLexer(label: String, source: String): Unit with {Console} = {
    Console.print("--- " + label + " ---");
    Console.print("  Input: \"" + source + "\"");
    let tokens = tokenize(source);
    printTokens(tokens)
 }
 fn main(): Unit with {Console} = {
    Console.print("=== Lux Self-Hosted Lexer ===");
    Console.print("");
    // Basic tokens
    testLexer("numbers", "42 3");
    Console.print("");
    // Identifiers and keywords
    testLexer("keywords", "fn main let x");
    Console.print("");
    // Operators
    testLexer("operators", "a + b == c");
    Console.print("");
    // String literal
    testLexer("string", "\"hello world\"");
    Console.print("");
    // Function declaration
    testLexer("function", "fn add(a: Int, b: Int): Int = a + b");
    Console.print("");
    // Behavioral properties
    testLexer("behavioral", "fn add(a: Int): Int is pure = a");
    Console.print("");
    // Complex expression
    testLexer("complex", "let result = if x > 0 then x else 0 - x");
    Console.print("");
    Console.print("=== Lexer test complete ===")
 }
 let _ = run main() with {}
--- a/scripts/release.sh
+++ b/scripts/release.sh
@@ -0,0 +1,213 @@
 #!/usr/bin/env bash
 set -euo pipefail
 # Lux Release Script
 # Builds a static binary, generates changelog, and creates a Gitea release.
 #
 # Usage:
 #   ./scripts/release.sh              # auto-bump patch (0.2.0 → 0.2.1)
 #   ./scripts/release.sh patch        # same as above
 #   ./scripts/release.sh minor        # bump minor (0.2.0 → 0.3.0)
 #   ./scripts/release.sh major        # bump major (0.2.0 → 1.0.0)
 #   ./scripts/release.sh v1.2.3       # explicit version
 #
 # Environment:
 #   GITEA_TOKEN - API token for git.qrty.ink (prompted if not set)
 #   GITEA_URL   - Gitea instance URL (default: https://git.qrty.ink)
 # cd to repo root (directory containing this script's parent)
 SCRIPT_DIR="$(cd "$(dirname "$0")" && pwd)"
 cd "$SCRIPT_DIR/.."
 GITEA_URL="${GITEA_URL:-https://git.qrty.ink}"
 REPO_OWNER="blu"
 REPO_NAME="lux"
 API_BASE="$GITEA_URL/api/v1"
 # Colors
 RED='\033[0;31m'
 GREEN='\033[0;32m'
 YELLOW='\033[1;33m'
 CYAN='\033[0;36m'
 BOLD='\033[1m'
 NC='\033[0m'
 info()  { printf "${CYAN}::${NC} %s\n" "$1"; }
 ok()    { printf "${GREEN}ok${NC} %s\n" "$1"; }
 warn()  { printf "${YELLOW}!!${NC} %s\n" "$1"; }
 err()   { printf "${RED}error:${NC} %s\n" "$1" >&2; exit 1; }
 # --- Determine version ---
 CURRENT=$(grep '^version' Cargo.toml | head -1 | sed 's/.*"\(.*\)".*/\1/')
 BUMP="${1:-patch}"
 bump_version() {
    local ver="$1" part="$2"
    IFS='.' read -r major minor patch <<< "$ver"
    case "$part" in
        major) echo "$((major + 1)).0.0" ;;
        minor) echo "$major.$((minor + 1)).0" ;;
        patch) echo "$major.$minor.$((patch + 1))" ;;
        *)     echo "$part" ;;  # treat as explicit version
    esac
 }
 case "$BUMP" in
    major|minor|patch)
        VERSION=$(bump_version "$CURRENT" "$BUMP")
        info "Bumping $BUMP: $CURRENT → $VERSION"
        ;;
    *)
        # Explicit version — strip v prefix if present
        VERSION="${BUMP#v}"
        info "Explicit version: $VERSION"
        ;;
 esac
 TAG="v$VERSION"
 # --- Check for clean working tree ---
 if [ -n "$(git status --porcelain)" ]; then
    warn "Working tree has uncommitted changes:"
    git status --short
    printf "\n"
    read -rp "Continue anyway? [y/N] " confirm
    [[ "$confirm" =~ ^[Yy]$ ]] || exit 1
 fi
 # --- Check if tag already exists ---
 if git rev-parse "$TAG" >/dev/null 2>&1; then
    err "Tag $TAG already exists. Choose a different version."
 fi
 # --- Update version in source files ---
 if [ "$VERSION" != "$CURRENT" ]; then
    info "Updating version in Cargo.toml and flake.nix..."
    sed -i "0,/^version = \"$CURRENT\"/s//version = \"$VERSION\"/" Cargo.toml
    sed -i "s/version = \"$CURRENT\";/version = \"$VERSION\";/g" flake.nix
    sed -i "s/v$CURRENT/v$VERSION/g" flake.nix
    git add Cargo.toml flake.nix
    git commit --no-gpg-sign -m "chore: bump version to $VERSION"
    ok "Version updated and committed"
 fi
 # --- Generate changelog ---
 info "Generating changelog..."
 LAST_TAG=$(git describe --tags --abbrev=0 2>/dev/null || echo "")
 if [ -n "$LAST_TAG" ]; then
    RANGE="$LAST_TAG..HEAD"
    info "Changes since $LAST_TAG:"
 else
    RANGE="HEAD"
    info "First release — summarizing recent commits:"
 fi
 CHANGELOG=$(git log "$RANGE" --pretty=format:"- %s" --no-merges 2>/dev/null | head -50 || true)
 if [ -z "$CHANGELOG" ]; then
    CHANGELOG="- Initial release"
 fi
 # --- Build static binary ---
 info "Building static binary (nix build .#static)..."
 nix build .#static
 BINARY="result/bin/lux"
 if [ ! -f "$BINARY" ]; then
    err "Static binary not found at $BINARY"
 fi
 BINARY_SIZE=$(ls -lh "$BINARY" | awk '{print $5}')
 BINARY_TYPE=$(file "$BINARY" | sed 's/.*: //')
 ok "Binary: $BINARY_SIZE, $BINARY_TYPE"
 # --- Prepare release artifact ---
 ARTIFACT="/tmp/lux-${TAG}-linux-x86_64"
 cp "$BINARY" "$ARTIFACT"
 chmod +x "$ARTIFACT"
 # --- Show release summary ---
 printf "\n"
 printf "${BOLD}═══ Release Summary ═══${NC}\n"
 printf "\n"
 printf "  ${BOLD}Tag:${NC}      %s\n" "$TAG"
 printf "  ${BOLD}Binary:${NC}   %s (%s)\n" "lux-${TAG}-linux-x86_64" "$BINARY_SIZE"
 printf "  ${BOLD}Commit:${NC}   %s\n" "$(git rev-parse --short HEAD)"
 printf "\n"
 printf "${BOLD}Changelog:${NC}\n"
 printf "%s\n" "$CHANGELOG"
 printf "\n"
 # --- Confirm ---
 read -rp "Create release $TAG? [y/N] " confirm
 [[ "$confirm" =~ ^[Yy]$ ]] || { info "Aborted."; exit 0; }
 # --- Get Gitea token ---
 if [ -z "${GITEA_TOKEN:-}" ]; then
    printf "\n"
    info "Gitea API token required (create at $GITEA_URL/user/settings/applications)"
    read -rsp "Token: " GITEA_TOKEN
    printf "\n"
 fi
 if [ -z "$GITEA_TOKEN" ]; then
    err "No token provided"
 fi
 # --- Create and push tag ---
 info "Creating tag $TAG..."
 git tag -a "$TAG" -m "Release $TAG" --no-sign
 ok "Tag created"
 info "Pushing tag to origin..."
 git push origin "$TAG"
 ok "Tag pushed"
 # --- Create Gitea release ---
 info "Creating release on Gitea..."
 RELEASE_BODY=$(printf "## Lux %s\n\n### Changes\n\n%s\n\n### Installation\n\n\`\`\`bash\ncurl -Lo lux %s/%s/%s/releases/download/%s/lux-linux-x86_64\nchmod +x lux\n./lux --version\n\`\`\`" \
    "$TAG" "$CHANGELOG" "$GITEA_URL" "$REPO_OWNER" "$REPO_NAME" "$TAG")
 RELEASE_JSON=$(jq -n \
    --arg tag "$TAG" \
    --arg name "Lux $TAG" \
    --arg body "$RELEASE_BODY" \
    '{tag_name: $tag, name: $name, body: $body, draft: false, prerelease: false}')
 RELEASE_RESPONSE=$(curl -s -X POST \
    "$API_BASE/repos/$REPO_OWNER/$REPO_NAME/releases" \
    -H "Authorization: token $GITEA_TOKEN" \
    -H "Content-Type: application/json" \
    -d "$RELEASE_JSON")
 RELEASE_ID=$(echo "$RELEASE_RESPONSE" | jq -r '.id // empty')
 if [ -z "$RELEASE_ID" ]; then
    echo "$RELEASE_RESPONSE" | jq . 2>/dev/null || echo "$RELEASE_RESPONSE"
    err "Failed to create release"
 fi
 ok "Release created (id: $RELEASE_ID)"
 # --- Upload binary ---
 info "Uploading binary..."
 UPLOAD_RESPONSE=$(curl -s -X POST \
    "$API_BASE/repos/$REPO_OWNER/$REPO_NAME/releases/$RELEASE_ID/assets?name=lux-linux-x86_64" \
    -H "Authorization: token $GITEA_TOKEN" \
    -H "Content-Type: application/octet-stream" \
    --data-binary "@$ARTIFACT")
 ASSET_NAME=$(echo "$UPLOAD_RESPONSE" | jq -r '.name // empty')
 if [ -z "$ASSET_NAME" ]; then
    echo "$UPLOAD_RESPONSE" | jq . 2>/dev/null || echo "$UPLOAD_RESPONSE"
    err "Failed to upload binary"
 fi
 ok "Binary uploaded: $ASSET_NAME"
 # --- Done ---
 printf "\n"
 printf "${GREEN}${BOLD}Release $TAG published!${NC}\n"
 printf "\n"
 printf "  ${BOLD}URL:${NC} %s/%s/%s/releases/tag/%s\n" "$GITEA_URL" "$REPO_OWNER" "$REPO_NAME" "$TAG"
 printf "  ${BOLD}Download:${NC} %s/%s/%s/releases/download/%s/lux-linux-x86_64\n" "$GITEA_URL" "$REPO_OWNER" "$REPO_NAME" "$TAG"
 printf "\n"
 # Cleanup
 rm -f "$ARTIFACT"
--- a/scripts/validate.sh
+++ b/scripts/validate.sh
@@ -0,0 +1,73 @@
 #!/usr/bin/env bash
 set -euo pipefail
 # Lux Full Validation Script
 # Runs all checks: Rust tests, package tests, type checking, formatting, linting.
 # Run after every committable change to ensure no regressions.
 # cd to repo root (directory containing this script's parent)
 SCRIPT_DIR="$(cd "$(dirname "$0")" && pwd)"
 cd "$SCRIPT_DIR/.."
 LUX="$(pwd)/target/release/lux"
 PACKAGES_DIR="$(pwd)/../packages"
 RED='\033[0;31m'
 GREEN='\033[0;32m'
 CYAN='\033[0;36m'
 BOLD='\033[1m'
 NC='\033[0m'
 FAILED=0
 TOTAL=0
 step() {
    TOTAL=$((TOTAL + 1))
    printf "${CYAN}[%d]${NC} %s... " "$TOTAL" "$1"
 }
 ok() { printf "${GREEN}ok${NC} %s\n" "${1:-}"; }
 fail() { printf "${RED}FAIL${NC} %s\n" "${1:-}"; FAILED=$((FAILED + 1)); }
 # --- Rust checks ---
 step "cargo check"
 if nix develop --command cargo check 2>&1 | grep -q "Finished"; then ok; else fail; fi
 step "cargo test"
 OUTPUT=$(nix develop --command cargo test 2>&1 || true)
 RESULT=$(echo "$OUTPUT" | grep "test result:" || echo "no result")
 if echo "$RESULT" | grep -q "0 failed"; then ok "$RESULT"; else fail "$RESULT"; fi
 # --- Build release binary ---
 step "cargo build --release"
 if nix develop --command cargo build --release 2>&1 | grep -q "Finished"; then ok; else fail; fi
 # --- Package tests ---
 for pkg in path frontmatter xml rss markdown; do
    PKG_DIR="$PACKAGES_DIR/$pkg"
    if [ -d "$PKG_DIR" ]; then
        step "lux test ($pkg)"
        OUTPUT=$(cd "$PKG_DIR" && "$LUX" test 2>&1 || true)
        RESULT=$(echo "$OUTPUT" | grep "passed" | tail -1 || echo "no result")
        if echo "$RESULT" | grep -q "passed"; then ok "$RESULT"; else fail "$RESULT"; fi
    fi
 done
 # --- Lux check on packages ---
 for pkg in path frontmatter xml rss markdown; do
    PKG_DIR="$PACKAGES_DIR/$pkg"
    if [ -d "$PKG_DIR" ]; then
        step "lux check ($pkg)"
        OUTPUT=$(cd "$PKG_DIR" && "$LUX" check 2>&1 || true)
        RESULT=$(echo "$OUTPUT" | grep "passed" | tail -1 || echo "no result")
        if echo "$RESULT" | grep -q "passed"; then ok; else fail "$RESULT"; fi
    fi
 done
 # --- Summary ---
 printf "\n${BOLD}═══ Validation Summary ═══${NC}\n"
 if [ $FAILED -eq 0 ]; then
    printf "${GREEN}All %d checks passed.${NC}\n" "$TOTAL"
 else
    printf "${RED}%d/%d checks failed.${NC}\n" "$FAILED" "$TOTAL"
    exit 1
 fi
--- a/src/ast.rs
+++ b/src/ast.rs
@@ -499,6 +499,12 @@ pub enum Expr {
        field: Ident,
        span: Span,
    },
    /// Tuple index access: tuple.0, tuple.1
    TupleIndex {
        object: Box<Expr>,
        index: usize,
        span: Span,
    },
    /// Lambda: fn(x, y) => x + y  or  fn(x: Int): Int => x + 1
    Lambda {
        params: Vec<Parameter>,
@@ -535,7 +541,9 @@ pub enum Expr {
        span: Span,
    },
    /// Record literal: { name: "Alice", age: 30 }
    /// With optional spread: { ...base, name: "Bob" }
    Record {
        spread: Option<Box<Expr>>,
        fields: Vec<(Ident, Expr)>,
        span: Span,
    },
@@ -563,6 +571,7 @@ impl Expr {
            Expr::Call { span, .. } => *span,
            Expr::EffectOp { span, .. } => *span,
            Expr::Field { span, .. } => *span,
            Expr::TupleIndex { span, .. } => *span,
            Expr::Lambda { span, .. } => *span,
            Expr::Let { span, .. } => *span,
            Expr::If { span, .. } => *span,
@@ -614,7 +623,8 @@ pub enum BinaryOp {
    And,
    Or,
    // Other
-    Pipe, // |>
+    Pipe,   // |>
    Concat, // ++
 }
 impl fmt::Display for BinaryOp {
@@ -634,6 +644,7 @@ impl fmt::Display for BinaryOp {
            BinaryOp::And => write!(f, "&&"),
            BinaryOp::Or => write!(f, "||"),
            BinaryOp::Pipe => write!(f, "|>"),
            BinaryOp::Concat => write!(f, "++"),
        }
    }
 }
--- a/src/codegen/c_backend.rs
+++ b/src/codegen/c_backend.rs
--- a/src/codegen/js_backend.rs
+++ b/src/codegen/js_backend.rs
@@ -909,13 +909,16 @@ impl JsBackend {
        let val = self.emit_expr(&let_decl.value)?;
        let var_name = &let_decl.name.name;
-        // Check if this is a run expression (often results in undefined)
+        if var_name == "_" {
-        // We still want to execute it for its side effects
+            // Wildcard binding: just execute for side effects
-        self.writeln(&format!("const {} = {};", var_name, val));
+            self.writeln(&format!("{};", val));
        } else {
            self.writeln(&format!("const {} = {};", var_name, val));
-        // Register the variable for future use
+            // Register the variable for future use
-        self.var_substitutions
+            self.var_substitutions
-            .insert(var_name.clone(), var_name.clone());
+                .insert(var_name.clone(), var_name.clone());
        }
        Ok(())
    }
@@ -954,12 +957,17 @@ impl JsBackend {
                let r = self.emit_expr(right)?;
                // Check for string concatenation
-                if matches!(op, BinaryOp::Add) {
+                if matches!(op, BinaryOp::Add | BinaryOp::Concat) {
                    if self.is_string_expr(left) || self.is_string_expr(right) {
                        return Ok(format!("({} + {})", l, r));
                    }
                }
                // ++ on lists: use .concat()
                if matches!(op, BinaryOp::Concat) {
                    return Ok(format!("{}.concat({})", l, r));
                }
                let op_str = match op {
                    BinaryOp::Add => "+",
                    BinaryOp::Sub => "-",
@@ -974,6 +982,7 @@ impl JsBackend {
                    BinaryOp::Ge => ">=",
                    BinaryOp::And => "&&",
                    BinaryOp::Or => "||",
                    BinaryOp::Concat => unreachable!("handled above"),
                    BinaryOp::Pipe => {
                        // Pipe operator: x |> f becomes f(x)
                        return Ok(format!("{}({})", r, l));
@@ -1034,18 +1043,26 @@ impl JsBackend {
                name, value, body, ..
            } => {
                let val = self.emit_expr(value)?;
                let var_name = format!("{}_{}", name.name, self.fresh_name());
-                self.writeln(&format!("const {} = {};", var_name, val));
+                if name.name == "_" {
                    // Wildcard binding: just execute for side effects
                    self.writeln(&format!("{};", val));
                } else {
                    let var_name = format!("{}_{}", name.name, self.fresh_name());
-                // Add substitution
+                    self.writeln(&format!("const {} = {};", var_name, val));
-                self.var_substitutions
+
-                    .insert(name.name.clone(), var_name.clone());
+                    // Add substitution
                    self.var_substitutions
                        .insert(name.name.clone(), var_name.clone());
                }
                let body_result = self.emit_expr(body)?;
                // Remove substitution
-                self.var_substitutions.remove(&name.name);
+                if name.name != "_" {
                    self.var_substitutions.remove(&name.name);
                }
                Ok(body_result)
            }
@@ -1066,6 +1083,10 @@ impl JsBackend {
                        let arg = self.emit_expr(&args[0])?;
                        return Ok(format!("String({})", arg));
                    }
                    if ident.name == "print" {
                        let arg = self.emit_expr(&args[0])?;
                        return Ok(format!("console.log({})", arg));
                    }
                }
                let arg_strs: Result<Vec<_>, _> = args.iter().map(|a| self.emit_expr(a)).collect();
@@ -1228,10 +1249,15 @@ impl JsBackend {
                        }
                        Statement::Let { name, value, .. } => {
                            let val = self.emit_expr(value)?;
-                            let var_name = format!("{}_{}", name.name, self.fresh_name());
+                            if name.name == "_" {
-                            self.writeln(&format!("const {} = {};", var_name, val));
+                                self.writeln(&format!("{};", val));
-                            self.var_substitutions
+                            } else {
-                                .insert(name.name.clone(), var_name.clone());
+                                let var_name =
                                    format!("{}_{}", name.name, self.fresh_name());
                                self.writeln(&format!("const {} = {};", var_name, val));
                                self.var_substitutions
                                    .insert(name.name.clone(), var_name.clone());
                            }
                        }
                    }
                }
@@ -1240,15 +1266,19 @@ impl JsBackend {
                self.emit_expr(result)
            }
-            Expr::Record { fields, .. } => {
+            Expr::Record {
-                let field_strs: Result<Vec<_>, _> = fields
+                spread, fields, ..
-                    .iter()
+            } => {
-                    .map(|(name, expr)| {
+                let mut parts = Vec::new();
-                        let val = self.emit_expr(expr)?;
+                if let Some(spread_expr) = spread {
-                        Ok(format!("{}: {}", name.name, val))
+                    let spread_code = self.emit_expr(spread_expr)?;
-                    })
+                    parts.push(format!("...{}", spread_code));
-                    .collect();
+                }
-                Ok(format!("{{ {} }}", field_strs?.join(", ")))
+                for (name, expr) in fields {
                    let val = self.emit_expr(expr)?;
                    parts.push(format!("{}: {}", name.name, val));
                }
                Ok(format!("{{ {} }}", parts.join(", ")))
            }
            Expr::Tuple { elements, .. } => {
@@ -1268,6 +1298,11 @@ impl JsBackend {
                Ok(format!("{}.{}", obj, field.name))
            }
            Expr::TupleIndex { object, index, .. } => {
                let obj = self.emit_expr(object)?;
                Ok(format!("{}[{}]", obj, index))
            }
            Expr::Run {
                expr, handlers, ..
            } => {
@@ -2333,7 +2368,7 @@ impl JsBackend {
                }
            }
            Expr::BinaryOp { op, left, right, .. } => {
-                matches!(op, BinaryOp::Add)
+                matches!(op, BinaryOp::Add | BinaryOp::Concat)
                    && (self.is_string_expr(left) || self.is_string_expr(right))
            }
            _ => false,
--- a/src/diagnostics.rs
+++ b/src/diagnostics.rs
@@ -224,10 +224,31 @@ pub mod colors {
    pub const BOLD: &str = "\x1b[1m";
    pub const DIM: &str = "\x1b[2m";
    pub const RED: &str = "\x1b[31m";
    pub const GREEN: &str = "\x1b[32m";
    pub const YELLOW: &str = "\x1b[33m";
    pub const BLUE: &str = "\x1b[34m";
    pub const MAGENTA: &str = "\x1b[35m";
    pub const CYAN: &str = "\x1b[36m";
    pub const WHITE: &str = "\x1b[37m";
    pub const GRAY: &str = "\x1b[90m";
 }
 /// Apply color to text, respecting NO_COLOR / TERM=dumb
 pub fn c(color: &str, text: &str) -> String {
    if supports_color() {
        format!("{}{}{}", color, text, colors::RESET)
    } else {
        text.to_string()
    }
 }
 /// Apply bold + color to text
 pub fn bc(color: &str, text: &str) -> String {
    if supports_color() {
        format!("{}{}{}{}", colors::BOLD, color, text, colors::RESET)
    } else {
        text.to_string()
    }
 }
 /// Severity level for diagnostics
--- a/src/formatter.rs
+++ b/src/formatter.rs
@@ -598,6 +598,9 @@ impl Formatter {
            Expr::Field { object, field, .. } => {
                format!("{}.{}", self.format_expr(object), field.name)
            }
            Expr::TupleIndex { object, index, .. } => {
                format!("{}.{}", self.format_expr(object), index)
            }
            Expr::If { condition, then_branch, else_branch, .. } => {
                format!(
                    "if {} then {} else {}",
@@ -685,15 +688,17 @@ impl Formatter {
                        .join(", ")
                )
            }
-            Expr::Record { fields, .. } => {
+            Expr::Record {
-                format!(
+                spread, fields, ..
-                    "{{ {} }}",
+            } => {
-                    fields
+                let mut parts = Vec::new();
-                        .iter()
+                if let Some(spread_expr) = spread {
-                        .map(|(name, val)| format!("{}: {}", name.name, self.format_expr(val)))
+                    parts.push(format!("...{}", self.format_expr(spread_expr)));
-                        .collect::<Vec<_>>()
+                }
-                        .join(", ")
+                for (name, val) in fields {
-                )
+                    parts.push(format!("{}: {}", name.name, self.format_expr(val)));
                }
                format!("{{ {} }}", parts.join(", "))
            }
            Expr::EffectOp { effect, operation, args, .. } => {
                format!(
@@ -728,7 +733,7 @@ impl Formatter {
        match &lit.kind {
            LiteralKind::Int(n) => n.to_string(),
            LiteralKind::Float(f) => format!("{}", f),
-            LiteralKind::String(s) => format!("\"{}\"", s.replace('\\', "\\\\").replace('"', "\\\"")),
+            LiteralKind::String(s) => format!("\"{}\"", s.replace('\\', "\\\\").replace('"', "\\\"").replace('{', "\\{").replace('}', "\\}")),
            LiteralKind::Char(c) => format!("'{}'", c),
            LiteralKind::Bool(b) => b.to_string(),
            LiteralKind::Unit => "()".to_string(),
@@ -750,6 +755,7 @@ impl Formatter {
            BinaryOp::Ge => ">=",
            BinaryOp::And => "&&",
            BinaryOp::Or => "||",
            BinaryOp::Concat => "++",
            BinaryOp::Pipe => "|>",
        }
    }
--- a/src/interpreter.rs
+++ b/src/interpreter.rs
@@ -74,6 +74,9 @@ pub enum BuiltinFn {
    MathFloor,
    MathCeil,
    MathRound,
    MathSin,
    MathCos,
    MathAtan2,
    // Additional List operations
    ListIsEmpty,
@@ -90,6 +93,14 @@ pub enum BuiltinFn {
    StringToLower,
    StringSubstring,
    StringFromChar,
    StringCharAt,
    StringIndexOf,
    StringLastIndexOf,
    StringRepeat,
    // Int/Float operations
    IntToString,
    FloatToString,
    // JSON operations
    JsonParse,
@@ -620,6 +631,14 @@ pub struct Interpreter {
    pg_connections: RefCell<HashMap<i64, PgClient>>,
    /// Next PostgreSQL connection ID
    next_pg_conn_id: RefCell<i64>,
    /// Concurrent tasks: task_id -> (thunk_value, result_option, is_cancelled)
    concurrent_tasks: RefCell<HashMap<i64, (Value, Option<Value>, bool)>>,
    /// Next task ID
    next_task_id: RefCell<i64>,
    /// Channels: channel_id -> (queue, is_closed)
    channels: RefCell<HashMap<i64, (Vec<Value>, bool)>>,
    /// Next channel ID
    next_channel_id: RefCell<i64>,
 }
 /// Results from running tests
@@ -664,6 +683,10 @@ impl Interpreter {
            next_sql_conn_id: RefCell::new(1),
            pg_connections: RefCell::new(HashMap::new()),
            next_pg_conn_id: RefCell::new(1),
            concurrent_tasks: RefCell::new(HashMap::new()),
            next_task_id: RefCell::new(1),
            channels: RefCell::new(HashMap::new()),
            next_channel_id: RefCell::new(1),
        }
    }
@@ -966,6 +989,22 @@ impl Interpreter {
                "parseFloat".to_string(),
                Value::Builtin(BuiltinFn::StringParseFloat),
            ),
            (
                "charAt".to_string(),
                Value::Builtin(BuiltinFn::StringCharAt),
            ),
            (
                "indexOf".to_string(),
                Value::Builtin(BuiltinFn::StringIndexOf),
            ),
            (
                "lastIndexOf".to_string(),
                Value::Builtin(BuiltinFn::StringLastIndexOf),
            ),
            (
                "repeat".to_string(),
                Value::Builtin(BuiltinFn::StringRepeat),
            ),
        ]));
        env.define("String", string_module);
@@ -1036,9 +1075,24 @@ impl Interpreter {
            ("floor".to_string(), Value::Builtin(BuiltinFn::MathFloor)),
            ("ceil".to_string(), Value::Builtin(BuiltinFn::MathCeil)),
            ("round".to_string(), Value::Builtin(BuiltinFn::MathRound)),
            ("sin".to_string(), Value::Builtin(BuiltinFn::MathSin)),
            ("cos".to_string(), Value::Builtin(BuiltinFn::MathCos)),
            ("atan2".to_string(), Value::Builtin(BuiltinFn::MathAtan2)),
        ]));
        env.define("Math", math_module);
        // Int module
        let int_module = Value::Record(HashMap::from([
            ("toString".to_string(), Value::Builtin(BuiltinFn::IntToString)),
        ]));
        env.define("Int", int_module);
        // Float module
        let float_module = Value::Record(HashMap::from([
            ("toString".to_string(), Value::Builtin(BuiltinFn::FloatToString)),
        ]));
        env.define("Float", float_module);
        // JSON module
        let json_module = Value::Record(HashMap::from([
            ("parse".to_string(), Value::Builtin(BuiltinFn::JsonParse)),
@@ -1067,11 +1121,50 @@ impl Interpreter {
    /// Execute a program
    pub fn run(&mut self, program: &Program) -> Result<Value, RuntimeError> {
        let mut last_value = Value::Unit;
        let mut has_main_let = false;
        for decl in &program.declarations {
            // Track if there's a top-level `let main = ...`
            if let Declaration::Let(let_decl) = decl {
                if let_decl.name.name == "main" {
                    has_main_let = true;
                }
            }
            last_value = self.eval_declaration(decl)?;
        }
        // Auto-invoke main if it was defined as a let binding with a function value
        if has_main_let {
            if let Some(main_val) = self.global_env.get("main") {
                if let Value::Function(ref closure) = main_val {
                    if closure.params.is_empty() {
                        let span = Span { start: 0, end: 0 };
                        let mut result = self.eval_call(main_val.clone(), vec![], span)?;
                        // Trampoline loop
                        loop {
                            match result {
                                EvalResult::Value(v) => {
                                    last_value = v;
                                    break;
                                }
                                EvalResult::Effect(req) => {
                                    last_value = self.handle_effect(req)?;
                                    break;
                                }
                                EvalResult::TailCall { func, args, span } => {
                                    result = self.eval_call(func, args, span)?;
                                }
                                EvalResult::Resume(v) => {
                                    last_value = v;
                                    break;
                                }
                            }
                        }
                    }
                }
            }
        }
        Ok(last_value)
    }
@@ -1383,6 +1476,34 @@ impl Interpreter {
                }
            }
            Expr::TupleIndex {
                object,
                index,
                span,
            } => {
                let obj_val = self.eval_expr(object, env)?;
                match obj_val {
                    Value::Tuple(elements) => {
                        if *index < elements.len() {
                            Ok(EvalResult::Value(elements[*index].clone()))
                        } else {
                            Err(RuntimeError {
                                message: format!(
                                    "Tuple index {} out of bounds for tuple with {} elements",
                                    index,
                                    elements.len()
                                ),
                                span: Some(*span),
                            })
                        }
                    }
                    _ => Err(RuntimeError {
                        message: format!("Cannot use tuple index on {}", obj_val.type_name()),
                        span: Some(*span),
                    }),
                }
            }
            Expr::Lambda { params, body, .. } => {
                let closure = Closure {
                    params: params.iter().map(|p| p.name.name.clone()).collect(),
@@ -1449,8 +1570,28 @@ impl Interpreter {
                self.eval_expr_tail(result, &block_env, tail)
            }
-            Expr::Record { fields, .. } => {
+            Expr::Record {
                spread, fields, ..
            } => {
                let mut record = HashMap::new();
                // If there's a spread, evaluate it and start with its fields
                if let Some(spread_expr) = spread {
                    let spread_val = self.eval_expr(spread_expr, env)?;
                    if let Value::Record(spread_fields) = spread_val {
                        record = spread_fields;
                    } else {
                        return Err(RuntimeError {
                            message: format!(
                                "Spread expression must evaluate to a record, got {}",
                                spread_val.type_name()
                            ),
                            span: Some(expr.span()),
                        });
                    }
                }
                // Override with explicit fields
                for (name, expr) in fields {
                    let val = self.eval_expr(expr, env)?;
                    record.insert(name.name.clone(), val);
@@ -1523,6 +1664,18 @@ impl Interpreter {
                    span: Some(span),
                }),
            },
            BinaryOp::Concat => match (left, right) {
                (Value::String(a), Value::String(b)) => Ok(Value::String(a + &b)),
                (Value::List(a), Value::List(b)) => {
                    let mut result = a;
                    result.extend(b);
                    Ok(Value::List(result))
                }
                (l, r) => Err(RuntimeError {
                    message: format!("Cannot concatenate {} and {}", l.type_name(), r.type_name()),
                    span: Some(span),
                }),
            },
            BinaryOp::Sub => match (left, right) {
                (Value::Int(a), Value::Int(b)) => Ok(Value::Int(a - b)),
                (Value::Float(a), Value::Float(b)) => Ok(Value::Float(a - b)),
@@ -1578,6 +1731,7 @@ impl Interpreter {
                (Value::Int(a), Value::Int(b)) => Ok(Value::Bool(a < b)),
                (Value::Float(a), Value::Float(b)) => Ok(Value::Bool(a < b)),
                (Value::String(a), Value::String(b)) => Ok(Value::Bool(a < b)),
                (Value::Char(a), Value::Char(b)) => Ok(Value::Bool(a < b)),
                (l, r) => Err(RuntimeError {
                    message: format!("Cannot compare {} and {}", l.type_name(), r.type_name()),
                    span: Some(span),
@@ -1587,6 +1741,7 @@ impl Interpreter {
                (Value::Int(a), Value::Int(b)) => Ok(Value::Bool(a <= b)),
                (Value::Float(a), Value::Float(b)) => Ok(Value::Bool(a <= b)),
                (Value::String(a), Value::String(b)) => Ok(Value::Bool(a <= b)),
                (Value::Char(a), Value::Char(b)) => Ok(Value::Bool(a <= b)),
                (l, r) => Err(RuntimeError {
                    message: format!("Cannot compare {} and {}", l.type_name(), r.type_name()),
                    span: Some(span),
@@ -1596,6 +1751,7 @@ impl Interpreter {
                (Value::Int(a), Value::Int(b)) => Ok(Value::Bool(a > b)),
                (Value::Float(a), Value::Float(b)) => Ok(Value::Bool(a > b)),
                (Value::String(a), Value::String(b)) => Ok(Value::Bool(a > b)),
                (Value::Char(a), Value::Char(b)) => Ok(Value::Bool(a > b)),
                (l, r) => Err(RuntimeError {
                    message: format!("Cannot compare {} and {}", l.type_name(), r.type_name()),
                    span: Some(span),
@@ -1605,6 +1761,7 @@ impl Interpreter {
                (Value::Int(a), Value::Int(b)) => Ok(Value::Bool(a >= b)),
                (Value::Float(a), Value::Float(b)) => Ok(Value::Bool(a >= b)),
                (Value::String(a), Value::String(b)) => Ok(Value::Bool(a >= b)),
                (Value::Char(a), Value::Char(b)) => Ok(Value::Bool(a >= b)),
                (l, r) => Err(RuntimeError {
                    message: format!("Cannot compare {} and {}", l.type_name(), r.type_name()),
                    span: Some(span),
@@ -2187,6 +2344,26 @@ impl Interpreter {
                Ok(EvalResult::Value(Value::String(result)))
            }
            BuiltinFn::IntToString => {
                if args.len() != 1 {
                    return Err(err("Int.toString requires 1 argument"));
                }
                match &args[0] {
                    Value::Int(n) => Ok(EvalResult::Value(Value::String(format!("{}", n)))),
                    v => Ok(EvalResult::Value(Value::String(format!("{}", v)))),
                }
            }
            BuiltinFn::FloatToString => {
                if args.len() != 1 {
                    return Err(err("Float.toString requires 1 argument"));
                }
                match &args[0] {
                    Value::Float(f) => Ok(EvalResult::Value(Value::String(format!("{}", f)))),
                    v => Ok(EvalResult::Value(Value::String(format!("{}", v)))),
                }
            }
            BuiltinFn::TypeOf => {
                if args.len() != 1 {
                    return Err(err("typeOf requires 1 argument"));
@@ -2363,6 +2540,45 @@ impl Interpreter {
                }
            }
            BuiltinFn::MathSin => {
                if args.len() != 1 {
                    return Err(err("Math.sin requires 1 argument"));
                }
                match &args[0] {
                    Value::Float(n) => Ok(EvalResult::Value(Value::Float(n.sin()))),
                    Value::Int(n) => Ok(EvalResult::Value(Value::Float((*n as f64).sin()))),
                    v => Err(err(&format!("Math.sin expects number, got {}", v.type_name()))),
                }
            }
            BuiltinFn::MathCos => {
                if args.len() != 1 {
                    return Err(err("Math.cos requires 1 argument"));
                }
                match &args[0] {
                    Value::Float(n) => Ok(EvalResult::Value(Value::Float(n.cos()))),
                    Value::Int(n) => Ok(EvalResult::Value(Value::Float((*n as f64).cos()))),
                    v => Err(err(&format!("Math.cos expects number, got {}", v.type_name()))),
                }
            }
            BuiltinFn::MathAtan2 => {
                if args.len() != 2 {
                    return Err(err("Math.atan2 requires 2 arguments: y, x"));
                }
                let y = match &args[0] {
                    Value::Float(n) => *n,
                    Value::Int(n) => *n as f64,
                    v => return Err(err(&format!("Math.atan2 expects number, got {}", v.type_name()))),
                };
                let x = match &args[1] {
                    Value::Float(n) => *n,
                    Value::Int(n) => *n as f64,
                    v => return Err(err(&format!("Math.atan2 expects number, got {}", v.type_name()))),
                };
                Ok(EvalResult::Value(Value::Float(y.atan2(x))))
            }
            // Additional List operations
            BuiltinFn::ListIsEmpty => {
                let list = Self::expect_arg_1::<Vec<Value>>(&args, "List.isEmpty", span)?;
@@ -2498,6 +2714,89 @@ impl Interpreter {
                Ok(EvalResult::Value(Value::String(c.to_string())))
            }
            BuiltinFn::StringCharAt => {
                if args.len() != 2 {
                    return Err(err("String.charAt requires 2 arguments: string, index"));
                }
                let s = match &args[0] {
                    Value::String(s) => s.clone(),
                    v => return Err(err(&format!("String.charAt expects String, got {}", v.type_name()))),
                };
                let idx = match &args[1] {
                    Value::Int(n) => *n as usize,
                    v => return Err(err(&format!("String.charAt expects Int for index, got {}", v.type_name()))),
                };
                let chars: Vec<char> = s.chars().collect();
                if idx < chars.len() {
                    Ok(EvalResult::Value(Value::String(chars[idx].to_string())))
                } else {
                    Ok(EvalResult::Value(Value::String(String::new())))
                }
            }
            BuiltinFn::StringIndexOf => {
                if args.len() != 2 {
                    return Err(err("String.indexOf requires 2 arguments: string, substring"));
                }
                let s = match &args[0] {
                    Value::String(s) => s.clone(),
                    v => return Err(err(&format!("String.indexOf expects String, got {}", v.type_name()))),
                };
                let sub = match &args[1] {
                    Value::String(s) => s.clone(),
                    v => return Err(err(&format!("String.indexOf expects String for substring, got {}", v.type_name()))),
                };
                match s.find(&sub) {
                    Some(idx) => Ok(EvalResult::Value(Value::Constructor {
                        name: "Some".to_string(),
                        fields: vec![Value::Int(idx as i64)],
                    })),
                    None => Ok(EvalResult::Value(Value::Constructor {
                        name: "None".to_string(),
                        fields: vec![],
                    })),
                }
            }
            BuiltinFn::StringLastIndexOf => {
                if args.len() != 2 {
                    return Err(err("String.lastIndexOf requires 2 arguments: string, substring"));
                }
                let s = match &args[0] {
                    Value::String(s) => s.clone(),
                    v => return Err(err(&format!("String.lastIndexOf expects String, got {}", v.type_name()))),
                };
                let sub = match &args[1] {
                    Value::String(s) => s.clone(),
                    v => return Err(err(&format!("String.lastIndexOf expects String for substring, got {}", v.type_name()))),
                };
                match s.rfind(&sub) {
                    Some(idx) => Ok(EvalResult::Value(Value::Constructor {
                        name: "Some".to_string(),
                        fields: vec![Value::Int(idx as i64)],
                    })),
                    None => Ok(EvalResult::Value(Value::Constructor {
                        name: "None".to_string(),
                        fields: vec![],
                    })),
                }
            }
            BuiltinFn::StringRepeat => {
                if args.len() != 2 {
                    return Err(err("String.repeat requires 2 arguments: string, count"));
                }
                let s = match &args[0] {
                    Value::String(s) => s.clone(),
                    v => return Err(err(&format!("String.repeat expects String, got {}", v.type_name()))),
                };
                let count = match &args[1] {
                    Value::Int(n) => (*n).max(0) as usize,
                    v => return Err(err(&format!("String.repeat expects Int for count, got {}", v.type_name()))),
                };
                Ok(EvalResult::Value(Value::String(s.repeat(count))))
            }
            // JSON operations
            BuiltinFn::JsonParse => {
                let s = Self::expect_arg_1::<String>(&args, "Json.parse", span)?;
@@ -3709,6 +4008,26 @@ impl Interpreter {
                }
                Ok(Value::Unit)
            }
            ("Test", "assertEqualMsg") => {
                let expected = request.args.first().cloned().unwrap_or(Value::Unit);
                let actual = request.args.get(1).cloned().unwrap_or(Value::Unit);
                let label = match request.args.get(2) {
                    Some(Value::String(s)) => s.clone(),
                    _ => "Values not equal".to_string(),
                };
                if Value::values_equal(&expected, &actual) {
                    self.test_results.borrow_mut().passed += 1;
                } else {
                    self.test_results.borrow_mut().failed += 1;
                    self.test_results.borrow_mut().failures.push(TestFailure {
                        message: label,
                        expected: Some(format!("{}", expected)),
                        actual: Some(format!("{}", actual)),
                    });
                }
                Ok(Value::Unit)
            }
            ("Test", "assertNotEqual") => {
                let a = request.args.first().cloned().unwrap_or(Value::Unit);
                let b = request.args.get(1).cloned().unwrap_or(Value::Unit);
@@ -4369,6 +4688,237 @@ impl Interpreter {
                }
            }
            // ===== Concurrent Effect =====
            ("Concurrent", "spawn") => {
                // For now, spawn just stores the thunk - it will be evaluated on await
                // In a real implementation, this would start a thread/fiber
                let thunk = match request.args.first() {
                    Some(v) => v.clone(),
                    _ => return Err(RuntimeError {
                        message: "Concurrent.spawn requires a thunk argument".to_string(),
                        span: None,
                    }),
                };
                let task_id = *self.next_task_id.borrow();
                *self.next_task_id.borrow_mut() += 1;
                // Store task: (thunk, None for result, not cancelled)
                self.concurrent_tasks.borrow_mut().insert(task_id, (thunk, None, false));
                Ok(Value::Int(task_id))
            }
            ("Concurrent", "await") => {
                let task_id = match request.args.first() {
                    Some(Value::Int(id)) => *id,
                    _ => return Err(RuntimeError {
                        message: "Concurrent.await requires a task ID".to_string(),
                        span: None,
                    }),
                };
                // Check if already computed or cancelled
                let task_info = {
                    let tasks = self.concurrent_tasks.borrow();
                    tasks.get(&task_id).cloned()
                };
                match task_info {
                    Some((_, Some(result), _)) => Ok(result),
                    Some((_, _, true)) => Err(RuntimeError {
                        message: format!("Task {} was cancelled", task_id),
                        span: None,
                    }),
                    Some((thunk, None, false)) => {
                        // For cooperative concurrency, we just need to signal
                        // that we're waiting on this task
                        // Return the thunk to be evaluated by the caller
                        // This is a simplification - real async would use fibers
                        Ok(thunk)
                    }
                    None => Err(RuntimeError {
                        message: format!("Unknown task ID: {}", task_id),
                        span: None,
                    }),
                }
            }
            ("Concurrent", "yield") => {
                // In cooperative concurrency, yield allows other tasks to run
                // For now, this is a no-op in our single-threaded model
                Ok(Value::Unit)
            }
            ("Concurrent", "sleep") => {
                // Non-blocking sleep (delegates to thread sleep for now)
                use std::thread;
                use std::time::Duration;
                let ms = match request.args.first() {
                    Some(Value::Int(n)) => *n as u64,
                    _ => 0,
                };
                thread::sleep(Duration::from_millis(ms));
                Ok(Value::Unit)
            }
            ("Concurrent", "cancel") => {
                let task_id = match request.args.first() {
                    Some(Value::Int(id)) => *id,
                    _ => return Err(RuntimeError {
                        message: "Concurrent.cancel requires a task ID".to_string(),
                        span: None,
                    }),
                };
                let mut tasks = self.concurrent_tasks.borrow_mut();
                if let Some((thunk, result, _)) = tasks.get(&task_id).cloned() {
                    tasks.insert(task_id, (thunk, result, true));
                    Ok(Value::Bool(true))
                } else {
                    Ok(Value::Bool(false))
                }
            }
            ("Concurrent", "isRunning") => {
                let task_id = match request.args.first() {
                    Some(Value::Int(id)) => *id,
                    _ => return Err(RuntimeError {
                        message: "Concurrent.isRunning requires a task ID".to_string(),
                        span: None,
                    }),
                };
                let tasks = self.concurrent_tasks.borrow();
                let is_running = match tasks.get(&task_id) {
                    Some((_, None, false)) => true,  // Not completed and not cancelled
                    _ => false,
                };
                Ok(Value::Bool(is_running))
            }
            ("Concurrent", "taskCount") => {
                let tasks = self.concurrent_tasks.borrow();
                let count = tasks.iter()
                    .filter(|(_, (_, result, cancelled))| result.is_none() && !cancelled)
                    .count();
                Ok(Value::Int(count as i64))
            }
            // ===== Channel Effect =====
            ("Channel", "create") => {
                let channel_id = *self.next_channel_id.borrow();
                *self.next_channel_id.borrow_mut() += 1;
                // Create empty channel queue, not closed
                self.channels.borrow_mut().insert(channel_id, (Vec::new(), false));
                Ok(Value::Int(channel_id))
            }
            ("Channel", "send") => {
                let channel_id = match request.args.first() {
                    Some(Value::Int(id)) => *id,
                    _ => return Err(RuntimeError {
                        message: "Channel.send requires a channel ID".to_string(),
                        span: None,
                    }),
                };
                let value = match request.args.get(1) {
                    Some(v) => v.clone(),
                    _ => return Err(RuntimeError {
                        message: "Channel.send requires a value".to_string(),
                        span: None,
                    }),
                };
                let mut channels = self.channels.borrow_mut();
                match channels.get_mut(&channel_id) {
                    Some((queue, false)) => {
                        queue.push(value);
                        Ok(Value::Unit)
                    }
                    Some((_, true)) => Err(RuntimeError {
                        message: format!("Channel {} is closed", channel_id),
                        span: None,
                    }),
                    None => Err(RuntimeError {
                        message: format!("Unknown channel ID: {}", channel_id),
                        span: None,
                    }),
                }
            }
            ("Channel", "receive") => {
                let channel_id = match request.args.first() {
                    Some(Value::Int(id)) => *id,
                    _ => return Err(RuntimeError {
                        message: "Channel.receive requires a channel ID".to_string(),
                        span: None,
                    }),
                };
                let mut channels = self.channels.borrow_mut();
                match channels.get_mut(&channel_id) {
                    Some((queue, _)) if !queue.is_empty() => {
                        Ok(queue.remove(0))
                    }
                    Some((_, true)) => Err(RuntimeError {
                        message: format!("Channel {} is closed and empty", channel_id),
                        span: None,
                    }),
                    Some((_, false)) => Err(RuntimeError {
                        message: format!("Channel {} is empty (blocking receive not supported yet)", channel_id),
                        span: None,
                    }),
                    None => Err(RuntimeError {
                        message: format!("Unknown channel ID: {}", channel_id),
                        span: None,
                    }),
                }
            }
            ("Channel", "tryReceive") => {
                let channel_id = match request.args.first() {
                    Some(Value::Int(id)) => *id,
                    _ => return Err(RuntimeError {
                        message: "Channel.tryReceive requires a channel ID".to_string(),
                        span: None,
                    }),
                };
                let mut channels = self.channels.borrow_mut();
                match channels.get_mut(&channel_id) {
                    Some((queue, _)) if !queue.is_empty() => {
                        Ok(Value::Constructor {
                            name: "Some".to_string(),
                            fields: vec![queue.remove(0)],
                        })
                    }
                    Some(_) => {
                        Ok(Value::Constructor {
                            name: "None".to_string(),
                            fields: vec![],
                        })
                    }
                    None => Err(RuntimeError {
                        message: format!("Unknown channel ID: {}", channel_id),
                        span: None,
                    }),
                }
            }
            ("Channel", "close") => {
                let channel_id = match request.args.first() {
                    Some(Value::Int(id)) => *id,
                    _ => return Err(RuntimeError {
                        message: "Channel.close requires a channel ID".to_string(),
                        span: None,
                    }),
                };
                let mut channels = self.channels.borrow_mut();
                if let Some((queue, closed)) = channels.get_mut(&channel_id) {
                    *closed = true;
                    Ok(Value::Unit)
                } else {
                    Err(RuntimeError {
                        message: format!("Unknown channel ID: {}", channel_id),
                        span: None,
                    })
                }
            }
            _ => Err(RuntimeError {
                message: format!(
                    "Unhandled effect operation: {}.{}",
@@ -4610,6 +5160,7 @@ mod tests {
        // Create a simple migration that adds a field
        // Migration: old.name -> { name: old.name, email: "unknown" }
        let migration_body = Expr::Record {
            spread: None,
            fields: vec![
                (
                    Ident::new("name", Span::default()),
--- a/src/lexer.rs
+++ b/src/lexer.rs
@@ -70,6 +70,7 @@ pub enum TokenKind {
    // Operators
    Plus,       // +
    PlusPlus,   // ++
    Minus,      // -
    Star,       // *
    Slash,      // /
@@ -89,6 +90,7 @@ pub enum TokenKind {
    Arrow,      // =>
    ThinArrow,  // ->
    Dot,        // .
    DotDotDot,  // ...
    Colon,      // :
    ColonColon, // ::
    Comma,      // ,
@@ -160,6 +162,7 @@ impl fmt::Display for TokenKind {
            TokenKind::True => write!(f, "true"),
            TokenKind::False => write!(f, "false"),
            TokenKind::Plus => write!(f, "+"),
            TokenKind::PlusPlus => write!(f, "++"),
            TokenKind::Minus => write!(f, "-"),
            TokenKind::Star => write!(f, "*"),
            TokenKind::Slash => write!(f, "/"),
@@ -179,6 +182,7 @@ impl fmt::Display for TokenKind {
            TokenKind::Arrow => write!(f, "=>"),
            TokenKind::ThinArrow => write!(f, "->"),
            TokenKind::Dot => write!(f, "."),
            TokenKind::DotDotDot => write!(f, "..."),
            TokenKind::Colon => write!(f, ":"),
            TokenKind::ColonColon => write!(f, "::"),
            TokenKind::Comma => write!(f, ","),
@@ -268,7 +272,14 @@ impl<'a> Lexer<'a> {
        let kind = match c {
            // Single-character tokens
-            '+' => TokenKind::Plus,
+            '+' => {
                if self.peek() == Some('+') {
                    self.advance();
                    TokenKind::PlusPlus
                } else {
                    TokenKind::Plus
                }
            }
            '*' => TokenKind::Star,
            '%' => TokenKind::Percent,
            '(' => TokenKind::LParen,
@@ -364,7 +375,22 @@ impl<'a> Lexer<'a> {
                    TokenKind::Pipe
                }
            }
-            '.' => TokenKind::Dot,
+            '.' => {
                if self.peek() == Some('.') {
                    // Check for ... (need to peek past second dot)
                    // We look at source directly since we can only peek one ahead
                    let next_next = self.source[self.pos..].chars().nth(1);
                    if next_next == Some('.') {
                        self.advance(); // consume second '.'
                        self.advance(); // consume third '.'
                        TokenKind::DotDotDot
                    } else {
                        TokenKind::Dot
                    }
                } else {
                    TokenKind::Dot
                }
            }
            ':' => {
                if self.peek() == Some(':') {
                    self.advance();
@@ -493,6 +519,8 @@ impl<'a> Lexer<'a> {
                                Some('"') => '"',
                                Some('0') => '\0',
                                Some('\'') => '\'',
                                Some('{') => '{',
                                Some('}') => '}',
                                Some('x') => {
                                    // Hex escape \xNN
                                    let h1 = self.advance().and_then(|c| c.to_digit(16));
--- a/src/linter.rs
+++ b/src/linter.rs
--- a/src/lsp.rs
+++ b/src/lsp.rs
--- a/src/main.rs
+++ b/src/main.rs
--- a/src/modules.rs
+++ b/src/modules.rs
@@ -305,6 +305,11 @@ impl ModuleLoader {
        self.cache.iter()
    }
    /// Get the module cache (for passing to C backend)
    pub fn module_cache(&self) -> &HashMap<String, Module> {
        &self.cache
    }
    /// Clear the module cache
    pub fn clear_cache(&mut self) {
        self.cache.clear();
--- a/src/package.rs
+++ b/src/package.rs
@@ -6,6 +6,618 @@ use std::collections::HashMap;
 use std::fs;
 use std::path::{Path, PathBuf};
 use std::io::{self, Write};
 use std::cmp::Ordering;
 // =============================================================================
 // Semantic Versioning
 // =============================================================================
 /// A semantic version (major.minor.patch)
 #[derive(Debug, Clone, PartialEq, Eq, Hash)]
 pub struct Version {
    pub major: u32,
    pub minor: u32,
    pub patch: u32,
    pub prerelease: Option<String>,
 }
 impl Version {
    pub fn new(major: u32, minor: u32, patch: u32) -> Self {
        Self { major, minor, patch, prerelease: None }
    }
    pub fn parse(s: &str) -> Result<Self, String> {
        let s = s.trim();
        // Handle prerelease suffix (e.g., "1.0.0-alpha")
        let (version_part, prerelease) = if let Some(pos) = s.find('-') {
            (&s[..pos], Some(s[pos + 1..].to_string()))
        } else {
            (s, None)
        };
        let parts: Vec<&str> = version_part.split('.').collect();
        if parts.len() < 2 || parts.len() > 3 {
            return Err(format!("Invalid version format: {}", s));
        }
        let major = parts[0].parse::<u32>()
            .map_err(|_| format!("Invalid major version: {}", parts[0]))?;
        let minor = parts[1].parse::<u32>()
            .map_err(|_| format!("Invalid minor version: {}", parts[1]))?;
        let patch = if parts.len() > 2 {
            parts[2].parse::<u32>()
                .map_err(|_| format!("Invalid patch version: {}", parts[2]))?
        } else {
            0
        };
        Ok(Self { major, minor, patch, prerelease })
    }
 }
 impl std::fmt::Display for Version {
    fn fmt(&self, f: &mut std::fmt::Formatter<'_>) -> std::fmt::Result {
        if let Some(ref pre) = self.prerelease {
            write!(f, "{}.{}.{}-{}", self.major, self.minor, self.patch, pre)
        } else {
            write!(f, "{}.{}.{}", self.major, self.minor, self.patch)
        }
    }
 }
 impl Ord for Version {
    fn cmp(&self, other: &Self) -> Ordering {
        match self.major.cmp(&other.major) {
            Ordering::Equal => {}
            ord => return ord,
        }
        match self.minor.cmp(&other.minor) {
            Ordering::Equal => {}
            ord => return ord,
        }
        match self.patch.cmp(&other.patch) {
            Ordering::Equal => {}
            ord => return ord,
        }
        // Prerelease versions are less than release versions
        match (&self.prerelease, &other.prerelease) {
            (None, None) => Ordering::Equal,
            (Some(_), None) => Ordering::Less,
            (None, Some(_)) => Ordering::Greater,
            (Some(a), Some(b)) => a.cmp(b),
        }
    }
 }
 impl PartialOrd for Version {
    fn partial_cmp(&self, other: &Self) -> Option<Ordering> {
        Some(self.cmp(other))
    }
 }
 /// Version constraint for dependencies
 #[derive(Debug, Clone)]
 pub enum VersionConstraint {
    /// Exact version: "1.2.3"
    Exact(Version),
    /// Caret: "^1.2.3" - compatible updates (>=1.2.3, <2.0.0)
    Caret(Version),
    /// Tilde: "~1.2.3" - patch updates only (>=1.2.3, <1.3.0)
    Tilde(Version),
    /// Greater than or equal: ">=1.2.3"
    GreaterEq(Version),
    /// Less than: "<2.0.0"
    Less(Version),
    /// Range: ">=1.0.0, <2.0.0"
    Range { min: Version, max: Version },
    /// Any version: "*"
    Any,
 }
 impl VersionConstraint {
    pub fn parse(s: &str) -> Result<Self, String> {
        let s = s.trim();
        if s == "*" {
            return Ok(VersionConstraint::Any);
        }
        // Check for range (comma-separated constraints)
        if s.contains(',') {
            let parts: Vec<&str> = s.split(',').collect();
            if parts.len() != 2 {
                return Err("Range must have exactly two constraints".to_string());
            }
            let first = VersionConstraint::parse(parts[0].trim())?;
            let second = VersionConstraint::parse(parts[1].trim())?;
            match (first, second) {
                (VersionConstraint::GreaterEq(min), VersionConstraint::Less(max)) => {
                    Ok(VersionConstraint::Range { min, max })
                }
                _ => Err("Range must be >=version, <version".to_string())
            }
        } else if let Some(rest) = s.strip_prefix('^') {
            Ok(VersionConstraint::Caret(Version::parse(rest)?))
        } else if let Some(rest) = s.strip_prefix('~') {
            Ok(VersionConstraint::Tilde(Version::parse(rest)?))
        } else if let Some(rest) = s.strip_prefix(">=") {
            Ok(VersionConstraint::GreaterEq(Version::parse(rest)?))
        } else if let Some(rest) = s.strip_prefix('<') {
            Ok(VersionConstraint::Less(Version::parse(rest)?))
        } else {
            // Try to parse as exact version
            Ok(VersionConstraint::Exact(Version::parse(s)?))
        }
    }
    /// Check if a version satisfies this constraint
    pub fn satisfies(&self, version: &Version) -> bool {
        match self {
            VersionConstraint::Exact(v) => version == v,
            VersionConstraint::Caret(v) => {
                // ^1.2.3 means >=1.2.3, <2.0.0 (if major > 0)
                // ^0.2.3 means >=0.2.3, <0.3.0 (if major == 0)
                if v.major == 0 {
                    version.major == 0 && version.minor == v.minor && version >= v
                } else {
                    version.major == v.major && version >= v
                }
            }
            VersionConstraint::Tilde(v) => {
                // ~1.2.3 means >=1.2.3, <1.3.0
                version.major == v.major && version.minor == v.minor && version >= v
            }
            VersionConstraint::GreaterEq(v) => version >= v,
            VersionConstraint::Less(v) => version < v,
            VersionConstraint::Range { min, max } => version >= min && version < max,
            VersionConstraint::Any => true,
        }
    }
 }
 impl std::fmt::Display for VersionConstraint {
    fn fmt(&self, f: &mut std::fmt::Formatter<'_>) -> std::fmt::Result {
        match self {
            VersionConstraint::Exact(v) => write!(f, "{}", v),
            VersionConstraint::Caret(v) => write!(f, "^{}", v),
            VersionConstraint::Tilde(v) => write!(f, "~{}", v),
            VersionConstraint::GreaterEq(v) => write!(f, ">={}", v),
            VersionConstraint::Less(v) => write!(f, "<{}", v),
            VersionConstraint::Range { min, max } => write!(f, ">={}, <{}", min, max),
            VersionConstraint::Any => write!(f, "*"),
        }
    }
 }
 // =============================================================================
 // Lock File
 // =============================================================================
 /// A lock file entry for a resolved package
 #[derive(Debug, Clone)]
 pub struct LockedPackage {
    pub name: String,
    pub version: Version,
    pub source: LockedSource,
    pub checksum: Option<String>,
    pub dependencies: Vec<String>,
 }
 /// Source of a locked package
 #[derive(Debug, Clone)]
 pub enum LockedSource {
    Registry,
    Git { url: String, rev: String },
    Path { path: PathBuf },
 }
 /// The lock file (lux.lock)
 #[derive(Debug, Clone, Default)]
 pub struct LockFile {
    pub packages: Vec<LockedPackage>,
 }
 impl LockFile {
    pub fn new() -> Self {
        Self { packages: Vec::new() }
    }
    /// Parse a lock file
    pub fn parse(content: &str) -> Result<Self, String> {
        let mut packages = Vec::new();
        let mut current_pkg: Option<LockedPackage> = None;
        let mut in_package = false;
        for line in content.lines() {
            let line = line.trim();
            if line.is_empty() || line.starts_with('#') {
                continue;
            }
            if line == "[[package]]" {
                if let Some(pkg) = current_pkg.take() {
                    packages.push(pkg);
                }
                current_pkg = Some(LockedPackage {
                    name: String::new(),
                    version: Version::new(0, 0, 0),
                    source: LockedSource::Registry,
                    checksum: None,
                    dependencies: Vec::new(),
                });
                in_package = true;
                continue;
            }
            if in_package {
                if let Some(ref mut pkg) = current_pkg {
                    if let Some(eq_pos) = line.find('=') {
                        let key = line[..eq_pos].trim();
                        let value = line[eq_pos + 1..].trim().trim_matches('"');
                        match key {
                            "name" => pkg.name = value.to_string(),
                            "version" => pkg.version = Version::parse(value)?,
                            "source" => {
                                if value == "registry" {
                                    pkg.source = LockedSource::Registry;
                                } else if value.starts_with("git:") {
                                    let parts: Vec<&str> = value[4..].splitn(2, '@').collect();
                                    pkg.source = LockedSource::Git {
                                        url: parts[0].to_string(),
                                        rev: parts.get(1).unwrap_or(&"HEAD").to_string(),
                                    };
                                } else if value.starts_with("path:") {
                                    pkg.source = LockedSource::Path {
                                        path: PathBuf::from(&value[5..]),
                                    };
                                }
                            }
                            "checksum" => pkg.checksum = Some(value.to_string()),
                            "dependencies" => {
                                // Parse array
                                let deps_str = value.trim_matches(|c| c == '[' || c == ']');
                                pkg.dependencies = deps_str
                                    .split(',')
                                    .map(|s| s.trim().trim_matches('"').to_string())
                                    .filter(|s| !s.is_empty())
                                    .collect();
                            }
                            _ => {}
                        }
                    }
                }
            }
        }
        if let Some(pkg) = current_pkg {
            packages.push(pkg);
        }
        Ok(Self { packages })
    }
    /// Format lock file as TOML
    pub fn format(&self) -> String {
        let mut output = String::new();
        output.push_str("# This file is auto-generated by lux pkg. Do not edit manually.\n\n");
        for pkg in &self.packages {
            output.push_str("[[package]]\n");
            output.push_str(&format!("name = \"{}\"\n", pkg.name));
            output.push_str(&format!("version = \"{}\"\n", pkg.version));
            let source_str = match &pkg.source {
                LockedSource::Registry => "registry".to_string(),
                LockedSource::Git { url, rev } => format!("git:{}@{}", url, rev),
                LockedSource::Path { path } => format!("path:{}", path.display()),
            };
            output.push_str(&format!("source = \"{}\"\n", source_str));
            if let Some(ref checksum) = pkg.checksum {
                output.push_str(&format!("checksum = \"{}\"\n", checksum));
            }
            if !pkg.dependencies.is_empty() {
                let deps: Vec<String> = pkg.dependencies.iter()
                    .map(|d| format!("\"{}\"", d))
                    .collect();
                output.push_str(&format!("dependencies = [{}]\n", deps.join(", ")));
            }
            output.push('\n');
        }
        output
    }
    /// Find a locked package by name
    pub fn find(&self, name: &str) -> Option<&LockedPackage> {
        self.packages.iter().find(|p| p.name == name)
    }
 }
 // =============================================================================
 // Dependency Resolution
 // =============================================================================
 /// Resolution error
 #[derive(Debug)]
 pub struct ResolutionError {
    pub message: String,
    pub package: Option<String>,
 }
 impl std::fmt::Display for ResolutionError {
    fn fmt(&self, f: &mut std::fmt::Formatter<'_>) -> std::fmt::Result {
        if let Some(ref pkg) = self.package {
            write!(f, "Resolution error for '{}': {}", pkg, self.message)
        } else {
            write!(f, "Resolution error: {}", self.message)
        }
    }
 }
 /// Dependency resolver with transitive dependency support
 pub struct Resolver {
    /// Available versions for each package (simulated for now)
    available_versions: HashMap<String, Vec<Version>>,
    /// Package dependencies cache (package@version -> dependencies)
    package_deps: HashMap<String, HashMap<String, Dependency>>,
    /// Packages directory for reading transitive deps
    packages_dir: Option<PathBuf>,
 }
 impl Resolver {
    pub fn new() -> Self {
        Self {
            available_versions: HashMap::new(),
            package_deps: HashMap::new(),
            packages_dir: None,
        }
    }
    /// Create resolver with packages directory for reading transitive deps
    pub fn with_packages_dir(packages_dir: &Path) -> Self {
        Self {
            available_versions: HashMap::new(),
            package_deps: HashMap::new(),
            packages_dir: Some(packages_dir.to_path_buf()),
        }
    }
    /// Add available versions for a package (for testing/registry integration)
    pub fn add_available_versions(&mut self, name: &str, versions: Vec<Version>) {
        self.available_versions.insert(name.to_string(), versions);
    }
    /// Add package dependencies (for testing or when loaded from registry)
    pub fn add_package_deps(&mut self, name: &str, version: &Version, deps: HashMap<String, Dependency>) {
        let key = format!("{}@{}", name, version);
        self.package_deps.insert(key, deps);
    }
    /// Resolve dependencies to a lock file (with transitive dependencies)
    pub fn resolve(
        &self,
        manifest: &Manifest,
        existing_lock: Option<&LockFile>,
    ) -> Result<LockFile, ResolutionError> {
        let mut lock = LockFile::new();
        let mut resolved: HashMap<String, (Version, LockedSource)> = HashMap::new();
        let mut to_resolve: Vec<(String, Dependency, Option<String>)> = Vec::new();
        // If we have an existing lock file, prefer those versions
        if let Some(existing) = existing_lock {
            for pkg in &existing.packages {
                resolved.insert(pkg.name.clone(), (pkg.version.clone(), pkg.source.clone()));
            }
        }
        // Queue direct dependencies for resolution
        for (name, dep) in &manifest.dependencies {
            to_resolve.push((name.clone(), dep.clone(), None));
        }
        // Process queue (breadth-first for better dependency order)
        while let Some((name, dep, required_by)) = to_resolve.pop() {
            // Skip if already resolved with compatible version
            if let Some((existing_version, _)) = resolved.get(&name) {
                let constraint = VersionConstraint::parse(&dep.version)
                    .map_err(|e| ResolutionError {
                        message: e,
                        package: Some(name.clone()),
                    })?;
                if constraint.satisfies(existing_version) {
                    continue; // Already have a compatible version
                } else {
                    // Version conflict
                    return Err(ResolutionError {
                        message: format!(
                            "Version conflict: {} requires {} {}, but {} is already resolved{}",
                            required_by.as_deref().unwrap_or("project"),
                            name,
                            dep.version,
                            existing_version,
                            if let Some(rb) = &required_by {
                                format!(" (required by {})", rb)
                            } else {
                                String::new()
                            }
                        ),
                        package: Some(name.clone()),
                    });
                }
            }
            let constraint = VersionConstraint::parse(&dep.version)
                .map_err(|e| ResolutionError {
                    message: e,
                    package: Some(name.clone()),
                })?;
            // Resolve the version
            let version = self.select_version(&name, &constraint, &dep.source)?;
            let source = match &dep.source {
                DependencySource::Registry => LockedSource::Registry,
                DependencySource::Git { url, branch } => LockedSource::Git {
                    url: url.clone(),
                    rev: branch.clone().unwrap_or_else(|| "HEAD".to_string()),
                },
                DependencySource::Path { path } => LockedSource::Path { path: path.clone() },
            };
            resolved.insert(name.clone(), (version.clone(), source.clone()));
            // Get transitive dependencies
            let transitive_deps = self.get_package_dependencies(&name, &version, &dep.source);
            for (trans_name, trans_dep) in transitive_deps {
                if !resolved.contains_key(&trans_name) {
                    to_resolve.push((trans_name, trans_dep, Some(name.clone())));
                }
            }
        }
        // Build lock file from resolved packages
        for (name, (version, source)) in &resolved {
            // Get the dependency list for this package
            let deps = self.get_package_dependencies(name, version, &match source {
                LockedSource::Registry => DependencySource::Registry,
                LockedSource::Git { url, rev } => DependencySource::Git {
                    url: url.clone(),
                    branch: Some(rev.clone()),
                },
                LockedSource::Path { path } => DependencySource::Path { path: path.clone() },
            });
            let dep_names: Vec<String> = deps.keys().cloned().collect();
            lock.packages.push(LockedPackage {
                name: name.clone(),
                version: version.clone(),
                source: source.clone(),
                checksum: None,
                dependencies: dep_names,
            });
        }
        // Sort packages by name for deterministic output
        lock.packages.sort_by(|a, b| a.name.cmp(&b.name));
        Ok(lock)
    }
    /// Get dependencies of a package
    fn get_package_dependencies(
        &self,
        name: &str,
        version: &Version,
        source: &DependencySource,
    ) -> HashMap<String, Dependency> {
        // First check our cache
        let key = format!("{}@{}", name, version);
        if let Some(deps) = self.package_deps.get(&key) {
            return deps.clone();
        }
        // Try to read from installed package
        if let Some(ref packages_dir) = self.packages_dir {
            let pkg_dir = packages_dir.join(name);
            let manifest_path = pkg_dir.join("lux.toml");
            if manifest_path.exists() {
                if let Ok(content) = fs::read_to_string(&manifest_path) {
                    if let Ok(manifest) = parse_manifest(&content) {
                        return manifest.dependencies;
                    }
                }
            }
        }
        // For path dependencies, read from the path
        if let DependencySource::Path { path } = source {
            let manifest_path = if path.is_absolute() {
                path.join("lux.toml")
            } else if let Some(ref packages_dir) = self.packages_dir {
                packages_dir.parent().unwrap_or(packages_dir).join(path).join("lux.toml")
            } else {
                path.join("lux.toml")
            };
            if manifest_path.exists() {
                if let Ok(content) = fs::read_to_string(&manifest_path) {
                    if let Ok(manifest) = parse_manifest(&content) {
                        return manifest.dependencies;
                    }
                }
            }
        }
        // No dependencies found
        HashMap::new()
    }
    /// Select the best version that satisfies the constraint
    fn select_version(
        &self,
        name: &str,
        constraint: &VersionConstraint,
        source: &DependencySource,
    ) -> Result<Version, ResolutionError> {
        match source {
            DependencySource::Git { .. } | DependencySource::Path { .. } => {
                // For git/path sources, use the version from the constraint or 0.0.0
                match constraint {
                    VersionConstraint::Exact(v) => Ok(v.clone()),
                    _ => Ok(Version::new(0, 0, 0)),
                }
            }
            DependencySource::Registry => {
                // Check available versions
                if let Some(versions) = self.available_versions.get(name) {
                    // Find the highest version that satisfies the constraint
                    let mut matching: Vec<&Version> = versions
                        .iter()
                        .filter(|v| constraint.satisfies(v))
                        .collect();
                    matching.sort();
                    matching.reverse();
                    if let Some(v) = matching.first() {
                        return Ok((*v).clone());
                    }
                }
                // No available versions - use the constraint's base version
                match constraint {
                    VersionConstraint::Exact(v) => Ok(v.clone()),
                    VersionConstraint::Caret(v) => Ok(v.clone()),
                    VersionConstraint::Tilde(v) => Ok(v.clone()),
                    VersionConstraint::GreaterEq(v) => Ok(v.clone()),
                    VersionConstraint::Range { min, .. } => Ok(min.clone()),
                    VersionConstraint::Less(_) | VersionConstraint::Any => {
                        // Can't determine version without registry
                        Ok(Version::new(0, 0, 0))
                    }
                }
            }
        }
    }
 }
 impl Default for Resolver {
    fn default() -> Self {
        Self::new()
    }
 }
 // =============================================================================
 // Manifest and Package Manager
 // =============================================================================
 /// Package manifest (lux.toml)
 #[derive(Debug, Clone)]
@@ -69,6 +681,43 @@ impl PackageManager {
        }
    }
    /// Load the lock file (lux.lock)
    pub fn load_lock(&self) -> Result<Option<LockFile>, String> {
        let lock_path = self.project_root.join("lux.lock");
        if !lock_path.exists() {
            return Ok(None);
        }
        let content = fs::read_to_string(&lock_path)
            .map_err(|e| format!("Failed to read lux.lock: {}", e))?;
        LockFile::parse(&content).map(Some)
    }
    /// Save the lock file (lux.lock)
    pub fn save_lock(&self, lock: &LockFile) -> Result<(), String> {
        let lock_path = self.project_root.join("lux.lock");
        let content = lock.format();
        fs::write(&lock_path, content)
            .map_err(|e| format!("Failed to write lux.lock: {}", e))
    }
    /// Resolve dependencies and generate/update lock file
    pub fn resolve(&self) -> Result<LockFile, String> {
        let manifest = self.load_manifest()?;
        let existing_lock = self.load_lock()?;
        // Use resolver with packages directory for transitive dep lookup
        let resolver = Resolver::with_packages_dir(&self.packages_dir);
        let lock = resolver.resolve(&manifest, existing_lock.as_ref())
            .map_err(|e| e.to_string())?;
        self.save_lock(&lock)?;
        Ok(lock)
    }
    /// Find the project root by looking for lux.toml
    pub fn find_project_root() -> Option<PathBuf> {
        let mut current = std::env::current_dir().ok()?;
@@ -154,19 +803,139 @@ impl PackageManager {
            return Ok(());
        }
        // Resolve dependencies and generate/update lock file
        let lock = self.resolve()?;
        // Create packages directory
        fs::create_dir_all(&self.packages_dir)
            .map_err(|e| format!("Failed to create packages directory: {}", e))?;
-        println!("Installing {} dependencies...", manifest.dependencies.len());
+        println!("Installing {} dependencies...", lock.packages.len());
        println!();
-        for (_name, dep) in &manifest.dependencies {
+        // Install from lock file for reproducibility
-            self.install_dependency(dep)?;
+        for locked_pkg in &lock.packages {
            self.install_locked_package(locked_pkg, &manifest)?;
        }
        println!();
-        println!("Done! Installed {} packages.", manifest.dependencies.len());
+        println!("Done! Installed {} packages.", lock.packages.len());
        println!("Lock file written to lux.lock");
        Ok(())
    }
    /// Install a package from the lock file
    fn install_locked_package(&self, locked: &LockedPackage, manifest: &Manifest) -> Result<(), String> {
        print!("  Installing {} v{}... ", locked.name, locked.version);
        io::stdout().flush().unwrap();
        let dest_dir = self.packages_dir.join(&locked.name);
        // Get the dependency info from manifest for source details
        let dep = manifest.dependencies.get(&locked.name);
        match &locked.source {
            LockedSource::Registry => {
                self.install_from_registry_locked(locked, &dest_dir)?;
            }
            LockedSource::Git { url, rev } => {
                self.install_from_git_locked(url, rev, &dest_dir)?;
            }
            LockedSource::Path { path } => {
                let source_path = if let Some(d) = dep {
                    match &d.source {
                        DependencySource::Path { path } => path.clone(),
                        _ => path.clone(),
                    }
                } else {
                    path.clone()
                };
                self.install_from_path(&source_path, &dest_dir)?;
            }
        }
        println!("done");
        Ok(())
    }
    fn install_from_registry_locked(&self, locked: &LockedPackage, dest: &Path) -> Result<(), String> {
        // Check if already installed with correct version
        let version_file = dest.join(".version");
        if version_file.exists() {
            let installed_version = fs::read_to_string(&version_file).unwrap_or_default();
            if installed_version.trim() == locked.version.to_string() {
                return Ok(());
            }
        }
        // Check cache first
        let cache_path = self.cache_dir.join(&locked.name).join(locked.version.to_string());
        if cache_path.exists() {
            // Copy from cache
            copy_dir_recursive(&cache_path, dest)?;
        } else {
            // Create placeholder package (in real impl, would download)
            fs::create_dir_all(dest)
                .map_err(|e| format!("Failed to create package directory: {}", e))?;
            // Create a lib.lux placeholder
            let lib_content = format!(
                "// Package: {} v{}\n// This is a placeholder - real package would be downloaded from registry\n\n",
                locked.name, locked.version
            );
            fs::write(dest.join("lib.lux"), lib_content)
                .map_err(|e| format!("Failed to create lib.lux: {}", e))?;
        }
        // Write version file
        fs::write(&version_file, locked.version.to_string())
            .map_err(|e| format!("Failed to write version file: {}", e))?;
        // Verify checksum if present
        if let Some(ref expected) = locked.checksum {
            // In a real implementation, verify the checksum here
            let _ = expected; // Placeholder
        }
        Ok(())
    }
    fn install_from_git_locked(&self, url: &str, rev: &str, dest: &Path) -> Result<(), String> {
        // Remove existing if present
        if dest.exists() {
            fs::remove_dir_all(dest)
                .map_err(|e| format!("Failed to remove existing directory: {}", e))?;
        }
        // Clone at specific revision
        let mut cmd = std::process::Command::new("git");
        cmd.arg("clone")
            .arg("--depth").arg("1");
        // If rev is not HEAD, we need to fetch the specific revision
        if rev != "HEAD" && !rev.is_empty() {
            cmd.arg("--branch").arg(rev);
        }
        cmd.arg(url).arg(dest);
        let output = cmd.output()
            .map_err(|e| format!("Failed to run git: {}", e))?;
        if !output.status.success() {
            return Err(format!(
                "Git clone failed: {}",
                String::from_utf8_lossy(&output.stderr)
            ));
        }
        // Remove .git directory to save space
        let git_dir = dest.join(".git");
        if git_dir.exists() {
            fs::remove_dir_all(&git_dir).ok();
        }
        Ok(())
    }
--- a/src/parser.rs
+++ b/src/parser.rs
@@ -244,6 +244,7 @@ impl Parser {
            TokenKind::Let => Ok(Declaration::Let(self.parse_let_decl(visibility, doc)?)),
            TokenKind::Trait => Ok(Declaration::Trait(self.parse_trait_decl(visibility, doc)?)),
            TokenKind::Impl => Ok(Declaration::Impl(self.parse_impl_decl()?)),
            TokenKind::Run => Err(self.error("Bare 'run' expressions are not allowed at top level. Use 'let _ = run ...' or 'let result = run ...'")),
            _ => Err(self.error("Expected declaration (fn, effect, handler, type, trait, impl, or let)")),
        }
    }
@@ -532,7 +533,14 @@ impl Parser {
        let start = self.current_span();
        self.expect(TokenKind::Let)?;
-        let name = self.parse_ident()?;
+        // Allow underscore as wildcard pattern (discards the value)
        let name = if self.check(TokenKind::Underscore) {
            let span = self.current_span();
            self.advance();
            Ident::new("_".to_string(), span)
        } else {
            self.parse_ident()?
        };
        let typ = if self.check(TokenKind::Colon) {
            self.advance();
@@ -871,7 +879,8 @@ impl Parser {
        Ok(effects)
    }
-    /// Parse behavioral properties: is pure, is total, is idempotent, etc.
+    /// Parse behavioral properties: is pure, total, idempotent, etc.
    /// Supports: `is pure`, `is pure is total`, `is pure, total`, `is pure, is total`
    fn parse_behavioral_properties(&mut self) -> Result<Vec<BehavioralProperty>, ParseError> {
        let mut properties = Vec::new();
@@ -893,9 +902,15 @@ impl Parser {
            let property = self.parse_single_property()?;
            properties.push(property);
-            // Optional comma for multiple properties: is pure, is total
+            // After first property, allow comma-separated list without repeating 'is'
-            if self.check(TokenKind::Comma) {
+            while self.check(TokenKind::Comma) {
-                self.advance();
+                self.advance(); // consume comma
                // Allow optional 'is' after comma: `is pure, is total` or `is pure, total`
                if self.check(TokenKind::Is) {
                    self.advance();
                }
                let property = self.parse_single_property()?;
                properties.push(property);
            }
        }
@@ -1543,6 +1558,7 @@ impl Parser {
        loop {
            let op = match self.peek_kind() {
                TokenKind::Plus => BinaryOp::Add,
                TokenKind::PlusPlus => BinaryOp::Concat,
                TokenKind::Minus => BinaryOp::Sub,
                _ => break,
            };
@@ -1631,6 +1647,20 @@ impl Parser {
            } else if self.check(TokenKind::Dot) {
                let start = expr.span();
                self.advance();
                // Check for tuple index access: expr.0, expr.1, etc.
                if let TokenKind::Int(n) = self.peek_kind() {
                    let index = n as usize;
                    self.advance();
                    let span = start.merge(self.previous_span());
                    expr = Expr::TupleIndex {
                        object: Box::new(expr),
                        index,
                        span,
                    };
                    continue;
                }
                let field = self.parse_ident()?;
                // Check if this is an effect operation: Effect.operation(args)
@@ -1666,11 +1696,14 @@ impl Parser {
    fn parse_args(&mut self) -> Result<Vec<Expr>, ParseError> {
        let mut args = Vec::new();
        self.skip_newlines();
        while !self.check(TokenKind::RParen) {
            args.push(self.parse_expr()?);
            self.skip_newlines();
            if !self.check(TokenKind::RParen) {
                self.expect(TokenKind::Comma)?;
                self.skip_newlines();
            }
        }
@@ -1872,6 +1905,14 @@ impl Parser {
                    span: token.span,
                }))
            }
            TokenKind::Char(c) => {
                let c = *c;
                self.advance();
                Ok(Pattern::Literal(Literal {
                    kind: LiteralKind::Char(c),
                    span: token.span,
                }))
            }
            TokenKind::Ident(name) => {
                // Check if it starts with uppercase (constructor) or lowercase (variable)
                if name.chars().next().map_or(false, |c| c.is_uppercase()) {
@@ -1962,7 +2003,14 @@ impl Parser {
        let start = self.current_span();
        self.expect(TokenKind::Let)?;
-        let name = self.parse_ident()?;
+        // Allow underscore as wildcard pattern (discards the value)
        let name = if self.check(TokenKind::Underscore) {
            let span = self.current_span();
            self.advance();
            Ident::new("_".to_string(), span)
        } else {
            self.parse_ident()?
        };
        let typ = if self.check(TokenKind::Colon) {
            self.advance();
@@ -2160,6 +2208,11 @@ impl Parser {
            }));
        }
        // Check for record spread: { ...expr, field: val }
        if matches!(self.peek_kind(), TokenKind::DotDotDot) {
            return self.parse_record_expr_rest(start);
        }
        // Check if it's a record (ident: expr) or block
        if matches!(self.peek_kind(), TokenKind::Ident(_)) {
            let lookahead = self.tokens.get(self.pos + 1).map(|t| &t.kind);
@@ -2174,6 +2227,20 @@ impl Parser {
    fn parse_record_expr_rest(&mut self, start: Span) -> Result<Expr, ParseError> {
        let mut fields = Vec::new();
        let mut spread = None;
        // Check for spread: { ...expr, ... }
        if self.check(TokenKind::DotDotDot) {
            self.advance(); // consume ...
            let spread_expr = self.parse_expr()?;
            spread = Some(Box::new(spread_expr));
            self.skip_newlines();
            if self.check(TokenKind::Comma) {
                self.advance();
            }
            self.skip_newlines();
        }
        while !self.check(TokenKind::RBrace) {
            let name = self.parse_ident()?;
@@ -2190,7 +2257,11 @@ impl Parser {
        self.expect(TokenKind::RBrace)?;
        let span = start.merge(self.previous_span());
-        Ok(Expr::Record { fields, span })
+        Ok(Expr::Record {
            spread,
            fields,
            span,
        })
    }
    fn parse_block_rest(&mut self, start: Span) -> Result<Expr, ParseError> {
@@ -2207,7 +2278,15 @@ impl Parser {
                // Let statement
                let let_start = self.current_span();
                self.advance();
-                let name = self.parse_ident()?;
+
                // Allow underscore as wildcard pattern (discards the value)
                let name = if self.check(TokenKind::Underscore) {
                    let span = self.current_span();
                    self.advance();
                    Ident::new("_".to_string(), span)
                } else {
                    self.parse_ident()?
                };
                let typ = if self.check(TokenKind::Colon) {
                    self.advance();
--- a/src/registry.rs
+++ b/src/registry.rs
@@ -0,0 +1,637 @@
 //! Package Registry Server for Lux
 //!
 //! Provides a central repository for sharing Lux packages.
 //! The registry serves package metadata and tarballs via HTTP.
 use std::collections::HashMap;
 use std::fs;
 use std::io::{Read, Write};
 use std::net::{TcpListener, TcpStream};
 use std::path::{Path, PathBuf};
 use std::sync::{Arc, RwLock};
 use std::thread;
 /// Package metadata stored in the registry
 #[derive(Debug, Clone)]
 pub struct PackageMetadata {
    pub name: String,
    pub version: String,
    pub description: String,
    pub authors: Vec<String>,
    pub license: Option<String>,
    pub repository: Option<String>,
    pub keywords: Vec<String>,
    pub dependencies: HashMap<String, String>,
    pub checksum: String,
    pub published_at: String,
 }
 /// A version entry for a package
 #[derive(Debug, Clone)]
 pub struct VersionEntry {
    pub version: String,
    pub checksum: String,
    pub published_at: String,
    pub yanked: bool,
 }
 /// Package index entry (all versions of a package)
 #[derive(Debug, Clone)]
 pub struct PackageIndex {
    pub name: String,
    pub description: String,
    pub versions: Vec<VersionEntry>,
    pub latest_version: String,
 }
 /// The package registry
 pub struct Registry {
    /// Base directory for storing packages
    storage_dir: PathBuf,
    /// In-memory index of all packages
    index: Arc<RwLock<HashMap<String, PackageIndex>>>,
 }
 impl Registry {
    /// Create a new registry with the given storage directory
    pub fn new(storage_dir: &Path) -> Self {
        let registry = Self {
            storage_dir: storage_dir.to_path_buf(),
            index: Arc::new(RwLock::new(HashMap::new())),
        };
        registry.load_index();
        registry
    }
    /// Load the package index from disk
    fn load_index(&self) {
        let index_path = self.storage_dir.join("index.json");
        if !index_path.exists() {
            return;
        }
        if let Ok(content) = fs::read_to_string(&index_path) {
            if let Ok(index) = parse_index_json(&content) {
                let mut idx = self.index.write().unwrap();
                *idx = index;
            }
        }
    }
    /// Save the package index to disk
    fn save_index(&self) {
        let index_path = self.storage_dir.join("index.json");
        let idx = self.index.read().unwrap();
        let json = format_index_json(&idx);
        fs::write(&index_path, json).ok();
    }
    /// Publish a new package version
    pub fn publish(&self, metadata: PackageMetadata, tarball: &[u8]) -> Result<(), String> {
        // Validate package name
        if !is_valid_package_name(&metadata.name) {
            return Err("Invalid package name. Use lowercase letters, numbers, and hyphens.".to_string());
        }
        // Create package directory
        let pkg_dir = self.storage_dir.join("packages").join(&metadata.name);
        fs::create_dir_all(&pkg_dir)
            .map_err(|e| format!("Failed to create package directory: {}", e))?;
        // Write tarball
        let tarball_path = pkg_dir.join(format!("{}-{}.tar.gz", metadata.name, metadata.version));
        fs::write(&tarball_path, tarball)
            .map_err(|e| format!("Failed to write package tarball: {}", e))?;
        // Write metadata
        let meta_path = pkg_dir.join(format!("{}-{}.json", metadata.name, metadata.version));
        let meta_json = format_metadata_json(&metadata);
        fs::write(&meta_path, meta_json)
            .map_err(|e| format!("Failed to write package metadata: {}", e))?;
        // Update index
        {
            let mut idx = self.index.write().unwrap();
            let entry = idx.entry(metadata.name.clone()).or_insert_with(|| PackageIndex {
                name: metadata.name.clone(),
                description: metadata.description.clone(),
                versions: Vec::new(),
                latest_version: String::new(),
            });
            // Check if version already exists
            if entry.versions.iter().any(|v| v.version == metadata.version) {
                return Err(format!("Version {} already exists", metadata.version));
            }
            entry.versions.push(VersionEntry {
                version: metadata.version.clone(),
                checksum: metadata.checksum.clone(),
                published_at: metadata.published_at.clone(),
                yanked: false,
            });
            // Update latest version (simple comparison for now)
            entry.latest_version = metadata.version.clone();
            entry.description = metadata.description.clone();
        }
        self.save_index();
        Ok(())
    }
    /// Get package metadata
    pub fn get_metadata(&self, name: &str, version: &str) -> Option<PackageMetadata> {
        let meta_path = self.storage_dir
            .join("packages")
            .join(name)
            .join(format!("{}-{}.json", name, version));
        if let Ok(content) = fs::read_to_string(&meta_path) {
            parse_metadata_json(&content)
        } else {
            None
        }
    }
    /// Get package tarball
    pub fn get_tarball(&self, name: &str, version: &str) -> Option<Vec<u8>> {
        let tarball_path = self.storage_dir
            .join("packages")
            .join(name)
            .join(format!("{}-{}.tar.gz", name, version));
        fs::read(&tarball_path).ok()
    }
    /// Search packages
    pub fn search(&self, query: &str) -> Vec<PackageIndex> {
        let idx = self.index.read().unwrap();
        let query_lower = query.to_lowercase();
        idx.values()
            .filter(|pkg| {
                pkg.name.to_lowercase().contains(&query_lower) ||
                pkg.description.to_lowercase().contains(&query_lower)
            })
            .cloned()
            .collect()
    }
    /// List all packages
    pub fn list_all(&self) -> Vec<PackageIndex> {
        let idx = self.index.read().unwrap();
        idx.values().cloned().collect()
    }
    /// Get package index entry
    pub fn get_package(&self, name: &str) -> Option<PackageIndex> {
        let idx = self.index.read().unwrap();
        idx.get(name).cloned()
    }
 }
 /// HTTP Registry Server
 pub struct RegistryServer {
    registry: Arc<Registry>,
    bind_addr: String,
 }
 impl RegistryServer {
    /// Create a new registry server
    pub fn new(storage_dir: &Path, bind_addr: &str) -> Self {
        Self {
            registry: Arc::new(Registry::new(storage_dir)),
            bind_addr: bind_addr.to_string(),
        }
    }
    /// Run the server
    pub fn run(&self) -> Result<(), String> {
        let listener = TcpListener::bind(&self.bind_addr)
            .map_err(|e| format!("Failed to bind to {}: {}", self.bind_addr, e))?;
        println!("Lux Package Registry running at http://{}", self.bind_addr);
        println!("Storage directory: {}", self.registry.storage_dir.display());
        println!();
        println!("Endpoints:");
        println!("  GET  /api/v1/packages              - List all packages");
        println!("  GET  /api/v1/packages/:name        - Get package info");
        println!("  GET  /api/v1/packages/:name/:ver   - Get version metadata");
        println!("  GET  /api/v1/download/:name/:ver   - Download package tarball");
        println!("  GET  /api/v1/search?q=query        - Search packages");
        println!("  POST /api/v1/publish               - Publish a package");
        println!();
        for stream in listener.incoming() {
            match stream {
                Ok(stream) => {
                    let registry = Arc::clone(&self.registry);
                    thread::spawn(move || {
                        handle_request(stream, &registry);
                    });
                }
                Err(e) => {
                    eprintln!("Connection error: {}", e);
                }
            }
        }
        Ok(())
    }
 }
 /// Handle an HTTP request
 fn handle_request(mut stream: TcpStream, registry: &Registry) {
    let mut buffer = [0; 8192];
    let bytes_read = match stream.read(&mut buffer) {
        Ok(n) => n,
        Err(_) => return,
    };
    let request = String::from_utf8_lossy(&buffer[..bytes_read]);
    let lines: Vec<&str> = request.lines().collect();
    if lines.is_empty() {
        return;
    }
    let parts: Vec<&str> = lines[0].split_whitespace().collect();
    if parts.len() < 2 {
        return;
    }
    let method = parts[0];
    let path = parts[1];
    // Parse path and query string
    let (path, query) = if let Some(q_pos) = path.find('?') {
        (&path[..q_pos], Some(&path[q_pos + 1..]))
    } else {
        (path, None)
    };
    let response = match (method, path) {
        ("GET", "/") => {
            html_response(200, r#"
                <!DOCTYPE html>
                <html>
                <head><title>Lux Package Registry</title></head>
                <body>
                    <h1>Lux Package Registry</h1>
                    <p>Welcome to the Lux package registry.</p>
                    <h2>API Endpoints</h2>
                    <ul>
                        <li>GET /api/v1/packages - List all packages</li>
                        <li>GET /api/v1/packages/:name - Get package info</li>
                        <li>GET /api/v1/packages/:name/:version - Get version metadata</li>
                        <li>GET /api/v1/download/:name/:version - Download package</li>
                        <li>GET /api/v1/search?q=query - Search packages</li>
                    </ul>
                </body>
                </html>
            "#)
        }
        ("GET", "/api/v1/packages") => {
            let packages = registry.list_all();
            let json = format_packages_list_json(&packages);
            json_response(200, &json)
        }
        ("GET", path) if path.starts_with("/api/v1/packages/") => {
            let rest = &path[17..]; // Remove "/api/v1/packages/"
            let parts: Vec<&str> = rest.split('/').collect();
            match parts.len() {
                1 => {
                    // Get package info
                    if let Some(pkg) = registry.get_package(parts[0]) {
                        let json = format_package_json(&pkg);
                        json_response(200, &json)
                    } else {
                        json_response(404, r#"{"error": "Package not found"}"#)
                    }
                }
                2 => {
                    // Get version metadata
                    if let Some(meta) = registry.get_metadata(parts[0], parts[1]) {
                        let json = format_metadata_json(&meta);
                        json_response(200, &json)
                    } else {
                        json_response(404, r#"{"error": "Version not found"}"#)
                    }
                }
                _ => json_response(400, r#"{"error": "Invalid path"}"#)
            }
        }
        ("GET", path) if path.starts_with("/api/v1/download/") => {
            let rest = &path[17..]; // Remove "/api/v1/download/"
            let parts: Vec<&str> = rest.split('/').collect();
            if parts.len() == 2 {
                if let Some(tarball) = registry.get_tarball(parts[0], parts[1]) {
                    tarball_response(&tarball)
                } else {
                    json_response(404, r#"{"error": "Package not found"}"#)
                }
            } else {
                json_response(400, r#"{"error": "Invalid path"}"#)
            }
        }
        ("GET", "/api/v1/search") => {
            let q = query
                .and_then(|qs| parse_query_string(qs).get("q").cloned())
                .unwrap_or_default();
            let results = registry.search(&q);
            let json = format_packages_list_json(&results);
            json_response(200, &json)
        }
        ("POST", "/api/v1/publish") => {
            // Find content length
            let content_length: usize = lines.iter()
                .find(|l| l.to_lowercase().starts_with("content-length:"))
                .and_then(|l| l.split(':').nth(1))
                .and_then(|s| s.trim().parse().ok())
                .unwrap_or(0);
            // Find body start
            let body_start = request.find("\r\n\r\n")
                .map(|i| i + 4)
                .unwrap_or(bytes_read);
            // For now, return a message about publishing
            // Real implementation would parse multipart form data
            json_response(200, &format!(
                r#"{{"message": "Publish endpoint ready", "content_length": {}}}"#,
                content_length
            ))
        }
        _ => {
            json_response(404, r#"{"error": "Not found"}"#)
        }
    };
    stream.write_all(response.as_bytes()).ok();
 }
 /// Create an HTML response
 fn html_response(status: u16, body: &str) -> String {
    let status_text = match status {
        200 => "OK",
        400 => "Bad Request",
        404 => "Not Found",
        500 => "Internal Server Error",
        _ => "Unknown",
    };
    format!(
        "HTTP/1.1 {} {}\r\nContent-Type: text/html\r\nContent-Length: {}\r\nConnection: close\r\n\r\n{}",
        status, status_text, body.len(), body
    )
 }
 /// Create a JSON response
 fn json_response(status: u16, body: &str) -> String {
    let status_text = match status {
        200 => "OK",
        400 => "Bad Request",
        404 => "Not Found",
        500 => "Internal Server Error",
        _ => "Unknown",
    };
    format!(
        "HTTP/1.1 {} {}\r\nContent-Type: application/json\r\nContent-Length: {}\r\nConnection: close\r\n\r\n{}",
        status, status_text, body.len(), body
    )
 }
 /// Create a tarball response
 fn tarball_response(data: &[u8]) -> String {
    format!(
        "HTTP/1.1 200 OK\r\nContent-Type: application/gzip\r\nContent-Length: {}\r\nConnection: close\r\n\r\n",
        data.len()
    )
 }
 /// Validate package name
 fn is_valid_package_name(name: &str) -> bool {
    !name.is_empty() &&
    name.len() <= 64 &&
    name.chars().all(|c| c.is_ascii_lowercase() || c.is_ascii_digit() || c == '-' || c == '_') &&
    name.chars().next().map(|c| c.is_ascii_lowercase()).unwrap_or(false)
 }
 /// Parse query string into key-value pairs
 fn parse_query_string(qs: &str) -> HashMap<String, String> {
    let mut params = HashMap::new();
    for part in qs.split('&') {
        if let Some(eq_pos) = part.find('=') {
            let key = &part[..eq_pos];
            let value = &part[eq_pos + 1..];
            params.insert(
                urlldecode(key),
                urlldecode(value),
            );
        }
    }
    params
 }
 /// Simple URL decoding
 fn urlldecode(s: &str) -> String {
    let mut result = String::new();
    let mut chars = s.chars().peekable();
    while let Some(c) = chars.next() {
        if c == '%' {
            let hex: String = chars.by_ref().take(2).collect();
            if let Ok(byte) = u8::from_str_radix(&hex, 16) {
                result.push(byte as char);
            }
        } else if c == '+' {
            result.push(' ');
        } else {
            result.push(c);
        }
    }
    result
 }
 // JSON formatting helpers
 fn format_metadata_json(meta: &PackageMetadata) -> String {
    let deps: Vec<String> = meta.dependencies.iter()
        .map(|(k, v)| format!(r#""{}": "{}""#, k, v))
        .collect();
    let authors: Vec<String> = meta.authors.iter()
        .map(|a| format!(r#""{}""#, a))
        .collect();
    let keywords: Vec<String> = meta.keywords.iter()
        .map(|k| format!(r#""{}""#, k))
        .collect();
    format!(
        r#"{{
  "name": "{}",
  "version": "{}",
  "description": "{}",
  "authors": [{}],
  "license": {},
  "repository": {},
  "keywords": [{}],
  "dependencies": {{{}}},
  "checksum": "{}",
  "published_at": "{}"
 }}"#,
        meta.name,
        meta.version,
        escape_json(&meta.description),
        authors.join(", "),
        meta.license.as_ref().map(|l| format!(r#""{}""#, l)).unwrap_or("null".to_string()),
        meta.repository.as_ref().map(|r| format!(r#""{}""#, r)).unwrap_or("null".to_string()),
        keywords.join(", "),
        deps.join(", "),
        meta.checksum,
        meta.published_at,
    )
 }
 fn format_package_json(pkg: &PackageIndex) -> String {
    let versions: Vec<String> = pkg.versions.iter()
        .map(|v| format!(
            r#"{{"version": "{}", "checksum": "{}", "published_at": "{}", "yanked": {}}}"#,
            v.version, v.checksum, v.published_at, v.yanked
        ))
        .collect();
    format!(
        r#"{{
  "name": "{}",
  "description": "{}",
  "latest_version": "{}",
  "versions": [{}]
 }}"#,
        pkg.name,
        escape_json(&pkg.description),
        pkg.latest_version,
        versions.join(", ")
    )
 }
 fn format_packages_list_json(packages: &[PackageIndex]) -> String {
    let items: Vec<String> = packages.iter()
        .map(|pkg| format!(
            r#"{{"name": "{}", "description": "{}", "latest_version": "{}"}}"#,
            pkg.name,
            escape_json(&pkg.description),
            pkg.latest_version
        ))
        .collect();
    format!(r#"{{"packages": [{}]}}"#, items.join(", "))
 }
 fn format_index_json(index: &HashMap<String, PackageIndex>) -> String {
    let items: Vec<String> = index.values()
        .map(|pkg| format_package_json(pkg))
        .collect();
    format!(r#"{{"packages": [{}]}}"#, items.join(",\n"))
 }
 fn parse_index_json(content: &str) -> Result<HashMap<String, PackageIndex>, String> {
    // Simple JSON parsing for the index
    // In production, would use serde_json
    let mut index = HashMap::new();
    // Basic parsing - find package names and latest versions
    // This is a simplified parser for the index format
    let content = content.trim();
    if !content.starts_with('{') || !content.ends_with('}') {
        return Err("Invalid JSON format".to_string());
    }
    // For now, return empty index if parsing fails
    // Real implementation would properly parse JSON
    Ok(index)
 }
 fn parse_metadata_json(content: &str) -> Option<PackageMetadata> {
    // Simple JSON parsing for metadata
    // In production, would use serde_json
    let mut name = String::new();
    let mut version = String::new();
    let mut description = String::new();
    let mut checksum = String::new();
    let mut published_at = String::new();
    for line in content.lines() {
        let line = line.trim();
        if line.contains("\"name\":") {
            name = extract_json_string(line);
        } else if line.contains("\"version\":") {
            version = extract_json_string(line);
        } else if line.contains("\"description\":") {
            description = extract_json_string(line);
        } else if line.contains("\"checksum\":") {
            checksum = extract_json_string(line);
        } else if line.contains("\"published_at\":") {
            published_at = extract_json_string(line);
        }
    }
    if name.is_empty() || version.is_empty() {
        return None;
    }
    Some(PackageMetadata {
        name,
        version,
        description,
        authors: Vec::new(),
        license: None,
        repository: None,
        keywords: Vec::new(),
        dependencies: HashMap::new(),
        checksum,
        published_at,
    })
 }
 fn extract_json_string(line: &str) -> String {
    // Extract string value from "key": "value" format
    if let Some(colon) = line.find(':') {
        let value = line[colon + 1..].trim();
        let value = value.trim_start_matches('"');
        if let Some(end) = value.find('"') {
            return value[..end].to_string();
        }
    }
    String::new()
 }
 fn escape_json(s: &str) -> String {
    s.replace('\\', "\\\\")
     .replace('"', "\\\"")
     .replace('\n', "\\n")
     .replace('\r', "\\r")
     .replace('\t', "\\t")
 }
 /// Run the registry server (called from main)
 pub fn run_registry_server(storage_dir: &str, bind_addr: &str) -> Result<(), String> {
    let storage_path = PathBuf::from(storage_dir);
    fs::create_dir_all(&storage_path)
        .map_err(|e| format!("Failed to create storage directory: {}", e))?;
    let server = RegistryServer::new(&storage_path, bind_addr);
    server.run()
 }
--- a/src/symbol_table.rs
+++ b/src/symbol_table.rs
@@ -0,0 +1,682 @@
 //! Symbol Table for Lux
 //!
 //! Provides semantic analysis infrastructure for IDE features like
 //! go-to-definition, find references, and rename refactoring.
 use crate::ast::*;
 use std::collections::HashMap;
 /// Unique identifier for a symbol
 #[derive(Debug, Clone, Copy, PartialEq, Eq, Hash)]
 pub struct SymbolId(pub u32);
 /// Kind of symbol
 #[derive(Debug, Clone, PartialEq, Eq)]
 pub enum SymbolKind {
    Function,
    Variable,
    Parameter,
    Type,
    TypeParameter,
    Variant,
    Effect,
    EffectOperation,
    Field,
    Module,
 }
 /// A symbol definition
 #[derive(Debug, Clone)]
 pub struct Symbol {
    pub id: SymbolId,
    pub name: String,
    pub kind: SymbolKind,
    pub span: Span,
    /// Type signature (for display)
    pub type_signature: Option<String>,
    /// Documentation comment
    pub documentation: Option<String>,
    /// Parent symbol (e.g., type for variants, effect for operations)
    pub parent: Option<SymbolId>,
    /// Is this symbol exported (public)?
    pub is_public: bool,
 }
 /// A reference to a symbol
 #[derive(Debug, Clone)]
 pub struct Reference {
    pub symbol_id: SymbolId,
    pub span: Span,
    pub is_definition: bool,
    pub is_write: bool,
 }
 /// A scope in the symbol table
 #[derive(Debug, Clone)]
 pub struct Scope {
    /// Parent scope (None for global scope)
    pub parent: Option<usize>,
    /// Symbols defined in this scope
    pub symbols: HashMap<String, SymbolId>,
    /// Span of this scope
    pub span: Span,
 }
 /// The symbol table
 #[derive(Debug, Clone)]
 pub struct SymbolTable {
    /// All symbols
    symbols: Vec<Symbol>,
    /// All references
    references: Vec<Reference>,
    /// Scopes (index 0 is always the global scope)
    scopes: Vec<Scope>,
    /// Mapping from position to references
    position_to_reference: HashMap<(u32, u32), usize>,
    /// Next symbol ID
    next_id: u32,
 }
 impl SymbolTable {
    pub fn new() -> Self {
        Self {
            symbols: Vec::new(),
            references: Vec::new(),
            scopes: vec![Scope {
                parent: None,
                symbols: HashMap::new(),
                span: Span { start: 0, end: 0 },
            }],
            position_to_reference: HashMap::new(),
            next_id: 0,
        }
    }
    /// Build symbol table from a program
    pub fn build(program: &Program) -> Self {
        let mut table = Self::new();
        table.visit_program(program);
        table
    }
    /// Add a symbol to the current scope
    fn add_symbol(&mut self, scope_idx: usize, symbol: Symbol) -> SymbolId {
        let id = symbol.id;
        self.scopes[scope_idx].symbols.insert(symbol.name.clone(), id);
        self.symbols.push(symbol);
        id
    }
    /// Create a new symbol
    fn new_symbol(
        &mut self,
        name: String,
        kind: SymbolKind,
        span: Span,
        type_signature: Option<String>,
        is_public: bool,
    ) -> Symbol {
        let id = SymbolId(self.next_id);
        self.next_id += 1;
        Symbol {
            id,
            name,
            kind,
            span,
            type_signature,
            documentation: None,
            parent: None,
            is_public,
        }
    }
    /// Add a reference
    fn add_reference(&mut self, symbol_id: SymbolId, span: Span, is_definition: bool, is_write: bool) {
        let ref_idx = self.references.len();
        self.references.push(Reference {
            symbol_id,
            span,
            is_definition,
            is_write,
        });
        // Index by start position
        self.position_to_reference.insert((span.start as u32, span.end as u32), ref_idx);
    }
    /// Look up a symbol by name in the given scope and its parents
    pub fn lookup(&self, name: &str, scope_idx: usize) -> Option<SymbolId> {
        let scope = &self.scopes[scope_idx];
        if let Some(&id) = scope.symbols.get(name) {
            return Some(id);
        }
        if let Some(parent) = scope.parent {
            return self.lookup(name, parent);
        }
        None
    }
    /// Get a symbol by ID
    pub fn get_symbol(&self, id: SymbolId) -> Option<&Symbol> {
        self.symbols.iter().find(|s| s.id == id)
    }
    /// Get the symbol at a position
    pub fn symbol_at_position(&self, offset: usize) -> Option<&Symbol> {
        // Find a reference that contains this offset
        for reference in &self.references {
            if offset >= reference.span.start && offset <= reference.span.end {
                return self.get_symbol(reference.symbol_id);
            }
        }
        None
    }
    /// Get the definition of a symbol at a position
    pub fn definition_at_position(&self, offset: usize) -> Option<&Symbol> {
        self.symbol_at_position(offset)
    }
    /// Find all references to a symbol
    pub fn find_references(&self, symbol_id: SymbolId) -> Vec<&Reference> {
        self.references
            .iter()
            .filter(|r| r.symbol_id == symbol_id)
            .collect()
    }
    /// Get all symbols of a given kind
    pub fn symbols_of_kind(&self, kind: SymbolKind) -> Vec<&Symbol> {
        self.symbols.iter().filter(|s| s.kind == kind).collect()
    }
    /// Get all symbols in the global scope
    pub fn global_symbols(&self) -> Vec<&Symbol> {
        self.scopes[0]
            .symbols
            .values()
            .filter_map(|&id| self.get_symbol(id))
            .collect()
    }
    /// Create a new scope
    fn push_scope(&mut self, parent: usize, span: Span) -> usize {
        let idx = self.scopes.len();
        self.scopes.push(Scope {
            parent: Some(parent),
            symbols: HashMap::new(),
            span,
        });
        idx
    }
    // =========================================================================
    // AST Visitors
    // =========================================================================
    fn visit_program(&mut self, program: &Program) {
        // First pass: collect all top-level declarations
        for decl in &program.declarations {
            self.visit_declaration(decl, 0);
        }
    }
    fn visit_declaration(&mut self, decl: &Declaration, scope_idx: usize) {
        match decl {
            Declaration::Function(f) => self.visit_function(f, scope_idx),
            Declaration::Type(t) => self.visit_type_decl(t, scope_idx),
            Declaration::Effect(e) => self.visit_effect(e, scope_idx),
            Declaration::Let(let_decl) => {
                let is_public = matches!(let_decl.visibility, Visibility::Public);
                let type_sig = let_decl.typ.as_ref().map(|t| self.type_expr_to_string(t));
                let mut symbol = self.new_symbol(
                    let_decl.name.name.clone(),
                    SymbolKind::Variable,
                    let_decl.span,
                    type_sig,
                    is_public,
                );
                symbol.documentation = let_decl.doc.clone();
                let id = self.add_symbol(scope_idx, symbol);
                self.add_reference(id, let_decl.name.span, true, true);
                // Visit the expression
                self.visit_expr(&let_decl.value, scope_idx);
            }
            Declaration::Handler(h) => self.visit_handler(h, scope_idx),
            Declaration::Trait(t) => self.visit_trait(t, scope_idx),
            Declaration::Impl(i) => self.visit_impl(i, scope_idx),
        }
    }
    fn visit_function(&mut self, f: &FunctionDecl, scope_idx: usize) {
        let is_public = matches!(f.visibility, Visibility::Public);
        // Build type signature
        let param_types: Vec<String> = f.params.iter()
            .map(|p| format!("{}: {}", p.name.name, self.type_expr_to_string(&p.typ)))
            .collect();
        let return_type = self.type_expr_to_string(&f.return_type);
        let effects = if f.effects.is_empty() {
            String::new()
        } else {
            format!(" with {{{}}}", f.effects.iter()
                .map(|e| e.name.clone())
                .collect::<Vec<_>>()
                .join(", "))
        };
        let properties = if f.properties.is_empty() {
            String::new()
        } else {
            format!(" is {}", f.properties.iter()
                .map(|p| match p {
                    crate::ast::BehavioralProperty::Pure => "pure",
                    crate::ast::BehavioralProperty::Total => "total",
                    crate::ast::BehavioralProperty::Idempotent => "idempotent",
                    crate::ast::BehavioralProperty::Deterministic => "deterministic",
                    crate::ast::BehavioralProperty::Commutative => "commutative",
                })
                .collect::<Vec<_>>()
                .join(", "))
        };
        let type_sig = format!("fn {}({}): {}{}{}", f.name.name, param_types.join(", "), return_type, properties, effects);
        let mut symbol = self.new_symbol(
            f.name.name.clone(),
            SymbolKind::Function,
            f.name.span,
            Some(type_sig),
            is_public,
        );
        symbol.documentation = f.doc.clone();
        let fn_id = self.add_symbol(scope_idx, symbol);
        self.add_reference(fn_id, f.name.span, true, false);
        // Create scope for function body
        let body_span = f.body.span();
        let fn_scope = self.push_scope(scope_idx, body_span);
        // Add type parameters
        for tp in &f.type_params {
            let symbol = self.new_symbol(
                tp.name.clone(),
                SymbolKind::TypeParameter,
                tp.span,
                None,
                false,
            );
            self.add_symbol(fn_scope, symbol);
        }
        // Add parameters
        for param in &f.params {
            let type_sig = self.type_expr_to_string(&param.typ);
            let symbol = self.new_symbol(
                param.name.name.clone(),
                SymbolKind::Parameter,
                param.name.span,
                Some(type_sig),
                false,
            );
            self.add_symbol(fn_scope, symbol);
        }
        // Visit body
        self.visit_expr(&f.body, fn_scope);
    }
    fn visit_type_decl(&mut self, t: &TypeDecl, scope_idx: usize) {
        let is_public = matches!(t.visibility, Visibility::Public);
        let type_sig = format!("type {}", t.name.name);
        let mut symbol = self.new_symbol(
            t.name.name.clone(),
            SymbolKind::Type,
            t.name.span,
            Some(type_sig),
            is_public,
        );
        symbol.documentation = t.doc.clone();
        let type_id = self.add_symbol(scope_idx, symbol);
        self.add_reference(type_id, t.name.span, true, false);
        // Add variants
        match &t.definition {
            TypeDef::Enum(variants) => {
                for variant in variants {
                    let mut var_symbol = self.new_symbol(
                        variant.name.name.clone(),
                        SymbolKind::Variant,
                        variant.name.span,
                        None,
                        is_public,
                    );
                    var_symbol.parent = Some(type_id);
                    self.add_symbol(scope_idx, var_symbol);
                }
            }
            TypeDef::Record(fields) => {
                for field in fields {
                    let mut field_symbol = self.new_symbol(
                        field.name.name.clone(),
                        SymbolKind::Field,
                        field.name.span,
                        Some(self.type_expr_to_string(&field.typ)),
                        is_public,
                    );
                    field_symbol.parent = Some(type_id);
                    self.add_symbol(scope_idx, field_symbol);
                }
            }
            TypeDef::Alias(_) => {}
        }
    }
    fn visit_effect(&mut self, e: &EffectDecl, scope_idx: usize) {
        let is_public = true; // Effects are typically public
        let type_sig = format!("effect {}", e.name.name);
        let mut symbol = self.new_symbol(
            e.name.name.clone(),
            SymbolKind::Effect,
            e.name.span,
            Some(type_sig),
            is_public,
        );
        symbol.documentation = e.doc.clone();
        let effect_id = self.add_symbol(scope_idx, symbol);
        // Add operations
        for op in &e.operations {
            let param_types: Vec<String> = op.params.iter()
                .map(|p| format!("{}: {}", p.name.name, self.type_expr_to_string(&p.typ)))
                .collect();
            let return_type = self.type_expr_to_string(&op.return_type);
            let op_sig = format!("fn {}({}): {}", op.name.name, param_types.join(", "), return_type);
            let mut op_symbol = self.new_symbol(
                op.name.name.clone(),
                SymbolKind::EffectOperation,
                op.name.span,
                Some(op_sig),
                is_public,
            );
            op_symbol.parent = Some(effect_id);
            self.add_symbol(scope_idx, op_symbol);
        }
    }
    fn visit_handler(&mut self, _h: &HandlerDecl, _scope_idx: usize) {
        // Handlers are complex - visit their implementations
    }
    fn visit_trait(&mut self, t: &TraitDecl, scope_idx: usize) {
        let is_public = matches!(t.visibility, Visibility::Public);
        let type_sig = format!("trait {}", t.name.name);
        let mut symbol = self.new_symbol(
            t.name.name.clone(),
            SymbolKind::Type, // Traits are like types
            t.name.span,
            Some(type_sig),
            is_public,
        );
        symbol.documentation = t.doc.clone();
        self.add_symbol(scope_idx, symbol);
    }
    fn visit_impl(&mut self, _i: &ImplDecl, _scope_idx: usize) {
        // Impl blocks add methods to types
    }
    fn visit_expr(&mut self, expr: &Expr, scope_idx: usize) {
        match expr {
            Expr::Var(ident) => {
                // Look up the identifier and add a reference
                if let Some(id) = self.lookup(&ident.name, scope_idx) {
                    self.add_reference(id, ident.span, false, false);
                }
            }
            Expr::Let { name, value, body, span, .. } => {
                // Visit the value first
                self.visit_expr(value, scope_idx);
                // Create a new scope for the let binding
                let let_scope = self.push_scope(scope_idx, *span);
                // Add the variable
                let symbol = self.new_symbol(
                    name.name.clone(),
                    SymbolKind::Variable,
                    name.span,
                    None,
                    false,
                );
                let var_id = self.add_symbol(let_scope, symbol);
                self.add_reference(var_id, name.span, true, true);
                // Visit the body
                self.visit_expr(body, let_scope);
            }
            Expr::Lambda { params, body, span, .. } => {
                let lambda_scope = self.push_scope(scope_idx, *span);
                for param in params {
                    let symbol = self.new_symbol(
                        param.name.name.clone(),
                        SymbolKind::Parameter,
                        param.name.span,
                        None,
                        false,
                    );
                    self.add_symbol(lambda_scope, symbol);
                }
                self.visit_expr(body, lambda_scope);
            }
            Expr::Call { func, args, .. } => {
                self.visit_expr(func, scope_idx);
                for arg in args {
                    self.visit_expr(arg, scope_idx);
                }
            }
            Expr::EffectOp { args, .. } => {
                for arg in args {
                    self.visit_expr(arg, scope_idx);
                }
            }
            Expr::Field { object, .. } | Expr::TupleIndex { object, .. } => {
                self.visit_expr(object, scope_idx);
            }
            Expr::If { condition, then_branch, else_branch, .. } => {
                self.visit_expr(condition, scope_idx);
                self.visit_expr(then_branch, scope_idx);
                self.visit_expr(else_branch, scope_idx);
            }
            Expr::Match { scrutinee, arms, .. } => {
                self.visit_expr(scrutinee, scope_idx);
                for arm in arms {
                    // Each arm may bind variables
                    let arm_scope = self.push_scope(scope_idx, arm.body.span());
                    self.visit_pattern(&arm.pattern, arm_scope);
                    if let Some(ref guard) = arm.guard {
                        self.visit_expr(guard, arm_scope);
                    }
                    self.visit_expr(&arm.body, arm_scope);
                }
            }
            Expr::Block { statements, result, .. } => {
                for stmt in statements {
                    self.visit_statement(stmt, scope_idx);
                }
                self.visit_expr(result, scope_idx);
            }
            Expr::BinaryOp { left, right, .. } => {
                self.visit_expr(left, scope_idx);
                self.visit_expr(right, scope_idx);
            }
            Expr::UnaryOp { operand, .. } => {
                self.visit_expr(operand, scope_idx);
            }
            Expr::List { elements, .. } => {
                for e in elements {
                    self.visit_expr(e, scope_idx);
                }
            }
            Expr::Tuple { elements, .. } => {
                for e in elements {
                    self.visit_expr(e, scope_idx);
                }
            }
            Expr::Record { spread, fields, .. } => {
                if let Some(spread_expr) = spread {
                    self.visit_expr(spread_expr, scope_idx);
                }
                for (_, e) in fields {
                    self.visit_expr(e, scope_idx);
                }
            }
            Expr::Run { expr, handlers, .. } => {
                self.visit_expr(expr, scope_idx);
                for (_effect, handler_expr) in handlers {
                    self.visit_expr(handler_expr, scope_idx);
                }
            }
            Expr::Resume { value, .. } => {
                self.visit_expr(value, scope_idx);
            }
            // Literals don't need symbol resolution
            Expr::Literal(_) => {}
        }
    }
    fn visit_statement(&mut self, stmt: &Statement, scope_idx: usize) {
        match stmt {
            Statement::Expr(e) => self.visit_expr(e, scope_idx),
            Statement::Let { name, value, .. } => {
                self.visit_expr(value, scope_idx);
                let symbol = self.new_symbol(
                    name.name.clone(),
                    SymbolKind::Variable,
                    name.span,
                    None,
                    false,
                );
                let id = self.add_symbol(scope_idx, symbol);
                self.add_reference(id, name.span, true, true);
            }
        }
    }
    fn visit_pattern(&mut self, pattern: &Pattern, scope_idx: usize) {
        match pattern {
            Pattern::Var(ident) => {
                let symbol = self.new_symbol(
                    ident.name.clone(),
                    SymbolKind::Variable,
                    ident.span,
                    None,
                    false,
                );
                let id = self.add_symbol(scope_idx, symbol);
                self.add_reference(id, ident.span, true, true);
            }
            Pattern::Constructor { fields, .. } => {
                for p in fields {
                    self.visit_pattern(p, scope_idx);
                }
            }
            Pattern::Tuple { elements, .. } => {
                for p in elements {
                    self.visit_pattern(p, scope_idx);
                }
            }
            Pattern::Record { fields, .. } => {
                for (_, p) in fields {
                    self.visit_pattern(p, scope_idx);
                }
            }
            Pattern::Wildcard(_) => {}
            Pattern::Literal(_) => {}
        }
    }
    fn type_expr_to_string(&self, typ: &TypeExpr) -> String {
        match typ {
            TypeExpr::Named(ident) => ident.name.clone(),
            TypeExpr::App(base, args) => {
                let base_str = self.type_expr_to_string(base);
                if args.is_empty() {
                    base_str
                } else {
                    let args_str: Vec<String> = args.iter()
                        .map(|a| self.type_expr_to_string(a))
                        .collect();
                    format!("{}<{}>", base_str, args_str.join(", "))
                }
            }
            TypeExpr::Function { params, return_type, .. } => {
                let params_str: Vec<String> = params.iter()
                    .map(|p| self.type_expr_to_string(p))
                    .collect();
                format!("fn({}): {}", params_str.join(", "), self.type_expr_to_string(return_type))
            }
            TypeExpr::Tuple(types) => {
                let types_str: Vec<String> = types.iter()
                    .map(|t| self.type_expr_to_string(t))
                    .collect();
                format!("({})", types_str.join(", "))
            }
            TypeExpr::Record(fields) => {
                let fields_str: Vec<String> = fields.iter()
                    .map(|f| format!("{}: {}", f.name, self.type_expr_to_string(&f.typ)))
                    .collect();
                format!("{{ {} }}", fields_str.join(", "))
            }
            TypeExpr::Unit => "Unit".to_string(),
            TypeExpr::Versioned { base, .. } => {
                format!("{}@versioned", self.type_expr_to_string(base))
            }
        }
    }
 }
 impl Default for SymbolTable {
    fn default() -> Self {
        Self::new()
    }
 }
 #[cfg(test)]
 mod tests {
    use super::*;
    use crate::parser::Parser;
    #[test]
    fn test_symbol_table_basic() {
        let source = r#"
            fn add(a: Int, b: Int): Int = a + b
            let x = 42
        "#;
        let program = Parser::parse_source(source).unwrap();
        let table = SymbolTable::build(&program);
        // Should have add function and x variable
        let globals = table.global_symbols();
        assert!(globals.iter().any(|s| s.name == "add"));
        assert!(globals.iter().any(|s| s.name == "x"));
    }
    #[test]
    fn test_symbol_lookup() {
        let source = r#"
            fn foo(x: Int): Int = x + 1
        "#;
        let program = Parser::parse_source(source).unwrap();
        let table = SymbolTable::build(&program);
        // Should be able to find foo
        assert!(table.lookup("foo", 0).is_some());
    }
 }
--- a/src/typechecker.rs
+++ b/src/typechecker.rs
@@ -335,11 +335,14 @@ fn references_params(expr: &Expr, params: &[&str]) -> bool {
                Statement::Expr(e) => references_params(e, params),
            }) || references_params(result, params)
        }
-        Expr::Field { object, .. } => references_params(object, params),
+        Expr::Field { object, .. } | Expr::TupleIndex { object, .. } => references_params(object, params),
        Expr::Lambda { body, .. } => references_params(body, params),
        Expr::Tuple { elements, .. } => elements.iter().any(|e| references_params(e, params)),
        Expr::List { elements, .. } => elements.iter().any(|e| references_params(e, params)),
-        Expr::Record { fields, .. } => fields.iter().any(|(_, e)| references_params(e, params)),
+        Expr::Record { spread, fields, .. } => {
            spread.as_ref().is_some_and(|s| references_params(s, params))
                || fields.iter().any(|(_, e)| references_params(e, params))
        }
        Expr::Match { scrutinee, arms, .. } => {
            references_params(scrutinee, params)
                || arms.iter().any(|a| references_params(&a.body, params))
@@ -516,10 +519,11 @@ fn has_recursive_calls(func_name: &str, body: &Expr) -> bool {
        Expr::Tuple { elements, .. } | Expr::List { elements, .. } => {
            elements.iter().any(|e| has_recursive_calls(func_name, e))
        }
-        Expr::Record { fields, .. } => {
+        Expr::Record { spread, fields, .. } => {
-            fields.iter().any(|(_, e)| has_recursive_calls(func_name, e))
+            spread.as_ref().is_some_and(|s| has_recursive_calls(func_name, s))
                || fields.iter().any(|(_, e)| has_recursive_calls(func_name, e))
        }
-        Expr::Field { object, .. } => has_recursive_calls(func_name, object),
+        Expr::Field { object, .. } | Expr::TupleIndex { object, .. } => has_recursive_calls(func_name, object),
        Expr::Let { value, body, .. } => {
            has_recursive_calls(func_name, value) || has_recursive_calls(func_name, body)
        }
@@ -672,6 +676,7 @@ fn generate_auto_migration_expr(
    // Build the record expression
    Some(Expr::Record {
        spread: None,
        fields: field_exprs,
        span,
    })
@@ -759,6 +764,17 @@ impl TypeChecker {
        self.env.bindings.get(name)
    }
    /// Get the inferred type of a binding as a display string (for LSP inlay hints)
    pub fn get_inferred_type(&self, name: &str) -> Option<String> {
        let scheme = self.env.bindings.get(name)?;
        let type_str = scheme.typ.to_string();
        // Skip unhelpful types
        if type_str == "<error>" || type_str.contains('?') {
            return None;
        }
        Some(type_str)
    }
    /// Get auto-generated migrations from type checking
    /// Returns: type_name -> from_version -> migration_body
    pub fn get_auto_migrations(&self) -> &HashMap<String, HashMap<u32, Expr>> {
@@ -1525,7 +1541,7 @@ impl TypeChecker {
        // Use the declared type if present, otherwise use inferred
        let final_type = if let Some(ref type_expr) = let_decl.typ {
            let declared = self.resolve_type(type_expr);
-            if let Err(e) = unify(&inferred, &declared) {
+            if let Err(e) = unify_with_env(&inferred, &declared, &self.env) {
                self.errors.push(TypeError {
                    message: format!(
                        "Variable '{}' has type {}, but declared type is {}: {}",
@@ -1662,6 +1678,42 @@ impl TypeChecker {
                span,
            } => self.infer_field(object, field, *span),
            Expr::TupleIndex {
                object,
                index,
                span,
            } => {
                let object_type = self.infer_expr(object);
                match &object_type {
                    Type::Tuple(types) => {
                        if *index < types.len() {
                            types[*index].clone()
                        } else {
                            self.errors.push(TypeError {
                                message: format!(
                                    "Tuple index {} out of bounds for tuple with {} elements",
                                    index,
                                    types.len()
                                ),
                                span: *span,
                            });
                            Type::Error
                        }
                    }
                    Type::Var(_) => Type::var(),
                    _ => {
                        self.errors.push(TypeError {
                            message: format!(
                                "Cannot use tuple index on non-tuple type {}",
                                object_type
                            ),
                            span: *span,
                        });
                        Type::Error
                    }
                }
            }
            Expr::Lambda {
                params,
                return_type,
@@ -1697,7 +1749,11 @@ impl TypeChecker {
                span,
            } => self.infer_block(statements, result, *span),
-            Expr::Record { fields, span } => self.infer_record(fields, *span),
+            Expr::Record {
                spread,
                fields,
                span,
            } => self.infer_record(spread.as_deref(), fields, *span),
            Expr::Tuple { elements, span } => self.infer_tuple(elements, *span),
@@ -1736,7 +1792,7 @@ impl TypeChecker {
        match op {
            BinaryOp::Add => {
                // Add supports both numeric types and string concatenation
-                if let Err(e) = unify(&left_type, &right_type) {
+                if let Err(e) = unify_with_env(&left_type, &right_type, &self.env) {
                    self.errors.push(TypeError {
                        message: format!("Operands of '{}' must have same type: {}", op, e),
                        span,
@@ -1757,9 +1813,32 @@ impl TypeChecker {
                }
            }
            BinaryOp::Concat => {
                // Concat (++) supports strings and lists
                if let Err(e) = unify_with_env(&left_type, &right_type, &self.env) {
                    self.errors.push(TypeError {
                        message: format!("Operands of '++' must have same type: {}", e),
                        span,
                    });
                }
                match &left_type {
                    Type::String | Type::List(_) | Type::Var(_) => left_type,
                    _ => {
                        self.errors.push(TypeError {
                            message: format!(
                                "Operator '++' requires String or List operands, got {}",
                                left_type
                            ),
                            span,
                        });
                        Type::Error
                    }
                }
            }
            BinaryOp::Sub | BinaryOp::Mul | BinaryOp::Div | BinaryOp::Mod => {
                // Arithmetic: both operands must be same numeric type
-                if let Err(e) = unify(&left_type, &right_type) {
+                if let Err(e) = unify_with_env(&left_type, &right_type, &self.env) {
                    self.errors.push(TypeError {
                        message: format!("Operands of '{}' must have same type: {}", op, e),
                        span,
@@ -1783,7 +1862,7 @@ impl TypeChecker {
            BinaryOp::Eq | BinaryOp::Ne => {
                // Equality: operands must have same type
-                if let Err(e) = unify(&left_type, &right_type) {
+                if let Err(e) = unify_with_env(&left_type, &right_type, &self.env) {
                    self.errors.push(TypeError {
                        message: format!("Operands of '{}' must have same type: {}", op, e),
                        span,
@@ -1794,7 +1873,7 @@ impl TypeChecker {
            BinaryOp::Lt | BinaryOp::Le | BinaryOp::Gt | BinaryOp::Ge => {
                // Comparison: operands must be same orderable type
-                if let Err(e) = unify(&left_type, &right_type) {
+                if let Err(e) = unify_with_env(&left_type, &right_type, &self.env) {
                    self.errors.push(TypeError {
                        message: format!("Operands of '{}' must have same type: {}", op, e),
                        span,
@@ -1805,13 +1884,13 @@ impl TypeChecker {
            BinaryOp::And | BinaryOp::Or => {
                // Logical: both must be Bool
-                if let Err(e) = unify(&left_type, &Type::Bool) {
+                if let Err(e) = unify_with_env(&left_type, &Type::Bool, &self.env) {
                    self.errors.push(TypeError {
                        message: format!("Left operand of '{}' must be Bool: {}", op, e),
                        span: left.span(),
                    });
                }
-                if let Err(e) = unify(&right_type, &Type::Bool) {
+                if let Err(e) = unify_with_env(&right_type, &Type::Bool, &self.env) {
                    self.errors.push(TypeError {
                        message: format!("Right operand of '{}' must be Bool: {}", op, e),
                        span: right.span(),
@@ -1825,7 +1904,7 @@ impl TypeChecker {
                // right must be a function that accepts left's type
                let result_type = Type::var();
                let expected_fn = Type::function(vec![left_type.clone()], result_type.clone());
-                if let Err(e) = unify(&right_type, &expected_fn) {
+                if let Err(e) = unify_with_env(&right_type, &expected_fn, &self.env) {
                    self.errors.push(TypeError {
                        message: format!(
                            "Pipe target must be a function accepting {}: {}",
@@ -1857,7 +1936,7 @@ impl TypeChecker {
                }
            },
            UnaryOp::Not => {
-                if let Err(e) = unify(&operand_type, &Type::Bool) {
+                if let Err(e) = unify_with_env(&operand_type, &Type::Bool, &self.env) {
                    self.errors.push(TypeError {
                        message: format!("Operator '!' requires Bool operand: {}", e),
                        span,
@@ -1908,7 +1987,7 @@ impl TypeChecker {
            self.current_effects.clone(),
        );
-        match unify(&func_type, &expected_fn) {
+        match unify_with_env(&func_type, &expected_fn, &self.env) {
            Ok(subst) => result_type.apply(&subst),
            Err(e) => {
                // Provide more detailed error message based on the type of mismatch
@@ -1985,7 +2064,7 @@ impl TypeChecker {
                    let result_type = Type::var();
                    let expected_fn = Type::function(arg_types, result_type.clone());
-                    if let Err(e) = unify(field_type, &expected_fn) {
+                    if let Err(e) = unify_with_env(field_type, &expected_fn, &self.env) {
                        self.errors.push(TypeError {
                            message: format!(
                                "Type mismatch in {}.{} call: {}",
@@ -2057,7 +2136,7 @@ impl TypeChecker {
                for (i, (arg_type, (_, param_type))) in
                    arg_types.iter().zip(op.params.iter()).enumerate()
                {
-                    if let Err(e) = unify(arg_type, param_type) {
+                    if let Err(e) = unify_with_env(arg_type, param_type, &self.env) {
                        self.errors.push(TypeError {
                            message: format!(
                                "Argument {} of '{}.{}' has type {}, expected {}: {}",
@@ -2090,6 +2169,7 @@ impl TypeChecker {
    fn infer_field(&mut self, object: &Expr, field: &Ident, span: Span) -> Type {
        let object_type = self.infer_expr(object);
        let object_type = self.env.expand_type_alias(&object_type);
        match &object_type {
            Type::Record(fields) => match fields.iter().find(|(n, _)| n == &field.name) {
@@ -2170,7 +2250,7 @@ impl TypeChecker {
        // Check return type if specified
        let ret_type = if let Some(rt) = return_type {
            let declared = self.resolve_type(rt);
-            if let Err(e) = unify(&body_type, &declared) {
+            if let Err(e) = unify_with_env(&body_type, &declared, &self.env) {
                self.errors.push(TypeError {
                    message: format!(
                        "Lambda body type {} doesn't match declared {}: {}",
@@ -2236,7 +2316,7 @@ impl TypeChecker {
        span: Span,
    ) -> Type {
        let cond_type = self.infer_expr(condition);
-        if let Err(e) = unify(&cond_type, &Type::Bool) {
+        if let Err(e) = unify_with_env(&cond_type, &Type::Bool, &self.env) {
            self.errors.push(TypeError {
                message: format!("If condition must be Bool, got {}: {}", cond_type, e),
                span: condition.span(),
@@ -2246,7 +2326,7 @@ impl TypeChecker {
        let then_type = self.infer_expr(then_branch);
        let else_type = self.infer_expr(else_branch);
-        match unify(&then_type, &else_type) {
+        match unify_with_env(&then_type, &else_type, &self.env) {
            Ok(subst) => then_type.apply(&subst),
            Err(e) => {
                self.errors.push(TypeError {
@@ -2287,7 +2367,7 @@ impl TypeChecker {
            // Check guard if present
            if let Some(ref guard) = arm.guard {
                let guard_type = self.infer_expr(guard);
-                if let Err(e) = unify(&guard_type, &Type::Bool) {
+                if let Err(e) = unify_with_env(&guard_type, &Type::Bool, &self.env) {
                    self.errors.push(TypeError {
                        message: format!("Match guard must be Bool: {}", e),
                        span: guard.span(),
@@ -2303,7 +2383,7 @@ impl TypeChecker {
            match &result_type {
                None => result_type = Some(body_type),
                Some(prev) => {
-                    if let Err(e) = unify(prev, &body_type) {
+                    if let Err(e) = unify_with_env(prev, &body_type, &self.env) {
                        self.errors.push(TypeError {
                            message: format!(
                                "Match arm has incompatible type: expected {}, got {}: {}",
@@ -2353,7 +2433,7 @@ impl TypeChecker {
            Pattern::Literal(lit) => {
                let lit_type = self.infer_literal(lit);
-                if let Err(e) = unify(&lit_type, expected) {
+                if let Err(e) = unify_with_env(&lit_type, expected, &self.env) {
                    self.errors.push(TypeError {
                        message: format!("Pattern literal type mismatch: {}", e),
                        span: lit.span,
@@ -2367,7 +2447,7 @@ impl TypeChecker {
                // For now, handle Option specially
                match name.name.as_str() {
                    "None" => {
-                        if let Err(e) = unify(expected, &Type::Option(Box::new(Type::var()))) {
+                        if let Err(e) = unify_with_env(expected, &Type::Option(Box::new(Type::var())), &self.env) {
                            self.errors.push(TypeError {
                                message: format!(
                                    "None pattern doesn't match type {}: {}",
@@ -2380,7 +2460,7 @@ impl TypeChecker {
                    }
                    "Some" => {
                        let inner_type = Type::var();
-                        if let Err(e) = unify(expected, &Type::Option(Box::new(inner_type.clone())))
+                        if let Err(e) = unify_with_env(expected, &Type::Option(Box::new(inner_type.clone())), &self.env)
                        {
                            self.errors.push(TypeError {
                                message: format!(
@@ -2409,7 +2489,7 @@ impl TypeChecker {
            Pattern::Tuple { elements, span } => {
                let element_types: Vec<Type> = elements.iter().map(|_| Type::var()).collect();
-                if let Err(e) = unify(expected, &Type::Tuple(element_types.clone())) {
+                if let Err(e) = unify_with_env(expected, &Type::Tuple(element_types.clone()), &self.env) {
                    self.errors.push(TypeError {
                        message: format!("Tuple pattern doesn't match type {}: {}", expected, e),
                        span: *span,
@@ -2459,7 +2539,7 @@ impl TypeChecker {
                    if let Some(type_expr) = typ {
                        let declared = self.resolve_type(type_expr);
-                        if let Err(e) = unify(&value_type, &declared) {
+                        if let Err(e) = unify_with_env(&value_type, &declared, &self.env) {
                            self.errors.push(TypeError {
                                message: format!(
                                    "Variable '{}' has type {}, but declared type is {}: {}",
@@ -2480,12 +2560,46 @@ impl TypeChecker {
        self.infer_expr(result)
    }
-    fn infer_record(&mut self, fields: &[(Ident, Expr)], _span: Span) -> Type {
+    fn infer_record(
-        let field_types: Vec<(String, Type)> = fields
+        &mut self,
        spread: Option<&Expr>,
        fields: &[(Ident, Expr)],
        span: Span,
    ) -> Type {
        // Start with spread fields if present
        let mut field_types: Vec<(String, Type)> = if let Some(spread_expr) = spread {
            let spread_type = self.infer_expr(spread_expr);
            match spread_type {
                Type::Record(spread_fields) => spread_fields,
                _ => {
                    self.errors.push(TypeError {
                        message: format!(
                            "Spread expression must be a record type, got {}",
                            spread_type
                        ),
                        span,
                    });
                    Vec::new()
                }
            }
        } else {
            Vec::new()
        };
        // Apply explicit field overrides
        let explicit_types: Vec<(String, Type)> = fields
            .iter()
            .map(|(name, expr)| (name.name.clone(), self.infer_expr(expr)))
            .collect();
        for (name, typ) in explicit_types {
            if let Some(existing) = field_types.iter_mut().find(|(n, _)| n == &name) {
                existing.1 = typ;
            } else {
                field_types.push((name, typ));
            }
        }
        Type::Record(field_types)
    }
@@ -2502,7 +2616,7 @@ impl TypeChecker {
        let first_type = self.infer_expr(&elements[0]);
        for elem in &elements[1..] {
            let elem_type = self.infer_expr(elem);
-            if let Err(e) = unify(&first_type, &elem_type) {
+            if let Err(e) = unify_with_env(&first_type, &elem_type, &self.env) {
                self.errors.push(TypeError {
                    message: format!("List elements must have same type: {}", e),
                    span,
@@ -2808,7 +2922,7 @@ impl TypeChecker {
            // Check return type matches if specified
            if let Some(ref return_type_expr) = impl_method.return_type {
                let return_type = self.resolve_type(return_type_expr);
-                if let Err(e) = unify(&body_type, &return_type) {
+                if let Err(e) = unify_with_env(&body_type, &return_type, &self.env) {
                    self.errors.push(TypeError {
                        message: format!(
                            "Method '{}' body has type {}, but declared return type is {}: {}",
--- a/src/types.rs
+++ b/src/types.rs
@@ -1146,6 +1146,15 @@ impl TypeEnv {
                        ],
                        return_type: Type::Unit,
                    },
                    EffectOpDef {
                        name: "assertEqualMsg".to_string(),
                        params: vec![
                            ("expected".to_string(), Type::Var(0)),
                            ("actual".to_string(), Type::Var(0)),
                            ("label".to_string(), Type::String),
                        ],
                        return_type: Type::Unit,
                    },
                    EffectOpDef {
                        name: "assertNotEqual".to_string(),
                        params: vec![
@@ -1173,6 +1182,110 @@ impl TypeEnv {
            },
        );
        // Add Concurrent effect for concurrent/parallel execution
        // Task is represented as Int (task ID)
        env.effects.insert(
            "Concurrent".to_string(),
            EffectDef {
                name: "Concurrent".to_string(),
                type_params: Vec::new(),
                operations: vec![
                    // Spawn a new concurrent task that returns a value
                    // Returns a Task<A> (represented as Int task ID)
                    EffectOpDef {
                        name: "spawn".to_string(),
                        params: vec![("thunk".to_string(), Type::Function {
                            params: Vec::new(),
                            return_type: Box::new(Type::Var(0)),
                            effects: EffectSet::empty(),
                            properties: PropertySet::empty(),
                        })],
                        return_type: Type::Int, // Task ID
                    },
                    // Wait for a task to complete and get its result
                    EffectOpDef {
                        name: "await".to_string(),
                        params: vec![("task".to_string(), Type::Int)],
                        return_type: Type::Var(0),
                    },
                    // Yield control to allow other tasks to run
                    EffectOpDef {
                        name: "yield".to_string(),
                        params: Vec::new(),
                        return_type: Type::Unit,
                    },
                    // Sleep for milliseconds (non-blocking to other tasks)
                    EffectOpDef {
                        name: "sleep".to_string(),
                        params: vec![("ms".to_string(), Type::Int)],
                        return_type: Type::Unit,
                    },
                    // Cancel a running task
                    EffectOpDef {
                        name: "cancel".to_string(),
                        params: vec![("task".to_string(), Type::Int)],
                        return_type: Type::Bool,
                    },
                    // Check if a task is still running
                    EffectOpDef {
                        name: "isRunning".to_string(),
                        params: vec![("task".to_string(), Type::Int)],
                        return_type: Type::Bool,
                    },
                    // Get the number of active tasks
                    EffectOpDef {
                        name: "taskCount".to_string(),
                        params: Vec::new(),
                        return_type: Type::Int,
                    },
                ],
            },
        );
        // Add Channel effect for concurrent communication
        env.effects.insert(
            "Channel".to_string(),
            EffectDef {
                name: "Channel".to_string(),
                type_params: Vec::new(),
                operations: vec![
                    // Create a new channel, returns channel ID
                    EffectOpDef {
                        name: "create".to_string(),
                        params: Vec::new(),
                        return_type: Type::Int, // Channel ID
                    },
                    // Send a value on a channel
                    EffectOpDef {
                        name: "send".to_string(),
                        params: vec![
                            ("channel".to_string(), Type::Int),
                            ("value".to_string(), Type::Var(0)),
                        ],
                        return_type: Type::Unit,
                    },
                    // Receive a value from a channel (blocks until available)
                    EffectOpDef {
                        name: "receive".to_string(),
                        params: vec![("channel".to_string(), Type::Int)],
                        return_type: Type::Var(0),
                    },
                    // Try to receive (non-blocking, returns Option)
                    EffectOpDef {
                        name: "tryReceive".to_string(),
                        params: vec![("channel".to_string(), Type::Int)],
                        return_type: Type::Option(Box::new(Type::Var(0))),
                    },
                    // Close a channel
                    EffectOpDef {
                        name: "close".to_string(),
                        params: vec![("channel".to_string(), Type::Int)],
                        return_type: Type::Unit,
                    },
                ],
            },
        );
        // Add Sql effect for database access
        // Connection is represented as Int (connection ID)
        let row_type = Type::Record(vec![]); // Dynamic record type
@@ -1495,6 +1608,14 @@ impl TypeEnv {
                "parseFloat".to_string(),
                Type::function(vec![Type::String], Type::Option(Box::new(Type::Float))),
            ),
            (
                "indexOf".to_string(),
                Type::function(vec![Type::String, Type::String], Type::Option(Box::new(Type::Int))),
            ),
            (
                "lastIndexOf".to_string(),
                Type::function(vec![Type::String, Type::String], Type::Option(Box::new(Type::Int))),
            ),
        ]);
        env.bind("String", TypeScheme::mono(string_module_type));
@@ -1766,9 +1887,39 @@ impl TypeEnv {
                "round".to_string(),
                Type::function(vec![Type::var()], Type::Int),
            ),
            (
                "sin".to_string(),
                Type::function(vec![Type::Float], Type::Float),
            ),
            (
                "cos".to_string(),
                Type::function(vec![Type::Float], Type::Float),
            ),
            (
                "atan2".to_string(),
                Type::function(vec![Type::Float, Type::Float], Type::Float),
            ),
        ]);
        env.bind("Math", TypeScheme::mono(math_module_type));
        // Int module
        let int_module_type = Type::Record(vec![
            (
                "toString".to_string(),
                Type::function(vec![Type::Int], Type::String),
            ),
        ]);
        env.bind("Int", TypeScheme::mono(int_module_type));
        // Float module
        let float_module_type = Type::Record(vec![
            (
                "toString".to_string(),
                Type::function(vec![Type::Float], Type::String),
            ),
        ]);
        env.bind("Float", TypeScheme::mono(float_module_type));
        env
    }
@@ -1928,7 +2079,9 @@ pub fn unify(t1: &Type, t2: &Type) -> Result<Substitution, String> {
            // Function's required effects (e1) must be a subset of available effects (e2)
            // A pure function (empty effects) can be called anywhere
            // A function requiring {Logger} can be called in context with {Logger} or {Logger, Console}
-            if !e1.is_subset(&e2) {
+            // When expected effects (e2) are empty, it means "no constraint" (e.g., callback parameter)
            // so we allow any actual effects through
            if !e2.is_empty() && !e1.is_subset(&e2) {
                return Err(format!(
                    "Effect mismatch: expected {{{}}}, got {{{}}}",
                    e1, e2
--- a/stdlib/http.lux
+++ b/stdlib/http.lux
@@ -42,6 +42,10 @@ fn httpNotFound(body: String): { status: Int, body: String } =
 fn httpServerError(body: String): { status: Int, body: String } =
    { status: 500, body: body }
 // Create a 429 Too Many Requests response
 fn httpTooManyRequests(body: String): { status: Int, body: String } =
    { status: 429, body: body }
 // ============================================================
 // Path Matching
 // ============================================================
@@ -84,6 +88,54 @@ fn getPathSegment(path: String, index: Int): Option<String> = {
    List.get(parts, index + 1)
 }
 // Extract path parameters from a matched route pattern
 // For path "/users/42/posts/5" and pattern "/users/:userId/posts/:postId"
 // returns [("userId", "42"), ("postId", "5")]
 fn getPathParams(path: String, pattern: String): List<(String, String)> = {
    let pathParts = String.split(path, "/")
    let patternParts = String.split(pattern, "/")
    extractParamsHelper(pathParts, patternParts, [])
 }
 fn extractParamsHelper(pathParts: List<String>, patternParts: List<String>, acc: List<(String, String)>): List<(String, String)> = {
    if List.length(pathParts) == 0 || List.length(patternParts) == 0 then
        List.reverse(acc)
    else {
        match List.head(pathParts) {
            None => List.reverse(acc),
            Some(p) => match List.head(patternParts) {
                None => List.reverse(acc),
                Some(pat) => {
                    let restPath = Option.getOrElse(List.tail(pathParts), [])
                    let restPattern = Option.getOrElse(List.tail(patternParts), [])
                    if String.startsWith(pat, ":") then {
                        let paramName = String.substring(pat, 1, String.length(pat))
                        let newAcc = List.concat([(paramName, p)], acc)
                        extractParamsHelper(restPath, restPattern, newAcc)
                    } else {
                        extractParamsHelper(restPath, restPattern, acc)
                    }
                }
            }
        }
    }
 }
 // Get a specific path parameter by name from a list of params
 fn getParam(params: List<(String, String)>, name: String): Option<String> = {
    if List.length(params) == 0 then None
    else {
        match List.head(params) {
            None => None,
            Some(pair) => match pair {
                (pName, pValue) =>
                    if pName == name then Some(pValue)
                    else getParam(Option.getOrElse(List.tail(params), []), name)
            }
        }
    }
 }
 // ============================================================
 // JSON Helpers
 // ============================================================
@@ -130,32 +182,483 @@ fn jsonMessage(text: String): String =
    jsonObject(jsonString("message", text))
 // ============================================================
-// Usage Example (copy into your file)
+// Header Helpers
 // ============================================================
 // Get a header value from request headers (case-insensitive)
 fn getHeader(headers: List<(String, String)>, name: String): Option<String> = {
    let lowerName = String.toLower(name)
    getHeaderHelper(headers, lowerName)
 }
 fn getHeaderHelper(headers: List<(String, String)>, lowerName: String): Option<String> = {
    if List.length(headers) == 0 then None
    else {
        match List.head(headers) {
            None => None,
            Some(header) => match header {
                (hName, hValue) =>
                    if String.toLower(hName) == lowerName then Some(hValue)
                    else getHeaderHelper(Option.getOrElse(List.tail(headers), []), lowerName)
            }
        }
    }
 }
 // ============================================================
 // Routing Helpers
 // ============================================================
 //
 // Route matching pattern:
 //
 // fn router(method: String, path: String, body: String): { status: Int, body: String } = {
-//     if method == "GET" && path == "/" then httpOk("Welcome!")
+//     if method == "GET" && path == "/" then httpOk("Home")
 //     else if method == "GET" && pathMatches(path, "/users/:id") then {
-//         match getPathSegment(path, 1) {
+//         let params = getPathParams(path, "/users/:id")
 //         match getParam(params, "id") {
 //             Some(id) => httpOk(jsonObject(jsonString("id", id))),
 //             None => httpNotFound(jsonErrorMsg("User not found"))
 //         }
 //     }
-//     else httpNotFound(jsonErrorMsg("Not found"))
+//     else if method == "POST" && path == "/users" then
 //         httpCreated(body)
 //     else
 //         httpNotFound(jsonErrorMsg("Not found"))
 // }
 // Helper to check if request is a GET to a specific path
 fn isGet(method: String, path: String, pattern: String): Bool =
    method == "GET" && pathMatches(path, pattern)
 // Helper to check if request is a POST to a specific path
 fn isPost(method: String, path: String, pattern: String): Bool =
    method == "POST" && pathMatches(path, pattern)
 // Helper to check if request is a PUT to a specific path
 fn isPut(method: String, path: String, pattern: String): Bool =
    method == "PUT" && pathMatches(path, pattern)
 // Helper to check if request is a DELETE to a specific path
 fn isDelete(method: String, path: String, pattern: String): Bool =
    method == "DELETE" && pathMatches(path, pattern)
 // ============================================================
 // Server Loop Patterns
 // ============================================================
 //
 // The server loop should be defined in your main file:
 //
 // fn serverLoop(): Unit with {HttpServer} = {
 //     let req = HttpServer.accept()
 //     let resp = router(req.method, req.path, req.body, req.headers)
 //     HttpServer.respond(resp.status, resp.body)
 //     serverLoop()
 // }
 //
-// fn main(): Unit with {Console, HttpServer} = {
+// For testing with a fixed number of requests:
 //     HttpServer.listen(8080)
 //     Console.print("Server running on port 8080")
 //     serveLoop(5)  // Handle 5 requests
 // }
 //
-// fn serveLoop(remaining: Int): Unit with {Console, HttpServer} = {
+// fn serverLoopN(remaining: Int): Unit with {HttpServer} = {
 //     if remaining <= 0 then HttpServer.stop()
 //     else {
 //         let req = HttpServer.accept()
-//         let resp = router(req.method, req.path, req.body)
+//         let resp = router(req.method, req.path, req.body, req.headers)
 //         HttpServer.respond(resp.status, resp.body)
-//         serveLoop(remaining - 1)
+//         serverLoopN(remaining - 1)
 //     }
 // }
 // ============================================================
 // Middleware Pattern
 // ============================================================
 //
 // Middleware wraps handlers to add cross-cutting concerns.
 // In Lux, middleware is implemented as function composition.
 //
 // Example logging middleware:
 //
 // fn withLogging(
 //     handler: fn(String, String, String): { status: Int, body: String }
 // ): fn(String, String, String): { status: Int, body: String } with {Console} = {
 //     fn(method: String, path: String, body: String): { status: Int, body: String } => {
 //         Console.print("[HTTP] " + method + " " + path)
 //         let response = handler(method, path, body)
 //         Console.print("[HTTP] " + toString(response.status))
 //         response
 //     }
 // }
 //
 // Usage:
 //   let myHandler = withLogging(router)
 // ============================================================
 // CORS Headers
 // ============================================================
 // Standard CORS headers for API responses
 fn corsHeaders(): List<(String, String)> = [
    ("Access-Control-Allow-Origin", "*"),
    ("Access-Control-Allow-Methods", "GET, POST, PUT, DELETE, OPTIONS"),
    ("Access-Control-Allow-Headers", "Content-Type, Authorization")
 ]
 // CORS headers for specific origin with credentials
 fn corsHeadersWithOrigin(origin: String): List<(String, String)> = [
    ("Access-Control-Allow-Origin", origin),
    ("Access-Control-Allow-Credentials", "true"),
    ("Access-Control-Allow-Methods", "GET, POST, PUT, DELETE, OPTIONS"),
    ("Access-Control-Allow-Headers", "Content-Type, Authorization")
 ]
 // ============================================================
 // Content Type Headers
 // ============================================================
 fn jsonHeaders(): List<(String, String)> = [
    ("Content-Type", "application/json")
 ]
 fn htmlHeaders(): List<(String, String)> = [
    ("Content-Type", "text/html; charset=utf-8")
 ]
 fn textHeaders(): List<(String, String)> = [
    ("Content-Type", "text/plain; charset=utf-8")
 ]
 // ============================================================
 // Query String Parsing
 // ============================================================
 // Parse query string from path (e.g., "/search?q=hello&page=1")
 // Returns the path without query string and a list of parameters
 pub fn parseQueryString(fullPath: String): (String, List<(String, String)>) = {
    match String.indexOf(fullPath, "?") {
        None => (fullPath, []),
        Some(idx) => {
            let path = String.substring(fullPath, 0, idx)
            let queryStr = String.substring(fullPath, idx + 1, String.length(fullPath))
            let params = parseQueryParams(queryStr)
            (path, params)
        }
    }
 }
 fn parseQueryParams(queryStr: String): List<(String, String)> = {
    let pairs = String.split(queryStr, "&")
    List.filterMap(pairs, fn(pair: String): Option<(String, String)> => {
        match String.indexOf(pair, "=") {
            None => None,
            Some(idx) => {
                let key = String.substring(pair, 0, idx)
                let value = String.substring(pair, idx + 1, String.length(pair))
                Some((urlDecode(key), urlDecode(value)))
            }
        }
    })
 }
 // Get a query parameter by name
 pub fn getQueryParam(params: List<(String, String)>, name: String): Option<String> =
    getParam(params, name)
 // Simple URL decoding (handles %XX and +)
 fn urlDecode(s: String): String = {
    // For now, just replace + with space
    // Full implementation would decode %XX sequences
    String.replace(s, "+", " ")
 }
 // ============================================================
 // Cookie Handling
 // ============================================================
 // Parse cookies from Cookie header value
 pub fn parseCookies(cookieHeader: String): List<(String, String)> = {
    let pairs = String.split(cookieHeader, "; ")
    List.filterMap(pairs, fn(pair: String): Option<(String, String)> => {
        match String.indexOf(pair, "=") {
            None => None,
            Some(idx) => {
                let name = String.trim(String.substring(pair, 0, idx))
                let value = String.trim(String.substring(pair, idx + 1, String.length(pair)))
                Some((name, value))
            }
        }
    })
 }
 // Get a cookie value by name from request headers
 pub fn getCookie(headers: List<(String, String)>, name: String): Option<String> = {
    match getHeader(headers, "Cookie") {
        None => None,
        Some(cookieHeader) => {
            let cookies = parseCookies(cookieHeader)
            getParam(cookies, name)
        }
    }
 }
 // Create a Set-Cookie header value
 pub fn setCookie(name: String, value: String): String =
    name + "=" + value
 // Create a Set-Cookie header with options
 pub fn setCookieWithOptions(
    name: String,
    value: String,
    maxAge: Option<Int>,
    path: Option<String>,
    httpOnly: Bool,
    secure: Bool
 ): String = {
    let base = name + "=" + value
    let withMaxAge = match maxAge {
        Some(age) => base + "; Max-Age=" + toString(age),
        None => base
    }
    let withPath = match path {
        Some(p) => withMaxAge + "; Path=" + p,
        None => withMaxAge
    }
    let withHttpOnly = if httpOnly then withPath + "; HttpOnly" else withPath
    if secure then withHttpOnly + "; Secure" else withHttpOnly
 }
 // ============================================================
 // Static File MIME Types
 // ============================================================
 // Get MIME type for a file extension
 pub fn getMimeType(path: String): String = {
    let ext = getFileExtension(path)
    match ext {
        "html" => "text/html; charset=utf-8",
        "htm" => "text/html; charset=utf-8",
        "css" => "text/css; charset=utf-8",
        "js" => "application/javascript; charset=utf-8",
        "json" => "application/json; charset=utf-8",
        "png" => "image/png",
        "jpg" => "image/jpeg",
        "jpeg" => "image/jpeg",
        "gif" => "image/gif",
        "svg" => "image/svg+xml",
        "ico" => "image/x-icon",
        "woff" => "font/woff",
        "woff2" => "font/woff2",
        "ttf" => "font/ttf",
        "pdf" => "application/pdf",
        "xml" => "application/xml",
        "txt" => "text/plain; charset=utf-8",
        "md" => "text/markdown; charset=utf-8",
        _ => "application/octet-stream"
    }
 }
 fn getFileExtension(path: String): String = {
    match String.lastIndexOf(path, ".") {
        None => "",
        Some(idx) => String.toLower(String.substring(path, idx + 1, String.length(path)))
    }
 }
 // ============================================================
 // Request Type
 // ============================================================
 // Standard request record for cleaner routing
 type Request = {
    method: String,
    path: String,
    query: List<(String, String)>,
    headers: List<(String, String)>,
    body: String
 }
 // Parse a raw request into a Request record
 pub fn parseRequest(
    method: String,
    fullPath: String,
    headers: List<(String, String)>,
    body: String
 ): Request = {
    let (path, query) = parseQueryString(fullPath)
    { method: method, path: path, query: query, headers: headers, body: body }
 }
 // ============================================================
 // Response Type with Headers
 // ============================================================
 // Response with headers support
 type Response = {
    status: Int,
    headers: List<(String, String)>,
    body: String
 }
 // Create a response with headers
 pub fn httpResponse(status: Int, body: String, headers: List<(String, String)>): Response =
    { status: status, headers: headers, body: body }
 // Create a JSON response
 pub fn jsonResponse(status: Int, body: String): Response =
    { status: status, headers: jsonHeaders(), body: body }
 // Create an HTML response
 pub fn htmlResponse(status: Int, body: String): Response =
    { status: status, headers: htmlHeaders(), body: body }
 // Create a redirect response
 pub fn httpRedirect(location: String): Response =
    { status: 302, headers: [("Location", location)], body: "" }
 // Create a permanent redirect response
 pub fn httpRedirectPermanent(location: String): Response =
    { status: 301, headers: [("Location", location)], body: "" }
 // ============================================================
 // Middleware Functions
 // ============================================================
 // Request type for middleware (simplified)
 type Handler = fn(Request): Response
 // Logging middleware - logs request method, path, and response status
 pub fn withLogging(handler: Handler): Handler with {Console} =
    fn(req: Request): Response => {
        Console.print("[HTTP] " + req.method + " " + req.path)
        let resp = handler(req)
        Console.print("[HTTP] " + toString(resp.status))
        resp
    }
 // CORS middleware - adds CORS headers to all responses
 pub fn withCors(handler: Handler): Handler =
    fn(req: Request): Response => {
        // Handle preflight
        if req.method == "OPTIONS" then
            { status: 204, headers: corsHeaders(), body: "" }
        else {
            let resp = handler(req)
            { status: resp.status, headers: List.concat(resp.headers, corsHeaders()), body: resp.body }
        }
    }
 // JSON content-type middleware - ensures JSON content type on responses
 pub fn withJson(handler: Handler): Handler =
    fn(req: Request): Response => {
        let resp = handler(req)
        { status: resp.status, headers: List.concat(resp.headers, jsonHeaders()), body: resp.body }
    }
 // Error handling middleware - catches failures and returns 500
 pub fn withErrorHandling(handler: Handler): Handler =
    fn(req: Request): Response => {
        // In a real implementation, this would use effect handling
        // For now, just call the handler
        handler(req)
    }
 // Rate limiting check (returns remaining requests or 0 if limited)
 // Note: Actual rate limiting requires state/effects
 pub fn checkRateLimit(key: String, limit: Int, window: Int): Int with {Time} = {
    // Placeholder - real implementation would track requests
    limit
 }
 // ============================================================
 // Router DSL
 // ============================================================
 // Route definition
 type Route = {
    method: String,
    pattern: String,
    handler: fn(Request): Response
 }
 // Create a GET route
 pub fn get(pattern: String, handler: fn(Request): Response): Route =
    { method: "GET", pattern: pattern, handler: handler }
 // Create a POST route
 pub fn post(pattern: String, handler: fn(Request): Response): Route =
    { method: "POST", pattern: pattern, handler: handler }
 // Create a PUT route
 pub fn put(pattern: String, handler: fn(Request): Response): Route =
    { method: "PUT", pattern: pattern, handler: handler }
 // Create a DELETE route
 pub fn delete(pattern: String, handler: fn(Request): Response): Route =
    { method: "DELETE", pattern: pattern, handler: handler }
 // Create a PATCH route
 pub fn patch(pattern: String, handler: fn(Request): Response): Route =
    { method: "PATCH", pattern: pattern, handler: handler }
 // Match request against a list of routes
 pub fn matchRoute(req: Request, routes: List<Route>): Option<(Route, List<(String, String)>)> = {
    matchRouteHelper(req, routes)
 }
 fn matchRouteHelper(req: Request, routes: List<Route>): Option<(Route, List<(String, String)>)> = {
    match List.head(routes) {
        None => None,
        Some(route) => {
            if route.method == req.method && pathMatches(req.path, route.pattern) then {
                let params = getPathParams(req.path, route.pattern)
                Some((route, params))
            } else {
                matchRouteHelper(req, Option.getOrElse(List.tail(routes), []))
            }
        }
    }
 }
 // Create a router from a list of routes
 pub fn router(routes: List<Route>, notFound: fn(Request): Response): Handler =
    fn(req: Request): Response => {
        match matchRoute(req, routes) {
            Some((route, _params)) => route.handler(req),
            None => notFound(req)
        }
    }
 // ============================================================
 // Example Usage
 // ============================================================
 //
 // fn main(): Unit with {Console, HttpServer} = {
 //     // Define routes
 //     let routes = [
 //         get("/", fn(req: Request): Response => jsonResponse(200, jsonMessage("Welcome!"))),
 //         get("/users/:id", fn(req: Request): Response => {
 //             let params = getPathParams(req.path, "/users/:id")
 //             match getParam(params, "id") {
 //                 Some(id) => jsonResponse(200, jsonObject(jsonString("id", id))),
 //                 None => jsonResponse(404, jsonErrorMsg("User not found"))
 //             }
 //         }),
 //         post("/users", fn(req: Request): Response => jsonResponse(201, jsonMessage("Created")))
 //     ]
 //
 //     // Create router with middleware
 //     let app = withLogging(withCors(router(routes, fn(req: Request): Response =>
 //         jsonResponse(404, jsonErrorMsg("Not found"))
 //     )))
 //
 //     // Start server
 //     HttpServer.listen(8080)
 //     Console.print("Server running on http://localhost:8080")
 //
 //     // Server loop
 //     fn serverLoop(): Unit with {HttpServer} = {
 //         let rawReq = HttpServer.accept()
 //         let req = parseRequest(rawReq.method, rawReq.path, rawReq.headers, rawReq.body)
 //         let resp = app(req)
 //         HttpServer.respond(resp.status, resp.body)
 //         serverLoop()
 //     }
 //     serverLoop()
 // }
--- a/stdlib/json.lux
+++ b/stdlib/json.lux
@@ -0,0 +1,473 @@
 // JSON Serialization and Deserialization for Lux
 //
 // Provides type-safe JSON encoding and decoding with support
 // for schema versioning and custom codecs.
 //
 // Usage:
 //   let json = Json.encode(user)  // Serialize to JSON string
 //   let user = Json.decode(json)  // Deserialize from JSON string
 // ============================================================
 // JSON Value Type
 // ============================================================
 // Represents any JSON value
 type JsonValue =
    | JsonNull
    | JsonBool(Bool)
    | JsonInt(Int)
    | JsonFloat(Float)
    | JsonString(String)
    | JsonArray(List<JsonValue>)
    | JsonObject(List<(String, JsonValue)>)
 // ============================================================
 // Encoding Primitives
 // ============================================================
 // Escape a string for JSON
 pub fn escapeString(s: String): String = {
    let escaped = String.replace(s, "\\", "\\\\")
    let escaped = String.replace(escaped, "\"", "\\\"")
    let escaped = String.replace(escaped, "\n", "\\n")
    let escaped = String.replace(escaped, "\r", "\\r")
    let escaped = String.replace(escaped, "\t", "\\t")
    escaped
 }
 // Encode a JsonValue to a JSON string
 pub fn encode(value: JsonValue): String = {
    match value {
        JsonNull => "null",
        JsonBool(b) => if b then "true" else "false",
        JsonInt(n) => toString(n),
        JsonFloat(f) => toString(f),
        JsonString(s) => "\"" + escapeString(s) + "\"",
        JsonArray(items) => {
            let encodedItems = List.map(items, encode)
            "[" + String.join(encodedItems, ",") + "]"
        },
        JsonObject(fields) => {
            let encodedFields = List.map(fields, fn(field: (String, JsonValue)): String => {
                match field {
                    (key, val) => "\"" + escapeString(key) + "\":" + encode(val)
                }
            })
            "{" + String.join(encodedFields, ",") + "}"
        }
    }
 }
 // Pretty-print a JsonValue with indentation
 pub fn encodePretty(value: JsonValue): String =
    encodePrettyIndent(value, 0)
 fn encodePrettyIndent(value: JsonValue, indent: Int): String = {
    let spaces = String.repeat("  ", indent)
    let nextSpaces = String.repeat("  ", indent + 1)
    match value {
        JsonNull => "null",
        JsonBool(b) => if b then "true" else "false",
        JsonInt(n) => toString(n),
        JsonFloat(f) => toString(f),
        JsonString(s) => "\"" + escapeString(s) + "\"",
        JsonArray(items) => {
            if List.length(items) == 0 then "[]"
            else {
                let encodedItems = List.map(items, fn(item: JsonValue): String =>
                    nextSpaces + encodePrettyIndent(item, indent + 1)
                )
                "[\n" + String.join(encodedItems, ",\n") + "\n" + spaces + "]"
            }
        },
        JsonObject(fields) => {
            if List.length(fields) == 0 then "{}"
            else {
                let encodedFields = List.map(fields, fn(field: (String, JsonValue)): String => {
                    match field {
                        (key, val) => nextSpaces + "\"" + escapeString(key) + "\": " + encodePrettyIndent(val, indent + 1)
                    }
                })
                "{\n" + String.join(encodedFields, ",\n") + "\n" + spaces + "}"
            }
        }
    }
 }
 // ============================================================
 // Type-specific Encoders
 // ============================================================
 // Encode primitives
 pub fn encodeNull(): JsonValue = JsonNull
 pub fn encodeBool(b: Bool): JsonValue = JsonBool(b)
 pub fn encodeInt(n: Int): JsonValue = JsonInt(n)
 pub fn encodeFloat(f: Float): JsonValue = JsonFloat(f)
 pub fn encodeString(s: String): JsonValue = JsonString(s)
 // Encode a list
 pub fn encodeList<A>(items: List<A>, encodeItem: fn(A): JsonValue): JsonValue =
    JsonArray(List.map(items, encodeItem))
 // Encode an optional value
 pub fn encodeOption<A>(opt: Option<A>, encodeItem: fn(A): JsonValue): JsonValue =
    match opt {
        None => JsonNull,
        Some(value) => encodeItem(value)
    }
 // Encode a Result
 pub fn encodeResult<T, E>(
    result: Result<T, E>,
    encodeOk: fn(T): JsonValue,
    encodeErr: fn(E): JsonValue
 ): JsonValue =
    match result {
        Ok(value) => JsonObject([
            ("ok", encodeOk(value))
        ]),
        Err(error) => JsonObject([
            ("error", encodeErr(error))
        ])
    }
 // ============================================================
 // Object Building Helpers
 // ============================================================
 // Create an empty JSON object
 pub fn object(): JsonValue = JsonObject([])
 // Add a field to a JSON object
 pub fn withField(obj: JsonValue, key: String, value: JsonValue): JsonValue =
    match obj {
        JsonObject(fields) => JsonObject(List.concat(fields, [(key, value)])),
        _ => obj  // Not an object, return unchanged
    }
 // Add a string field
 pub fn withString(obj: JsonValue, key: String, value: String): JsonValue =
    withField(obj, key, JsonString(value))
 // Add an int field
 pub fn withInt(obj: JsonValue, key: String, value: Int): JsonValue =
    withField(obj, key, JsonInt(value))
 // Add a bool field
 pub fn withBool(obj: JsonValue, key: String, value: Bool): JsonValue =
    withField(obj, key, JsonBool(value))
 // Add an optional field (only adds if Some)
 pub fn withOptional<A>(obj: JsonValue, key: String, opt: Option<A>, encodeItem: fn(A): JsonValue): JsonValue =
    match opt {
        None => obj,
        Some(value) => withField(obj, key, encodeItem(value))
    }
 // ============================================================
 // Decoding
 // ============================================================
 // Result type for parsing
 type ParseResult<A> = Result<A, String>
 // Get a field from a JSON object
 pub fn getField(obj: JsonValue, key: String): Option<JsonValue> =
    match obj {
        JsonObject(fields) => findField(fields, key),
        _ => None
    }
 fn findField(fields: List<(String, JsonValue)>, key: String): Option<JsonValue> =
    match List.head(fields) {
        None => None,
        Some(field) => match field {
            (k, v) => if k == key then Some(v)
                     else findField(Option.getOrElse(List.tail(fields), []), key)
        }
    }
 // Get a string field
 pub fn getString(obj: JsonValue, key: String): Option<String> =
    match getField(obj, key) {
        Some(JsonString(s)) => Some(s),
        _ => None
    }
 // Get an int field
 pub fn getInt(obj: JsonValue, key: String): Option<Int> =
    match getField(obj, key) {
        Some(JsonInt(n)) => Some(n),
        _ => None
    }
 // Get a bool field
 pub fn getBool(obj: JsonValue, key: String): Option<Bool> =
    match getField(obj, key) {
        Some(JsonBool(b)) => Some(b),
        _ => None
    }
 // Get an array field
 pub fn getArray(obj: JsonValue, key: String): Option<List<JsonValue>> =
    match getField(obj, key) {
        Some(JsonArray(items)) => Some(items),
        _ => None
    }
 // Get an object field
 pub fn getObject(obj: JsonValue, key: String): Option<JsonValue> =
    match getField(obj, key) {
        Some(JsonObject(_) as obj) => Some(obj),
        _ => None
    }
 // ============================================================
 // Simple JSON Parser
 // ============================================================
 // Parse a JSON string into a JsonValue
 // Note: This is a simplified parser for common cases
 pub fn parse(json: String): Result<JsonValue, String> =
    parseValue(String.trim(json), 0).mapResult(fn(r: (JsonValue, Int)): JsonValue => {
        match r { (value, _) => value }
    })
 fn parseValue(json: String, pos: Int): Result<(JsonValue, Int), String> = {
    let c = String.charAt(json, pos)
    match c {
        "n" => parseNull(json, pos),
        "t" => parseTrue(json, pos),
        "f" => parseFalse(json, pos),
        "\"" => parseString(json, pos),
        "[" => parseArray(json, pos),
        "{" => parseObject(json, pos),
        "-" => parseNumber(json, pos),
        _ => if isDigit(c) then parseNumber(json, pos)
             else if c == " " || c == "\n" || c == "\r" || c == "\t" then
                 parseValue(json, pos + 1)
             else Err("Unexpected character at position " + toString(pos))
    }
 }
 fn parseNull(json: String, pos: Int): Result<(JsonValue, Int), String> =
    if String.substring(json, pos, pos + 4) == "null" then Ok((JsonNull, pos + 4))
    else Err("Expected 'null' at position " + toString(pos))
 fn parseTrue(json: String, pos: Int): Result<(JsonValue, Int), String> =
    if String.substring(json, pos, pos + 4) == "true" then Ok((JsonBool(true), pos + 4))
    else Err("Expected 'true' at position " + toString(pos))
 fn parseFalse(json: String, pos: Int): Result<(JsonValue, Int), String> =
    if String.substring(json, pos, pos + 5) == "false" then Ok((JsonBool(false), pos + 5))
    else Err("Expected 'false' at position " + toString(pos))
 fn parseString(json: String, pos: Int): Result<(JsonValue, Int), String> = {
    // Skip opening quote
    let start = pos + 1
    let result = parseStringContent(json, start, "")
    result.mapResult(fn(r: (String, Int)): (JsonValue, Int) => {
        match r { (s, endPos) => (JsonString(s), endPos) }
    })
 }
 fn parseStringContent(json: String, pos: Int, acc: String): Result<(String, Int), String> = {
    let c = String.charAt(json, pos)
    if c == "\"" then Ok((acc, pos + 1))
    else if c == "\\" then {
        let nextC = String.charAt(json, pos + 1)
        let escaped = match nextC {
            "n" => "\n",
            "r" => "\r",
            "t" => "\t",
            "\"" => "\"",
            "\\" => "\\",
            _ => nextC
        }
        parseStringContent(json, pos + 2, acc + escaped)
    }
    else if c == "" then Err("Unterminated string")
    else parseStringContent(json, pos + 1, acc + c)
 }
 fn parseNumber(json: String, pos: Int): Result<(JsonValue, Int), String> = {
    let result = parseNumberDigits(json, pos, "")
    result.mapResult(fn(r: (String, Int)): (JsonValue, Int) => {
        match r {
            (numStr, endPos) => {
                // Check if it's a float
                if String.contains(numStr, ".") then
                    (JsonFloat(parseFloat(numStr)), endPos)
                else
                    (JsonInt(parseInt(numStr)), endPos)
            }
        }
    })
 }
 fn parseNumberDigits(json: String, pos: Int, acc: String): Result<(String, Int), String> = {
    let c = String.charAt(json, pos)
    if isDigit(c) || c == "." || c == "-" || c == "e" || c == "E" || c == "+" then
        parseNumberDigits(json, pos + 1, acc + c)
    else if acc == "" then Err("Expected number at position " + toString(pos))
    else Ok((acc, pos))
 }
 fn parseArray(json: String, pos: Int): Result<(JsonValue, Int), String> = {
    // Skip opening bracket and whitespace
    let startPos = skipWhitespace(json, pos + 1)
    if String.charAt(json, startPos) == "]" then Ok((JsonArray([]), startPos + 1))
    else parseArrayItems(json, startPos, [])
 }
 fn parseArrayItems(json: String, pos: Int, acc: List<JsonValue>): Result<(JsonValue, Int), String> = {
    match parseValue(json, pos) {
        Err(e) => Err(e),
        Ok((value, nextPos)) => {
            let newAcc = List.concat(acc, [value])
            let afterWhitespace = skipWhitespace(json, nextPos)
            let c = String.charAt(json, afterWhitespace)
            if c == "]" then Ok((JsonArray(newAcc), afterWhitespace + 1))
            else if c == "," then parseArrayItems(json, skipWhitespace(json, afterWhitespace + 1), newAcc)
            else Err("Expected ',' or ']' at position " + toString(afterWhitespace))
        }
    }
 }
 fn parseObject(json: String, pos: Int): Result<(JsonValue, Int), String> = {
    // Skip opening brace and whitespace
    let startPos = skipWhitespace(json, pos + 1)
    if String.charAt(json, startPos) == "}" then Ok((JsonObject([]), startPos + 1))
    else parseObjectFields(json, startPos, [])
 }
 fn parseObjectFields(json: String, pos: Int, acc: List<(String, JsonValue)>): Result<(JsonValue, Int), String> = {
    // Parse key
    match parseString(json, pos) {
        Err(e) => Err(e),
        Ok((keyValue, afterKey)) => {
            match keyValue {
                JsonString(key) => {
                    let colonPos = skipWhitespace(json, afterKey)
                    if String.charAt(json, colonPos) != ":" then
                        Err("Expected ':' at position " + toString(colonPos))
                    else {
                        let valuePos = skipWhitespace(json, colonPos + 1)
                        match parseValue(json, valuePos) {
                            Err(e) => Err(e),
                            Ok((value, afterValue)) => {
                                let newAcc = List.concat(acc, [(key, value)])
                                let afterWhitespace = skipWhitespace(json, afterValue)
                                let c = String.charAt(json, afterWhitespace)
                                if c == "}" then Ok((JsonObject(newAcc), afterWhitespace + 1))
                                else if c == "," then parseObjectFields(json, skipWhitespace(json, afterWhitespace + 1), newAcc)
                                else Err("Expected ',' or '}' at position " + toString(afterWhitespace))
                            }
                        }
                    }
                },
                _ => Err("Expected string key at position " + toString(pos))
            }
        }
    }
 }
 fn skipWhitespace(json: String, pos: Int): Int = {
    let c = String.charAt(json, pos)
    if c == " " || c == "\n" || c == "\r" || c == "\t" then
        skipWhitespace(json, pos + 1)
    else pos
 }
 fn isDigit(c: String): Bool =
    c == "0" || c == "1" || c == "2" || c == "3" || c == "4" ||
    c == "5" || c == "6" || c == "7" || c == "8" || c == "9"
 // ============================================================
 // Codec Type (for automatic serialization)
 // ============================================================
 // A codec can both encode and decode a type
 type Codec<A> = {
    encode: fn(A): JsonValue,
    decode: fn(JsonValue): Result<A, String>
 }
 // Create a codec from encode/decode functions
 pub fn codec<A>(
    enc: fn(A): JsonValue,
    dec: fn(JsonValue): Result<A, String>
 ): Codec<A> =
    { encode: enc, decode: dec }
 // Built-in codecs
 pub fn stringCodec(): Codec<String> =
    codec(
        encodeString,
        fn(json: JsonValue): Result<String, String> => match json {
            JsonString(s) => Ok(s),
            _ => Err("Expected string")
        }
    )
 pub fn intCodec(): Codec<Int> =
    codec(
        encodeInt,
        fn(json: JsonValue): Result<Int, String> => match json {
            JsonInt(n) => Ok(n),
            _ => Err("Expected int")
        }
    )
 pub fn boolCodec(): Codec<Bool> =
    codec(
        encodeBool,
        fn(json: JsonValue): Result<Bool, String> => match json {
            JsonBool(b) => Ok(b),
            _ => Err("Expected bool")
        }
    )
 pub fn listCodec<A>(itemCodec: Codec<A>): Codec<List<A>> =
    codec(
        fn(items: List<A>): JsonValue => encodeList(items, itemCodec.encode),
        fn(json: JsonValue): Result<List<A>, String> => match json {
            JsonArray(items) => decodeAll(items, itemCodec.decode),
            _ => Err("Expected array")
        }
    )
 fn decodeAll<A>(items: List<JsonValue>, decode: fn(JsonValue): Result<A, String>): Result<List<A>, String> = {
    match List.head(items) {
        None => Ok([]),
        Some(item) => match decode(item) {
            Err(e) => Err(e),
            Ok(decoded) => match decodeAll(Option.getOrElse(List.tail(items), []), decode) {
                Err(e) => Err(e),
                Ok(rest) => Ok(List.concat([decoded], rest))
            }
        }
    }
 }
 pub fn optionCodec<A>(itemCodec: Codec<A>): Codec<Option<A>> =
    codec(
        fn(opt: Option<A>): JsonValue => encodeOption(opt, itemCodec.encode),
        fn(json: JsonValue): Result<Option<A>, String> => match json {
            JsonNull => Ok(None),
            _ => itemCodec.decode(json).mapResult(fn(a: A): Option<A> => Some(a))
        }
    )
 // ============================================================
 // Helper for Result.mapResult
 // ============================================================
 fn mapResult<A, B>(result: Result<A, String>, f: fn(A): B): Result<B, String> =
    match result {
        Ok(a) => Ok(f(a)),
        Err(e) => Err(e)
    }
--- a/website/docs/index.html
+++ b/website/docs/index.html
@@ -0,0 +1,175 @@
 <!DOCTYPE html>
 <html lang="en">
 <head>
    <meta charset="UTF-8">
    <meta name="viewport" content="width=device-width, initial-scale=1.0">
    <title>Documentation - Lux</title>
    <meta name="description" content="Lux language documentation and API reference.">
    <link rel="icon" href="data:image/svg+xml,<svg xmlns='http://www.w3.org/2000/svg' viewBox='0 0 100 100'><text y='.9em' font-size='90'>&#10024;</text></svg>">
    <link rel="preconnect" href="https://fonts.googleapis.com">
    <link rel="preconnect" href="https://fonts.gstatic.com" crossorigin>
    <link href="https://fonts.googleapis.com/css2?family=JetBrains+Mono:wght@400;600&family=Playfair+Display:wght@400;600;700&family=Source+Serif+4:opsz,wght@8..60,400;8..60,500;8..60,600&display=swap" rel="stylesheet">
    <link rel="stylesheet" href="../static/style.css">
    <style>
        .docs-container {
            max-width: 1000px;
            margin: 0 auto;
            padding: var(--space-xl) var(--space-lg);
        }
        .docs-header {
            text-align: center;
            margin-bottom: var(--space-2xl);
        }
        .docs-header p {
            font-size: 1.1rem;
        }
        .docs-sections {
            display: grid;
            grid-template-columns: repeat(auto-fit, minmax(280px, 1fr));
            gap: var(--space-lg);
        }
        .docs-section {
            background: var(--bg-glass);
            border: 1px solid var(--border-subtle);
            border-radius: 8px;
            padding: var(--space-lg);
        }
        .docs-section h2 {
            font-size: 1.25rem;
            margin-bottom: var(--space-md);
            text-align: left;
        }
        .docs-section ul {
            list-style: none;
        }
        .docs-section li {
            margin-bottom: var(--space-sm);
        }
        .docs-section a {
            display: block;
            color: var(--text-secondary);
            padding: var(--space-xs) 0;
            transition: color 0.2s ease;
        }
        .docs-section a:hover {
            color: var(--gold);
        }
        .docs-section p {
            font-size: 0.95rem;
            color: var(--text-muted);
            margin-bottom: var(--space-md);
        }
    </style>
 </head>
 <body>
    <nav>
        <a href="/" class="logo">Lux</a>
        <ul class="nav-links" id="nav-links">
            <li><a href="/install">Install</a></li>
            <li><a href="/tour/">Tour</a></li>
            <li><a href="/examples/">Examples</a></li>
            <li><a href="/docs/" class="active">Docs</a></li>
            <li><a href="/play">Play</a></li>
            <li><a href="https://git.qrty.ink/blu/lux" class="nav-source">Source</a></li>
        </ul>
    </nav>
    <main class="docs-container">
        <header class="docs-header">
            <h1>Documentation</h1>
            <p>Complete reference for the Lux programming language.</p>
        </header>
        <div class="docs-sections">
            <div class="docs-section">
                <h2>Standard Library</h2>
                <p>Core types and functions.</p>
                <ul>
                    <li><a href="stdlib/list.html">List</a></li>
                    <li><a href="stdlib/string.html">String</a></li>
                    <li><a href="stdlib/option.html">Option</a></li>
                    <li><a href="stdlib/result.html">Result</a></li>
                    <li><a href="stdlib/math.html">Math</a></li>
                    <li><a href="stdlib/json.html">Json</a></li>
                </ul>
            </div>
            <div class="docs-section">
                <h2>Effects</h2>
                <p>Built-in effect types and operations.</p>
                <ul>
                    <li><a href="effects/console.html">Console</a></li>
                    <li><a href="effects/file.html">File</a></li>
                    <li><a href="effects/process.html">Process</a></li>
                    <li><a href="effects/http.html">Http</a></li>
                    <li><a href="effects/http-server.html">HttpServer</a></li>
                    <li><a href="effects/time.html">Time</a></li>
                    <li><a href="effects/random.html">Random</a></li>
                    <li><a href="effects/state.html">State</a></li>
                    <li><a href="effects/fail.html">Fail</a></li>
                    <li><a href="effects/sql.html">Sql</a></li>
                    <li><a href="effects/postgres.html">Postgres</a></li>
                    <li><a href="effects/concurrent.html">Concurrent</a></li>
                    <li><a href="effects/channel.html">Channel</a></li>
                    <li><a href="effects/test.html">Test</a></li>
                </ul>
            </div>
            <div class="docs-section">
                <h2>Language Reference</h2>
                <p>Syntax, types, and semantics.</p>
                <ul>
                    <li><a href="spec/grammar.html">Grammar (EBNF)</a></li>
                    <li><a href="spec/types.html">Type System</a></li>
                    <li><a href="spec/effects.html">Effect System</a></li>
                    <li><a href="spec/operators.html">Operators</a></li>
                    <li><a href="spec/keywords.html">Keywords</a></li>
                </ul>
            </div>
            <div class="docs-section">
                <h2>Guides</h2>
                <p>In-depth explanations of key concepts.</p>
                <ul>
                    <li><a href="../learn/effects.html">Effects Guide</a></li>
                    <li><a href="../learn/behavioral-types.html">Behavioral Types</a></li>
                    <li><a href="../learn/compilation.html">Compilation</a></li>
                    <li><a href="../learn/performance.html">Performance</a></li>
                </ul>
            </div>
            <div class="docs-section">
                <h2>Coming From</h2>
                <p>Lux for developers of other languages.</p>
                <ul>
                    <li><a href="../learn/from-rust.html">Rust</a></li>
                    <li><a href="../learn/from-haskell.html">Haskell</a></li>
                    <li><a href="../learn/from-typescript.html">TypeScript</a></li>
                    <li><a href="../learn/from-python.html">Python</a></li>
                </ul>
            </div>
            <div class="docs-section">
                <h2>Tooling</h2>
                <p>CLI, LSP, and editor integration.</p>
                <ul>
                    <li><a href="tools/cli.html">CLI Reference</a></li>
                    <li><a href="tools/lsp.html">LSP Setup</a></li>
                    <li><a href="tools/vscode.html">VS Code</a></li>
                    <li><a href="tools/neovim.html">Neovim</a></li>
                </ul>
            </div>
        </div>
    </main>
 </body>
 </html>
--- a/website/examples/http-server.html
+++ b/website/examples/http-server.html
@@ -0,0 +1,239 @@
 <!DOCTYPE html>
 <html lang="en">
 <head>
    <meta charset="UTF-8">
    <meta name="viewport" content="width=device-width, initial-scale=1.0">
    <title>HTTP Server - Lux by Example</title>
    <meta name="description" content="Build an HTTP server in Lux.">
    <link rel="icon" href="data:image/svg+xml,<svg xmlns='http://www.w3.org/2000/svg' viewBox='0 0 100 100'><text y='.9em' font-size='90'>&#10024;</text></svg>">
    <link rel="preconnect" href="https://fonts.googleapis.com">
    <link rel="preconnect" href="https://fonts.gstatic.com" crossorigin>
    <link href="https://fonts.googleapis.com/css2?family=JetBrains+Mono:wght@400;600&family=Playfair+Display:wght@400;600;700&family=Source+Serif+4:opsz,wght@8..60,400;8..60,500;8..60,600&display=swap" rel="stylesheet">
    <link rel="stylesheet" href="../static/style.css">
    <style>
        .example-container {
            max-width: 900px;
            margin: 0 auto;
            padding: var(--space-xl) var(--space-lg);
        }
        .example-header {
            margin-bottom: var(--space-xl);
        }
        .example-header h1 {
            margin-bottom: var(--space-sm);
        }
        .example-header p {
            font-size: 1.1rem;
        }
        .code-block {
            background: var(--code-bg);
            border: 1px solid var(--code-border);
            border-radius: 8px;
            margin: var(--space-lg) 0;
            overflow: hidden;
        }
        .code-header {
            display: flex;
            justify-content: space-between;
            align-items: center;
            padding: var(--space-sm) var(--space-md);
            background: var(--bg-secondary);
            border-bottom: 1px solid var(--code-border);
        }
        .code-filename {
            color: var(--text-muted);
            font-family: var(--font-code);
            font-size: 0.85rem;
        }
        .code-content {
            padding: var(--space-md);
            overflow-x: auto;
        }
        .code-content pre {
            font-family: var(--font-code);
            font-size: 0.9rem;
            line-height: 1.6;
            margin: 0;
        }
        .explanation {
            background: var(--bg-glass);
            border: 1px solid var(--border-subtle);
            border-radius: 8px;
            padding: var(--space-lg);
            margin: var(--space-lg) 0;
        }
        .explanation h2 {
            font-size: 1.25rem;
            margin-bottom: var(--space-md);
            text-align: left;
        }
        .explanation ul {
            padding-left: var(--space-lg);
        }
        .explanation li {
            color: var(--text-secondary);
            margin-bottom: var(--space-sm);
        }
        .explanation code {
            background: var(--code-bg);
            padding: 0.1em 0.3em;
            border-radius: 3px;
            font-size: 0.9em;
            color: var(--gold);
        }
        .run-it {
            background: var(--bg-tertiary);
            border: 1px solid var(--border-subtle);
            border-radius: 8px;
            padding: var(--space-lg);
            margin: var(--space-lg) 0;
        }
        .run-it h3 {
            margin-bottom: var(--space-md);
        }
        .run-it pre {
            background: var(--code-bg);
            padding: var(--space-md);
            border-radius: 6px;
            font-family: var(--font-code);
            font-size: 0.9rem;
        }
        .example-nav {
            display: flex;
            justify-content: space-between;
            margin-top: var(--space-xl);
            padding-top: var(--space-lg);
            border-top: 1px solid var(--border-subtle);
        }
        .example-nav a {
            color: var(--text-secondary);
            padding: var(--space-sm) var(--space-md);
            border: 1px solid var(--border-subtle);
            border-radius: 4px;
        }
        .example-nav a:hover {
            color: var(--gold);
            border-color: var(--border-gold);
        }
    </style>
 </head>
 <body>
    <nav>
        <a href="/" class="logo">Lux</a>
        <ul class="nav-links" id="nav-links">
            <li><a href="/install">Install</a></li>
            <li><a href="/tour/">Tour</a></li>
            <li><a href="/examples/">Examples</a></li>
            <li><a href="/docs/">Docs</a></li>
            <li><a href="/play">Play</a></li>
            <li><a href="https://git.qrty.ink/blu/lux" class="nav-source">Source</a></li>
        </ul>
    </nav>
    <main class="example-container">
        <header class="example-header">
            <h1>HTTP Server</h1>
            <p>Build a simple HTTP server with effect-tracked I/O. The type signature tells you exactly what side effects this code performs.</p>
        </header>
        <div class="code-block">
            <div class="code-header">
                <span class="code-filename">server.lux</span>
            </div>
            <div class="code-content">
                <pre><code><span class="cm">// A simple HTTP server in Lux</span>
 <span class="cm">// Notice the effect signature: {HttpServer, Console}</span>
 <span class="kw">fn</span> <span class="fn">handleRequest</span>(req: <span class="ty">Request</span>): <span class="ty">Response</span> = {
    <span class="kw">match</span> req.path {
        <span class="st">"/"</span> => Response {
            status: <span class="num">200</span>,
            body: <span class="st">"Welcome to Lux!"</span>
        },
        <span class="st">"/api/hello"</span> => Response {
            status: <span class="num">200</span>,
            body: Json.stringify({ message: <span class="st">"Hello, World!"</span> })
        },
        _ => Response {
            status: <span class="num">404</span>,
            body: <span class="st">"Not Found"</span>
        }
    }
 }
 <span class="kw">fn</span> <span class="fn">main</span>(): <span class="ty">Unit</span> <span class="kw">with</span> {<span class="ef">HttpServer</span>, <span class="ef">Console</span>} = {
    <span class="ef">HttpServer</span>.listen(<span class="num">8080</span>)
    <span class="ef">Console</span>.print(<span class="st">"Server listening on http://localhost:8080"</span>)
    <span class="kw">loop</span> {
        <span class="kw">let</span> req = <span class="ef">HttpServer</span>.accept()
        <span class="ef">Console</span>.print(req.method + <span class="st">" "</span> + req.path)
        <span class="kw">let</span> response = handleRequest(req)
        <span class="ef">HttpServer</span>.respond(response.status, response.body)
    }
 }
 <span class="kw">run</span> main() <span class="kw">with</span> {}</code></pre>
            </div>
        </div>
        <div class="explanation">
            <h2>Key Concepts</h2>
            <ul>
                <li><code>with {HttpServer, Console}</code> - The function signature declares exactly which effects this code uses</li>
                <li><code>HttpServer.listen(port)</code> - Start listening on a port</li>
                <li><code>HttpServer.accept()</code> - Wait for and return the next request</li>
                <li><code>HttpServer.respond(status, body)</code> - Send a response</li>
                <li>Pattern matching on <code>req.path</code> for routing</li>
            </ul>
        </div>
        <div class="explanation">
            <h2>Why Effects Matter Here</h2>
            <ul>
                <li>The type signature <code>with {HttpServer, Console}</code> tells you this function does network I/O and console output</li>
                <li>Pure functions like <code>handleRequest</code> have no effects - they're easy to test</li>
                <li>For testing, you can swap the <code>HttpServer</code> handler to simulate requests without a real network</li>
            </ul>
        </div>
        <div class="run-it">
            <h3>Run It</h3>
            <pre>$ lux run server.lux
 Server listening on http://localhost:8080
 # In another terminal:
 $ curl http://localhost:8080/
 Welcome to Lux!
 $ curl http://localhost:8080/api/hello
 {"message":"Hello, World!"}</pre>
        </div>
        <nav class="example-nav">
            <a href="index.html">&larr; All Examples</a>
            <a href="rest-api.html">REST API &rarr;</a>
        </nav>
    </main>
 </body>
 </html>
--- a/website/examples/index.html
+++ b/website/examples/index.html
@@ -0,0 +1,255 @@
 <!DOCTYPE html>
 <html lang="en">
 <head>
    <meta charset="UTF-8">
    <meta name="viewport" content="width=device-width, initial-scale=1.0">
    <title>Lux by Example</title>
    <meta name="description" content="Learn Lux through annotated example programs.">
    <link rel="icon" href="data:image/svg+xml,<svg xmlns='http://www.w3.org/2000/svg' viewBox='0 0 100 100'><text y='.9em' font-size='90'>&#10024;</text></svg>">
    <link rel="preconnect" href="https://fonts.googleapis.com">
    <link rel="preconnect" href="https://fonts.gstatic.com" crossorigin>
    <link href="https://fonts.googleapis.com/css2?family=JetBrains+Mono:wght@400;600&family=Playfair+Display:wght@400;600;700&family=Source+Serif+4:opsz,wght@8..60,400;8..60,500;8..60,600&display=swap" rel="stylesheet">
    <link rel="stylesheet" href="../static/style.css">
    <style>
        .examples-container {
            display: grid;
            grid-template-columns: 250px 1fr;
            max-width: 1200px;
            margin: 0 auto;
            min-height: calc(100vh - 80px);
        }
        .examples-sidebar {
            background: var(--bg-secondary);
            padding: var(--space-lg);
            border-right: 1px solid var(--border-subtle);
            position: sticky;
            top: 60px;
            height: calc(100vh - 60px);
            overflow-y: auto;
        }
        .examples-sidebar h2 {
            font-size: 1rem;
            color: var(--text-muted);
            margin-bottom: var(--space-md);
            text-transform: uppercase;
            letter-spacing: 0.05em;
            text-align: left;
        }
        .examples-sidebar ul {
            list-style: none;
            margin-bottom: var(--space-lg);
        }
        .examples-sidebar li {
            margin-bottom: var(--space-xs);
        }
        .examples-sidebar a {
            display: block;
            padding: var(--space-xs) var(--space-sm);
            color: var(--text-secondary);
            border-radius: 4px;
            transition: all 0.2s ease;
        }
        .examples-sidebar a:hover {
            color: var(--gold);
            background: var(--bg-glass);
        }
        .examples-sidebar a.active {
            color: var(--gold);
            background: var(--bg-glass);
            border-left: 2px solid var(--gold);
        }
        .examples-content {
            padding: var(--space-xl);
        }
        .examples-content h1 {
            margin-bottom: var(--space-md);
        }
        .examples-intro {
            max-width: 700px;
            margin-bottom: var(--space-xl);
        }
        .examples-grid {
            display: grid;
            grid-template-columns: repeat(auto-fill, minmax(300px, 1fr));
            gap: var(--space-lg);
        }
        .example-card {
            background: var(--bg-glass);
            border: 1px solid var(--border-subtle);
            border-radius: 8px;
            padding: var(--space-lg);
            transition: all 0.2s ease;
        }
        .example-card:hover {
            border-color: var(--border-gold);
            transform: translateY(-2px);
        }
        .example-card h3 {
            color: var(--gold);
            margin-bottom: var(--space-sm);
        }
        .example-card p {
            font-size: 0.95rem;
            margin-bottom: var(--space-md);
        }
        .example-card a {
            font-size: 0.9rem;
        }
        @media (max-width: 768px) {
            .examples-container {
                grid-template-columns: 1fr;
            }
            .examples-sidebar {
                position: static;
                height: auto;
            }
        }
    </style>
 </head>
 <body>
    <nav>
        <a href="/" class="logo">Lux</a>
        <ul class="nav-links" id="nav-links">
            <li><a href="/install">Install</a></li>
            <li><a href="/tour/">Tour</a></li>
            <li><a href="/examples/" class="active">Examples</a></li>
            <li><a href="/docs/">Docs</a></li>
            <li><a href="/play">Play</a></li>
            <li><a href="https://git.qrty.ink/blu/lux" class="nav-source">Source</a></li>
        </ul>
    </nav>
    <main class="examples-container">
        <aside class="examples-sidebar">
            <h2>Basics</h2>
            <ul>
                <li><a href="hello-world.html">Hello World</a></li>
                <li><a href="values.html">Values</a></li>
                <li><a href="variables.html">Variables</a></li>
                <li><a href="functions.html">Functions</a></li>
                <li><a href="closures.html">Closures</a></li>
                <li><a href="recursion.html">Recursion</a></li>
            </ul>
            <h2>Types</h2>
            <ul>
                <li><a href="records.html">Records</a></li>
                <li><a href="variants.html">Variants</a></li>
                <li><a href="generics.html">Generics</a></li>
                <li><a href="option.html">Option</a></li>
                <li><a href="result.html">Result</a></li>
            </ul>
            <h2>Effects</h2>
            <ul>
                <li><a href="console-io.html">Console I/O</a></li>
                <li><a href="file-operations.html">File Operations</a></li>
                <li><a href="http-requests.html">HTTP Requests</a></li>
                <li><a href="random-numbers.html">Random Numbers</a></li>
                <li><a href="time-sleep.html">Time & Sleep</a></li>
                <li><a href="state-management.html">State Management</a></li>
                <li><a href="error-handling.html">Error Handling</a></li>
            </ul>
            <h2>Data</h2>
            <ul>
                <li><a href="json-parsing.html">JSON Parsing</a></li>
                <li><a href="json-generation.html">JSON Generation</a></li>
                <li><a href="string-processing.html">String Processing</a></li>
                <li><a href="list-operations.html">List Operations</a></li>
                <li><a href="sqlite-database.html">SQLite Database</a></li>
                <li><a href="postgresql.html">PostgreSQL</a></li>
            </ul>
            <h2>Concurrent</h2>
            <ul>
                <li><a href="spawning-tasks.html">Spawning Tasks</a></li>
                <li><a href="channels.html">Channels</a></li>
                <li><a href="producer-consumer.html">Producer/Consumer</a></li>
                <li><a href="parallel-map.html">Parallel Map</a></li>
            </ul>
            <h2>Web</h2>
            <ul>
                <li><a href="http-server.html">HTTP Server</a></li>
                <li><a href="rest-api.html">REST API</a></li>
                <li><a href="middleware.html">Middleware</a></li>
                <li><a href="routing.html">Routing</a></li>
            </ul>
            <h2>Projects</h2>
            <ul>
                <li><a href="/projects/">All Projects</a></li>
                <li><a href="/projects/#rest-api">REST API</a></li>
                <li><a href="/projects/#todo-app">Todo App</a></li>
                <li><a href="/projects/#json-parser">JSON Parser</a></li>
            </ul>
        </aside>
        <div class="examples-content">
            <h1>Lux by Example</h1>
            <div class="examples-intro">
                <p>Learn Lux through annotated example programs. Each example is self-contained and demonstrates a specific concept or pattern.</p>
                <p>Click any example to see the full code with explanations.</p>
            </div>
            <div class="examples-grid">
                <div class="example-card">
                    <h3>Hello World</h3>
                    <p>Your first Lux program. Learn about the main function and Console effect.</p>
                    <a href="hello-world.html">View example &rarr;</a>
                </div>
                <div class="example-card">
                    <h3>Effects Basics</h3>
                    <p>Understand how effects make side effects explicit in type signatures.</p>
                    <a href="console-io.html">View example &rarr;</a>
                </div>
                <div class="example-card">
                    <h3>Pattern Matching</h3>
                    <p>Destructure data with exhaustive pattern matching.</p>
                    <a href="variants.html">View example &rarr;</a>
                </div>
                <div class="example-card">
                    <h3>HTTP Server</h3>
                    <p>Build a simple web server with effect-tracked I/O.</p>
                    <a href="http-server.html">View example &rarr;</a>
                </div>
                <div class="example-card">
                    <h3>JSON Processing</h3>
                    <p>Parse and generate JSON data with type safety.</p>
                    <a href="json-parsing.html">View example &rarr;</a>
                </div>
                <div class="example-card">
                    <h3>Concurrency</h3>
                    <p>Spawn tasks and communicate via channels.</p>
                    <a href="spawning-tasks.html">View example &rarr;</a>
                </div>
            </div>
        </div>
    </main>
 </body>
 </html>
--- a/website/index.html
+++ b/website/index.html
@@ -0,0 +1,266 @@
 <!DOCTYPE html>
 <html lang="en">
 <head>
    <meta charset="UTF-8">
    <meta name="viewport" content="width=device-width, initial-scale=1.0">
    <title>Lux - Side Effects Can't Hide</title>
    <meta name="description" content="Lux is a functional programming language with first-class effects. See what your code does. Test without mocks. Ship with confidence.">
    <meta name="keywords" content="programming language, functional programming, algebraic effects, effect system, type system, native compilation">
    <!-- Open Graph -->
    <meta property="og:type" content="website">
    <meta property="og:title" content="Lux - Side Effects Can't Hide">
    <meta property="og:description" content="A functional programming language with first-class effects. See what your code does, test without mocks, ship with confidence.">
    <meta property="og:url" content="https://lux-lang.org">
    <meta name="twitter:card" content="summary_large_image">
    <link rel="icon" href="data:image/svg+xml,<svg xmlns='http://www.w3.org/2000/svg' viewBox='0 0 100 100'><text y='.9em' font-size='90'>&#10024;</text></svg>">
    <!-- Fonts -->
    <link rel="preconnect" href="https://fonts.googleapis.com">
    <link rel="preconnect" href="https://fonts.gstatic.com" crossorigin>
    <link href="https://fonts.googleapis.com/css2?family=JetBrains+Mono:wght@400;600&family=Playfair+Display:wght@400;600;700&family=Source+Serif+4:opsz,wght@8..60,400;8..60,500;8..60,600&display=swap" rel="stylesheet">
    <link rel="stylesheet" href="static/style.css">
 </head>
 <body>
    <nav>
        <a href="/" class="logo">Lux</a>
        <button class="mobile-menu-btn" id="mobile-menu-btn" aria-label="Toggle menu">
            <span id="menu-icon">&#9776;</span>
        </button>
        <ul class="nav-links" id="nav-links">
            <li><a href="/install">Install</a></li>
            <li><a href="/tour/">Tour</a></li>
            <li><a href="/examples/">Examples</a></li>
            <li><a href="/docs/">Docs</a></li>
            <li><a href="/play">Play</a></li>
            <li><a href="https://git.qrty.ink/blu/lux" class="nav-source">Source</a></li>
        </ul>
    </nav>
    <!-- Hero Section -->
    <header class="hero">
        <h1>Side Effects Can't Hide</h1>
        <p class="tagline">See what your code does. Test without mocks. Ship with confidence.</p>
        <div class="hero-cta">
            <a href="#playground" class="btn btn-primary">Try Now</a>
            <a href="/install" class="btn btn-secondary">Install</a>
            <a href="/tour/" class="btn btn-tertiary">Take the Tour</a>
        </div>
        <div class="hero-code">
            <pre><code><span class="kw">fn</span> <span class="fn">processOrder</span>(order: <span class="ty">Order</span>): <span class="ty">Receipt</span> <span class="kw">with</span> {<span class="ef">Database</span>, <span class="ef">Email</span>} = {
    <span class="kw">let</span> saved = <span class="ef">Database</span>.save(order)
    <span class="ef">Email</span>.send(order.customer, <span class="st">"Order confirmed!"</span>)
    Receipt(saved.id)
 }
 <span class="cm">// The signature tells you EVERYTHING this function does</span></code></pre>
        </div>
        <div class="badges">
            <span class="badge">MIT Licensed</span>
            <span class="badge">372+ Tests</span>
            <span class="badge">Native Performance</span>
        </div>
    </header>
    <!-- Problem/Solution Section -->
    <section class="problem-section">
        <h2>The Problem with Side Effects</h2>
        <p class="section-subtitle">In most languages, functions can do <em>anything</em>. You can't tell from the signature.</p>
        <div class="comparison">
            <div class="comparison-card bad">
                <h3>Other Languages</h3>
                <div class="comparison-code">
                    <pre><code><span class="fn">fetchUser</span>(id: <span class="ty">Int</span>): <span class="ty">User</span>
 <span class="cm">// Does this call the network?
 // Touch the database?
 // Write to a file?
 // Who knows!</span></code></pre>
                </div>
            </div>
            <div class="comparison-card good">
                <h3>Lux</h3>
                <div class="comparison-code">
                    <pre><code><span class="kw">fn</span> <span class="fn">fetchUser</span>(id: <span class="ty">Int</span>): <span class="ty">User</span>
    <span class="kw">with</span> {<span class="ef">Http</span>, <span class="ef">Database</span>}
 <span class="cm">// You KNOW this touches
 // network and database</span></code></pre>
                </div>
            </div>
        </div>
    </section>
    <!-- Three Pillars Section -->
    <section class="pillars-section">
        <h2>The Lux Solution</h2>
        <p class="section-subtitle">Three pillars that make functional programming practical.</p>
        <div class="pillars">
            <div class="pillar">
                <h3>Effects You Can See</h3>
                <p>Every function declares its effects in the type signature. No hidden surprises. Refactor with confidence.</p>
                <div class="pillar-code">
                    <pre><code><span class="kw">fn</span> <span class="fn">sendNotification</span>(
    user: <span class="ty">User</span>,
    msg: <span class="ty">String</span>
 ): <span class="ty">Unit</span> <span class="kw">with</span> {<span class="ef">Email</span>, <span class="ef">Log</span>} = {
    <span class="ef">Log</span>.info(<span class="st">"Sending to "</span> + user.email)
    <span class="ef">Email</span>.send(user.email, msg)
 }</code></pre>
                </div>
            </div>
            <div class="pillar">
                <h3>Testing Without Mocks</h3>
                <p>Swap effect handlers at runtime. Test database code without a database. Test HTTP code without network.</p>
                <div class="pillar-code">
                    <pre><code><span class="cm">// Production</span>
 <span class="kw">run</span> app() <span class="kw">with</span> {
    <span class="ef">Database</span> = postgres,
    <span class="ef">Email</span> = smtp
 }
 <span class="cm">// Test - same code!</span>
 <span class="kw">run</span> app() <span class="kw">with</span> {
    <span class="ef">Database</span> = inMemory,
    <span class="ef">Email</span> = collect
 }</code></pre>
                </div>
            </div>
            <div class="pillar">
                <h3>Native Performance</h3>
                <p>Compiles to C via gcc/clang. Reference counting with FBIP optimization. Matches Rust and C speed.</p>
                <div class="pillar-code">
                    <pre><code>Benchmark      Lux     Rust    Go
 ───────────────────────────────────
 fibonacci(40)  <span class="hl">0.015s</span>  0.018s  0.041s
 ackermann      <span class="hl">0.020s</span>  0.029s  0.107s
 primes 1M      <span class="hl">0.012s</span>  0.014s  0.038s
 quicksort 1M   0.089s  0.072s  0.124s</code></pre>
                </div>
            </div>
        </div>
    </section>
    <!-- Interactive Playground -->
    <section class="playground-section" id="playground">
        <h2>Try It Now</h2>
        <p class="section-subtitle">Edit the code and click Run. No installation required.</p>
        <div class="playground">
            <div class="playground-tabs">
                <button class="playground-tab active" data-tab="hello">Hello World</button>
                <button class="playground-tab" data-tab="effects">Effects</button>
                <button class="playground-tab" data-tab="patterns">Patterns</button>
                <button class="playground-tab" data-tab="handlers">Handlers</button>
                <button class="playground-tab" data-tab="behavioral">Behavioral</button>
            </div>
            <div class="playground-content">
                <div class="playground-editor">
                    <textarea id="code-input" spellcheck="false">fn main(): Unit with {Console} = {
    Console.print("Hello, Lux!")
 }
 run main() with {}</textarea>
                </div>
                <div class="playground-output">
                    <div class="output-header">Output</div>
                    <pre id="code-output"><span class="cm">// Click "Run" to execute</span></pre>
                </div>
            </div>
            <div class="playground-toolbar">
                <span class="version">Lux v0.1.0</span>
                <div class="toolbar-actions">
                    <button class="btn btn-run" id="run-btn">Run</button>
                </div>
            </div>
        </div>
    </section>
    <!-- Getting Started Section -->
    <section class="install-section" id="install">
        <h2>Get Started</h2>
        <p class="section-subtitle">One command to try Lux. Two to build from source.</p>
        <div class="install-options">
            <div class="install-option">
                <h3>With Nix (Recommended)</h3>
                <div class="install-code">
                    <pre><code>nix run git+https://git.qrty.ink/blu/lux</code></pre>
                    <button class="copy-btn" data-copy="nix run git+https://git.qrty.ink/blu/lux">Copy</button>
                </div>
                <p class="install-note">One command. Zero dependencies. Works on Linux and macOS.</p>
            </div>
            <div class="install-option">
                <h3>From Source</h3>
                <div class="install-code">
                    <pre><code>git clone https://git.qrty.ink/blu/lux
 cd lux && cargo build --release</code></pre>
                    <button class="copy-btn" data-copy="git clone https://git.qrty.ink/blu/lux && cd lux && cargo build --release">Copy</button>
                </div>
            </div>
        </div>
        <div class="next-steps">
            <h4>Then</h4>
            <div class="next-steps-grid">
                <a href="/tour/" class="next-step">
                    <span class="next-step-icon">&#10140;</span>
                    <span>Take the Tour</span>
                </a>
                <a href="/examples/" class="next-step">
                    <span class="next-step-icon">&#10697;</span>
                    <span>Browse Examples</span>
                </a>
                <a href="/docs/" class="next-step">
                    <span class="next-step-icon">&#128214;</span>
                    <span>Read the Docs</span>
                </a>
            </div>
        </div>
    </section>
    <!-- Footer -->
    <footer>
        <div class="footer-content">
            <div class="footer-section">
                <h4>Lux</h4>
                <p>Functional programming with first-class effects.</p>
            </div>
            <div class="footer-section">
                <h4>Learn</h4>
                <ul>
                    <li><a href="/tour/">Language Tour</a></li>
                    <li><a href="/examples/">Examples</a></li>
                    <li><a href="/docs/">Documentation</a></li>
                </ul>
            </div>
            <div class="footer-section">
                <h4>Community</h4>
                <ul>
                    <li><a href="https://git.qrty.ink/blu/lux">Source Code</a></li>
                    <li><a href="https://git.qrty.ink/blu/lux/issues">Issues</a></li>
                    <li><a href="/community/code-of-conduct.html">Code of Conduct</a></li>
                </ul>
            </div>
        </div>
        <div class="footer-bottom">
            <p>MIT License. Built with Lux.</p>
        </div>
    </footer>
    <script src="static/app.js"></script>
 </body>
 </html>
--- a/Show More
+++ b/Show More
Author	SHA1	Message	Date
Brandon Lucas	6a2e4a7ac1	chore: bump version to 0.1.3	2026-02-18 23:06:10 -05:00
Brandon Lucas	3d706cb32b	feat: add record spread syntax { ...base, field: val } Adds spread operator for records, allowing concise record updates: let p2 = { ...p, x: 5.0 } Changes across the full pipeline: - Lexer: new DotDotDot (...) token - AST: optional spread field on Record variant - Parser: detect ... at start of record expression - Typechecker: merge spread record fields with explicit overrides - Interpreter: evaluate spread, overlay explicit fields - JS backend: emit native JS spread syntax - C backend: copy spread into temp, assign overrides - Formatter, linter, LSP, symbol table: propagate spread Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>	2026-02-18 23:05:27 -05:00
Brandon Lucas	7c3bfa9301	feat: add Math.sin, Math.cos, Math.atan2 trig functions Adds trigonometric functions to the Math module across interpreter, type system, and C backend. JS backend already supported them. Also adds #include <math.h> to C preamble and handles Math module calls through both Call and EffectOp paths in C backend. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>	2026-02-18 23:05:12 -05:00
Brandon Lucas	b56c5461f1	fix: JS const _ duplication and hardcoded version string - JS backend now emits wildcard let bindings as side-effect statements instead of const _ declarations, fixing SyntaxError on multiple let _ = ... - Version string now uses env!("CARGO_PKG_VERSION") to auto-sync with Cargo.toml - Add -lm linker flag for math library support Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>	2026-02-18 23:05:03 -05:00
Brandon Lucas	61e1469845	feat: add ++ concat operator and auto-invoke main BUG-004: Add ++ operator for string and list concatenation across all backends (interpreter, C, JS) with type checking and formatting support. BUG-001: Auto-invoke top-level `let main = fn () => ...` when main is a zero-parameter function, instead of just printing the function value. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>	2026-02-18 22:01:41 -05:00
Brandon Lucas	bb0a288210	chore: bump version to 0.1.2	2026-02-18 21:16:44 -05:00
Brandon Lucas	5d7f4633e1	docs: add explicit commit instructions to CLAUDE.md Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>	2026-02-18 21:09:27 -05:00
Brandon Lucas	d05b13d840	fix: JS backend compiles print() to console.log() Bare `print()` calls in Lux now emit `console.log()` in JS output instead of undefined `print()`. Fixes BUG-006. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>	2026-02-18 21:09:07 -05:00
Brandon Lucas	0ee3050704	chore: bump version to 0.1.1	2026-02-18 20:41:43 -05:00
Brandon Lucas	80b1276f9f	fix: release script auto-bumps patch by default Release script now supports: patch (default), minor, major, or explicit version. Auto-updates Cargo.toml and flake.nix before building. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>	2026-02-18 20:41:29 -05:00
Brandon Lucas	bd843d2219	fix: record type aliases now work for unification and field access Expand type aliases via unify_with_env() everywhere in the type checker, not just in a few places. This fixes named record types like `type Vec2 = { x: Float, y: Float }` — they now properly unify with anonymous records and support field access (v.x, v.y). Also adds scripts/validate.sh for automated full-suite regression testing (Rust tests + all 5 package test suites + type checking). Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>	2026-02-18 20:21:29 -05:00
Brandon Lucas	d76aa17b38	feat: static binary builds and automated release script Switch reqwest from native-tls (openssl) to rustls-tls for a pure-Rust TLS stack, enabling fully static musl builds. Add `nix build .#static` for portable Linux binaries and `scripts/release.sh` for automated Gitea releases with changelog generation. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>	2026-02-18 19:09:32 -05:00
Brandon Lucas	c23d9c7078	fix: test runner now supports module imports The `lux test` command used Parser::parse_source() and check_program() directly, which meant test files with `import` statements would fail with type errors. Now uses ModuleLoader and check_program_with_modules() to properly resolve imports, and run_with_modules() for execution. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>	2026-02-18 17:11:16 -05:00
Brandon Lucas	fffacd2467	feat: C backend module import support, Int/Float.toString, Test.assertEqualMsg The C backend can now compile programs that import user-defined modules. Module-qualified calls like `mymodule.func(args)` are resolved to prefixed C functions (e.g., `mymodule_func_lux`), with full support for transitive imports and effect-passing. Also adds Int.toString/Float.toString to type system, interpreter, and C backend, and Test.assertEqualMsg for labeled test assertions. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>	2026-02-18 16:35:24 -05:00
Brandon Lucas	2ae2c132e5	docs: add language philosophy document and compiler integration Write comprehensive PHILOSOPHY.md covering Lux's six core principles (explicit over implicit, composition over configuration, safety without ceremony, practical over academic, one right way, tools are the language) with detailed comparisons against JS/TS, Python, Rust, Go, Java/C#, Haskell/Elm, and Gleam/Elixir. Includes tooling audit and improvement suggestions. Add `lux philosophy` command to the compiler, update help screen with abbreviated philosophy, and link from README. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>	2026-02-18 10:19:29 -05:00
Brandon Lucas	4909ff9fff	docs: add package ecosystem plan and error documentation workflow Add PACKAGES.md analyzing the Lux package ecosystem gaps vs stdlib, with prioritized implementation plans for markdown, xml, rss, frontmatter, path, and sitemap packages. Add CLAUDE.md instructions for documenting Lux language errors in ISSUES.md during every major task. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>	2026-02-18 10:01:56 -05:00
Brandon Lucas	8e788c8a9f	fix: embed C compiler path at build time for self-contained binary build.rs captures the absolute path to cc/gcc/clang during compilation and bakes it into the binary. On Nix systems this embeds the full /nix/store path so `lux compile` works without cc on PATH. Lookup order: $CC env var > embedded build-time path > PATH search. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>	2026-02-18 08:12:18 -05:00
Brandon Lucas	dbdd3cca57	chore: move blu-site to its own repo at ~/src/blu-site Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>	2026-02-18 07:57:55 -05:00
Brandon Lucas	3ac022c04a	chore: gitignore build output (_site/, docs/) Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>	2026-02-18 07:48:51 -05:00
Brandon Lucas	6bedd37ac7	fix: show help menu when running lux with no arguments Previously `lux` with no args entered the REPL. Now it shows the help menu. Use `lux repl` to start the REPL explicitly. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>	2026-02-18 07:34:09 -05:00
Brandon Lucas	2909bf14b6	fix: eliminate all non-json C backend errors (79→0) Second round of C backend fixes, building on `d8871ac` which reduced errors from 286 to 111. This eliminates all 79 non-json errors: - Fix function references as values (wrap in LuxClosure) - Fix fold/map/filter with type-aware calling conventions - Add String.indexOf/lastIndexOf emission and C runtime functions - Add File.readDir with dirent.h implementation - Fix string concat in closure bodies - Exclude ADT constructors from closure free variable capture - Fix match result type inference (prioritize pattern binding types) - Fix Option inner type inference (usage-based for List.head) - Fix void to struct cast (dereference through pointer) - Handle constructors in emit_expr_with_env Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>	2026-02-18 05:56:21 -05:00
Brandon Lucas	d8871acf7e	fix: improve C backend robustness, reduce compilation errors by 61% - Fix closure captured variable types: look up actual types from var_types instead of hardcoding LuxInt for all captured variables - Register function parameters in var_types so closures can find their types - Replace is_string_expr() with infer_expr_type() for more accurate string detection in binary ops (concat, comparison) - Add missing String operations to infer_expr_type (substring, indexOf, etc.) - Add module method call type inference (String., List., Int., Float.) - Add built-in Result type (Ok/Err) to C prelude alongside Option - Register Ok/Err/Some/None in variant_to_type and variant_field_types - Fix variable scoping: use if-statement pattern instead of ternary when branches emit statements (prevents redefinition of h2/h3 etc.) - Add RC scope management for if-else branches and match arms to prevent undeclared variable errors from cleanup code - Add infer_pattern_binding_type for better match result type inference - Add expr_emits_statements helper to detect statement-emitting expressions - Add infer_option_inner_type for String.indexOf (returns Option<Int>) Reduces blu-site compilation errors from 286 to 111 (remaining are mostly unsupported json effect and function-as-value references). Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>	2026-02-17 17:56:27 -05:00
Brandon Lucas	73b5eee664	docs: add commit-after-every-piece-of-work instruction to CLAUDE.md Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>	2026-02-17 16:21:54 -05:00
Brandon Lucas	542255780d	feat: add tuple index access, multiline args, and effect unification fix - Tuple index: `pair.0`, `pair.1` syntax across parser, typechecker, interpreter, C/JS backends, formatter, linter, and symbol table - Multi-line function args: allow newlines inside argument lists - Fix effect unification for callback parameters (empty expected effects means "no constraint", not "must be pure") Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>	2026-02-17 16:21:48 -05:00
Brandon Lucas	bac63bab2a	feat: add blu-site static site generator and fix language issues Build a complete static site generator in Lux that faithfully clones blu.cx (elmstatic). Generates 14 post pages, section indexes, tag pages, and a home page with snippets grid from markdown content. Language fixes discovered during development: - Add \{ and \} escape sequences in string literals (lexer) - Register String.indexOf and String.lastIndexOf in type checker - Fix formatter to preserve brace escapes in string literals - Improve LSP hover to show documentation for let bindings and functions ISSUES.md documents 15 Lux language limitations found during the project. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>	2026-02-17 15:43:05 -05:00
Brandon Lucas	db82ca1a1c	fix: improve LSP hover to show function info when cursor is on `fn` keyword When hovering on declaration keywords (fn, type, effect, let, trait), look ahead to find the declaration name and show that symbol's full info from the symbol table instead of generic keyword documentation. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>	2026-02-17 08:32:01 -05:00
Brandon Lucas	98605d2b70	feat: add self-hosted Lux lexer as first step toward bootstrapping The lexer tokenizes Lux source code written entirely in Lux itself. Supports all token types: keywords, operators, literals, behavioral properties, doc comments, and delimiters. This is the first component of the Lux-in-Lux compiler, demonstrating that Lux's pattern matching, recursion, and string handling are sufficient for compiler construction. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>	2026-02-17 08:25:22 -05:00
Brandon Lucas	e3b6f4322a	fix: add Char pattern matching and Char comparison operators - Parser: support Char literals in match patterns (e.g., 'x' => ...) - Interpreter: add Char comparison for <, <=, >, >= operators Previously only Int, Float, and String supported ordering comparisons. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>	2026-02-17 08:25:15 -05:00
Brandon Lucas	d26fd975d1	feat: enhance LSP with inlay hints, parameter hints, and improved hover Add inlay type hints for let bindings, parameter name hints at call sites, behavioral property documentation in hover, and long signature wrapping. - Inlay hints: show inferred types for let bindings without annotations - Parameter hints: show param names at call sites for multi-arg functions - Hover: wrap long signatures, show behavioral property docs (pure, total, etc.) - Rich docs: detailed hover for keywords like pure, total, idempotent, run, with - TypeChecker: expose get_inferred_type() for LSP consumption - Symbol table: include behavioral properties in function type signatures Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>	2026-02-17 08:06:36 -05:00
Brandon Lucas	1fa599f856	fix: support comma-separated behavioral properties without repeating 'is' Allows `is pure, commutative` syntax in addition to `is pure is commutative`. After the initial `is`, comma-separated properties no longer require repeating the `is` keyword (though it's still accepted for compatibility). Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>	2026-02-17 07:44:18 -05:00
Brandon Lucas	c2404a5ec1	docs: update CLAUDE.md with post-work checklist and CLI aliases table Adds the post-work checklist (cargo check, cargo test, lux check, lux fmt, lux lint) and documents all CLI command aliases. Updates test count to 381. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>	2026-02-17 07:36:09 -05:00
Brandon Lucas	19068ead96	feat: add lux lint command with Lux-specific static analysis Implements a linter with 21 lint rules across 6 categories (correctness, suspicious, idiom, performance, style, pedantic). Lux-specific lints include could-be-pure, could-be-total, unnecessary-effect-decl, and single-arm-match. Integrates lints into `lux check` for unified type+lint checking. Available standalone via `lux lint` (alias: `lux l`) with --explain for detailed help. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>	2026-02-17 07:35:36 -05:00
Brandon Lucas	44ea1eebb0	style: auto-format example files with lux fmt Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>	2026-02-17 06:52:44 -05:00
Brandon Lucas	8c90d5a8dc	feat: CLI UX overhaul with colored output, timing, shorthands, and fuzzy suggestions Add polished CLI output across all commands: colored help text, green/red pass/fail indicators (✓/✗), elapsed timing on compile/check/test/fmt, command shorthands (c/t/f/s/k), fuzzy "did you mean?" on typos, and smart port-in-use suggestions for serve. Respects NO_COLOR/TERM=dumb. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>	2026-02-17 06:52:36 -05:00
Brandon Lucas	bc60f1c8f1	fix: improve error message for bare 'run' expressions at top level When users write `run main() with {}` at top level instead of `let _ = run main() with {}`, provide a helpful error message explaining the correct syntax instead of the generic "Expected declaration" error. Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>	2026-02-16 23:40:15 -05:00
Brandon Lucas	52e3876b81	feat: add projects showcase and Lux-powered static file server - Add website/serve.lux: static file server using HttpServer effect - Demonstrates serving the Lux website with Lux itself - Handles index files, clean URLs, and 404 responses - Add website/projects/index.html: example projects showcase - Features 6 real project cards (REST API, Todo App, JSON Parser, etc.) - Highlights Task Manager API demonstrating all 3 killer features - Links to full source code in the repository - Update examples sidebar with Projects section Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>	2026-02-16 23:34:17 -05:00
Brandon Lucas	7e76acab18	feat: rebuild website with full learning funnel Website rebuilt from scratch based on analysis of 11 beloved language websites (Elm, Zig, Gleam, Swift, Kotlin, Haskell, OCaml, Crystal, Roc, Rust, Go). New website structure: - Homepage with hero, playground, three pillars, install guide - Language Tour with interactive lessons (hello world, types, effects) - Examples cookbook with categorized sidebar - API documentation index - Installation guide (Nix and source) - Sleek/noble design (black/gold, serif typography) Also includes: - New stdlib/json.lux module for JSON serialization - Enhanced stdlib/http.lux with middleware and routing - New string functions (charAt, indexOf, lastIndexOf, repeat) - LSP improvements (rename, signature help, formatting) - Package manager transitive dependency resolution - Updated documentation for effects and stdlib - New showcase example (task_manager.lux) Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>	2026-02-16 23:05:35 -05:00
Brandon Lucas	5a853702d1	fix: C backend string comparison, underscore patterns, and list memory 1. String == comparison now uses strcmp instead of pointer comparison - Added check in emit_expr() for BinaryOp::Eq/Ne on strings - Also fixed in emit_expr_with_env() for closures 2. Support `let _ = expr` pattern to discard values - Parser now accepts underscore in let bindings (both blocks and expressions) - C backend emits (void)expr; for underscore patterns 3. Fix list head/tail/get memory management - Added lux_incref() when extracting elements from lists - Prevents use-after-free when original list is freed 4. String.startsWith was already implemented (verified working) Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>	2026-02-16 07:14:02 -05:00