Files
lux/website/lux-site/LUX_WEAKNESSES.md
Brandon Lucas 4b553031fd fix: parser newline handling in lists and stdlib exports
- Fix parse_list_expr to skip newlines between list elements
- Add `pub` keyword to all exported functions in stdlib/html.lux
- Change List.foldl to List.fold (matching built-in name)
- Update weaknesses document with fixed issues

The module import system now works correctly. This enables:
- import stdlib/html to work as expected
- html.div(), html.render() etc. to be accessible
- Multi-line list expressions in Lux source files

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
2026-02-16 06:51:37 -05:00

174 lines
4.9 KiB
Markdown

# Lux Weaknesses Discovered During Website Development
This document tracks issues and limitations discovered while building the Lux website in Lux.
## Fixed Issues
### 1. Module Import System Not Working (FIXED)
**Description:** The `import` statement wasn't working for importing standard library modules.
**Root Cause:** Two issues were found:
1. Parser didn't skip newlines inside list expressions, causing parse errors in multi-line lists
2. Functions in stdlib modules weren't marked as `pub` (public), so they weren't exported
**Fix:**
1. Added `skip_newlines()` calls to `parse_list_expr()` in parser.rs
2. Added `pub` keyword to all exported functions in stdlib/html.lux
**Status:** FIXED
---
### 2. Parse Error in html.lux (FIXED)
**Description:** When trying to load files that import the html module, there was a parse error at line 196-197.
**Error Message:**
```
Module error: Module error in '<main>': Parse error: Parse error at 196-197: Unexpected token: \n
```
**Root Cause:** The parser's `parse_list_expr()` function didn't handle newlines between list elements.
**Fix:** Added `skip_newlines()` calls after `[`, after each element, and after commas in list expressions.
**Status:** FIXED
---
### 3. List.foldl Renamed to List.fold (FIXED)
**Description:** The html.lux file used `List.foldl` but the built-in List module exports `List.fold`.
**Fix:** Changed `List.foldl` to `List.fold` in stdlib/html.lux.
**Status:** FIXED
---
## Minor Issues
### 4. String.replace Verified Working
**Description:** The `escapeHtml` function in html.lux uses `String.replace`.
**Status:** VERIFIED WORKING - String.replace is implemented in the interpreter.
---
### 5. FileSystem Effect Not Fully Implemented
**Description:** For static site generation, we need `FileSystem.mkdir`, `FileSystem.write`, `FileSystem.copy` operations. These may not be fully implemented.
**Workaround:** Use Bash commands via scripts.
**Status:** Needs verification
---
### 6. List.concat Verified Working
**Description:** The `List.concat` function is used in html.lux document generation.
**Status:** VERIFIED WORKING - List.concat is implemented in the interpreter.
---
## Feature Gaps
### 7. No Built-in Static Site Generation
**Description:** There's no built-in way to generate static HTML files from Lux. A static site generator effect or module would be helpful.
**Recommendation:** Add a `FileSystem` effect with:
- `mkdir(path: String): Unit`
- `write(path: String, content: String): Unit`
- `copy(src: String, dst: String): Unit`
- `readDir(path: String): List<String>`
---
### 8. No Template String Support
**Description:** Multi-line strings and template literals (like JavaScript's backticks) would make HTML generation much easier.
**Current Approach:**
```lux
let html = "<div class=\"" + className + "\">" + content + "</div>"
```
**Better Approach (not available):**
```lux
let html = `<div class="${className}">${content}</div>`
```
**Status:** Feature request
---
### 9. No Markdown Parser
**Description:** A built-in Markdown parser would be valuable for documentation sites.
**Status:** Feature request
---
## Working Features
These features work correctly and can be used for website generation:
1. **String concatenation** - Works with `+` operator
2. **Conditional expressions** - `if/then/else` works
3. **Console output** - `Console.print()` works
4. **Basic types** - `String`, `Int`, `Bool` work
5. **Let bindings** - Variable assignment works
6. **Functions** - Function definitions work
7. **Module imports** - Works with `import stdlib/module`
8. **Html module** - Fully functional for generating HTML strings
---
## Recommendations
### For Website MVP
The module system now works! The website can be:
1. **Generated from Lux** using the html module for HTML rendering
2. **CSS separate file** (already done in `static/style.css`)
3. **Lux code examples** embedded as text in HTML
### For Future
1. **Add FileSystem effect** for writing generated HTML to files
2. **Markdown-based documentation** parsed and rendered by Lux
3. **Live playground** using WASM compilation
---
## Test Results
| Feature | Status | Notes |
|---------|--------|-------|
| String concatenation | Works | `"<" + tag + ">"` |
| Conditionals | Works | `if x then y else z` |
| Console.print | Works | Basic output |
| Module imports | Works | `import stdlib/html` |
| Html module | Works | Full HTML generation |
| List.fold | Works | Fold over lists |
| List.concat | Works | Concatenate list of lists |
| String.replace | Works | String replacement |
| FileSystem | Unknown | Not tested |
---
## Date Log
| Date | Finding |
|------|---------|
| 2026-02-16 | Module import system not working, parse error at line 196-197 in html.lux |
| 2026-02-16 | Fixed: Parser newline handling in list expressions |
| 2026-02-16 | Fixed: Added `pub` to stdlib/html.lux exports |
| 2026-02-16 | Fixed: Changed List.foldl to List.fold |