feat: add rust lexer

.changeset/perky-plums-dig.md

@@ -0,0 +1,5 @@
+---
+'@fuzdev/fuz_code': minor
+---
+
+feat: add rust lexer

CLAUDE.md

@@ -44,7 +44,7 @@ fuz_code is a **syntax highlighting library**:
 
 - Runtime HTML generation with CSS classes
 - Hand-written single-pass lexers (zero regex), one per language
-- 8 built-in languages (TS, JS, CSS, HTML, JSON, Svelte, Markdown, Bash)
+- 9 built-in languages (TS, JS, CSS, HTML, JSON, Svelte, Markdown, Shell, Rust)
 - Extensible by writing a lexer (`SyntaxLang`)
 - Optional Svelte component (`Code.svelte`)
 
@@ -69,7 +69,7 @@ src/
 │   ├── syntax_styler.ts        # SyntaxStyler class: registry + lex/stylize facade
 │   ├── syntax_styler_global.ts # pre-configured global instance
 │   ├── lexer.ts                # lexer substrate: Lexer, TokenTypeRegistry, flat events, HTML render
-│   ├── lexer_*.ts              # hand-written lexers (json, ts, css, bash, markup, svelte, md)
+│   ├── lexer_*.ts              # hand-written lexers (json, ts, css, bash, markup, svelte, md, rust)
 │   ├── Code.svelte             # main Svelte component
 │   ├── CodeHighlight.svelte    # experimental CSS Highlight API
 │   ├── CodeTextarea.svelte     # experimental live-highlighted textarea
@@ -141,6 +141,11 @@ One `SyntaxLang` lexer per language, registered via `add_lang`:
   info matching and any-length closers, headings, blockquotes, lists, hr)
   with a per-block inline scan (emphasis, inline code, links, entities, raw
   markup via the markup scanner); fences embed their languages
+- `lexer_rust.ts` - Rust, registered as `rust` with `rs` as an alias: one
+  flat scan loop (nested block comments and raw strings resolve with counters;
+  attribute interiors lex inline under an `[`/`]` depth counter — no frame
+  machine), lifetime-vs-char `'` disambiguation, `name!` macros, doc-vs-plain
+  comment split, the r/b/c string prefixes
 
 Embedded languages resolve lazily by name through the registry (markdown
 fences → any language, markup `<script>`/`<style>`/`style=`/`on*=`, svelte
@@ -174,7 +179,7 @@ can't overflow the call stack; past the cap a region stays plain text.
 
 ## Supported languages
 
-`ts`, `js`, `css`, `html`, `json`, `svelte`, `md`, `sh`
+`ts`, `js`, `css`, `html`, `json`, `svelte`, `md`, `sh`, `rust`/`rs`
 
 ## Testing
 
@@ -296,7 +301,7 @@ Theme uses CSS variables from fuz_css:
 3. **Progress discipline** - every scan loop advances position or emits+advances
 4. **Nesting discipline** - no JS-call-stack recursion that scales with input
    nesting. Flat single-pass loops wherever the grammar allows (css, markup,
-   svelte, json, and md need no stacks at all); an explicit pooled frame
+   svelte, json, md, and rust need no stacks at all); an explicit pooled frame
    machine only where a construct's interior re-enters the same grammar
    (`lexer_ts.ts` templates/generics/annotations, `lexer_bash.ts`
    substitutions — use these as the template, including their inline fast

README.md

@@ -17,7 +17,7 @@ Highlights:
 - a minimal, explicit API to generate stylized HTML — `stylize(code, lang)`
 - stateless ES modules, instead of globals with side effects
 - written in TypeScript, with no runtime dependencies
-- eight built-in languages (see below), extensible by writing a lexer
+- nine built-in languages (see below), extensible by writing a lexer
 
 Two optional integrations:
 
@@ -112,6 +112,7 @@ Registered by default in `syntax_styler_global` — one hand-written lexer each:
 - [`css`](src/lib/lexer_css.ts)
 - [`json`](src/lib/lexer_json.ts) (with comments — jsonc)
 - [`sh`](src/lib/lexer_bash.ts) (POSIX/bash family — also serves `bash`/`shell`)
+- [`rust`](src/lib/lexer_rust.ts) (also serves `rs`)
 
 Add a language by writing a `SyntaxLang` lexer and registering it with
 `add_lang` — see the existing `lexer_*.ts` modules.

src/lib/code_sample.ts

@@ -5,6 +5,6 @@ export interface CodeSample {
 }
 
 // Languages ordered from simple to complex
-export const sample_langs = ['json', 'css', 'ts', 'html', 'svelte', 'md', 'sh'] as const;
+export const sample_langs = ['json', 'css', 'ts', 'rs', 'html', 'svelte', 'md', 'sh'] as const;
 
 export type SampleLang = (typeof sample_langs)[number];

src/lib/lexer.ts

@@ -480,6 +480,8 @@ export const matches_ci = (text: string, from: number, word: string): boolean =>
 
 export const is_digit = (c: number): boolean => c >= 48 && c <= 57;
 
+export const is_upper = (c: number): boolean => c >= 65 && c <= 90;
+
 export const is_ascii_alnum = (c: number): boolean =>
 	(c >= 48 && c <= 57) || ((c | 0x20) >= 97 && (c | 0x20) <= 122);
 

src/lib/lexer_md.ts

@@ -53,6 +53,7 @@ const T_LANG_JSON = token_type('lang_json');
 const T_LANG_SVELTE = token_type('lang_svelte');
 const T_LANG_BASH = token_type('lang_bash');
 const T_LANG_MD = token_type('lang_md');
+const T_LANG_RUST = token_type('lang_rust');
 const T_HEADING = token_type('heading');
 const T_HEADING_PUNCTUATION = token_type('punctuation', 'heading_punctuation');
 const T_BLOCKQUOTE = token_type('blockquote');
@@ -96,6 +97,7 @@ add_fence_lang('json', 'json', T_LANG_JSON);
 add_fence_lang('svelte', 'svelte', T_LANG_SVELTE);
 add_fence_lang('sh bash shell', 'sh', T_LANG_BASH);
 add_fence_lang('md markdown', 'md', T_LANG_MD);
+add_fence_lang('rust rs', 'rust', T_LANG_RUST);
 
 // line-internal whitespace — `is_space` from the toolkit includes newlines
 const is_line_space = (c: number): boolean => c === 32 || c === 9;