feat: Errors are now tracked, so duplicate errors don't report more than once.

chore: Updated TODOS.md.
feat: help for misspelled pipes is now shown to the user.
2026-10-25 18:21:54 -04:00 · 2026-07-31 10:15:33 -04:00 · 2026-07-31 09:33:15 -04:00 · 2026-07-31 09:25:38 -04:00 · 2026-07-31 09:24:01 -04:00 · 2026-07-31 08:57:36 -04:00
118 changed files with 1371 additions and 7407 deletions
@@ -1,78 +0,0 @@
-# Custom Pages build: compiles Thor from source via the Nix flake, uses it to
-# generate thor/site/, and publishes the result. Replaces the default Jekyll build.
-name: Deploy site to GitHub Pages
-
-on:
-  push:
-    branches: [main]
-    # Only rebuild when something that affects the output changes.
-    paths:
-      - "site/**"
-      - "**/*.odin"
-      - "defaults/**"
-      - "flake.nix"
-      - "flake.lock"
-      - ".github/workflows/pages.yml"
-  workflow_dispatch:
-
-permissions:
-  contents: read
-  pages: write
-  id-token: write
-
-# Allow one concurrent deployment; don't cancel an in-progress publish.
-concurrency:
-  group: pages
-  cancel-in-progress: false
-
-jobs:
-  build:
-    runs-on: ubuntu-latest
-    steps:
-      - name: Checkout
-        uses: actions/checkout@v4
-
-      - name: Install Nix
-        uses: cachix/install-nix-action@v30
-        with:
-          extra_nix_config: |
-            experimental-features = nix-command flakes
-
-      # Free cache: snapshots /nix/store into GitHub's Actions cache. The key is
-      # hashed over the inputs that change Thor's output, so a site-content-only
-      # change is a full hit (Thor not recompiled), while an Odin source change
-      # restores the nearest store and rebuilds just the delta.
-      - name: Cache /nix/store
-        uses: nix-community/cache-nix-action@v6
-        with:
-          primary-key: nix-${{ runner.os }}-${{ hashFiles('flake.lock', '**/*.nix', '**/*.odin') }}
-          restore-prefixes-first-match: nix-${{ runner.os }}-
-          gc-max-store-size-linux: 2G
-          purge: true
-          purge-prefixes: nix-${{ runner.os }}-
-          purge-created: 604800 # drop entries older than 7 days
-
-      # First run compiles Thor + the tree-sitter grammar libs; later runs
-      # restore them from the cache above and skip the compile.
-      - name: Build Thor
-        run: nix build .#default --print-build-logs
-
-      - name: Generate site
-        working-directory: site
-        run: ../result/bin/thor -base-url=https://oneidiotdeveloper.com/thor -minify
-
-      - name: Upload Pages artifact
-        uses: actions/upload-pages-artifact@v3
-        with:
-          path: site/public
-
-  deploy:
-    needs: build
-    runs-on: ubuntu-latest
-    environment:
-      name: github-pages
-      url: ${{ steps.deployment.outputs.page_url }}
-    steps:
-      - name: Deploy to GitHub Pages
-        id: deployment
-        uses: actions/deploy-pages@v4
@@ -2,18 +2,6 @@

 Thor is a static site generator written in [Odin](https://odin-lang.org), replacing Hugo for the `sbrow.github.io` blog. It lives at `./thor/` as a git subtree with its own `flake.nix`.

-## 🚫 DO NOT EDIT THE DOCS — BY HUMANS, FOR HUMANS
-
-> **THE DOCUMENTATION UNDER `thor/site/` IS HANDWRITTEN, BY HUMANS, FOR HUMANS.**
->
-> **NO AI, AGENT, BOT, ASSISTANT, OR OTHER NON-HUMAN MAY EDIT, REWRITE,
-> REPHRASE, REFORMAT, "IMPROVE," SUMMARIZE, OR GENERATE ANY FILE UNDER
-> `thor/site/` — EVER.**
->
-> These are not machine artifacts. A human wrote every word. AI may be
-> consulted as a sanity check, but the prose stays human. If you are not a
-> human, do not touch these files. See `thor/site/content/ai.md`.
-
 ## Architecture

 ```
@@ -32,7 +20,6 @@ thor/
 ├── treesitter/         # FFI types + grammar management (standalone package)
 ├── markdown/           # Content transformation pipeline (imports ../treesitter)
 ├── mustache/           # Template engine with lambdas + pipe filters + diagnostics
-├── diagnostics/        # Shared diagnostic utilities (suggest_correction)
 ├── content.odin        # Page struct, Pending_File, scan_content_files, collect_languages, load_page
 ├── render.odin         # Template rendering, Template_Context, sort_pages, RSS, sitemap
 ├── menus.odin          # Menu_Entry, DEFAULT_WEIGHT, build_menus, collect_auto_menus, merge_page_menus, parse_page_menus, parse_config_menus
@@ -54,10 +41,10 @@ thor/

 | File | Responsibility |
 |---|---|
-| `main.odin` | Entry point. Parses CLI flags via `core:flags`, sets logger level from `-verbose`/`-quiet`, calls `init_site`, `build_vfs`, wires `treesitter.grammar_dir`/`query_dir` from config, `site_load_content`, `render_site`. Optional Spall profiling via `SPALL` config flag. |
-| `site.odin` | `Flags` (CLI, includes `-verbose`/`-quiet`), `Config_File` (thor.json), `Site_Context` (template-facing: `title`, `description`, `base_url`, `params`, `og`, `menus`), `Site` (runtime state + arena + VFS + pages + `og`). `Feature` enum. `init_site(site, flags)` — takes pre-parsed `Flags`. Config menu parsing in `site_apply_config`. |
-| `content.odin` | `Page` struct (includes `weight`, `menus`, `params: json.Value`, `toc: string`, `og`), `Pending_File` struct, `scan_content_files` (section-aware walk that handles leaf bundles), `collect_languages` (pre-scan for code fence languages), `load_page` (falls back to file mtime, generates TOC via `md.generate_toc` when frontmatter `"toc": true`), `infer_layout`. Calls `md.process()` for the markdown pipeline. |
-| `render.odin` | Template rendering: `render_site`, `render_page_html`, `render_home_html`, `render_section`. `Template_Context` (unified render struct with `site: Site_Context`, `page: Page`, `menus`, `params`, `posts`, `pages`). 3-frame context stack via `[]any{ctx.site, ctx.page, ctx}`. `merge_params(site, page)` — shallow merge of site + page params. Error deduplication via `seen: ^map[string]bool` passed through render chain. `sort_pages` (weight primary, date secondary). `to_title_case` for section display names. VFS-based template loading with fallback chain (`get_template`). |
+| `main.odin` | Entry point. Sets `context.logger`, calls `init_site`, `build_vfs`, wires `treesitter.grammar_dir`/`query_dir` from config, `site_load_content`, `render_site`. Optional Spall profiling via `SPALL` config flag. |
+| `site.odin` | `Flags` (CLI), `Config_File` (thor.json), `Site_Context` (template-facing: `title`, `description`, `base_url`, `params`, `og`, `menus`), `Site` (runtime state + arena + VFS + pages + `og`). `Feature` enum. 5-step `init_site`. Config menu parsing in `site_apply_config`. |
+| `content.odin` | `Page` struct (includes `weight`, `menus: map[string]Menu_Entry`, `og`), `Pending_File` struct, `scan_content_files` (section-aware walk that handles leaf bundles), `collect_languages` (pre-scan for code fence languages), `load_page` (falls back to file mtime when no frontmatter date), `infer_layout`. Calls `md.process()` for the markdown pipeline. |
+| `render.odin` | Template rendering: `render_site`, `render_page_html`, `render_home_html`, `render_section`. `Template_Context` (unified render struct with `site: Site_Context`, `page: Page`, `menus`, `posts`, `pages`). 3-frame context stack via `[]any{ctx.site, ctx.page, ctx}`. `sort_pages` (weight primary, date secondary). `to_title_case` for section display names. VFS-based template loading with fallback chain (`get_template`). |
 | `menus.odin` | Menu system: `Menu_Entry {name, url, weight: Maybe(int)}`, `DEFAULT_WEIGHT = 10`. `build_menus` (priority chain: config → auto + page frontmatter, then `warn_all_duplicate_weights`). `collect_auto_menus` (sections + root-level pages, skips pages with explicit `"menus": "main"` frontmatter). `merge_page_menus` (frontmatter entries with effective weight fallback via nil check). `parse_page_menus` (string/array/object forms). `parse_config_menus` (from thor.json). `sort_menu_entries` / `compare_menu_entries` (weight primary via `.? or_else DEFAULT_WEIGHT`, name secondary). `warn_duplicate_weights` / `warn_all_duplicate_weights` (log when two entries in same menu have same explicitly-set weight). |
 | `minify.odin` | HTML/CSS minification via tree-sitter. Imports `ts "treesitter"`. |
 | `feed.odin` | RSS feed + sitemap XML. Uses `page.url` for canonical URLs. |
@@ -65,7 +52,7 @@ thor/
 | `assets.odin` | `copy_assets_dir` — iterates VFS entries with `assets/` prefix, minifies CSS, copies verbatim or via `os.copy_file`. |
 | `html.odin` | `strip_html_tags`, `unescape_html`, `generate_summary` (word-count truncation, zero-alloc), `generate_description` (HTML→plain text: strip tags, decode entities, collapse whitespace). |
 | `opengraph.odin` | `Open_Graph` struct (fields ordered per OGP spec, `is_article: Maybe(bool)`). `og_for_site(site)` for site defaults (from config + derived), `og_for_page(site_og, page)` for page-specific (overlay page.og + derive from page data). Description falls back to `generate_description(generate_summary(body_html))`. |
-| `frontmatter.odin` | JSON frontmatter parser (`{ }` delimited). Supports `layout`, `lastmod`, `weight: Maybe(int)`, `menus`, `params: json.Value`, `toc: bool`, and nested `og` object (via `json_get_open_graph`). Helpers: `json_get_string`, `json_get_bool`, `json_get_int` (returns `Maybe(int)`, nil for absent/invalid). |
+| `frontmatter.odin` | JSON frontmatter parser (`{ }` delimited). Supports `layout`, `lastmod`, `weight: Maybe(int)`, `menus`, and nested `og` object (via `json_get_open_graph`). Helpers: `json_get_string`, `json_get_bool`, `json_get_int` (returns `Maybe(int)`, nil for absent/invalid). |
 | `defaults.odin` | `DEFAULTS_PATH` constant, resolved at compile time via `#directory` so bundled templates ship in the binary. |

 ### Subpackages
@@ -73,17 +60,14 @@ thor/
 | Package | Files | Responsibility |
 |---|---|---|
 | `treesitter/` | `treesitter.odin` | FFI types (`Parser`, `Node`, `Query`, etc.), `@(link_prefix="ts_")` foreign bindings, grammar management (`Grammar_Store` with persistent allocator, `load_language`/`compile_query` building blocks, `ensure_parser`/`load_grammar` lazy loading, `preload_grammar`/`preload_grammars` for parallel loading with `sync.Mutex` cache protection), statically-linked HTML/CSS grammars |
-| `markdown/` | `markdown.odin` | `Extension` enum, `DEFAULT_EXTENSIONS`, `EXTENSION_NAMES`, `process(body, ext, file_path, allocator)` — full pipeline (clones cmark output, frees original), `parse_extension_list` (with "did you mean?" hints via `diagnostics.suggest_correction`), `apply_extension_config` (same), `resolve_extension_conflicts` (Footnotes wins over Sidenotes) |
-| | `footnotes.odin` | `strip_definitions` (pre-cmark, shared by `.Sidenotes` + `.Footnotes`), `inject_notes` (post-cmark sidenote rendering), `inject_footnotes` (post-cmark standard footnote rendering — numbered `<sup>` links + `<section class="footnotes"><ol>` at bottom). `.Sidenotes` and `.Footnotes` are mutually exclusive; `resolve_extension_conflicts` in `markdown.odin` picks `.Footnotes` if both are set. |
+| `markdown/` | `markdown.odin` | `Extension` enum, `DEFAULT_EXTENSIONS`, `process(body, ext, file_path)` — full pipeline, `parse_extension_list`, `apply_extension_config` |
+| | `footnotes.odin` | `strip_definitions` (pre-cmark), `inject_notes` (post-cmark) |
 | | `alerts.odin` | `inject_alerts` — GitHub alert blocks (`> [!NOTE]`) → styled blockquotes with semantic class names (`alert-note` etc.) |
 | | `emoji.odin` | `expand_emoji` — `:shortcode:` → unicode emoji |
 | | `sectionate.odin` | `wrap_sections` — splits HTML at `<h2>` into `<section>` wrappers |
 | | `highlight.odin` | Syntax highlighting via tree-sitter. Imports `../treesitter`. |
 | | `heading_ids.odin` | `inject_heading_ids` — adds `id` attributes to `<h1>`-`<h6>` from heading text. Slug-based, deduplicated. |
-| | `deflists.odin` | `convert_deflists` — pre-cmark pass. Scans for definition list patterns (`term\n\n: definition`) and converts to `<dl><dt><dd>` HTML blocks. Terms and definitions rendered through cmark individually for inline markdown. Consecutive pairs grouped into single `<dl>`. |
-| | `toc.odin` | `generate_toc(html, allocator)` — page-level feature (not a pipeline extension). Scans `<h1>`-`<h6>` for IDs (after `inject_heading_ids`), builds nested `<ul>` with `<a href="#id">` links. Called from `load_page` when frontmatter `"toc": true`. Depends on `.HeadingIDs` being enabled. |
 | `mustache/` | See [Mustache engine](#mustache-engine) below | Template engine |
-| `diagnostics/` | `diagnostics.odin` | Shared diagnostic utilities: `suggest_correction` (Levenshtein "did you mean?"), `format_error` (rust-style multi-line error formatter with `caret_start`/`caret_end` params), `line_col`, `line_text`, `count_lines`, `digit_count`, `display_width`, `should_colorize`, `write_gutter`. Used by `mustache` (tag-aware errors via `format_tag_error` wrapper), `markdown` (extension name suggestions), and `render.odin` (parse/render error display). |
 | `bench/` | `bench.odin` + `templates/` | Standalone template rendering benchmark. Generates 500 posts + 100 comments, renders with indented partials + inheritance + pipes. `--dump <path>` for output validation, positional arg for iteration count (default 250). |

 Icon SVGs live as HTML partials in `layouts/partials/icons/` (home, github, rss, chevron_up, star).
@@ -117,15 +101,13 @@ Page :: struct {
    weight:      Maybe(int),  // page ordering (nil = unset, defaults to DEFAULT_WEIGHT at comparison time)
    lastmod:     string,
    menus:       map[string]Menu_Entry,  // frontmatter menu assignments
-    params:      json.Value,  // per-page params (merged with site params at render time)
    content:     string,      // rendered HTML body
    og:          Open_Graph,
    draft:       bool,
-    toc:         string,      // generated table of contents HTML (empty if not requested)
+    starred:     bool,
    _is_index:   bool `private`,
 }
 ```
-```

 No `Page_Type` enum — page type is inferred from section + `_is_index`. Layout is inferred via `infer_layout(section, is_index)`:

@@ -174,13 +156,13 @@ All weight fields use `Maybe(int)` — nil means "unset," `some(v)` means explic

 Config is split into three structs with a clear 5-step initialization flow:

- **`Flags`** — CLI args only. Parsed once in `main.odin` via `core:flags`, passed to `init_site`. Includes path overrides (`--content`, `--assets`, `--output`, `--layouts`), build-mode toggles (`-drafts`, `-watch`, `-minify`), log level (`-verbose` → Debug, `-quiet` → Warning), and `-ext`/`-no-ext` for markdown extension overrides.
+- **`Flags`** — CLI args only. Parsed by `core:flags`. Includes path overrides (`--content`, `--assets`, `--output`, `--layouts`), build-mode toggles (`-drafts`, `-watch`, `-minify`), and `-ext`/`-no-ext` for markdown extension overrides.
 - **`Config_File`** — parsed from `thor.json` via `json.unmarshal_string`. Holds title, paths, `markdown_extensions` (JSON), `params` (JSON), `modules` (JSON array of relative paths), `og` (`Open_Graph` struct for site-level OG defaults).
 - **`Site`** — runtime state: arena, pages, modules, VFS, `features: bit_set[Feature]`, `markdown_extensions: bit_set[md.Extension]`, `og: Open_Graph` (resolved site-level OG).

 **`Feature` enum** — `Drafts`, `Minify`, `Watch`. Checked with `.Minify in site.features`.

-**`markdown.Extension` enum** (in the `markdown` package, not main) — `Emoji`, `Sidenotes`, `Alerts`, `Highlight`, `Sections`, `HeadingIDs`, `DefLists`, `Footnotes`. Default is `md.DEFAULT_EXTENSIONS` (currently `.Emoji, .Sidenotes, .Alerts, .HeadingIDs, .DefLists`). Configurable via:
+**`markdown.Extension` enum** (in the `markdown` package, not main) — `Emoji`, `Sidenotes`, `Alerts`, `Highlight`, `Sections`, `HeadingIDs`. Default is `md.DEFAULT_EXTENSIONS` (currently `.Emoji, .Sidenotes, .Alerts, .HeadingIDs`). Configurable via:
 - `thor.json`: `"markdown_extensions": { "emoji": true, "highlight": false, ... }`
 - CLI: `-ext:highlight,sections` (enable) / `-no-ext:emoji` (disable). Comma-separated, case-insensitive.

@@ -260,12 +242,10 @@ Lives in the `markdown` package. Entry point: `md.process(body, ext, file_path)`

 ```
 raw markdown
-  → md.strip_definitions     (if .Sidenotes || .Footnotes — pre-cmark)
-  → md.convert_deflists      (if .DefLists — pre-cmark)
+  → md.strip_definitions     (if .Sidenotes — pre-cmark)
  → cmark markdown_to_html   (Unsafe mode for HTML passthrough)
  → md.expand_emoji          (if .Emoji — post-cmark)
-  → md.inject_notes          (if .Sidenotes — post-cmark, sidenote rendering)
-  → md.inject_footnotes      (if .Footnotes — post-cmark, standard footnote rendering)
+  → md.inject_notes          (if .Sidenotes — post-cmark)
  → md.inject_alerts         (if .Alerts — post-cmark)
  → md.highlight_code        (if .Highlight — post-cmark)
  → md.inject_heading_ids    (if .HeadingIDs — post-cmark, pre-sections)
@@ -297,17 +277,17 @@ Template_Context :: struct {
    site:        Site_Context,   // site-level data (title, description, base_url, params, og)
    menus:       map[string][]Menu_Entry,  // generated menu data (copied from site, resolves above Page.menus)
    now:         string,          // UTC ISO 8601 build timestamp
+    title:       string,          // computed browser title ("Page | Site")
    date_format: string,          // from site.date.format (thor.json)
    timezone:    ^datetime.TZ_Region,  // for format pipe
    og:          Open_Graph,      // computed per-page OG
-    params:      json.Value,      // merged site + page params (resolves above Page.params)
    page:        Page,            // current page
    pages:       [dynamic]Page,   // home page list
    posts:       [dynamic]Page,   // section post list
 }
 ```

-`Site_Context` is embedded in `Site` via `using site_context`. Fields like `site.title`, `site.menus`, `site.params` are accessed directly on `Site` through promotion. `Template_Context.menus` is copied from `site.menus` to resolve above `Page.menus` (frontmatter assignments) on the context stack. `Template_Context.params` is set per-page via `merge_params(site.params, page.params)` — site params overlaid with page params. Browser title is handled by the `{{> title}}` partial (not a computed field).
+`Site_Context` is embedded in `Site` via `using site_context`. Fields like `site.title`, `site.menus`, `site.params` are accessed directly on `Site` through promotion. `Template_Context.menus` is copied from `site.menus` to resolve above `Page.menus` (frontmatter assignments) on the context stack.

 `render_site` pre-parses all partials and the base layout once (via `mustache.parse`), then per-layout templates are cached in `get_template`. Year-based grouping on section index pages is done in the template via `{{#posts | group_by year}}` (see Pipes extension below).

@@ -419,11 +399,12 @@ Spec-compliant implementation at `mustache/`. See `mustache/SPEC.md` for the imp
 |---|---|
 | `mustache.odin` | Public API (`parse`, `render`, `Template`), parser (`parse_section`), renderer (`render_nodes` with `Indent_State` for partial indentation), template inheritance (`merge_block_overrides`), `delete_template`/`delete_partials`. Pipe support in Variable/Unescaped/Section/Inverted tags. |
 | `tokenizer.odin` | Tokenizer (template string → `[]Token`), standalone whitespace detection |
-| `data.odin` | Reflection-based data model: `base_value` (peels union/any/nested-any layers), `lookup_in` (structs + maps, handles `Type_Info_Any` value kind in maps), `resolve_name`, `is_truthy`, `any_to_string`, `write_value`, `list_info`, `extract_list_element`, `collect_map_keys` |
-| `pipes.odin` | Pipes extension: `Pipe_Op` enum (`.Format`, `.Group_By`), `pipe_op_from_string`/`pipe_op_candidates` (reflection-based enum name lookup), `Pipe_Filter` AST (with `op_pos` for diagnostics), `parse_pipeline` (tracks byte offsets via `strings.index`), `apply_pipeline`, `apply_filter` (exhaustive enum switch), `apply_group_by`, `apply_format`. Misspelled pipe op suggestions via `suggest_correction`. Stored on `Node.filters`; render-scoped results in temp allocator. |
-| `diagnostic.odin` | Mustache-specific diagnostics: `context_extent` (scans `{{ }}` tags for caret underline), `format_tag_error` (wraps `diags.format_error` + `context_extent`), `format_render_error` (formats `Error` using `Template` source/path). Generic error formatting lives in `diagnostics/` package. |
+| `data.odin` | Reflection-based data model: `base_value` (peels union/any/nested-any layers), `lookup_in` (structs + maps, handles `Type_Info_Any` value kind in maps), `resolve_name`, `is_truthy`, `any_to_string`, `list_info`, `extract_list_element`, `call_interp_lambda`/`call_section_lambda` |
+| `pipes.odin` | Pipes extension: `Pipe_Filter` AST, `parse_pipeline` (takes `pos`), `apply_pipeline`, `apply_filter` (switch dispatch: `group_by` + `format`), `apply_group_by`, `apply_format`. Stored on `Node.filters`; render-scoped results in temp allocator. |
+| `diagnostic.odin` | Rust-style error formatter: `format_error` (multi-line context, ANSI colors via `core:terminal/ansi`, `colorize` param), `format_render_error` (formats `Error`), `line_col`, `line_text`, `context_extent`, `count_lines`, `digit_count`, `should_colorize`. |
 | `suggest.odin` | Strict-warning helpers: `validate_key_path` (walks dotted path, crosses maps silently), `suggest_correction` (Levenshtein via `core:strings/levenshtein_distance`), `collect_struct_keys` (via reflection, recurses into `using`), `struct_has_field` (distinguishes missing field from nil value — needed for `Maybe(bool)`), `collect_partial_names`, `collect_block_names`. |
 | `spec_test.odin` | JSON spec test runner — loads `spec/specs/*.json`, runs each test case. Uses `log.nil_logger()` to suppress expected warnings. |
+| `lambda_test.odin` | Spec lambda tests |
 | `pipes_test.odin` | Pipe filter tests (`group_by` + `format`) |
 | `diagnostic_test.odin` | Golden-output tests for `format_error` (multi-line context, edge cases, alignment, caret position, hint) + parser error message brace-escaping |
 | `suggest_test.odin` | Tests for `validate_key_path`, `suggest_correction`, `struct_has_field` with `Maybe(bool)` and `using`-promoted fields |
@@ -451,11 +432,7 @@ render(tmpl, data, partials) → render_nodes (walks flat node array against con

 Rust-style error messages with multi-line source context, caret underlines, and Levenshtein suggestions. ANSI colors via `core:terminal/ansi`, gated on `should_colorize()` (TTY detection on stderr).

-**Error types**: `Error_Body{msg, pos, kind, source, path, span, hint}` where `kind` is `Error_Kind.Syntax` (parse-time) or `Error_Kind.Data` (render-time). `source`/`path` carry the template the error originated in (set by `tag_error` — enables correct file/line for errors inside partials). `span` controls caret underline width (used by pipe op diagnostics). `hint` carries "did you mean?" suggestions. `Error` is a single-variant union wrapping `Error_Body` (nilable for `!= nil` / `or_return`).
-
-**Error deduplication**: `render_template` takes `seen: ^map[string]bool`. Duplicate errors (same formatted diagnostic) are suppressed across page renders within a single build.
-
-**Partial source tracking**: `tag_error(err, current)` stamps render-time errors with `current.source`/`current.path` at the 4 `apply_pipeline` call sites in `render_nodes`. Ensures errors inside partials point at the partial file, not the top-level template.
+**Error types**: `Error_Body{msg, pos, kind}` where `kind` is `Error_Kind.Syntax` (parse-time) or `Error_Kind.Data` (render-time). `Error` is a single-variant union wrapping `Error_Body` (nilable for `!= nil` / `or_return`).

 **Strict-by-default warnings** — `render_nodes` emits `log.warnf` diagnostics for:
 - Unknown keys in `{{k}}`, `{{{k}}}`, `{{#k}}`, `{{^k}}` (via `validate_key_path` + `suggest_correction`)
@@ -465,12 +442,18 @@ Rust-style error messages with multi-line source context, caret underlines, and

 **Exceptions** (no warning):
 - `{{.}}` and dot-prefixed names (current context)
- Paths that cross a map (e.g., `params.*`) — validated for typos via Levenshtein: close matches warn with a suggestion, genuinely absent keys are suppressed silently
+- Paths that cross a map (e.g., `params.*` — user-defined namespace)
 - `Maybe(bool)` fields with nil value (field exists, value is nil — distinguished via `struct_has_field`)
- Found fields whose value is nil/empty (e.g., nil `json.Value` union) — the field exists, looking up sub-keys is valid "not found" behavior

 **Block override source tracking**: `Block_Override.source: Template` ensures warnings inside block overrides point at the override's source file (e.g., `page.html`), not the parent template (`base.html`).

+### Lambdas
+
+Spec-compliant. Stored as `any` values in the data context.
+
+- **Interpolation lambdas**: `proc() -> string`, `proc() -> int`, `proc() -> bool` — called via `call_interp_lambda`, result stringified and escaped.
+- **Section lambdas**: `proc(string) -> string`, `proc(string) -> int`, `proc(string) -> bool` — called via `call_section_lambda` with the raw section text (`node.content`). String result is re-parsed as mustache and rendered against the current context stack.
+
 ### Pipes

 `{{key | op args…}}` for interpolation, `{{#key | op args…}}…{{/key}}` for sections. Stored as `[dynamic; MAX_PIPES]Pipe_Filter` on each `Node`. Applied in the renderer via `apply_pipeline` before truthiness/interpolation. Implemented filters:
@@ -487,13 +470,12 @@ See `mustache/EXTENSIONS.md`.

 ## Known limitations

+- cmark allocates via C malloc, not the arena. HTML output leaks until process exit (problematic in watch mode — see `TODOS.md`).
 - CSS/JS cache busting uses manual `?v=N` query params instead of content hashing.
 - Tree-sitter grammar/query paths must be configured manually via `thor.json` (`grammars`, `queries`) — no auto-discovery. HTML/CSS are statically linked.
 - `map[string]any` only works through `lookup_in`'s special-case handling; thor otherwise uses structs.
 - `format_f64` in mustache brute-forces shortest float representation.
 - Content directory not mounted in VFS (modules can ship templates/assets but not content packs yet).
- Per-page params rendering is incomplete — `merge_params` produces correct data, but `base_value` may return nil for `json.Value` fields accessed through the `[]any` context stack via reflection in some cases. Warning suppression masks this; actual rendering may not work for all param values.
- Lambda support removed. Mustache lambdas (`proc() -> string` in data context) are not supported. Pipes cover data transformation.

 ## Design decisions

@@ -501,8 +483,6 @@ You may never, *ever* remove `TODO:` or `FIXME:` comments. Those are for humans,
 See `HUGO.md` for analysis of why thor doesn't need Hugo's shortcode context isolation.
 See `mustache/SPEC.md` for the original implementation specification.
 See `mustache/EXTENSIONS.md` for non-standard extensions (pipes).
-See `DIAGNOSTICS.md` for the two-tier diagnostic system design.
-See `mustache/TOKENIZERS.md` for tokenizer architecture comparison (Go, Liquid, Thor).

 ## Odin language facts

@@ -517,7 +497,6 @@ These are things that are easy to get wrong:
 - **`Maybe(T)` unwrap syntax:** `value.? or_else default`. Not `value or_else default` — `or_else` works on the `?T` returned by `.?`, not on `Maybe(T)` directly.
 - **`Maybe(T)` equality:** `a == b` works directly between two `Maybe(T)` values (nil == nil → true, some(5) == some(5) → true, nil == some(5) → false). Also `a == 5` works (int coerces to `Maybe(int)`).
 - **File logger in tests:** `log.create_file_logger(&f)` + `context.logger = logger` captures log output. Must be set inline in the test proc (not via a helper proc) for context propagation. Clean up with `log.destroy_file_logger(logger)` then `os.read_entire_file_from_path` to verify output.
- **`fmt.sbprintf` writes directly to a `strings.Builder`.** Prefer `fmt.sbprintf(&sb, fmt, args...)` over `fmt.aprintf(fmt, args...)` + `defer delete` + `strings.write_string`. The `aprintf` pattern allocates an intermediate string, requires manual cleanup, and queues a `defer delete` per loop iteration. `sbprintf` avoids all of this.

 ## TODO

@@ -1,135 +0,0 @@
-# Asset Pipeline Plan
-
-Design for letting thor users bring their own CSS toolchain (Sass, PostCSS,
-Lightning CSS, Tailwind, …).
-
-## Thesis
-
-Let thor users bring their own CSS toolchain by **orchestrating the tool they
-already installed** — never bundling or reimplementing one. This is expressed
-through thor's existing mustache pipe grammar with two new ops: `tool` and
-`fingerprint`.
-
-The earlier research design (content-addressed registry + deferred token
-resolution + a separate post-render path for "whole-output scanners") is
-**discarded**. It was overbuilt. Every tool — Tailwind included — scans
-*source*, which exists at render time, so every transform runs **eagerly,
-synchronously, and memoized**. No placeholder tokens, no post-render
-string-replace pass, no second integration surface.
-
-## The two ops
-
-```html
-<link href="{{ "css/main.scss" | tool sass | fingerprint | rel_url }}">
-```
-
-```json5
-// thor.json
-{
-  "tools": {
-    "sass":         "sass $in $out",
-    "postcss":      "postcss $in -o $out",
-    "lightningcss": "lightningcss --minify $in -o $out",
-    "tailwind":     "tailwindcss -i $in -o $out"
-  }
-}
-```
-
-## Core model
-
-The value flowing through the pipe is **bytes held by thor**, with a logical
-name attached (`css/main.css`). Each stage transforms the bytes; the **final
-stage publishes** one file to the output dir and returns its URL.
-
- A literal (`"css/main.scss"`) or context key names the **source** file on disk.
- `| tool <name>` runs a subprocess, replacing the carried bytes with its output.
- `| fingerprint` writes the content-addressed file and returns its hashed URL.
- The chain composes with the existing `rel_url` for base_url subpaths.
-
-**Memoization:** results are cached per build, keyed by (op-chain, source). The
-first page to reference an asset runs the tool; every other page hits the cache.
-This is what stops N pages from shelling out N times.
-
-## `| tool <name>`
-
-Looks up `<name>` in the `thor.json` `tools` table and runs its command
-template. `$in` and `$out` are **optional per-tool I/O adapters**:
-
-| Template has | Behavior |
-|---|---|
-| `$in` | thor writes current bytes to a scratch path, substitutes it |
-| no `$in` | thor pipes current bytes to the tool's **stdin** |
-| `$out` | thor substitutes a scratch path, reads it back after |
-| no `$out` | thor captures the tool's **stdout** |
-
-Defaulting to stdin/stdout avoids scratch-file overhead when the tool supports
-it. Tools that can't stream (Lightning CSS wants a path; Tailwind v3 needs an
-input file) just get `$in`/`$out` in their template. Either way the bytes are
-thor's, so chaining works regardless of each tool's preference:
-`| tool sass | tool postcss` composes even if sass streams and postcss wants
-files.
-
- **`$out` is always thor-controlled scratch** — the tool never picks the
-  published filename.
- **Extension derivation:** thor owns the output name; `main.scss` through
-  `tool sass` becomes `main.css`.
- **Errors:** thor gates on exit code and surfaces the tool's stderr through one
-  code path. If the tool is missing from PATH, fail with its stderr — never
-  bundle, never assume a runtime.
- **A bare `tool` (no `fingerprint`) still publishes** — to the logical unhashed
-  path (`output/css/main.css`), returning `/css/main.css`.
-
-## `| fingerprint`
-
-Content-addresses whatever bytes it currently holds. It has **two outputs**:
-
- **Side effect:** writes `css/main.<hash>.css` to the output dir.
- **Return value:** the hashed URL string `/css/main.<hash>.css` (feeds
-  `rel_url`).
-
-Properties:
-
- **Works standalone**, no `tool` needed: `{{ "css/main.css" | fingerprint }}`
-  is plain cache-busting of a hand-written file — probably the most common use.
- **Position decides what's hashed** — it hashes current bytes, so it belongs
-  **last** in the chain. No terminal-only special-casing; the rule is just "put
-  it last."
- **Hash format:** `name.<hash>.ext` (CDN-friendly), 8 hex chars of SHA-256.
-
-## Tailwind is not special
-
-It scans your **source** (`layouts/`, `content/`, partials) — which exists at
-render time — so it runs synchronously like sass. Its `thor.json` entry just
-needs content globs pointing at the source tree (or v4 auto-detection). The
-**one** asymmetry is watch invalidation: sass re-runs when `main.scss` changes;
-Tailwind must re-run when *any* scanned source changes. That's a cache-key
-breadth difference, not an architectural one. The cost of scanning source vs.
-output is losing dynamically-assembled class tokens (`col-{{n}}`) — a Tailwind
-anti-pattern whose sanctioned fix is a safelist, not warping the build.
-
-## Interaction: `copy_assets_dir`
-
-Today thor copies everything under `assets/` to the output dir. Any source
-consumed by a pipe must be **excluded from that plain copy** — otherwise you'd
-ship the raw `main.scss` (or a dead unhashed `main.css` alongside the hashed
-one). The pipe owns the write for assets it touches.
-
-## Implementation steps (separable)
-
-1. **`| fingerprint`** — no subprocess, no config. Hash + write + return URL,
-   plus the `copy_assets_dir` exclusion. Smallest blast radius, independently
-   useful, ships first.
-2. **`| tool <name>`** — `thor.json` tools table, the `$in`/`$out`/stdin/stdout
-   adapter, subprocess + exit-code/stderr handling, extension derivation,
-   memoization. Composes with step 1.
-
-## Open questions for implementation
-
- **Watch invalidation:** simplest correct v1 recomputes the memo each build
-  (cheap for the common single-asset case). A cross-build skip keyed on source
-  content/mtime is a later optimization — and Tailwind's broad content-glob key
-  is where it matters most.
- **Source-map sidecars:** sass/lightning emit `.map` files next to `$out`;
-  disable in the template or ignore the sidecar. Minor.
- **Error policy default:** hard-fail the build vs. warn-and-skip on tool
-  failure/absence. Lean fail; possibly configurable later.
@@ -1,96 +0,0 @@
-# Visualizing & Managing `thor` Binary Size
-
-How to inspect what makes up the compiled `thor` binary, and how to keep that
-inspection useful once we bundle layout/asset files into the binary.
-
-## Quick inspection (tools already available)
-
-The binary is a not-stripped ELF, so it ships symbols we can break down.
-
-```bash
-# Section totals (text / data / bss)
-size ./thor
-
-# Top symbols by size
-nm --print-size --size-sort --radix=d ./thor \
-  | awk '{sz=$2+0; sub(/^[0-9a-fA-F]+ +[0-9]+ +[A-Za-z] +/,""); printf "%8d  %s\n", sz, $0}' \
-  | sort -rn | head -25
-
-# Aggregate symbol sizes by Odin module namespace (foo::bar -> "foo")
-nm --print-size --size-sort --radix=d ./thor \
-  | awk '{sz=$2+0; rest=$0; sub(/^[0-9a-fA-F]+ +[0-9]+ +[A-Za-z] +/,"",rest);
-          ns=rest; if (rest ~ /::/){sub(/::.*/,"",ns)}
-          else if (rest ~ /^ts_/){ns="treesitter (ts_*)"} else {ns="<other/runtime>"};
-          s[ns]+=sz} END{for (k in s) printf "%9d  %s\n", s[k], k}' \
-  | sort -rn | head -25
-```
-
-As of this writing the top contributors were: tree-sitter tables (~118 KB),
-`main` (~96 KB), runtime/other (~93 KB), `mustache` (~67 KB), then `fmt` /
-`encoding_json` from the stdlib.
-
-## Better visualization: `bloaty`
-
-`bloaty` (Google's binary-size profiler) is the right tool. Not installed, but
-runnable via Nix:
-
-```bash
-bloaty ./thor                      # by section
-bloaty ./thor -d symbols -n 30     # by symbol, hierarchical
-bloaty ./thor -d sections,symbols  # section then symbols within each
-# via nix (no install):
-nix run nixpkgs#bloaty -- ./thor -d symbols -n 30
-```
-
-For a stripped-size estimate:
-
-```bash
-readelf -S ./thor | grep -E 'debug|symtab|strtab'   # strippable overhead
-strip -s ./thor -o thor.stripped && size thor.stripped
-```
-
-Odin also supports smaller builds: `odin build . -o:size -no-bounds-check`.
-
-## IMPORTANT: keeping bloaty useful once we bundle layouts into the binary
-
-Today the VFS (`vfs.odin`) reads layout/asset files lazily from disk at runtime
-(`VFS_Entry.fs_path` + `os.read_entire_file_from_path`); `VFS_Entry.data` is
-never populated at build time. So embedded content is **not** in `./thor` yet.
-
-When we switch to embedding via Odin `#load(...)`, the file bytes land in the
-binary's `.rodata` section. Whether bloaty can show *per-file* sizes depends on
-how we emit the embed:
-
- **Section total — always visible.** `.rodata` grows by ~the total bundled
-  size. `bloaty ./thor -d sections` (or `size`) shows the aggregate, and we can
-  diff before/after. This always works.
-
- **Per-file breakdown — only if each file is a NAMED symbol.**
-  - A generated `map[string]VFS_Entry` literal full of inline `#load(...)`
-    expressions makes each blob an **anonymous** `.rodata` constant. Bloaty
-    lumps them together — no per-file breakdown.
-  - Assigning each file to its own **package-level variable** makes each a named
-    global symbol, so `bloaty ./thor -d symbols` (and the `nm` recipe above)
-    show each file individually:
-
-    ```odin
-    layout_base_html   := #load("defaults/layouts/base.html")
-    layout_single_html := #load("defaults/layouts/single.html")
-    // ...then reference these in the VFS map instead of inlining #load.
-    ```
-
-### Decision to bake into the bundling codegen
-
-When we build the codegen that replaces the runtime `mount_dir` disk walk, have
-it emit **one named package-level variable per embedded file** (not an anonymous
-map literal). That single choice is what makes bloaty / `nm` per-file analysis
-possible — cheap to do up front, painful to retrofit.
-
-Additional options that help regardless of symbolization:
-
-1. Build with Odin `-debug` (DWARF) and use `bloaty ./thor -d compileunits` to
-   attribute embedded bytes to the generated source file.
-2. Since the generator already knows every file and its size, emit a size
-   manifest at bundle time (`path -> bytes`, sorted / treemap). This is the most
-   reliable per-file view and can't drift from reality; bloaty then just
-   confirms the `.rodata` total matches.
@@ -1,38 +0,0 @@
-## Code Review
-
-A human should manually review every file in the project. AI cannot complete
-these tasks.
-
- [ ] Review every file in thor
-  - [ ] Review assets.odin
-  - [ ] Review content.odin
-  - [ ] Review defaults.odin
-  - [ ] Review feed.odin
-  - [ ] Review frontmatter.odin
-  - [ ] Review main.odin
-  - [ ] Review minify.odin
-  - [ ] Review `markdown/`
-    - [x] Review alerts.odin
-    - [x] Review alerts_test.odin
-    - [x] Review emoji.odin
-    - [x] Review emoji_test.odin
-    - [ ] Review footnotes.odin
-    - [ ] Review footnotes_test.odin
-    - [ ] Review deflists.odin
-    - [ ] Review deflists_test.odin
-    - [ ] Review highlight.odin
-    - [ ] Review markdown.odin
-    - [x] Review sectionate.odin
-    - [x] Review sectionate_test.odin
-    - [ ] Review suggest.odin
-    - [ ] Review suggest_test.odin
-  - [ ] Review `diagnostics/`
-    - [ ] Review diagnostics.odin
-    - [ ] Review diagnostics_test.odin
-  - [x] Review opengraph.odin
-  - [ ] Review render.odin
-  - [x] Review site.odin
-  - [ ] Review treesitter/treesitter.odin
-  - [ ] Review vfs.odin
- [ ] Review procs
-  - [ ] markdown.transform_alert
@@ -1,379 +0,0 @@
-# This file contains tests for all known mustache syntax errors and their 
-# desired error messages. Each one is converted into an Odin test.
-name: Missing 2nd closing brace
-enable: false # TODO: Re-enable when passing
-should: error
-input: |
-  <menu>
-    <a class="goto-top opacity-0" href="#">{{> icons/chevron_up}</a>
-    {{#params.social}}
-    <a href="{{url}}" target="_blank" rel="noopener noreferrer me" title="{{name}}">{{>* icon}}</a>
-    {{/params.social}}
-  </menu>
-  <p>Proudly built with <a href="https://github.com/sbrow/thor/">Thor</a> and <a
-expected: |
-  Missing closing brace
-   --> <test>:1:1
-  1 | <menu>
-  2 | <a class="goto-top opacity-0" href="#">{{> icons/chevron_up}</a>
-    |                                        ^^ opened           ^ never closed
-  3 | {{#params.social}}
-  4 |   <a href="{{url}}" target="_blank" rel="noopener noreferrer me" title="{{name}}">{{>* icon}}</a>
-  5 | {{/params.social}}
-    |
---
-name: Missing 2nd opening brace
-enable: false # TODO: Re-enable when passing
-should: error
-input: |
-  <menu>
-    <a class="goto-top opacity-0" href="#">{> icons/chevron_up}}</a>
-    {{#params.social}}
-    <a href="{{url}}" target="_blank" rel="noopener noreferrer me" title="{{name}}">{{>* icon}}</a>
-    {{/params.social}}
-  </menu>
-  <p>Proudly built with <a href="https://github.com/sbrow/thor/">Thor</a> and <a
-expected: |
-  Missing opening brace
-   --> <test>:1:1
-  1 | <menu>
-  2 | <a class="goto-top opacity-0" href="#">{> icons/chevron_up}}</a>
-    |                                         ^ opened          ^^ closed
-  3 | {{#params.social}}
-  4 |   <a href="{{url}}" target="_blank" rel="noopener noreferrer me" title="{{name}}">{{>* icon}}</a>
-  5 | {{/params.social}}
-    |
---
-# TODO: make it work
-name: Invalid timezone
-enable: false
-should: warn
-input: |
-  {
-    "date": {
-      "timezone": "America/New_Yorkside"
-    }
-  }
-expected: "Did you mean 'America/New_York'? pointing to line+col no in the config"
-actual: |
-  [WARN ] --- [148:init_site()] unable to load timezone 'America/New_Yorkside'
---
-name: Mismatched section tags
-enable: false # TODO: Re-enable when passing
-should: error
-input: |
-  <footer>
-    <menu>
-      <a class="goto-top opacity-0" href="#">{{> icons/chevron_up}}</a>
-      {{#params.ocial}}
-      <a href="{{url}}" target="_blank" rel="noopener noreferrer me" title="{{name}}">{{>* icon}}</a>
-      {{/params.social}}
-    </menu>
-    <p>Proudly built with <a href="https://github.com/sbrow/thor/">Thor</a> and <a
-        href="https://edwardtufte.github.io/tufte-css/">Tufte CSS</a>
-    </p>
-    <p><small>&copy;</small> {{now | format "2006" }} {{params.author.name}}</p>
-  </footer>
-expected: |
-  Mismatched section: 'ocial' opened, 'social' closed
-   --> <test>:4:5
-    |
-  4 |     {{#params.ocial}}
-    |               ^^^^^ opened — did you mean 'social'?
-  5 |     <a href="{{url}}" target="_blank" rel="noopener noreferrer me" title="{{name}}">{{>* icon}}</a>
-  6 |     {{/params.social}}
-    |               ^^^^^^ expected {{/params.ocial}}, got {{/params.social}}
-    |
---
-name: Missing closing section
-enable: false # TODO: Re-enable when passing
-should: error
-input: |
-  <footer>
-    <menu>
-      <a class="goto-top opacity-0" href="#">{{> icons/chevron_up}}</a>
-      {{#params.social}}
-      <a href="{{url}}" target="_blank" rel="noopener noreferrer me" title="{{name}}">{{>* icon}}</a>
-      <!-- {{/params.social}} -->
-    </menu>
-    <p>Proudly built with <a href="https://github.com/sbrow/thor/">Thor</a> and <a
-        href="https://edwardtufte.github.io/tufte-css/">Tufte CSS</a>
-    </p>
-    <p><small>&copy;</small> {{now | format "2006" }} {{params.author.name}}</p>
-  </footer>
-expected: |
-  Missing closing tag '{{/params.social}}'
-   --> <test>:4:5
-    |
-  4 |     {{#params.social}}
-    |     ^^^^^^^^^^^^^^^^^ opened
-  5 |     <a href="{{url}}" target="_blank" rel="noopener noreferrer me" title="{{name}}">{{>* icon}}</a>
-  6 |     <!-- {{/params.social}} -->
-    |          ^^^^^^^^^^^^^^^^^^ closing tag is commented out
-    |
---
-name: empty tag
-enable: true
-should: error
-input: |
-  <footer>
-    <menu>
-      <a class="goto-top opacity-0" href="#">{{> icons/chevron_up}}</a>
-      {{#params.social}}
-      <a href="{{url}}" target="_blank" rel="noopener noreferrer me" title="{{name}}">{{>* icon}}</a>
-      {{/params.social}}
-    </menu>
-    <p>Proudly built with <a href="https://github.com/sbrow/thor/">Thor</a> and <a
-        href="https://edwardtufte.github.io/tufte-css/">Tufte CSS</a>
-    </p>
-    <p><small>&copy;</small> {{}} {{params.author.name}}</p>
-  </footer>
-expected: |
-  Empty tag
-    --> <test>:11:28
-     |
-   9 |       href="https://edwardtufte.github.io/tufte-css/">Tufte CSS</a>
-  10 |   </p>
-  11 |   <p><small>&copy;</small> {{}} {{params.author.name}}</p>
-     |                            ^^^^ tag must contain a key
-  12 | </footer>
-     |
---
-name: empty raw tag
-enable: true
-should: error
-input: |
-  Hello, {{&}}
-expected: |
-  Empty tag
-   --> <test>:1:8
-    |
-  1 | Hello, {{&}}
-    |        ^^^^^ tag must contain a key
-    |
---
-name: Missing pipe arguement
-enable: true
-should: error
-input: |
-  <footer>
-    <menu>
-      <a class="goto-top opacity-0" href="#">{{> icons/chevron_up}}</a>
-      {{#params.social}}
-      <a href="{{url}}" target="_blank" rel="noopener noreferrer me" title="{{name}}">{{>* icon}}</a>
-      {{/params.social}}
-    </menu>
-    <p>Proudly built with <a href="https://github.com/sbrow/thor/">Thor</a> and <a
-        href="https://edwardtufte.github.io/tufte-css/">Tufte CSS</a>
-    </p>
-    <p><small>&copy;</small> {{now | group_by }} {{params.author.name}}</p>
-  </footer>
-expected: |
-  group_by expects 1 argument, got 0
-    --> <test>:11:36
-     |
-   9 |       href="https://edwardtufte.github.io/tufte-css/">Tufte CSS</a>
-  10 |   </p>
-  11 |   <p><small>&copy;</small> {{now | group_by }} {{params.author.name}}</p>
-     |                                    ^^^^^^^^ missing argument
-  12 | </footer>
-     |
---
-name: Double dot access
-enable: false # TODO: Re-enable when passing
-should: error
-input: |
-  <footer>
-    <menu>
-      <a class="goto-top opacity-0" href="#">{{> icons/chevron_up}}</a>
-      {{#params.social}}
-      <a href="{{url}}" target="_blank" rel="noopener noreferrer me" title="{{name}}">{{>* icon}}</a>
-      {{/params.social}}
-    </menu>
-    <p>Proudly built with <a href="https://github.com/sbrow/thor/">Thor</a> and <a
-        href="https://edwardtufte.github.io/tufte-css/">Tufte CSS</a>
-    </p>
-    <p><small>&copy;</small> {{now | format "2006" }} {{params..name}}</p>
-  </footer>
-expected: |
-  Invalid key path: consecutive dots
-   --> <test>:11:60
-    |
-   9 |       href="https://edwardtufte.github.io/tufte-css/">Tufte CSS</a>
-  10 |   </p>
-  11 |   <p><small>&copy;</small> {{now | format "2006" }} {{params..name}}</p>
-     |                                                             ^^ missing path name
-  12 | </footer>
-     |
---
-name: Too many opening braces
-enable: false # TODO: Re-enable when passing
-should: error
-input: |
-  <footer>
-    <menu>
-      <a class="goto-top opacity-0" href="#">{{> icons/chevron_up}}</a>
-      {{#params.social}}
-      <a href="{{url}}" target="_blank" rel="noopener noreferrer me" title="{{name}}">{{>* icon}}</a>
-      {{/params.social}}
-    </menu>
-    <p>Proudly built with <a href="https://github.com/sbrow/thor/">Thor</a> and <a
-        href="https://edwardtufte.github.io/tufte-css/">Tufte CSS</a>
-    </p>
-    <p><small>&copy;</small> {{{{now | format "2006" }}} {{params.author.name}}</p>
-  </footer>
-expected: |
-  Too many opening braces
-   --> <test>:11:28
-     |
-   9 |       href="https://edwardtufte.github.io/tufte-css/">Tufte CSS</a>
-  10 |   </p>
-  11 |   <p><small>&copy;</small> {{{{now | format "2006" }}} {{params.author.name}}</p>
-     |                            ^^^^ expected {{{, got {{{{
-  12 | </footer>
-     |
---
-name: Too many closing braces
-enable: false # TODO: Re-enable when passing
-should: error
-input: |
-  <footer>
-    <menu>
-      <a class="goto-top opacity-0" href="#">{{> icons/chevron_up}}</a>
-      {{#params.social}}
-      <a href="{{url}}" target="_blank" rel="noopener noreferrer me" title="{{name}}">{{>* icon}}</a>
-      {{/params.social}}
-    </menu>
-    <p>Proudly built with <a href="https://github.com/sbrow/thor/">Thor</a> and <a
-        href="https://edwardtufte.github.io/tufte-css/">Tufte CSS</a>
-    </p>
-    <p><small>&copy;</small> {{now | format "2006" }}}} {{params.author.name}}</p>
-  </footer>
-expected: |
-  Too many closing braces
-   --> <test>:11:50
-    |
-   9 |       href="https://edwardtufte.github.io/tufte-css/">Tufte CSS</a>
-  10 |   </p>
-  11 |   <p><small>&copy;</small> {{now | format "2006" }}}} {{params.author.name}}</p>
-     |                                                  ^^^^ expected }}, got }}}
-  12 | </footer>
-     |
---
-# TODO: Implement
-name: Frontmatter missing closing quote on key
-enable: false
-should: warn
-input: |
-  {
-    "title: Docs"
-  }
-
-  [TOC]
-
-  ## Introduction
-expected: 'TODO:'
-actual: |
-  [ERROR] --- [42:parse_frontmatter()] failed to parse frontmatter JSON: Expected_Colon_After_Key
---
-# TODO: Implement
-name: Fronmatter missing opening quote on value
-should: warn
-enable: false
-input: |
-  {
-    "title": Docs"
-  }
-
-  [TOC]
-
-  ## Introduction
-expected: 'detailed diagnostic pointing to the point of failure and suggesting the addition of a quote'
-actual: Nothing. The page compiles with no warnings or errors
---
-name: No error when missing index page
-enable: false # TODO: Re-enable when passing
-should: error
-expected: 'Your site has no index page! Please create an index.md or index.html file in ./content/'
---
-# FIXME: Doesn't  work properly
-name: format pipe with no date format set
-enable: false # TODO: Re-enable when passing
-should: warn
-input: '{{ now | format }}'
-expected: |
-  no date format configured — set date.format in thor.json or specify a format, e.g. {{key | format "2 Jan 2006"}}
-   --> <test>:1:10
-    |
-  1 | {{now | format}}
-    |         ^^^^^^ defaulting to "2 Jan 2006"
---
-# TODO: called First/last, but only tests first
-name: first/last with zero argument
-enable: true
-should: error
-input: |
-  <ul>{{#items | first 0}}<li>{{.}}</li>{{/items}}</ul>
-expected: |
-  argument must not be zero
-   --> <test>:1:16
-    |
-  1 | <ul>{{#items | first 0}}<li>{{.}}</li>{{/items}}</ul>
-    |                ^^^^^^^
-    |
---
-# TODO: called First/last, but only tests first
-name: first/last with negative argument
-enable: true
-should: warn
-input: |
-  <ul>{{#items | first -3}}<li>{{.}}</li>{{/items}}</ul>
-expected: |
-  first -3 is equivalent to last 3
-   --> <test>:1:16
-    |
-  1 | <ul>{{#items | first -3}}<li>{{.}}</li>{{/items}}</ul>
-    |                ^^^^^^^^ consider using the clearer form
-    |
---
-# TODO: called First/last, but only tests first
-name: first/last on string with no argument
-# enable: true
-input: '{{word | first}}'
-should: warn
-expected: |
-  defaulting to n=1 for string
-   --> <test>:1:1
-    |
-  1 | {{word | first}}
-    |          ^^^^^ consider specifying n explicitly
-    |
---
-name: "first succeeds on params.author.name"
-enable: true
-input: '{{ params.author.name | first 3 }}'
-should: ok
-expected: "foo"
---
-# TODO: called First/last, but only tests first
-# TODO: Include | last 3 }} in the suggestion.
-name: "first/last in partial"
-enable: true
-input: "{{> footer }}"
-should: warn
-expected: |
-  first -3 is equivalent to last 3
-    --> fixtures/layouts/partials/footer.html:11:78
-     |
-   9 |       href="https://edwardtufte.github.io/tufte-css/">Tufte CSS</a>
-  10 |   </p>
-  11 |   <p><small>&copy;</small> {{! now | format "2006" }} {{params.author.name | first -3 }}</p>
-     |                                                                              ^^^^^^^^ consider using the clearer form
-  12 | </footer>
-     |
---
-name: "Pipes are ignored in comments"
-enable: true
-should: ok
-input: 'Hello, {{! now | format "2006" }}'
-expected: 'Hello, '
@@ -1,171 +0,0 @@
-# Grammar cache thread-safety & allocator lifetime
-
-Status: **resolved.** All three issues below are fixed. The cache was redesigned
-so the library caches only the immutable, shareable objects and callers own the
-per-thread ones. The rest of this document is kept as the rationale.
-
-## Resolution (what shipped)
-
-The `Grammar_Cache` (which mixed shareable `language`/`query` with per-thread
-`parser`/`cursor`) was split:
-
- **`Grammar`** holds only the immutable `language` + `query`. It lives in a
-  process-lifetime **`registry`** (`map[string]^Grammar`) guarded by one mutex.
- **`grammar(lang)`** is the thread-safe, self-initializing, lazy loader — fixes
-  **#1a** (all map access under the lock) and lets tests run with no setup.
- The registry allocator is pinned with `os.heap_allocator()` — fixes **#2**; the
-  cache no longer captures a transient `context.allocator`.
- Parsers/cursors are **caller-owned**: `open_parser(g)` mints one, the caller
-  `defer`s `parser_delete` / `query_cursor_delete`. No parser is shared, so
-  **#1b** is gone. `open_parser` is the seam where a future pool / thread-local
-  strategy plugs in without touching call sites.
- Deleted: `ensure_parser`, `load_grammar`, `preload_grammar`, `grammar_worker`,
-  `cache_mutex`; `preload_grammars` is now a serial registry warm-up. The minify
-  tests dropped their `parse_mu` serialization and preload scaffolding entirely.
-
-The **later** decision — pool vs per-thread parser reuse for genuine parallel
-parsing — is still open, but intentionally so: it only matters once we parse in
-parallel, and it now changes only `open_parser`, not the rest of the library.
-The original investigation follows.
-
-## How we got here
-
-While adding a feature to minify inline `<style>` blocks (route the CSS body
-through `minify_css`), the new unit tests exercised `minify_html` / `minify_css`
-for the first time. Those are the first tests to touch the tree-sitter parser
-cache, and under the multi-threaded test runner (default 16 threads) they failed
-**intermittently** with a rotating cast of symptoms:
-
- `panic: unable to insert into a map`
- tree-sitter internal aborts: `ts_subtree_retain: Assertion 'self.ptr->ref_count > 0' failed`, `_array__erase: Assertion 'index < *size' failed`
- `malloc(): unaligned tcache chunk detected`
- assertions passing on the wrong data (e.g. minified output missing `<style>`)
-
-The flakiness comes from **three distinct problems**, two thread-safety and one
-allocator-lifetime. They are orthogonal — fixing one does not fix the others.
-
-## The three issues
-
-### #1a — concurrent cache **writes** (data race)
-
-`ensure_parser` (`treesitter.odin`) reads and writes `grammar_store.cache`
-with **no lock**. `cache_mutex` exists but is only taken on the parallel preload
-path (`grammar_worker`), never in the lazy `ensure_parser` / `load_grammar`
-paths. Two threads first-touching `"html"`/`"css"` at once corrupt the map.
-
-### #1b — concurrent parser **use** (data race)
-
-There is exactly **one cached `Grammar_Cache` (and one `parser`) per language**.
-When two threads both call `minify_html`, they fetch the *same* `gc.parser` and
-call `parser_parse_string` on it simultaneously. A tree-sitter `TSParser`
-mutates its internal subtree pool during a parse, so concurrent parses corrupt
-it — that's the `ts_subtree_retain` / `_array__erase` aborts.
-
-This is **not fixed by preloading/"loading html+css statically."** Loading gives
-you *one shared* parser; the race is about *using* that one parser from two
-threads. It is fixed only by serializing the parse or giving each thread its own
-parser.
-
-### #2 — allocator lifetime (not a race)
-
-`init_persistent` captures `context.allocator` into the process-lifetime
-`grammar_store`. In `main` that happens to be the heap (it runs before the site
-arena is installed at `main.odin`), so production is fine. But whatever
-allocator is live at that instant becomes the cache's owner forever:
-
- Under the **test runner**, `context.allocator` is a **per-test tracking
-  allocator** that is torn down when that test ends → the global cache is left
-  pointing at dead memory (→ `unable to insert into a map`, garbage reads).
- A **lazy** first call during rendering would capture the **site arena**
-  (`context.allocator` after `main.odin` swaps it in), which is destroyed every
-  watch-mode rebuild → dangling parsers across builds.
-
-This would break even single-threaded; threading just decides *which* test wins
-the init race.
-
-## tree-sitter's threading model (the root fact)
-
-tree-sitter objects have specific, documented threading rules:
-
-| Object | Shareable across threads? | Rule |
-|---|---|---|
-| `TSLanguage` (`tree_sitter_html()` etc.) | Yes | Immutable; share freely. |
-| `TSParser` (`gc.parser`) | **No** | One per thread; never parse on one parser from two threads. |
-| `TSTree` | Only via `ts_tree_copy` | Copy before using on another thread. |
-| `TSQuery` (`gc.query`) | Yes | Immutable after compile. |
-| `TSQueryCursor` (`gc.cursor`) | **No** | One per thread, like the parser. |
-
-This explains the existing design:
-
- The preload fan-out (`preload_grammar` / `grammar_worker`) creates a parser
-  **per worker** and only publishes the finished `Grammar_Cache` under
-  `cache_mutex` — because parsers can't be shared during construction/use.
- `ensure_parser` + the shared cached parser is fine in production **only
-  because rendering is single-threaded**.
-
-A cached `Grammar_Cache` mixes shareable (`language`, `query`) and per-thread
-(`parser`, `cursor`) objects in one shared struct, which makes it
-single-threaded-use by construction — a fact that was previously undocumented.
-
-## Where this actually bites
-
-**Only in tests.** Confirmed by grep:
-
- Production parses on one thread: every real caller — `minify.odin` (html/css)
-  and `markdown/highlight.odin` — runs inside `main`'s single-threaded render
-  loop. No concurrent `ensure_parser` / `parser_parse_string`.
- The one threaded path in production, `preload_grammars`, is already safe (own
-  parser per worker + `cache_mutex` on the write) and only handles non-builtin
-  grammars.
- Only `minify_test.odin` touches tree-sitter from tests, and the test runner
-  runs those on many threads.
-
-So production is correct **by convention** (single-threaded-from-heap usage),
-not **by construction** — nothing stops a future parallel render from breaking
-it, and until now nothing documented the constraint.
-
-## Chosen direction
-
-**For now:** keep the single global grammar cache and make it safe by
-**serializing the non-thread-safe behavior with a mutex** — only one thread uses
-a parser at a time. This is simple, matches how production already behaves, and
-unblocks the tests.
-
-**Later (TODO):** decide whether true multi-threaded parsing is worth it for
-performance. If yes, give each thread its own parser + cursor (as the preload
-workers already do) rather than sharing one behind a lock.
-
-## Implementation options (deferred — pick when we resume)
-
-All assume the allocator fix (#2) below, which is independent and unambiguous.
-
- **#2 allocator (do regardless):** pin the heap allocator explicitly —
-  `grammar_store.allocator = runtime.heap_allocator()` (or `os.heap_allocator()`)
-  instead of `context.allocator`. Makes the cache persistent from any context;
-  behavior-identical in production. Optionally make `init_persistent` idempotent
-  and/or an `@(init)` so callers need no ordering.
-
-Then, for the races:
-
- **Option A — library mutex (sustainable):** take `cache_mutex` in
-  `ensure_parser` (fixes #1a) and add a locking `parse` helper that wraps
-  `parser_parse_string` (fixes #1b), wired into the three call sites
-  (`minify.odin` ×2, `markdown/highlight.odin` ×1). Tests then need no
-  scaffolding. Serializes all parsing.
-
- **Option B — test-side only (minimal):** leave the library single-threaded by
-  contract; make the tests not violate it. Either fold the minify tests into one
-  `@(test)` (no concurrency at all → no locks), or wrap the `minify_*` calls in a
-  test-local mutex. Production stays lock-free.
-
-Recommendation captured during the discussion: the allocator one-liner is
-worth doing immediately; for the races, Option A is the "correct" long-term
-answer but only necessary if/when we want concurrent parsing — otherwise
-Option B is enough.
-
-## Pointers
-
- `treesitter/treesitter.odin`: doc comments on `Grammar_Cache`, `cache_mutex`,
-  `init_persistent`, `ensure_parser` reference this file.
- Original feature that surfaced this: minify inline `<style>` bodies via
-  `minify_css` (reverted; to be re-landed alongside whichever fix we choose).
@@ -1,11 +1,54 @@
-# Hugo Notes
+# Hugo vs Thor: Why Thor Doesn't Need Shortcode Context Isolation

-from the [Hugo docs](https://gohugo.io/quick-reference/glossary/#default-sort-order)
+## Hugo's Shortcode Isolation

-default sort order
+Hugo strictly separates shortcode context from layout template context. Shortcodes run during markdown rendering, before layouts. They get a limited context (`.Page`, `.Site`, `.Params`), not the full layout rendering state.

-: The default sort order for page collections, used when no other criteria are set, follows this priority:
-  1. weight (ascending)
-  2. date (descending)
-  3. linkTitle falling back to title (ascending)
-  4. logical path (ascending)
+### Why Hugo does this
+
+1. **Circular dependencies** — Hugo shortcodes output into `.Content`. Layouts read `.Content` and compute derived properties (`.TableOfContents`, `.WordCount`, `.ReadingTime`). If shortcodes could access these computed properties, you'd get circular dependencies (shortcode output → computed property → shortcode reads it).
+
+2. **Multiple layouts/themes** — Hugo supports user-selectable themes and multiple layouts per section type. A shortcode must produce identical output regardless of which layout renders it. Isolating context guarantees portability.
+
+3. **Shared mutable page object** — Hugo builds a rich `Page` object that both shortcodes and layouts access. Context isolation prevents shortcodes from mutating layout-visible state.
+
+4. **Security** — Shortcodes live in content files (potentially user-authored). Full template access would let content authors inject arbitrary logic, access sensitive config, or break site structure.
+
+5. **Caching** — Hugo caches shortcode output independently. Predictable context = predictable cache behavior.
+
+## Why this doesn't apply to thor
+
+- **No computed page properties** — The content Mustache pass runs on raw markdown (before cmark). The layout Mustache pass runs on finished HTML (after cmark). There's no shared `.Content` object or derived properties. The two passes are at different pipeline stages with no shared mutable state.
+
+- **Single layout set** — Thor has one set of templates (`base.html` + page templates). No theme switching, no multiple layout variants per section. Portability across layouts isn't a concern.
+
+- **Explicit data passing** — Each `mustache.render` call receives its own data struct. Content pass gets frontmatter + params. Layout pass gets page data + site config. No shared mutable object between them.
+
+- **Single author** — Content is authored by the site owner. No untrusted user-generated content.
+
+- **No caching** — Thor rebuilds from scratch every time. No cache invalidation concerns.
+
+## Thor's approach
+
+Thor runs Mustache on content **before** cmark as a pre-processing step, then runs Mustache on layout templates **after** cmark as a post-processing step. These are two independent `mustache.render` calls:
+
+```
+content (markdown)
+  → mustache.render(content, content_data, partials)   ← content pass
+  → cmark markdown_to_html
+  → inject_sidenotes / inject_alerts / highlight_code
+  → mustache.render_in_layout(template, page_data, layout, partials)  ← layout pass
+  → final HTML
+```
+
+The content pass can use Mustache variables, partials (`{{> ./file}}`), sections, and conditionals freely. The layout pass wraps the rendered content in the page chrome. They share nothing except the data we explicitly choose to pass.
+
+## When this WOULD matter for thor
+
+If thor ever adds:
+- Computed page properties (table of contents, reading time, word count)
+- Multiple selectable themes/layouts
+- User-generated content / multi-author support
+- Partial template caching
+
+...then context isolation between the content and layout passes would become relevant. Until then, it's unnecessary complexity.
@@ -1,131 +0,0 @@
-## v0.2.0
-
- [ ] [aliases](https://gohugo.io/methods/page/aliases/#redirects)?
- [ ] consider sites with date based urls.
- [ ] Create a json schema file for `thor.json`.
- [ ] Content-hash fingerprinting for CSS and JS cache busting
- [ ] Mount content in VFS
- [ ] `-watch` flag
-  - [x] basic poll loop
-  - [ ] filesystem poll loop
-  - [ ] event based
- [ ] if no `html` tag detected in output, re-render output with base template
-      (or whatever template is next in the chain). (Hugo seems to do this)
- [ ] Markdown
-  - [ ] Block attributes extension — hello-world.md
-  - [ ] add tables extension
-  - [ ] enable template level rendering of TOCs
-  - [ ] Write css for toc sidebar and figure out where to put it.
-  - [ ] Add [hugo style configuration](https://gohugo.io/configuration/markup/#table-of-contents)
- [ ] implicit page titles (Set when missing?)
-  -  guess from h1 tag or first h2 tag?  
- [ ] Syntax highlighting
-  - [ ] grammars
-      - [ ] load grammars dynamically
-      - [ ] Search in multiple places
-      - [ ] Download missing grammars.
-      - [ ] Durable highlight paths: read `GRAPHS_PATH`/`QUERIES_PATH` from env vars set by the flake instead of hardcoded nix store hashes, so they survive `nix flake update` and let the grammar/query version-mismatch detector fire automatically.
-  - [ ] Theme selector 
-    - use http://github.com/helix-editor/helix/tree/master/runtime/themes) as a
-      guide
-  - [ ] Unit tests for highlighting helpers: `capture_name_to_css`, `escape_html`, `unescape_html`, `extract_query_token`, `helix_version_from_path`.
-  - [ ] `<pre><code>` blocks need to set background to theme background,
-    regardless of prefers-dark. (or use a different theme)
- [ ] Avoid `json.Value` / `json.Object` where possible.
- [ ] come up with scrapers / scrape sources to harvest site data
-  - we'll use this to help us sculpt defaults.
- [ ] merge `render_{section,home_html,page_html}` procs.
- [ ] try to combine render_page_html and render_home_html?
- [ ] make `parse` an overload of `parse_text/parse_inline` and `parse_file`, or something.
- [ ] refactor `mustache.load_partials` to accept an iterator so we don't copy `main.load_partials`
-
-### Menus
-
- [ ] Nav items should be active when the current page is selected.
- [ ] configure opt-out of automatic sections being added to menu.
- [ ] nested menus (i.e. `parent` support)
-
-### Strict Mode 
-
- [ ] opt-in "strict_keys" mode. in this mode, key lookups may not view parent objects.
-  - [ ] warn/error when unknown key used in mustache.
- [ ] enforce heading structure.
-  - [ ] Either frontmatter.title set, or 1 h1 tag at top, not both
-  - [ ] No skipping. 
-
-### Dates
- [ ] display an error when no part of the date appears in the output.
- [ ] Handle 0 and whitespace padding i.e. "_2" -> " 2"
- [ ] Do we *need* mustache.Date_Components, or can we use core:time/datetime.DateTime?
-
-## Beyond
-
- [ ] consider logging the number of times an error occurred. (diagnostics)
- [ ] Need to be careful about diagnostics across module boundaries.
-  - we don't necessarily want to warn users about theme designers mistakes. (though perhaps we do)
- [ ] consider reporting duplicate weights outside of menus (diagnostic)
- [ ] Do mustache's whitespace rules actually suit us, or should we make our own? 
- [ ] create a default `head.html`?
- [ ] improve home link customization.
-  - [ ] currently an accessibility issue.
- [ ] get rid of the global variables in the `treesitter` package.
- [ ] site-wide Integrity hash
-  - Allows users to verify their output didn't change after upgrading to a new version
- [ ] Add `-production` flag
-  - sets `-minify`
- [ ] manually pass `site_allocator` to `load_page`
- [ ] Import/export packages. Hugo, jekyll, WordPress, etc.
- [ ] shortcodes?
-  - include-code shortcode (`{{< include-code ... >}}`) — i-ported-fd-to-odin
- [ ] follow symlinks in `scan_content`?
- [ ] Add overloads for every md extension? i.e. accept ^strings.Builder.
- [ ] content link checker?
-  - Checks all links on each page to make sure they are valid.
- [ ] Peruse [GitHub's](https://docs.github.com/en/get-started/writing-on-github/getting-started-with-writing-and-formatting-on-github/basic-writing-and-formatting-syntax#alerts)
-      docs for any juicy nuggets we may have missed.
- [ ] Avoid using `render_inline_md` if possible.
- [ ] don't use bullshit "sub-tokens", add filters and pipes as proper tokens.
- [ ] Create a Location struct that somewhat matches Odin's [Source_Code_Location](https://pkg.odin-lang.org/base/runtime/#Source_Code_Location)?
-  - Note that odin's version doesn't contain the stack trace.
- [ ] Link checker / linter that warns if links don't resolve. ("please use {{whatever | rel_url }}").
-
-### Memory Management
-
- [ ] Leaks in highlighter code.
- [ ] Not sure whether to use temp allocator or site_allocator in opengraph.odin.
- [ ] Not sure whether to use temp allocator or site_allocator in `site_load_content`.
- [ ] Might not need to allocate in `strip_html_tags`
- [ ] Fix `apply_filter`'s `format` case (`mustache/pipes.odin`) boxing `apply_format`'s
-      `string` result into `any` via bare `return`, which materializes a hidden
-      header temp in `apply_filter`'s own stack frame. Dangling once the frame
-      returns; caused the `-o:speed` segfault in `write_value`. Fix: box explicitly
-      with `any{new_clone(formatted, context.temp_allocator), typeid_of(string)}`.
- [ ] Same pattern in `apply_group_by` (`mustache/pipes.odin`): `return groups, nil`
-      boxes a freshly-built `[dynamic]Group` as bare `any` — same latent
-      stack-temp UB, hasn't crashed yet but should get the same treatment.
-
-### Performance
-
- [ ] See if we can disable bounds checks in `write_indented` and elsewhere.
- [ ] Instead of loading the site fresh each time in watch mode, create a
-      `reload_site` proc, that just updates changed resources.
- [ ] Only publish referenced assets.
- [ ] Split `load_page` into frontmatter-parse + body-process phases so draft pages can skip the markdown pipeline entirely
- [ ] Use spall to find ways to reduce run time.
- [ ] too many `write_string` calls in `highlight_block`
- [ ] return `src: cstring` from `load_query`.
- [ ] Improve `unescape_html` with simd.
- [ ] generate summary before syntax highlighting.
- [ ] generate summary before markdown to html conversion.
- [ ] mount_recursive is pretty significant
- [ ] thread pool for grammar loading is unbounded.
- [ ] load grammars async.
- [ ] during `load_page`:
-  - pass each code block to the treesitter queue
-  - continue working on the page,
-  - `await` the highlighted code.
- [ ] can markdown extensions run in parallel?
- [ ] enforce MAX_SLUG_LENGTH
- [ ] enforce MAX_CONTEXT_DEPTH
- [ ] ensure struct fields are ordered correctly
-
@@ -1,143 +1,249 @@
-# TODOs
+## High priority

-## Priority Unclear (Evaluate Later)
+- Polish existing features before moving on to new ones.
+- [ ] Improve diagnostics
+  - [x] keep track of every error and don't report them more than once.
+  - [ ] All Diagnostics should show:
+    - [ ] *What* went wrong
+    - [ ] *where* (in the file)
+    - [ ] *where* (in the stack trace)
+    - [ ] *how* you can fix it (if applicable)
+  - [ ] Create a Location struct that somewhat matches Odin's [Source_Code_Location](https://pkg.odin-lang.org/base/runtime/#Source_Code_Location)?
+    - Note that odin's version doesn't contain the stack trace.
+  - [ ] show "stack traces" in template error diagnostics
+  - [ ] better diagnostics for syntax errors in treesitter.
+  - [ ] Ensure diagnostics for MAX_CONTEXT_DEPTH are good.
+  - [ ] improve matching weights message.
+  - [ ] show a proper diagnostic for timezones
+    - currently "unable to load timezone 'America/New_Yorkskie'"
+    - want rust style diagnostic and better message, maybe "unknown timezone 'America/New_Yorkskie'"
+  - [ ] Test menu diagnostics
+  - [ ] Honestly, Test **all** diagnostics
+  - [ ] Need to be careful about diagnostics across module boundaries.
+    - we don't necessarily want to warn users about theme designers mistakes. (though perhaps we do)
+  - [ ] consider reporting duplicate weights outside of menus
+  - [ ] Extend `tag_error` to all render-time errors, not just pipe errors.
+    Currently only pipe errors (4 sites in `render_nodes`) get stamped with
+    the correct template source/path. Other render errors still use the
+    content template's source/path, which can point at the wrong file.
+  - [ ] try to make file paths clickable links.
+  - [ ] centralize diagnostics to one place.
+  - [ ] consider logging the number of times an error occurred.
+- [ ] Load grammars dynamically
+- [ ] starred must be a param.
+- [ ] Documentation
+  - [ ] talk about the context stack (and its limit).
+  - [ ] highlight the differences in the way menus are handled.
+- [ ] consider sites with date based urls.
+- [ ] Don't show annoying log output in tests.
+- [ ] improve home link customization.
+  - [ ] currently an accessibility issue.
+- [ ] support JSON5 in in frontmatter
+- [ ] Create a json schema file for `thor.json`.
+- [ ] cleanup `#partial switch`es.
+- [ ] improve json diagnostics.
+  - i.e. "Missing quotes around string", etc.
+- [ ] don't use bullshit "sub-tokens", add filters and pipes as proper tokens.

- [ ] Decide the parallel-parsing strategy for the grammar cache (pool vs
-      thread-local parsers) — deferred until the parallel usage shape is clear.
-  - Resolved already: the registry (`grammar()`) is thread-safe and self-
-    initializing, its allocator is pinned to the heap, and parsers/cursors are now
-    caller-owned (`open_parser`) instead of cached — so the library holds no
-    single-thread-use state and the multi-threaded tests pass without locks.
-  - Remaining choice only matters once we actually parse in parallel: reuse
-    parsers via a pool or per-thread storage, plugged in behind `open_parser`
-    without touching call sites. A `TSParser`/`TSQueryCursor` is one-per-thread.
-  - Full write-up: `GRAMMAR_CACHE_THREADING.md`.
- [x] Restore parallel grammar preload (serialized in `edb2e64`); `preload_grammars`
-      now fans `build_grammar` across a bounded `core:thread.Pool`.
-  - [ ] Add tests: pool fan-out correctness + `-sanitize:thread` for race-freedom.
-  - [ ] Concurrent-read path: an `RW_Mutex` on `grammar()` only helps a future
-        multi-threaded render, not the load side. Revisit if render goes parallel.
- [ ] instead of `warnings: [dynamic]Error` we should use `warnings: [dynamic; 8]Error`
-  - when limit reached, the template fails and stops rendering.
- [ ] Brainstorm plugin/tool/(postccs/less/tailwind/sass) support.
- [ ] `fingerprint` pipe + contents-based composition. The `tool` pipe currently
-      carries a URL string and each stage publishes a file to the output dir, so
-      `| tool sass | fingerprint` would leave a dead unhashed intermediate. When
-      fingerprint lands, move the pipe value from a URL string to a contents/
-      Resource (bytes + logical name) where only the terminal stage publishes;
-      a Resource must stringify to its URL so existing bare-`tool` templates keep
-      working. stdin/stdout tools compose best there (bytes stay in memory), which
-      is why stdin/stdout is the documented preferred form over `{{in}}`/`{{out}}`.
- [ ] Reconsider `"menus": {}` opt-out: it currently discards all page frontmatter
-      menus (with only a warning). Decide whether empty config should instead just
-      suppress the auto-generated menu while still honoring frontmatter.
- [ ] asset css files not  automatically  included in site.
-  - should we at least warn about them?
- [ ] Review the whole `copy_assets_dir` pipeline. It copies every file under
-      `assets/` to the output dir blindly, whether or not anything references it,
-      and now also skips dests that already exist (a blunt collision guard so the
-      raw copy doesn't clobber `tool`-pipe output written earlier in the build).
-      Reconsider tracking which assets are actually used/published instead.
- [ ] warn if stylesheet url starts with /assets  (and is therefore invalid).
- [ ] warn when heading levels are skipped in a toc.
- [ ] extract `[TOC]` from `.md` body (or at least from  the start.)
- [ ] add `--production` flag
-  - set `minify` to true
-  - warn when `base_url` unset
+## Performance
+
+- [ ] See if we can disable bounds checks in `write_indented` and elsewhere.
+- [ ] Instead of loading the site fresh each time in watch mode, create a
+      `reload_site` proc, that just updates changed resources.
+- [ ] Only publish referenced assets.
+- [ ] Split `load_page` into frontmatter-parse + body-process phases so draft pages can skip the markdown pipeline entirely
+- [ ] Use spall to find ways to reduce run time.
+- [ ] too many `write_string` calls in `highlight_block`
+- [ ] return `src: cstring` from `load_query`.
+- [ ] Improve `unescape_html` with simd.
+- [ ] generate summary before syntax highlighting.
+- [ ] generate summary before markdown to html conversion.
+- [ ] mount_recursive is pretty significant
+- [ ] thread pool for grammar loading is unbounded.
+- [ ] load grammars async.
+- [ ] during `load_page`:
+  - pass each code block to the treesitter queue
+  - continue working on the page,
+  - `await` the highlighted code.
+- [ ] can markdown extensions run in parallel?
+- [ ] enforce MAX_SLUG_LENGTH
+- [ ] enforce MAX_CONTEXT_DEPTH
+- [ ] ensure struct fields are ordered correctly
+
+## Remove Privileged content
+
+- [ ] `group_by` currently requires a computed `year` field on the page.
+  - We should replace this with `{{ pages | group_by (date | "2006") }}` or similar
+
+
+## Memory Management
+
+- [ ] Leaks in highlighter code.
+- [ ] Not sure whether to use temp allocator or site_allocator in opengraph.odin.
+- [ ] Not sure whether to use temp allocator or site_allocator in `site_load_content`.
+- [ ] Might not need to allocate in `strip_html_tags`
+- [ ] Fix `apply_filter`'s `format` case (`mustache/pipes.odin`) boxing `apply_format`'s
+      `string` result into `any` via bare `return`, which materializes a hidden
+      header temp in `apply_filter`'s own stack frame. Dangling once the frame
+      returns; caused the `-o:speed` segfault in `write_value`. Fix: box explicitly
+      with `any{new_clone(formatted, context.temp_allocator), typeid_of(string)}`.
+- [ ] Same pattern in `apply_group_by` (`mustache/pipes.odin`): `return groups, nil`
+      boxes a freshly-built `[dynamic]Group` as bare `any` — same latent
+      stack-temp UB, hasn't crashed yet but should get the same treatment.
+
+## Markdown
+- [ ] Add overloads for every extension - accept ^strings.Builder.
+- [ ] Add conventional (Hugo style) footnotes option.
+- [ ] Add opt-in deflist support.
+- [ ] Decide if lambdas actually provide any value.
+
+## Dates
+- [ ] display an error when no part of the date appears in the output.
+- [ ] Handle 0 and whitespace padding i.e. "_2" -> " 2"
+- [ ] Do we *need* mustache.Date_Components, or can we use core:time/datetime.DateTime?

 ## General 
+- [ ] Menus
+  - [ ] configure opt-out of automatic sections being added to menu.
+  - [ ] nested menus (i.e. `parent` support)
+- [ ] get rid of the global variables in the `treesitter` package.
+- [ ] Consider using `or_else` when applying default values to structs. i.e.
+```odin
+package main

-Remember to polish existing features before moving on to new ones.
+X :: struct {
+  foo: string
+}

- [ ] Bake the default `layouts` into the binary at compile time via
-      `#load_directory` so the binary is self-contained and doesn't read
-      `DEFAULTS_PATH` from disk at runtime.
-  - `#load_directory` is NOT recursive and its `name` is the base filename
-    only, so `defaults/layouts` and `defaults/layouts/partials` must be loaded
-    separately and mounted into the VFS with `data` set (not `fs_path`).
-  - Removes the runtime dependency on `THOR_DEFAULTS_PATH` / `#directory` and
-    the need to ship `defaults/` alongside the binary (see `flake.nix`).
- [ ] `build` command alias of default
- [ ] `new site` command to set up new project
- [ ] Some render errors blank the entire page instead of scoping to the failing tag.
+main :: proc () {
+  x: X  
+
+  x.foo = x.foo or_else "bar"
+}
+```
+- [ ] Integrity hash
+  - Allows users to verify their output didn't change after upgrading to a new version
+- [ ] Content-hash fingerprinting for CSS and JS cache busting
+- [ ] come up with scrapers / scrape sources to harvest site data
+  - we'll use this to help us sculpt defaults.
+- [ ] merge `render_{section,home_html,page_html}` procs.
+- [ ] try to combine render_page_html and render_home_html?
+- [ ] Debug log stats. (analytics)
+  - [ ] final Context_Stack cap
+  - [ ] highest PIPE args used 
+  - [ ] longest slug length + name that generated it
+  - [ ] number of pages
+  - [ ] number of blocks
+  - [ ] enabled features / extensions
+  - [ ] etc
+- [ ] Avoid `json.Value` / `json.Object` where possible.
+- [ ] make `parse` an overload of `parse_text/parse_inline` and `parse_file`, or something.
+- [ ] Add page params
 - [ ] We must remove all mention of `posts` from the odin code.
      At present, "posts" are a user-level construct defined as pages in a
      particular collection.
- [ ] ? Clean up the default layouts (no idea what this means)
- [ ] `group_by` currently requires a computed `year` field on the page.
-  - We should replace this with `{{ pages | group_by (date | "2006") }}` or similar
- [ ] does it make sense for partials to be inside layouts?
-  - current: `layouts/partials`
-  - alt1: `layouts`, `partials`
-  - alt2: `templates/layouts`, `templates/partials`
- [ ] add `sort_by` pipe
- [ ] pipes
-  - [x] first/last
-    - [ ] write test for error when n == 0
-    - [x] warn when using negative numbers
-    - [x] warn when using default arg on string
-    - [x] add the above warnings to DIAGNOSTIC_TESTS.md
-    - [ ] Instead of "clearer form", be exact: "last 5"
-    - [ ] write tests for `first 0` & `last 0`
-    - [ ] update tests to assert diagnostics when appropriate
-      - do we want to do this or just use DIAGNOSTIC_TESTS.md as the tests?
-      - [ ] also update to use file logger so they don't clog up the output
-    - [ ] narrow highlight in diagnostic to relevent part of line.
- [ ] page.url and page.permalink are confusing names. Switch to url + page_url
-  - There is confusion over how to do this and whethor or not `| rel_url` should be needed to render these.
-    Requires **careful thought**.
-
-## Documentation
-
- [ ] talk about the context stack (and its limit).
- [ ] highlight the differences in the way menus are handled.
- [ ] explain that html/css/js comments are rendered, except in minify mode.
-
-## Investigate
-
 - [ ] running ./thor/thor still logs the debug message: using config /home/spencer/github.com/sbrow.github.io/thor.json
  - wrong cwd?
+- [ ] Clean up the default layouts
+- [ ] Menus
+  - [x] Detailed frontmatter menu form ("menu": {"main": {"weight": 5}})
+  - [ ] Menu active state (pre-compute is_active based on page.permalink prefix match)
+  - [x] Page.weight field for general-purpose page ordering (menus, lists, related posts)
+- [ ] if no `html` tag detected in output, re-render output with base template
+      (or whatever template is next in the chain)
+- [ ] Add `-production` flag
+  - sets `-minify`
+- [ ] Mustache diagnostics
+  - [ ] Partial invocation stack in diagnostics: when an error fires inside a partial, show "invoked from" chain through `{{> name}}` calls. Currently warnings inside partials point at the partial (correct file) but don't show the invocation site.
+  - [ ] Could be better error message when missing a closing (or opening) brace
+  - [x] Error message doesn't show position of faulty pipe name correctly.
+  - [ ] `render_template` (`render.odin`) blanks the *entire page* to `""` on any
+        mustache render error and only `log.errorf`s it — a single bad tag/pipe
+        anywhere on the page silently kills the whole output with no visible
+        signal outside the terminal log. Should at least be scoped to the
+        failing tag/section, or surfaced somewhere the person building the
+        site will actually see it.
+  ```bash
+  [ERROR] --- [138:render_template()] unknown pipe op 'formats'
+ --> /home/spencer/github.com/sbrow.github.io/layouts/home.html:1:1
+  |
+1 | {{<base}}
+  | ^^^^^^^^^
+2 | {{$content}}
+3 | <main>
+```
+- [ ] Block attributes on code fences (`{ #ex-1 }`) — hello-world.md
+- [ ] include-code shortcode (`{{< include-code ... >}}`) — i-ported-fd-to-odin
+- [ ] follow symlinks in `scan_content`?
+- [ ] ensure sidenote numbers render in display order and not in declaration order.
+- [ ] Table of contents support.
+- [ ] Nav items should be active when the current page is selected.
+- [ ] Theme selector for syntax highlighting.
+  - use http://github.com/helix-editor/helix/tree/master/runtime/themes) as a
+    guide
+- [ ] grammars
+  - [ ] Search in multiple places
+  - [ ] Download missing grammars.
+  - [ ] Durable highlight paths: read `GRAPHS_PATH`/`QUERIES_PATH` from env vars set by the flake instead of hardcoded nix store hashes, so they survive `nix flake update` and let the grammar/query version-mismatch detector fire automatically.
+- [ ] CI
+  - [ ] Syntax highlighting in production: CI (`nix build` on ubuntu-latest) has no grammar `.so`s and a machine-specific `QUERIES_PATH` nix store hash, so the deployed site renders unhighlighted. Provide grammars + queries as nix build inputs and pass paths to thor at runtime (env vars/flags).
+- [ ] Unit tests for highlighting helpers: `capture_name_to_css`, `escape_html`, `unescape_html`, `extract_query_token`, `helix_version_from_path`.
+- [ ] `<pre><code>` blocks need to set background to theme background,
+  regardless of prefers-dark. (or use a different theme)
+- [ ] `-watch` flag
+  - [x] basic poll loop
+  - [ ] filesystem poll loop
+  - [ ] event based
+- [ ] Free cmark HTML output (`body_html`) — cmark allocates via C malloc, not the arena, so it leaks per iteration in watch mode
+- [ ] Mount content in VFS
+- [ ] commands
+  - [ ] `build` alias of default
+  - [ ] `new site` set up new project
+- [ ] warn/error when unknown key used in mustache.
+- [ ] Import/export packages. Hugo, jekyll, WordPress, etc.
+- [ ] opt-in "strict_keys" mode. in this mode, key lookups may not view parent objects.

-## Diagnostics
+## Notes

-All diagnostics (errors) should show
+from the [Hugo docs](https://gohugo.io/quick-reference/glossary/#default-sort-order)

- *What* went wrong
- *where* (in the file)
- *where* (in the stack trace)
- *how* you can fix it (if applicable)
+default sort order
+: The default sort order for page collections, used when no other criteria are set, follows this priority:
+  1. weight (ascending)
+  2. date (descending)
+  3. linkTitle falling back to title (ascending)
+  4. logical path (ascending)

- [ ] try to make file paths clickable links.
- [ ] make sure the frontmatter parser has good diagnostics.
- [ ] add all known diagnostics to [DIAGNOSTIC TODOS](./DIAGNOSTIC_TESTS.yaml).
- [ ] fix the broken diagnostics in [DIAGNOSTIC TODOS](./DIAGNOSTIC_TESTS.yaml)
- [ ] show *stack trace* for template partials?
- [ ] show where the typo occurred (config file path for `apply_extension_config`, full `-ext:value` for `parse_extension_list`)
- [ ] **all** diagnostics in [DIAGNOSTIC TODOS](./DIAGNOSTIC_TESTS.yaml) must be tested.
- [ ] only report format errors once.
- [ ] only report missing partial errors once.
- [ ] better diagnostics for syntax errors in treesitter.
- [ ] Ensure diagnostics for MAX_CONTEXT_DEPTH are good.
- [ ] improve matching weights message.
- [ ] Test menu diagnostics
- [ ] Extend `tag_error` to all render-time errors, not just pipe errors.
-  Currently only pipe errors (4 sites in `render_nodes`) get stamped with
-  the correct template source/path. Other render errors still use the
-  content template's source/path, which can point at the wrong file.
+## Code Review

-## Analytics
+A human should manually review every file in the project. AI cannot complete
+these tasks.

-Debug log stats:
-
- [ ] final Context_Stack cap
- [ ] highest PIPE args used 
- [ ] longest slug length + name that generated it
- [ ] number of pages
- [ ] number of blocks
- [ ] enabled features / extensions
- [ ] etc
-
-## sbrow.github.io
- [ ] The home file was accidentally deleted in commit 9a9f731
- [ ] Ideas is now in the wrong spot. Date is wrong, and it is showing date
-      when it shouldn't be.
+- [ ] Review every file in thor
+  - [ ] Review assets.odin
+  - [ ] Review content.odin
+  - [ ] Review defaults.odin
+  - [ ] Review feed.odin
+  - [ ] Review frontmatter.odin
+  - [ ] Review main.odin
+  - [ ] Review minify.odin
+  - [ ] Review `markdown/`
+    - [x] Review alerts.odin
+    - [x] Review alerts_test.odin
+    - [x] Review emoji.odin
+    - [x] Review emoji_test.odin
+    - [ ] Review footnotes.odin
+    - [ ] Review footnotes_test.odin
+    - [ ] Review highlight.odin
+    - [ ] Review markdown.odin
+    - [x] Review sectionate.odin
+    - [x] Review sectionate_test.odin
+    - [ ] Review suggest.odin
+    - [ ] Review suggest_test.odin
+  - [x] Review opengraph.odin
+  - [ ] Review render.odin
+  - [x] Review site.odin
+  - [ ] Review treesitter/treesitter.odin
+  - [ ] Review vfs.odin
+- [ ] Review procs
+  - [ ] markdown.transform_alert
@@ -1,109 +0,0 @@
-# Warning Counter Plan
-
-## Goal
-
-When the same warning fires multiple times across a single page render, show it once with a `[×N]` count prefix instead of N identical diagnostic blocks.
-
-```
-[×5] first -3 is equivalent to last 3 — consider using the clearer form
- --> layouts/home.html:12:1
-  |
-12 | <ul>{{#items | first -3}}<li>{{.}}</li>{{/items}}</ul>
-   | ^^^^^^^^^^^^^^^^^^^^^ consider using the clearer form
-   |
-```
-
-## Approach
-
-Module-level buffer in `pipes.odin`. Warnings are collected during render, deduplicated + counted + logged after the page completes. No parameter threading through `render_nodes`.
-
-## Changes
-
-### `mustache/pipes.odin`
-
-**New module-level state:**
-```odin
-_pending_warnings: [dynamic]Error
-```
-
-**New procs:**
-```odin
-reset_warnings :: proc() {
-    clear(&_pending_warnings)
-}
-
-flush_warnings :: proc(colorize: bool = false) {
-    if len(_pending_warnings) == 0 do return
-
-    counts := make(map[string]int, context.temp_allocator)
-    first: map[string]Error = make(map[string]Error, context.temp_allocator)
-    defer delete(counts)
-    defer delete(first)
-
-    for w in _pending_warnings {
-        b := body(w)
-        if _, seen := first[b.msg]; !seen {
-            first[b.msg] = w
-        }
-        counts[b.msg] += 1
-    }
-
-    for msg, w in first {
-        c := counts[msg]
-        b := body(w)
-        path := b.path != "" ? b.path : "<input>"
-        prefix := c > 1 ? fmt.tprintf("[×%d] ", c) : ""
-        diag := diags.format_error(
-            path, b.source, b.pos, b.msg,
-            hint = b.hint,
-            colorize = colorize,
-        )
-        log.warnf("%s%s", prefix, diag)
-    }
-
-    clear(&_pending_warnings)
-}
-```
-
-**`apply_pipeline` change:**
-
-Replace the current warning handling (which logs immediately):
-```odin
-if warnings != nil {
-    append(warnings, ferr)
-} else {
-    log.warnf("%s", format_render_error(ferr, tmpl))
-}
-```
-
-With:
-```odin
-append(&_pending_warnings, ferr)
-```
-
-Drop the `warnings: ^[dynamic]Error` parameter — no longer needed.
-
-### `render.odin`
-
-In `render_template`, wrap the render call:
-```odin
-mustache.reset_warnings()
-result, err := mustache.render(content_tpl, []any{ctx.site, ctx.page, ctx}, partials)
-mustache.flush_warnings(colorize = diags.should_colorize())
-```
-
-### `mustache/pipes.odin` cleanup
-
- Remove `warnings` parameter from `apply_pipeline` signature
- Remove `format_render_error` import/call from `apply_pipeline` (formatting moves to `flush_warnings`)
- Remove the `log` import if no longer used directly in `apply_pipeline`
-
-## Scope
-
-~25 lines new code (`reset_warnings` + `flush_warnings`), ~3 lines changed in `apply_pipeline`, ~2 lines added in `render.odin`. Total ~30 lines.
-
-## Limitations
-
- Module-level state means warnings don't survive across page renders (by design — `reset_warnings` is called per page)
- Not thread-safe (fine for sequential rendering)
- Counting is by exact `msg` string — warnings with different messages but same root cause are counted separately
@@ -5,20 +5,7 @@ import "core:log"
 import "core:os"
 import "core:strings"

-import ts "treesitter"
-
 copy_assets_dir :: proc(vfs: ^VFS, output_dir: string, features: bit_set[Feature]) {
-	// One CSS parser, reused for every .css asset (single-threaded). Owned here.
-	css_parser: ts.Parser
-	if .Minify in features {
-		if g, ok := ts.grammar("css"); ok {
-			css_parser = ts.open_parser(g)
-		}
-	}
-	defer if css_parser != nil {
-		ts.parser_delete(css_parser)
-	}
-
 	for virtual_path, entry in vfs.files {
 		if !strings.has_prefix(virtual_path, "assets/") {
 			continue
@@ -27,12 +14,6 @@ copy_assets_dir :: proc(vfs: ^VFS, output_dir: string, features: bit_set[Feature
 		rel := virtual_path[len("assets/"):]
 		dest := fmt.tprintf("%s/%s", output_dir, rel)

-		// Don't clobber a file already written this build (e.g. the output of a
-		// `tool` pipe, which publishes to the same dest path as its source).
-		if os.exists(dest) {
-			continue
-		}
-
 		if idx := strings.last_index(dest, "/"); idx >= 0 {
 			if err := os.make_directory_all(dest[:idx]); err != nil && err != .Exist {
 				log.warnf("cannot create %s: %v", dest[:idx], err)
@@ -43,11 +24,7 @@ copy_assets_dir :: proc(vfs: ^VFS, output_dir: string, features: bit_set[Feature
 		if .Minify in features && strings.has_suffix(rel, ".css") {
 			data, ok := vfs_get(vfs, virtual_path)
 			if ok {
-				if m, mok := minify_css(css_parser, string(data)); mok {
-					write_file(dest, m)
-				} else {
-					write_file(dest, string(data))
-				}
+				write_file(dest, minify_css(string(data)))
 			}
 		} else if entry.data != nil {
 			if err := os.write_entire_file(dest, entry.data); err != nil {
@@ -3,7 +3,6 @@ package main
 import md "markdown"
 import ts "treesitter"

-import "core:encoding/json"
 import "core:fmt"
 import "core:log"
 import "core:os"
@@ -24,11 +23,10 @@ Page :: struct {
 	weight:      Maybe(int),
 	lastmod:     string,
 	menus:       map[string]Menu_Entry,
-	params:      json.Value,
 	content:     string,
 	og:          Open_Graph,
 	draft:       bool,
-	toc:         string,
+	starred:     bool,
 	_is_index:   bool `private`,
 }

@@ -71,9 +69,6 @@ site_load_content :: proc(site: ^Site) {

 	for &page in site.pages {
 		page.url = fmt.tprintf("%s%s", site.base_url, page.permalink)
-		if site.base_path != "" {
-			page.content = rewrite_root_urls(page.content, site.base_path, site_allocator(site))
-		}
 	}

 	build_menus(site)
@@ -262,7 +257,7 @@ load_page :: proc(
 	page.weight = fm.weight
 	page.lastmod = fm.lastmod
 	page.draft = fm.draft
-	page.params = fm.params
+	page.starred = fm.isStarred

 	if section == "" && is_index {
 		page.permalink = "/"
@@ -279,70 +274,15 @@ load_page :: proc(
 	page.og = fm.og

 	if strings.has_suffix(file_path, ".html") {
-		page.content = strings.clone(body, context.allocator)
+		page.content = strings.clone(body)
 	} else {
-		page.content = md.process(body, ext, file_path, context.allocator)
-	}
-
-	if fm.toc {
-		page.toc = md.generate_toc(page.content, context.allocator)
+		page.content = md.process(body, ext, file_path)
 	}

 	ok = true
 	return
 }

-// rewrite_root_urls prefixes root-relative href/src attribute values with
-// base_path so links authored in content (e.g. [x](/ideas)) resolve under a
-// subpath deployment. Only values beginning with a single '/' are rewritten;
-// absolute ("http://…"), protocol-relative ("//cdn…"), fragment ("#x"),
-// mailto:, and document-relative ("foo/bar", "./foo") values all fail that
-// test and pass through untouched. Assumes cmark's double-quoted output.
-rewrite_root_urls :: proc(html, base_path: string, allocator := context.allocator) -> string {
-	b := strings.builder_make(allocator)
-	strings.builder_grow(&b, len(html))
-
-	i := 0
-	for i < len(html) {
-		n := attr_opener_at(html, i) // ` href="` or ` src="`
-		if n == 0 {
-			strings.write_byte(&b, html[i])
-			i += 1
-			continue
-		}
-		strings.write_string(&b, html[i:i + n]) // copy the opener verbatim
-		i += n
-
-		end := strings.index_byte(html[i:], '"')
-		if end < 0 { 	// malformed; bail cleanly
-			strings.write_string(&b, html[i:])
-			break
-		}
-		value := html[i:i + end]
-		if url_is_root_relative(value) {
-			strings.write_string(&b, base_path)
-		}
-		strings.write_string(&b, value) // closing quote copied next loop
-		i += end
-	}
-	return strings.to_string(b)
-}
-
-// url_is_root_relative reports whether value is a site-root path ("/faq/") as
-// opposed to protocol-relative ("//cdn") or anything not starting with '/'.
-url_is_root_relative :: proc(value: string) -> bool {
-	return len(value) >= 1 && value[0] == '/' && !strings.has_prefix(value, "//")
-}
-
-// attr_opener_at returns the length of a ` href="` / ` src="` attribute opener
-// at s[i:], or 0. The leading space anchors the match to a real HTML attribute
-// so we don't rewrite the strings inside link text or code blocks.
-attr_opener_at :: proc(s: string, i: int) -> int {
-	if strings.has_prefix(s[i:], " href=\"") do return 7
-	if strings.has_prefix(s[i:], " src=\"") do return 6
-	return 0
-}
-
 is_content_file :: proc(name: string) -> bool {
 	return strings.has_suffix(name, ".md") || strings.has_suffix(name, ".html")
 }
@@ -2,17 +2,4 @@ package main

 import "core:os"

-// Path to the built-in default layouts, resolved at runtime.
-//
-// The default points at the `defaults` directory next to the source tree (via
-// `#directory`), which works for local `odin run`/`odin build` from the repo.
-// Packaged builds install `defaults/` to a fixed location and override this
-// with `-define:THOR_DEFAULTS_PATH=<path>` (see flake.nix).
-//
-// TODO: bake these into the binary at compile time via `#load_directory` so the
-// binary is self-contained and this runtime path goes away entirely. See
-// TODOS.md.
-DEFAULTS_PATH :: #config(
-	THOR_DEFAULTS_PATH,
-	#directory + os.Path_Separator_String + "defaults",
-)
+DEFAULTS_PATH :: #directory + os.Path_Separator_String + "defaults"
@@ -4,10 +4,9 @@
 <head>
  <meta charset="utf-8">
  <meta name="viewport" content="width=device-width, initial-scale=1">
-  <title>{{> title}}</title>
-  {{> opengraph}}
-  {{> styles}}
-  {{> scripts}}
+  <title>{{title}}</title>
+  {{> head}}
+  <link rel="stylesheet" href="/css/main.css">
 </head>

 <body>
@@ -5,8 +5,8 @@
    {{&content}}
  </header>
  <ul>
-    {{#pages}} <li><a href="{{permalink | rel_url}}">{{&title}}</a><span>{{#date}}<time
-          datetime="{{date}}">{{date | format}}</time>{{/date}}</span>
+    {{#pages}} <li><a href="{{permalink}}">{{&title}}</a><span>{{#date_iso}}<time
+          datetime="{{date_iso}}">{{date_display}}</time>{{/date_iso}}</span>
    </li>
    {{/pages}}
  </ul>
@@ -3,13 +3,8 @@
 <main>
  <article>
    <h1>{{page.title}}</h1>
-    {{#date}}<time class="subtitle" datetime="{{date}}">{{ date | format}}</time>{{/date}}
-    {{#page.toc}}
-    <nav class="toc">
-    {{&page.toc}}
-    </nav>
-    {{/page.toc}}
-    {{&content}}
+    {{#date_iso}} <time class="subtitle" datetime="{{date_iso}}">{{date_display}}</time>
+    {{/date_iso}} {{&content}}
  </article>
 </main>
 {{/main}}
@@ -1,3 +1,3 @@
 <footer>
-  <p>&copy; {{now | format "2006"}} {{params.author.name}}</p>
+  <p>&copy; {{now.year}} {{params.author.name}}</p>
 </footer>
@@ -1 +1 @@
-{{site.title}}{{^site.title}}Home{{/site.title}}
+{{site.title}}
@@ -1,9 +1,9 @@
 <header>
  <nav>
    <ul>
-      <li><a href="{{site.base_path}}/">{{> home-link}}</a></li>
+      <li><a href="/">{{> home-link}}</a></li>
      {{#menus.main}}
-      <li><a href="{{url | rel_url}}">{{name}}</a></li>
+      <li><a href="{{url}}">{{name}}</a></li>
      {{/menus.main}}
    </ul>
  </nav>
@@ -1 +0,0 @@
-{{params.scripts}}
@@ -1,294 +0,0 @@
-{{^params.stylesheets}}
-<style>
-/* Nav */
-header nav ul {
-  list-style: none;
-  display: flex;
-  gap: 1rem;
-  padding: 0;
-}
-
-/* Base layout */
-body {
-  margin-left: auto;
-  margin-right: auto;
-}
-
-main {
-  max-width: 760px;
-  margin: 0 auto;
-}
-
-/* Sidenote layout — activates only when sidenote elements are present */
-body:has(.sidenote, .marginnote) {
-  counter-reset: sidenote-counter;
-  padding-left: 12.5%;
-  width: 87.5%;
-}
-
-body:has(.sidenote, .marginnote) main {
-  max-width: none;
-  width: 60%;
-  margin: 0;
-}
-
-/* Images */
-img {
-  max-width: 100%;
-  height: auto;
-}
-
-/* Superscript (footnote refs) */
-sup {
-  line-height: 0;
-}
-
-/* Code blocks */
-code,
-pre>code {
-  font-family: Consolas, "Liberation Mono", Menlo, Courier, monospace;
-  font-size: 1.0rem;
-  line-height: 1.42;
-  -webkit-text-size-adjust: 100%;
-}
-
-pre>code {
-  font-size: 0.9rem;
-  overflow-x: auto;
-  display: block;
-}
-
-/* Alerts (GitHub-style) */
-.alert {
-  border-left: 4px solid;
-  border-radius: 0 4px 4px 0;
-  padding: 0.5rem 1rem;
-  margin-inline-start: 0;
-}
-
-.alert-title {
-  font-weight: bold;
-  margin-bottom: 0.25rem;
-  margin-top: 0.25rem;
-}
-
-.alert-note {
-  border-left-color: #3b82f6;
-}
-
-.alert-tip {
-  border-left-color: #22c55e;
-}
-
-.alert-important {
-  border-left-color: #a855f7;
-}
-
-.alert-warning {
-  border-left-color: #eab308;
-}
-
-.alert-caution {
-  border-left-color: #ef4444;
-}
-
-/* Definition lists */
-dl,
-ol,
-ul {
-  font-size: 1.4rem;
-  line-height: 2rem;
-}
-
-dt:not(:first-child) {
-  margin-top: 0.25rem;
-}
-
-dd {
-  margin-left: 0;
-}
-
-/* Sidenotes, margin notes */
-.sidenote,
-.marginnote {
-  float: right;
-  clear: right;
-  margin-right: -45%;
-  width: 40%;
-  margin-top: 0.3rem;
-  margin-bottom: 0;
-  font-size: 0.85em;
-  line-height: 1.5;
-  vertical-align: baseline;
-  position: relative;
-}
-
-.sidenote-number {
-  counter-increment: sidenote-counter;
-}
-
-.sidenote-number:after,
-.sidenote:before {
-  position: relative;
-  vertical-align: baseline;
-  color: var(--color-accent, currentColor);
-}
-
-.sidenote-number:after {
-  content: counter(sidenote-counter);
-  font-size: 0.8rem;
-  top: -0.5rem;
-  left: 0.1rem;
-}
-
-.sidenote:before {
-  content: counter(sidenote-counter) " ";
-  font-size: 0.8rem;
-  top: -0.5rem;
-}
-
-blockquote .sidenote,
-blockquote .marginnote {
-  margin-right: -82%;
-  min-width: 59%;
-  text-align: left;
-}
-
-.marginnote>code,
-.sidenote>code {
-  font-size: 1rem;
-}
-
-input.margin-toggle {
-  display: none;
-}
-
-label.sidenote-number {
-  display: inline-block;
-  max-height: 2rem;
-}
-
-label.margin-toggle:not(.sidenote-number) {
-  display: none;
-}
-
-/* Responsive */
-@media (max-width: 1200px) {
-  body {
-    padding-left: 1rem;
-    padding-right: 1rem;
-  }
-
-  body:has(.sidenote, .marginnote) {
-    padding-left: 8%;
-    padding-right: 8%;
-    width: 84%;
-  }
-
-  body:has(.sidenote, .marginnote) main {
-    width: 100%;
-    max-width: 760px;
-    margin: 0 auto;
-  }
-
-  pre>code {
-    width: 97%;
-  }
-
-  img {
-    width: 100%;
-  }
-
-  label.margin-toggle:not(.sidenote-number) {
-    display: inline;
-  }
-
-  label.margin-toggle:not(.sidenote-number)::after {
-    content: "\2295";
-  }
-
-  .sidenote,
-  .marginnote {
-    display: none;
-  }
-
-  .margin-toggle:checked+.sidenote,
-  .margin-toggle:checked+.marginnote {
-    display: block;
-    float: left;
-    left: 1rem;
-    clear: both;
-    width: 95%;
-    margin: 1rem 2.5%;
-    vertical-align: baseline;
-    position: relative;
-  }
-
-  label {
-    cursor: pointer;
-  }
-}
-</style>
-{{/params.stylesheets}}
-<style>
-/* Syntax highlighting. Build-time <span class="hl-*"> wrappers only appear
-   inside fenced code blocks. The base class (e.g. hl-keyword) is always
-   emitted alongside any dotted variant (hl-keyword-control), so styling the
-   base names covers every capture. Re-theme by overriding the --hl-* vars. */
-:root {
-  --hl-bg:        #282c34;
-  --hl-fg:        #abb2bf;
-  --hl-comment:   #5c6370;
-  --hl-keyword:   #c678dd;
-  --hl-string:    #98c379;
-  --hl-escape:    #56b6c2;
-  --hl-number:    #d19a66;
-  --hl-function:  #61afef;
-  --hl-type:      #e5c07b;
-  --hl-variable:  #e06c75;
-  --hl-attribute: #d19a66;
-}
-
-pre:has(> code [class^="hl-"]) {
-  background: var(--hl-bg);
-  color: var(--hl-fg);
-  padding: 0.75rem 1rem;
-  border-radius: 4px;
-}
-
-.hl-comment { color: var(--hl-comment); font-style: italic; }
-.hl-keyword,
-.hl-conditional,
-.hl-repeat,
-.hl-exception,
-.hl-include,
-.hl-storageclass { color: var(--hl-keyword); }
-.hl-operator,
-.hl-punctuation { color: var(--hl-fg); }
-.hl-string,
-.hl-character { color: var(--hl-string); }
-.hl-escape,
-.hl-string-escape,
-.hl-string-special { color: var(--hl-escape); }
-.hl-number,
-.hl-float,
-.hl-boolean,
-.hl-constant { color: var(--hl-number); }
-.hl-function,
-.hl-method,
-.hl-constructor { color: var(--hl-function); }
-.hl-type,
-.hl-namespace,
-.hl-module { color: var(--hl-type); }
-.hl-variable,
-.hl-property,
-.hl-field,
-.hl-parameter,
-.hl-tag,
-.hl-symbol { color: var(--hl-variable); }
-.hl-attribute,
-.hl-label,
-.hl-preproc { color: var(--hl-attribute); }
-</style>
-{{#params.stylesheets}}<link rel="stylesheet" href="{{.|rel_url}}">
-{{/params.stylesheets}}
@@ -1 +0,0 @@
-{{#site.title}}{{.}}{{#page.title}} | {{/page.title}}{{/site.title}}{{page.title}}
@@ -7,8 +7,8 @@
  <section>
    <h2>{{key}}</h2>
    <ul>
-      {{#items}} <li><a href="{{permalink | rel_url}}">{{&title}}</a><span>{{#date}}<time
-            datetime="{{date}}">{{date | format}}</time>{{/date}}</span>
+      {{#items}} <li><a href="{{permalink}}">{{&title}}</a><span>{{#date_iso}}<time
+            datetime="{{date_iso}}">{{date_display}}</time>{{/date_iso}}</span>
      </li>
      {{/items}}
    </ul>
@@ -1,279 +0,0 @@
-package diagnostics
-
-import "core:fmt"
-import "core:os"
-import "core:strings"
-import "core:terminal/ansi"
-import "core:unicode/utf8"
-
-// suggest_correction returns the closest match from `available` to `missing`
-// using Levenshtein distance, or "" if no good match exists. The threshold
-// scales with the length of the missing key.
-suggest_correction :: proc(available: []string, missing: string) -> string {
-	if len(available) == 0 || len(missing) == 0 {
-		return ""
-	}
-	threshold := max(2, len(missing) / 3)
-
-	best: string
-	best_dist := threshold + 1
-	for candidate in available {
-		if abs(len(candidate) - len(missing)) > threshold {
-			continue
-		}
-		d := strings.levenshtein_distance(missing, candidate)
-		if d <= threshold && d < best_dist {
-			best = candidate
-			best_dist = d
-		}
-	}
-	return best
-}
-
-// line_col returns the 1-indexed line and column for a byte offset in source.
-line_col :: proc(source: string, pos_in: int) -> (line: int, col: int) {
-	pos := pos_in
-	if pos < 0 {
-		return 1, 1
-	}
-	if pos > len(source) {
-		pos = len(source)
-	}
-	line = 1
-	col = 1
-	for i := 0; i < pos; i += 1 {
-		if source[i] == '\n' {
-			line += 1
-			col = 1
-		} else {
-			col += 1
-		}
-	}
-	return
-}
-
-// line_text returns the Nth (1-indexed) line of source, without the trailing
-// newline. Returns "" if line is out of range.
-line_text :: proc(source: string, line: int) -> string {
-	if line < 1 {
-		return ""
-	}
-	current := 1
-	start := 0
-	for i := 0; i < len(source); i += 1 {
-		if current == line {
-			end := i
-			for end < len(source) && source[end] != '\n' {
-				end += 1
-			}
-			return source[start:end]
-		}
-		if source[i] == '\n' {
-			current += 1
-			start = i + 1
-		}
-	}
-	if current == line {
-		return source[start:]
-	}
-	return ""
-}
-
-// should_colorize returns true if stderr is a TTY and color output is wanted.
-should_colorize :: proc() -> bool {
-	return os.is_tty(os.stderr)
-}
-
-// count_lines returns the number of '\n'-separated lines in source.
-count_lines :: proc(source: string) -> int {
-	if len(source) == 0 {
-		return 1
-	}
-	n := 1
-	for c in source {
-		if c == '\n' {
-			n += 1
-		}
-	}
-	if len(source) > 0 && source[len(source) - 1] == '\n' {
-		n -= 1
-	}
-	return n
-}
-
-// digit_count returns the number of decimal digits in n (min 1).
-digit_count :: proc(n: int) -> int {
-	if n <= 0 {
-		return 1
-	}
-	c := 0
-	x := n
-	for x > 0 {
-		c += 1
-		x /= 10
-	}
-	return c
-}
-
-// display_width returns the number of terminal cells `s` occupies.
-display_width :: proc(s: string) -> int {
-	return utf8.rune_count_in_string(s)
-}
-
-// line_start_at returns the byte offset of the start of the line containing pos.
-line_start_at :: proc(source: string, pos_in: int) -> int {
-	pos := pos_in
-	if pos < 0 {
-		return 0
-	}
-	if pos >= len(source) {
-		pos = len(source) - 1
-	}
-	for pos > 0 && source[pos - 1] != '\n' {
-		pos -= 1
-	}
-	return pos
-}
-
-// format_error produces a rust-style multi-line diagnostic string.
-//
-//   <msg>
-//    --> <path>:<line>:<col>
-//     |
-//  N | <source line N-2>
-//  N | <source line N-1>
-//  N | <source line N — the error line>
-//     |     ^^^^^^^^^^^ <hint>
-//  N | <source line N+1>
-//  N | <source line N+2>
-//     |
-//
-// `caret_start`/`caret_end` control the byte range of the caret underline.
-// If not provided (or negative), defaults to a single-rune underline at `pos`.
-format_error :: proc(
-	path: string,
-	source: string,
-	pos: int,
-	msg: string,
-	hint: string = "",
-	context_before: int = 2,
-	context_after: int = 2,
-	colorize: bool = false,
-	caret_start: int = -1,
-	caret_end: int = -1,
-) -> string {
-	line, col := line_col(source, pos)
-	total_lines := count_lines(source)
-
-	start_line := line - context_before
-	if start_line < 1 {
-		start_line = 1
-	}
-	end_line := line + context_after
-	if end_line > total_lines {
-		end_line = total_lines
-	}
-
-	width := digit_count(end_line)
-	if width < 1 {
-		width = 1
-	}
-
-	cs := caret_start >= 0 ? caret_start : pos
-	ce := caret_end > cs ? caret_end : cs + 1
-
-	ls := line_start_at(source, pos)
-	caret_start_col := cs - ls + 1
-	caret_end_col := ce - ls + 1
-
-	sb := strings.builder_make(context.temp_allocator)
-	defer strings.builder_destroy(&sb)
-
-	red, faint, reset := "", "", ""
-	if colorize {
-		red = ansi.CSI + ansi.FG_RED + ansi.SGR
-		faint = ansi.CSI + ansi.FAINT + ansi.SGR
-		reset = ansi.CSI + ansi.RESET + ansi.SGR
-	}
-
-	// Header
-	strings.write_string(&sb, msg)
-	strings.write_byte(&sb, '\n')
-
-	// Location line
-	strings.write_string(&sb, faint)
-	for _ in 0 ..< width {
-		strings.write_byte(&sb, ' ')
-	}
-	strings.write_string(&sb, "--> ")
-	strings.write_string(&sb, reset)
-	fmt.sbprintf(&sb, "%s:%d:%d\n", path, line, col)
-
-	// Top gutter
-	write_gutter(&sb, width, faint, reset)
-
-	// Context lines
-	for n in start_line ..= end_line {
-		num_str := fmt.tprintf("%d", n)
-		strings.write_string(&sb, faint)
-		for _ in 0 ..< width - len(num_str) {
-			strings.write_byte(&sb, ' ')
-		}
-		strings.write_string(&sb, num_str)
-		strings.write_string(&sb, " | ")
-		strings.write_string(&sb, reset)
-
-		strings.write_string(&sb, line_text(source, n))
-		strings.write_byte(&sb, '\n')
-
-		// Caret row after error line
-		if n == line {
-			strings.write_string(&sb, faint)
-			for _ in 0 ..< width + 1 {
-				strings.write_byte(&sb, ' ')
-			}
-			strings.write_string(&sb, "| ")
-			strings.write_string(&sb, reset)
-
-			for _ in 1 ..< caret_start_col {
-				strings.write_byte(&sb, ' ')
-			}
-			if colorize {
-				strings.write_string(&sb, red)
-			}
-			for _ in 0 ..< caret_end_col - caret_start_col {
-				strings.write_byte(&sb, '^')
-			}
-			if colorize {
-				strings.write_string(&sb, reset)
-			}
-			if hint != "" {
-				strings.write_byte(&sb, ' ')
-				if colorize {
-					strings.write_string(&sb, faint)
-				}
-				strings.write_string(&sb, hint)
-				if colorize {
-					strings.write_string(&sb, reset)
-				}
-			}
-			strings.write_byte(&sb, '\n')
-		}
-	}
-
-	// Trailing gutter
-	write_gutter(&sb, width, faint, reset)
-
-	return strings.to_string(sb)
-}
-
-// write_gutter emits a faint pipe-only gutter line: `<width+1 spaces> |`.
-write_gutter :: proc(sb: ^strings.Builder, width: int, faint: string, reset: string) {
-	strings.write_string(sb, faint)
-	for _ in 0 ..< width + 1 {
-		strings.write_byte(sb, ' ')
-	}
-	strings.write_string(sb, "|")
-	strings.write_string(sb, reset)
-	strings.write_byte(sb, '\n')
-}
@@ -1,111 +0,0 @@
-#+test
-package diagnostics
-
-import "core:fmt"
-import "core:testing"
-
-@(test)
-test_suggest_correction_exact :: proc(t: ^testing.T) {
-	available := []string{"title", "page.title", "body"}
-	testing.expect_value(t, suggest_correction(available, "page_titel"), "page.title")
-}
-
-@(test)
-test_suggest_correction_close :: proc(t: ^testing.T) {
-	available := []string{"title", "body", "now"}
-	testing.expect_value(t, suggest_correction(available, "titel"), "title")
-}
-
-@(test)
-test_suggest_correction_no_match :: proc(t: ^testing.T) {
-	available := []string{"completely_different", "unrelated"}
-	testing.expect_value(t, suggest_correction(available, "page_titel"), "")
-}
-
-@(test)
-test_suggest_correction_empty :: proc(t: ^testing.T) {
-	testing.expect_value(t, suggest_correction([]string{}, "anything"), "")
-	testing.expect_value(t, suggest_correction([]string{"a"}, ""), "")
-}
-
-// ---------------------------------------------------------------------------
-// line_col / line_text / count_lines / digit_count — primitive helpers
-// ---------------------------------------------------------------------------
-
-@(test)
-test_line_col_basic :: proc(t: ^testing.T) {
-	src := "abc\ndef\nghi"
-	cases := [?]struct {
-		pos:  int,
-		line: int,
-		col:  int,
-	}{{0, 1, 1}, {2, 1, 3}, {3, 1, 4}, {4, 2, 1}, {6, 2, 3}, {7, 2, 4}, {8, 3, 1}}
-	for c in cases {
-		l, col := line_col(src, c.pos)
-		testing.expect(t, l == c.line, fmt.tprintf("pos %d: line %d, want %d", c.pos, l, c.line))
-		testing.expect(t, col == c.col, fmt.tprintf("pos %d: col %d, want %d", c.pos, col, c.col))
-	}
-}
-
-@(test)
-test_line_col_empty :: proc(t: ^testing.T) {
-	l, col := line_col("", 0)
-	testing.expect_value(t, l, 1)
-	testing.expect_value(t, col, 1)
-}
-
-@(test)
-test_line_col_negative :: proc(t: ^testing.T) {
-	l, col := line_col("abc", -1)
-	testing.expect_value(t, l, 1)
-	testing.expect_value(t, col, 1)
-}
-
-@(test)
-test_line_col_past_end :: proc(t: ^testing.T) {
-	l, col := line_col("abc", 100)
-	testing.expect_value(t, l, 1)
-	testing.expect_value(t, col, 4)
-}
-
-@(test)
-test_line_text_first :: proc(t: ^testing.T) {
-	src := "first\nsecond\nthird"
-	testing.expect_value(t, line_text(src, 1), "first")
-	testing.expect_value(t, line_text(src, 2), "second")
-	testing.expect_value(t, line_text(src, 3), "third")
-}
-
-@(test)
-test_line_text_trailing_newline :: proc(t: ^testing.T) {
-	src := "first\nsecond\n"
-	testing.expect_value(t, line_text(src, 1), "first")
-	testing.expect_value(t, line_text(src, 2), "second")
-	testing.expect_value(t, line_text(src, 3), "")
-}
-
-@(test)
-test_line_text_out_of_range :: proc(t: ^testing.T) {
-	testing.expect_value(t, line_text("abc", 5), "")
-	testing.expect_value(t, line_text("abc", 0), "")
-}
-
-@(test)
-test_count_lines :: proc(t: ^testing.T) {
-	testing.expect_value(t, count_lines(""), 1)
-	testing.expect_value(t, count_lines("abc"), 1)
-	testing.expect_value(t, count_lines("a\nb"), 2)
-	testing.expect_value(t, count_lines("a\nb\n"), 2)
-	testing.expect_value(t, count_lines("a\nb\nc"), 3)
-}
-
-@(test)
-test_digit_count :: proc(t: ^testing.T) {
-	testing.expect_value(t, digit_count(0), 1)
-	testing.expect_value(t, digit_count(1), 1)
-	testing.expect_value(t, digit_count(9), 1)
-	testing.expect_value(t, digit_count(10), 2)
-	testing.expect_value(t, digit_count(99), 2)
-	testing.expect_value(t, digit_count(100), 3)
-	testing.expect_value(t, digit_count(-5), 1)
-}
@@ -1,6 +1,5 @@
 package main

-import "base:runtime"
 import "core:fmt"
 import "core:strings"
 import "core:time"
@@ -8,9 +7,10 @@ import "core:time"
 generate_rss :: proc(site: ^Site) -> string {
 	sb := strings.builder_make()

-	fmt.sbprintf(
+	strings.write_string(
 		&sb,
-		`<?xml version="1.0" encoding="utf-8" standalone="yes"?>
+		fmt.aprintf(
+			`<?xml version="1.0" encoding="utf-8" standalone="yes"?>
 <rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
 <channel>
 <title>%s</title>
@@ -18,10 +18,11 @@ generate_rss :: proc(site: ^Site) -> string {
 <description>%s</description>
 <language>en-us</language>
 <atom:link href="%s/index.xml" rel="self" type="application/rss+xml"/>`,
-		xml_escape(site.title),
-		site.base_url,
-		xml_escape(site.description),
-		site.base_url,
+			xml_escape(site.title),
+			site.base_url,
+			xml_escape(site.description),
+			site.base_url,
+		),
 	)

 	for page in site.pages {
@@ -34,9 +35,10 @@ generate_rss :: proc(site: ^Site) -> string {
 			pub_date = format_rfc822(page.date)
 		}

-		fmt.sbprintf(
+		strings.write_string(
 			&sb,
-			`<item>
+			fmt.aprintf(
+				`<item>
 <title>%s</title>
 <link>%s</link>
 <pubDate>%s</pubDate>
@@ -44,11 +46,12 @@ generate_rss :: proc(site: ^Site) -> string {
 <description>%s</description>
 </item>
 `,
-			xml_escape(page.title),
-			page.url,
-			pub_date,
-			page.url,
-			xml_escape(page.content),
+				xml_escape(page.title),
+				page.url,
+				pub_date,
+				page.url,
+				xml_escape(page.content),
+			),
 		)
 	}

@@ -67,11 +70,11 @@ generate_sitemap :: proc(site: ^Site) -> string {
 	)

 	for page in site.pages {
-		fmt.sbprintf(&sb, "<url><loc>%s</loc>", page.url)
+		lastmod := ""
 		if page.date != "" {
-			fmt.sbprintf(&sb, "<lastmod>%s</lastmod>", page.date)
+			lastmod = fmt.aprintf("<lastmod>%s</lastmod>", page.date)
 		}
-		fmt.sbprintf(&sb, "</url>\n")
+		strings.write_string(&sb, fmt.aprintf("<url><loc>%s</loc>%s</url>\n", page.url, lastmod))
 	}

 	// Section index pages (for sections without an index in content)
@@ -100,18 +103,22 @@ generate_sitemap :: proc(site: ^Site) -> string {
 				section_lastmod = page.date
 			}
 		}
-		fmt.sbprintf(&sb, "<url><loc>%s/%s/</loc>", site.base_url, section)
+		lm := ""
 		if section_lastmod != "" {
-			fmt.sbprintf(&sb, "<lastmod>%s</lastmod>", section_lastmod)
+			lm = fmt.aprintf("<lastmod>%s</lastmod>", section_lastmod)
 		}
-		fmt.sbprintf(&sb, "</url>\n")
+		strings.write_string(
+			&sb,
+			fmt.aprintf("<url><loc>%s/%s/</loc>%s</url>\n", site.base_url, section, lm),
+		)
 	}

 	strings.write_string(&sb, "</urlset>")
 	return strings.to_string(sb)
 }

-format_rfc822 :: proc(iso: string, allocator := context.temp_allocator) -> string {
+// TODO: Leaks
+format_rfc822 :: proc(iso: string) -> string {
 	if len(iso) < 19 {
 		// TODO: should indicate error somehow
 		return iso
@@ -124,66 +131,21 @@ format_rfc822 :: proc(iso: string, allocator := context.temp_allocator) -> strin
 	buf: [8]byte
 	t := time.to_string_hms(date, buf[:])

-	// RFC-822 zone: sign followed by two-digit hours and minutes (e.g. +0000,
-	// +0530, -0800). offset is in minutes.
-	sign := '+'
-	off := offset
-	if off < 0 {
-		sign = '-'
-		off = -off
-	}
-
 	return fmt.aprintf(
-		"%s, %02d %s %d %s %c%02d%02d",
+		"%s, %02d %s %d %s %3d%2d",
 		weekday[:3],
 		time.day(date),
 		month[:3],
 		time.year(date),
 		t,
-		sign,
-		off / 60,
-		off % 60,
-		allocator = allocator,
+		offset / 60,
+		offset % 60,
 	)
 }

-// xml_escape escapes the XML metacharacters '&', '<' and '>' in s. Only these
-// ASCII characters are touched; all other bytes (including UTF-8 multibyte
-// sequences) pass through verbatim, so s must already be valid UTF-8.
-//
-// The result is allocated in `allocator`, which defaults to
-// context.temp_allocator — the returned string is only valid until the next
-// temp allocator reset, so callers must not retain it across frames.
-xml_escape :: proc(
-	s: string,
-	allocator := context.temp_allocator,
-) -> (
-	escaped: string,
-	err: runtime.Allocator_Error,
-) #optional_allocator_error {
-	sb := strings.builder_make_len_cap(0, len(s), allocator) or_return
-
-	// Single pass over the bytes, copying runs of ordinary text in bulk and
-	// emitting an entity for each metacharacter. The switch is both the
-	// membership test and the replacement lookup, so no per-match scan.
-	start := 0
-	for i := 0; i < len(s); i += 1 {
-		esc: string
-		switch s[i] {
-		case '&':
-			esc = "&amp;"
-		case '<':
-			esc = "&lt;"
-		case '>':
-			esc = "&gt;"
-		case:
-			continue // ordinary byte: copied as part of the next run
-		}
-		strings.write_string(&sb, s[start:i]) // run before the metacharacter
-		strings.write_string(&sb, esc)
-		start = i + 1
-	}
-	strings.write_string(&sb, s[start:]) // tail
-	return strings.to_string(sb), nil
+xml_escape :: proc(s: string) -> string {
+	r, _ := strings.replace_all(s, "&", "&amp;")
+	r, _ = strings.replace_all(r, "<", "&lt;")
+	r, _ = strings.replace_all(r, ">", "&gt;")
+	return r
 }
-
@@ -1,166 +0,0 @@
-#+test
-package main
-
-import "core:strings"
-import "core:testing"
-
-// ---------------------------------------------------------------------------
-// xml_escape
-// ---------------------------------------------------------------------------
-
-@(test)
-test_xml_escape_passthrough :: proc(t: ^testing.T) {
-	testing.expect_value(t, xml_escape(""), "")
-	testing.expect_value(t, xml_escape("hello world"), "hello world")
-	// UTF-8 multibyte bytes are not metacharacters and pass through untouched.
-	testing.expect_value(t, xml_escape("héllo — 世界"), "héllo — 世界")
-}
-
-@(test)
-test_xml_escape_entities :: proc(t: ^testing.T) {
-	testing.expect_value(t, xml_escape("&"), "&amp;")
-	testing.expect_value(t, xml_escape("<"), "&lt;")
-	testing.expect_value(t, xml_escape(">"), "&gt;")
-	// Adjacent specials, and specials at the very start/end of the string.
-	testing.expect_value(t, xml_escape("&<>"), "&amp;&lt;&gt;")
-	testing.expect_value(t, xml_escape("<a>"), "&lt;a&gt;")
-}
-
-@(test)
-test_xml_escape_mixed :: proc(t: ^testing.T) {
-	testing.expect_value(t, xml_escape("a<b>c&d"), "a&lt;b&gt;c&amp;d")
-	testing.expect_value(
-		t,
-		xml_escape("<p>Tom & Jerry</p>"),
-		"&lt;p&gt;Tom &amp; Jerry&lt;/p&gt;",
-	)
-}
-
-// ---------------------------------------------------------------------------
-// format_rfc822
-// ---------------------------------------------------------------------------
-
-@(test)
-test_format_rfc822_short_passthrough :: proc(t: ^testing.T) {
-	// Inputs shorter than a full ISO-8601 timestamp are returned unchanged
-	// (and are the input slice itself, so they must not be deleted).
-	testing.expect_value(t, format_rfc822("2024"), "2024")
-	testing.expect_value(t, format_rfc822(""), "")
-}
-
-@(test)
-test_format_rfc822_full_date :: proc(t: ^testing.T) {
-	// Results are temp-allocated (reset by the test runner), so no delete.
-	// 2024-01-15 is a Monday; wall-clock time is preserved and the zone is
-	// rendered as an RFC-822 ±HHMM offset.
-	testing.expect_value(
-		t,
-		format_rfc822("2024-01-15T10:30:00Z"),
-		"Mon, 15 Jan 2024 10:30:00 +0000",
-	)
-	testing.expect_value(
-		t,
-		format_rfc822("2024-01-15T10:30:00+05:30"),
-		"Mon, 15 Jan 2024 10:30:00 +0530",
-	)
-	testing.expect_value(
-		t,
-		format_rfc822("2024-01-15T10:30:00-08:00"),
-		"Mon, 15 Jan 2024 10:30:00 -0800",
-	)
-}
-
-// ---------------------------------------------------------------------------
-// generate_rss
-// ---------------------------------------------------------------------------
-
-@(test)
-test_generate_rss_structure :: proc(t: ^testing.T) {
-	site: Site
-	site.title = "My & Site"
-	site.description = "Desc <here>"
-	site.base_url = "https://example.com"
-	site.pages = make(#soa[dynamic]Page)
-	defer delete_soa(site.pages)
-
-	// A real content page.
-	append(
-		&site.pages,
-		Page {
-			section = "posts",
-			title = "First <Post>",
-			url = "https://example.com/posts/first/",
-			content = "<p>Hello & welcome</p>",
-		},
-	)
-	// The home index page (section == "" && _is_index) must be skipped.
-	append(
-		&site.pages,
-		Page {
-			section = "",
-			_is_index = true,
-			title = "HOME_SHOULD_BE_SKIPPED",
-			url = "https://example.com/",
-			content = "home",
-		},
-	)
-	// A second content page.
-	append(
-		&site.pages,
-		Page {
-			section = "posts",
-			title = "Second Post",
-			url = "https://example.com/posts/second/",
-			content = "plain body",
-		},
-	)
-
-	rss := generate_rss(&site)
-	defer delete(rss)
-
-	// Channel header: site title/description are XML-escaped, link gets a
-	// trailing slash, and the atom self-link points at index.xml.
-	testing.expect(t, strings.contains(rss, "<title>My &amp; Site</title>"), rss)
-	testing.expect(
-		t,
-		strings.contains(rss, "<description>Desc &lt;here&gt;</description>"),
-		rss,
-	)
-	testing.expect(t, strings.contains(rss, "<link>https://example.com/</link>"), rss)
-	testing.expect(
-		t,
-		strings.contains(rss, `<atom:link href="https://example.com/index.xml"`),
-		rss,
-	)
-
-	// Exactly two <item>s — the home index page is skipped.
-	testing.expect_value(t, strings.count(rss, "<item>"), 2)
-	testing.expect(t, !strings.contains(rss, "HOME_SHOULD_BE_SKIPPED"), rss)
-
-	// First item: title and body are XML-escaped, link is verbatim.
-	testing.expect(t, strings.contains(rss, "<title>First &lt;Post&gt;</title>"), rss)
-	testing.expect(
-		t,
-		strings.contains(
-			rss,
-			"<description>&lt;p&gt;Hello &amp; welcome&lt;/p&gt;</description>",
-		),
-		rss,
-	)
-	testing.expect(
-		t,
-		strings.contains(rss, "<link>https://example.com/posts/first/</link>"),
-		rss,
-	)
-
-	// Undated pages get the sentinel pubDate (no date formatting/allocation).
-	testing.expect(
-		t,
-		strings.contains(rss, "<pubDate>Mon, 01 Jan 0001 00:00:00 +0000</pubDate>"),
-		rss,
-	)
-
-	// Well-formed at the boundaries.
-	testing.expect(t, strings.has_prefix(rss, `<?xml version="1.0"`), rss)
-	testing.expect(t, strings.has_suffix(rss, "</channel>\n</rss>"), rss)
-}
@@ -141,20 +141,13 @@

            buildPhase = ''
              runHook preBuild
-              odin build . -o:speed -no-bounds-check \
-                -define:THOR_DEFAULTS_PATH=$out/share/thor/defaults \
-                -out:${pname}-keep
+              odin build . -o:speed -no-bounds-check -out:${pname}-keep
              runHook postBuild
            '';

            installPhase = ''
              runHook preInstall
              install -Dm755 ${pname}-keep $out/bin/${pname}
-              # Ship the built-in default layouts alongside the binary. The
-              # binary reads these at runtime from THOR_DEFAULTS_PATH (baked in
-              # above). Keep in sync with defaults.odin.
-              mkdir -p $out/share/thor
-              cp -r defaults $out/share/thor/defaults
              runHook postInstall
            '';
          };
@@ -165,17 +158,9 @@
              tree-sitter

              gdb
-              perf
-
-              bloaty # du for binaries?
-
-              # temp
-              hugo
-              pandoc

              # IDE
              unstable.helix
-              claude-code
              typescript-language-server
              vscode-langservers-extracted
            ]);
@@ -12,11 +12,10 @@ Frontmatter :: struct {
 	publishDate: string,
 	weight:      Maybe(int),
 	menus:       json.Value,
-	params:      json.Value,
 	layout:      string,
 	og:          Open_Graph,
 	draft:       bool,
-	toc:         bool,
+	isStarred:   bool,
 }

 // parse_frontmatter splits raw file content into a Frontmatter struct and the
@@ -38,7 +37,7 @@ parse_frontmatter :: proc(content: string) -> (fm: Frontmatter, body: string, ok
 	json_str := content[:end + 1]
 	body = strings.trim_left(content[end + 1:], " \t\r\n")

-	value, err := json.parse_string(json_str, spec = .JSON5)
+	value, err := json.parse_string(json_str, spec = .JSON)
 	if err != nil {
 		log.errorf("failed to parse frontmatter JSON: %v", err)
 		return
@@ -57,13 +56,12 @@ parse_frontmatter :: proc(content: string) -> (fm: Frontmatter, body: string, ok
 	fm.publishDate = json_get_string(obj, "publishDate")
 	fm.weight = json_get_int(obj, "weight")
 	fm.draft = json_get_bool(obj, "draft")
-	fm.toc = json_get_bool(obj, "toc")
+	fm.isStarred = json_get_bool(obj, "isStarred")
 	if v, ok := obj["menus"]; ok {
 		fm.menus = v
 	}
 	fm.layout = json_get_string(obj, "layout")
 	fm.og = json_get_open_graph(obj, "og")
-	if v, ok := obj["params"]; ok {fm.params = v}

 	ok = true
 	return
@@ -119,4 +117,3 @@ json_get_open_graph :: proc(obj: json.Object, key: string) -> Open_Graph {
 	}
 	return og
 }
-
@@ -1,9 +1,7 @@
 package main

 import "base:runtime"
-import "core:flags"
 import "core:log"
-import "core:mem"
 import "core:os"
 import "core:prof/spall"
 import "core:sync"
@@ -45,21 +43,9 @@ main :: proc() {
 		defer spall.buffer_destroy(&spall_ctx, &spall_buffer)
 	}

-	cli_flags: Flags
-	flags.parse_or_exit(&cli_flags, os.args, .Odin)
-
-	level: log.Level
-	switch {
-	case cli_flags.quiet:
-		level = .Warning
-	case cli_flags.verbose:
-		level = .Debug
-	case:
-		level = .Info
-	}
 	logger_opts: log.Options =
 		(log.Default_Console_Logger_Opts - log.Full_Timestamp_Opts - {.Short_File_Path})
-	console_logger := log.create_console_logger(level, logger_opts)
+	console_logger := log.create_console_logger(.Info, logger_opts)
 	context.logger = console_logger
 	defer log.destroy_console_logger(console_logger)

@@ -71,21 +57,12 @@ main :: proc() {

 	for {
 		defer free_all(context.temp_allocator)
-		defer log.debugf(
-			"max_temp_allocator_size=%M",
-			(cast(^runtime.Default_Temp_Allocator)context.temp_allocator.data)^.arena.total_used,
-		)
 		tick := time.tick_now()
 		site: Site
-		init_site(&site, cli_flags)
+		init_site(&site, os.args)
 		defer destroy_site(&site)
 		// TODO: Make it so this isn't necessary
 		context.allocator = site_allocator(&site)
-		defer log.debugf(
-			"current_site_allocator_size=%M",
-			site.arena.block_size * (len(site.arena.used_blocks) + len(site.arena.unused_blocks)) -
-			site.arena.bytes_left,
-		)
 		build_vfs(&site)

 		treesitter.grammar_dir = site.grammars
@@ -93,7 +70,7 @@ main :: proc() {

 		site_load_content(&site)
 		render_site(&site)
-		log.infof("Built site in %M", time.tick_since(tick))
+		log.infof("Built site in %s", time.tick_since(tick))

 		(.Watch in site.features) or_break
 		time.sleep(5 * time.Second)
@@ -117,4 +94,3 @@ when SPALL {
 		spall._buffer_end(&spall_ctx, &spall_buffer)
 	}
 }
-
@@ -1,324 +0,0 @@
-package markdown
-
-import cm "vendor:commonmark"
-
-import "core:strings"
-
-// convert_deflists scans markdown text for definition list patterns and
-// converts them to <dl><dt><dd> HTML blocks before cmark processing.
-//
-// A definition line starts with fewer than four columns of leading whitespace
-// followed by a colon and a space (or end-of-line). The term is the nearest
-// preceding non-blank line (immediately or within one blank line). Consecutive
-// term+definition pairs are grouped into a single <dl> block, and consecutive
-// definition lines under one term become multiple <dd> elements.
-//
-// The scanner walks the body one line at a time using byte offsets: it never
-// splits the input, allocates nothing per line, and copies non-deflist text
-// through verbatim. Matching is disabled inside fenced code blocks. Terms and
-// definitions are rendered through cmark individually so that inline markdown
-// (code, links, emphasis) is processed.
-// Line_Kind classifies a line for convert_deflists's group-finding scan. Other
-// covers fenced-code delimiters and their contents, where definition syntax is
-// inert.
-Line_Kind :: enum {
-	None,
-	Blank,
-	Term,
-	Def,
-	Other,
-}
-
-convert_deflists :: proc(body: string, allocator := context.allocator) -> string {
-	// Fast path: a body with no definition line at all cannot contain a
-	// definition list, so return it untouched — skipping the builder allocation
-	// and the classifying scan below. This is the common case (most pages have
-	// no definition lists), and the guard scan is leaner than the full loop.
-	if !has_def_line(body) {
-		return body
-	}
-
-	sb := strings.builder_make(context.temp_allocator)
-
-	// emitted_up_to is the byte offset up to which `body` has been copied into
-	// `sb`. It stays at 0 until the first deflist is emitted; if it is still 0
-	// at the end, nothing matched and we can return the body unchanged.
-	emitted_up_to := 0
-
-	in_fence := false
-	fence_marker: u8 = 0
-
-	// The scan is definition-driven: it walks line by line, remembering the two
-	// previous lines' kinds and offsets, and only does group-finding work when it
-	// reaches a (rare) definition line — whose term is the immediately preceding
-	// line, or the one before a single blank line. This avoids a per-line forward
-	// lookahead: each line is classified once, with no rescanning.
-	prev1_kind, prev2_kind := Line_Kind.None, Line_Kind.None
-	prev1_pos, prev2_pos := 0, 0
-
-	pos := 0
-	for {
-		line, next, ok := line_at(body, pos)
-		if !ok {
-			break
-		}
-
-		// Classify the line. Definition-list syntax is inert inside fenced code
-		// blocks, so fence delimiters and their contents are classified Other.
-		kind: Line_Kind
-		if in_fence {
-			if is_fence_line(line, fence_marker) {
-				in_fence = false
-			}
-			kind = .Other
-		} else if marker, opened := fence_open(line); opened {
-			in_fence = true
-			fence_marker = marker
-			kind = .Other
-		} else if is_def_line(line) {
-			kind = .Def
-		} else if is_blank_line(line) {
-			kind = .Blank
-		} else {
-			kind = .Term
-		}
-
-		// A definition-list group begins at the term preceding this def line
-		// (immediately, or across a single blank line). Flush pending passthrough
-		// text verbatim, then render the whole group directly into the builder.
-		if kind == .Def {
-			term_pos := -1
-			if prev1_kind == .Term {
-				term_pos = prev1_pos
-			} else if prev1_kind == .Blank && prev2_kind == .Term {
-				term_pos = prev2_pos
-			}
-			if term_pos >= 0 {
-				strings.write_string(&sb, body[emitted_up_to:term_pos])
-				group_end := write_deflist_group(&sb, body, term_pos)
-				// write_deflist_group emits a raw <dl> HTML block with no
-				// trailing newline, so the blank line the author left after the
-				// list is consumed as the </dl> line's terminator rather than a
-				// separating blank line. CommonMark (type-6 HTML block) then
-				// swallows the following block — e.g. a fenced code example —
-				// as raw, unescaped HTML, leaking its markup (a template's
-				// <h1>) into the document. Re-emit a newline so </dl> stands on
-				// its own line and the author's blank line still separates the
-				// block. Only when a blank line actually follows: fused
-				// non-blank fall-through content is left as-is.
-				if line, _, ok := line_at(body, group_end); ok && is_blank_line(line) {
-					strings.write_byte(&sb, '\n')
-				}
-				emitted_up_to = group_end
-				pos = group_end
-				prev1_kind, prev2_kind = .None, .None
-				in_fence = false
-				continue
-			}
-		}
-
-		prev2_kind, prev2_pos = prev1_kind, prev1_pos
-		prev1_kind, prev1_pos = kind, pos
-		pos = next
-	}
-
-	if emitted_up_to == 0 {
-		// No definition list anywhere; the body passes through unchanged.
-		return body
-	}
-	strings.write_string(&sb, body[emitted_up_to:])
-	return strings.clone(strings.to_string(sb), allocator)
-}
-
-// line_at returns the line beginning at byte offset `pos` (excluding its
-// trailing newline) along with the offset of the following line. `pos` must be
-// 0 or one past a '\n'. ok is false once `pos` reaches the end of `body`.
-line_at :: proc(body: string, pos: int) -> (line: string, next: int, ok: bool) {
-	if pos >= len(body) {
-		return "", pos, false
-	}
-	nl := strings.index_byte(body[pos:], '\n')
-	if nl < 0 {
-		return body[pos:], len(body), true
-	}
-	return body[pos:pos + nl], pos + nl + 1, true
-}
-
-// has_def_line reports whether any line in `body` is a definition line (see
-// is_def_line). It is a cheap guard for convert_deflists: a body with no
-// definition line cannot contain a definition list. There are no false
-// negatives (every definition list requires a definition line); a false
-// positive — e.g. a colon line that only appears inside a code fence — simply
-// falls through to the full scan, which then finds no match and returns `body`.
-has_def_line :: proc(body: string) -> bool {
-	pos := 0
-	for {
-		line, next := line_at(body, pos) or_break
-		if is_def_line(line) {
-			return true
-		}
-		pos = next
-	}
-	return false
-}
-
-// write_deflist_group renders the definition-list group beginning at `start`
-// directly into `sb`, wrapped in <dl>...</dl>, and returns the byte offset one
-// past the group (the start of the first line not consumed). The caller must
-// have confirmed via term_has_def that a group starts at `start`, so the group
-// always contains at least one term+definition pair. A trailing blank line that
-// separates the group from following content is left unconsumed.
-write_deflist_group :: proc(sb: ^strings.Builder, body: string, start: int) -> (end: int) {
-	strings.write_string(sb, "<dl>")
-
-	pos := start
-	for {
-		term_line, term_next, term_ok := line_at(body, pos)
-		if !term_ok || is_blank_line(term_line) || is_def_line(term_line) {
-			break
-		}
-
-		// The definition follows immediately or across a single blank line.
-		def_pos := term_next
-		def_line, def_next, def_ok := line_at(body, def_pos)
-		if def_ok && is_blank_line(def_line) {
-			def_pos = def_next
-			def_line, def_next, def_ok = line_at(body, def_pos)
-		}
-		if !def_ok || !is_def_line(def_line) {
-			break
-		}
-
-		// Emit the term with its first definition, then any consecutive
-		// definition lines as additional <dd>s under the same term.
-		strings.write_string(sb, "<dt>")
-		strings.write_string(sb, render_inline_md(strings.trim_space(term_line)))
-		strings.write_string(sb, "</dt><dd>")
-		strings.write_string(sb, render_inline_md(def_content(def_line)))
-		strings.write_string(sb, "</dd>")
-		pos = def_next
-		for {
-			cont, cont_next, cont_ok := line_at(body, pos)
-			if !cont_ok || !is_def_line(cont) {
-				break
-			}
-			strings.write_string(sb, "<dd>")
-			strings.write_string(sb, render_inline_md(def_content(cont)))
-			strings.write_string(sb, "</dd>")
-			pos = cont_next
-		}
-
-		// Continue the group if another term+definition pair follows,
-		// optionally separated by one blank line. Otherwise stop here without
-		// consuming a trailing blank line.
-		blank_line, after_blank, blank_ok := line_at(body, pos)
-		if blank_ok && is_blank_line(blank_line) {
-			if term_has_def(body, after_blank) {
-				pos = after_blank
-				continue
-			}
-			break
-		}
-		// No blank line: loop and let the top re-test this line as a term.
-	}
-
-	strings.write_string(sb, "</dl>")
-	return pos
-}
-
-// term_has_def reports whether the line at `pos` is a term that is immediately
-// (or across one blank line) followed by a definition line.
-term_has_def :: proc(body: string, pos: int) -> bool {
-	line, next, ok := line_at(body, pos)
-	if !ok || is_blank_line(line) || is_def_line(line) {
-		return false
-	}
-	def_pos := next
-	def_line, def_next, def_ok := line_at(body, def_pos)
-	if def_ok && is_blank_line(def_line) {
-		def_pos = def_next
-		def_line, def_next, def_ok = line_at(body, def_pos)
-	}
-	return def_ok && is_def_line(def_line)
-}
-
-// fence_open reports whether `line` opens a fenced code block and, if so, which
-// marker character it uses. Leading spaces are ignored.
-fence_open :: proc(line: string) -> (marker: u8, ok: bool) {
-	trimmed := strings.trim_left(line, " ")
-	if strings.has_prefix(trimmed, "```") {
-		return '`', true
-	}
-	if strings.has_prefix(trimmed, "~~~") {
-		return '~', true
-	}
-	return 0, false
-}
-
-// is_fence_line reports whether `line` is a fence using the given marker
-// character (used to detect the closing fence of an open block).
-is_fence_line :: proc(line: string, marker: u8) -> bool {
-	trimmed := strings.trim_left(line, " ")
-	if marker == '~' {
-		return strings.has_prefix(trimmed, "~~~")
-	}
-	return strings.has_prefix(trimmed, "```")
-}
-
-// is_def_line returns true if the line is a definition line: fewer than four
-// columns of leading whitespace, a colon, then whitespace or end-of-line. Four
-// or more columns of indentation is a code block, not a definition.
-is_def_line :: proc(line: string) -> bool {
-	indent := 0
-	n := 0
-	for n < len(line) && (line[n] == ' ' || line[n] == '\t') {
-		indent += line[n] == '\t' ? 4 : 1
-		n += 1
-	}
-	if indent >= 4 {
-		return false
-	}
-	rest := line[n:]
-	if len(rest) < 1 || rest[0] != ':' {
-		return false
-	}
-	if len(rest) == 1 {
-		return true
-	}
-	return rest[1] == ' ' || rest[1] == '\t'
-}
-
-// def_content extracts the definition text from a definition line,
-// stripping the leading colon and surrounding whitespace.
-def_content :: proc(line: string) -> string {
-	trimmed := strings.trim_left(line, " \t")
-	content := trimmed[1:]
-	content = strings.trim_left(content, " \t")
-	return content
-}
-
-// is_blank_line returns true for empty or whitespace-only lines.
-is_blank_line :: proc(line: string) -> bool {
-	return strings.trim_space(line) == ""
-}
-
-// render_inline_md renders a snippet of markdown through cmark and strips
-// the surrounding <p> tags. Result lives in context.temp_allocator.
-render_inline_md :: proc(text: string) -> string {
-	raw := cm.markdown_to_html_from_string(text, {.Unsafe})
-	defer cm.free_string(raw)
-	return strings.clone(strip_p_tags(raw), context.temp_allocator)
-}
-
-// strip_p_tags removes surrounding <p></p> if the HTML is a single paragraph.
-strip_p_tags :: proc(html: string) -> string {
-	s := html
-	if len(s) > 0 && s[len(s) - 1] == '\n' {
-		s = s[:len(s) - 1]
-	}
-	if strings.has_prefix(s, "<p>") && strings.has_suffix(s, "</p>") {
-		return s[3:len(s) - 4]
-	}
-	return s
-}
-
@@ -1,92 +0,0 @@
-#+test
-package markdown
-
-import "core:testing"
-
-// Fixture-driven tests for convert_deflists. Every fixture is exact-matched
-// against its oracle in testdata/deflists/expected/. See
-// testdata/deflists/reference/README.md for how each expected file was decided
-// (cross-checked against Pandoc and Hugo/Goldmark).
-//
-// The fixture inputs are the definition-list examples from
-// https://www.markdownlang.com/extended/definition-lists.html
-//
-// Provenance of each expected file:
-//   01,02,08,09  Pandoc and Hugo agree; oracle = that agreed HTML.
-//   06           Both agree it is NOT a deflist; convert_deflists passes the
-//                markdown through unchanged, so expected = the raw input.
-//   07           CommonMark's <4-space marker rule: we accept the 2-space colon
-//                (Pandoc does; Hugo does not).
-//   05           Block nesting is unsupported; our flatten happens to match
-//                Pandoc's output exactly.
-//   03,04        Block content in definitions is unsupported; expected captures
-//                our current degraded behavior (a <dl> for the term/first line,
-//                the rest falling through as markdown) as a regression guard.
-Deflist_Fixture :: struct {
-	name:     string,
-	input:    string,
-	expected: string,
-}
-
-@(test)
-test_deflist_fixtures :: proc(t: ^testing.T) {
-	cases := []Deflist_Fixture {
-		{
-			"01-basic",
-			#load(#directory + "testdata/deflists/01-basic.md", string),
-			#load(#directory + "testdata/deflists/expected/01-basic.html", string),
-		},
-		{
-			"02-multiple-defs",
-			#load(#directory + "testdata/deflists/02-multiple-defs.md", string),
-			#load(#directory + "testdata/deflists/expected/02-multiple-defs.html", string),
-		},
-		{
-			"03-multiline-def",
-			#load(#directory + "testdata/deflists/03-multiline-def.md", string),
-			#load(#directory + "testdata/deflists/expected/03-multiline-def.html", string),
-		},
-		{
-			"04-markdown-in-def",
-			#load(#directory + "testdata/deflists/04-markdown-in-def.md", string),
-			#load(#directory + "testdata/deflists/expected/04-markdown-in-def.html", string),
-		},
-		{
-			"05-nested",
-			#load(#directory + "testdata/deflists/05-nested.md", string),
-			#load(#directory + "testdata/deflists/expected/05-nested.html", string),
-		},
-		{
-			"06-compact",
-			#load(#directory + "testdata/deflists/06-compact.md", string),
-			#load(#directory + "testdata/deflists/expected/06-compact.html", string),
-		},
-		{
-			"07-extra-space",
-			#load(#directory + "testdata/deflists/07-extra-space.md", string),
-			#load(#directory + "testdata/deflists/expected/07-extra-space.html", string),
-		},
-		{
-			"08-glossary",
-			#load(#directory + "testdata/deflists/08-glossary.md", string),
-			#load(#directory + "testdata/deflists/expected/08-glossary.html", string),
-		},
-		{
-			"09-api-params",
-			#load(#directory + "testdata/deflists/09-api-params.md", string),
-			#load(#directory + "testdata/deflists/expected/09-api-params.html", string),
-		},
-	}
-
-	for c in cases {
-		result := convert_deflists(c.input, context.temp_allocator)
-		testing.expectf(
-			t,
-			result == c.expected,
-			"[%s]\n  expected: %q\n  got:      %q",
-			c.name,
-			c.expected,
-			result,
-		)
-	}
-}
@@ -1,115 +0,0 @@
-#+test
-package markdown
-
-import "core:strings"
-import "core:testing"
-
-@(test)
-test_single_entry :: proc(t: ^testing.T) {
-	input := "term\n\n: definition"
-	result := convert_deflists(input, context.temp_allocator)
-	expected := "<dl><dt>term</dt><dd>definition</dd></dl>"
-	testing.expect_value(t, result, expected)
-}
-
-@(test)
-test_single_entry_no_blank :: proc(t: ^testing.T) {
-	input := "term\n: definition"
-	result := convert_deflists(input, context.temp_allocator)
-	expected := "<dl><dt>term</dt><dd>definition</dd></dl>"
-	testing.expect_value(t, result, expected)
-}
-
-@(test)
-test_indented_variant :: proc(t: ^testing.T) {
-	input := " term\n : definition"
-	result := convert_deflists(input, context.temp_allocator)
-	expected := "<dl><dt>term</dt><dd>definition</dd></dl>"
-	testing.expect_value(t, result, expected)
-}
-
-@(test)
-test_multiple_entries :: proc(t: ^testing.T) {
-	input := "t1\n\n: d1\n\nt2\n\n: d2"
-	result := convert_deflists(input, context.temp_allocator)
-	expected := "<dl><dt>t1</dt><dd>d1</dd><dt>t2</dt><dd>d2</dd></dl>"
-	testing.expect_value(t, result, expected)
-}
-
-@(test)
-test_mixed_indented_and_non_indented :: proc(t: ^testing.T) {
-	input := "t1\n\n: d1\n\n t2\n : d2\n\nt3\n\n: d3"
-	result := convert_deflists(input, context.temp_allocator)
-	expected := "<dl><dt>t1</dt><dd>d1</dd><dt>t2</dt><dd>d2</dd><dt>t3</dt><dd>d3</dd></dl>"
-	testing.expect_value(t, result, expected)
-}
-
-@(test)
-test_inline_markdown_in_term :: proc(t: ^testing.T) {
-	input := "`code`\n\n: def"
-	result := convert_deflists(input, context.temp_allocator)
-	expected := "<dl><dt><code>code</code></dt><dd>def</dd></dl>"
-	testing.expect_value(t, result, expected)
-}
-
-@(test)
-test_inline_markdown_in_definition :: proc(t: ^testing.T) {
-	input := "term\n\n: see [Content](#content) here"
-	result := convert_deflists(input, context.temp_allocator)
-	testing.expect(t, strings.contains(result, `<a href="#content">Content</a>`))
-	testing.expect(t, strings.contains(result, "<dd>see "))
-	testing.expect(t, strings.contains(result, "</dd>"))
-}
-
-@(test)
-test_regular_text_passes_through :: proc(t: ^testing.T) {
-	input := "This is: not a deflist"
-	result := convert_deflists(input, context.temp_allocator)
-	testing.expect_value(t, result, input)
-}
-
-@(test)
-test_colon_inside_paragraph_no_false_positive :: proc(t: ^testing.T) {
-	input := "First paragraph.\n\nSecond paragraph."
-	result := convert_deflists(input, context.temp_allocator)
-	testing.expect_value(t, result, input)
-}
-
-@(test)
-test_deflist_between_paragraphs :: proc(t: ^testing.T) {
-	input := "Before.\n\nterm\n\n: def\n\nAfter."
-	result := convert_deflists(input, context.temp_allocator)
-	testing.expect(t, strings.has_prefix(result, "Before."))
-	testing.expect(t, strings.contains(result, "<dl><dt>term</dt><dd>def</dd></dl>"))
-	testing.expect(t, strings.has_suffix(result, "After."))
-}
-
-@(test)
-test_empty_body :: proc(t: ^testing.T) {
-	result := convert_deflists("", context.temp_allocator)
-	testing.expect_value(t, result, "")
-}
-
-// Regression: a fenced code block following a definition list must stay a
-// separate block. Without a blank line after the emitted </dl>, CommonMark
-// treats the fence as raw HTML continuation of the <dl> block and passes its
-// contents through unescaped (leaking e.g. an example <h1> into the document).
-@(test)
-test_deflist_before_code_fence :: proc(t: ^testing.T) {
-	input := "term\n\n: def\n\n```\n<h1>example</h1>\n```"
-	result := convert_deflists(input, context.temp_allocator)
-	testing.expect(t, strings.contains(result, "</dl>\n\n```"))
-}
-
-@(test)
-test_docs_md_pattern :: proc(t: ^testing.T) {
-	input := "content\n\n: `content` holds your pages.\n\n assets\n : `assets` contains files."
-	result := convert_deflists(input, context.temp_allocator)
-	testing.expect(t, strings.contains(result, "<dl>"))
-	testing.expect(t, strings.contains(result, "<dt>content</dt>"))
-	testing.expect(t, strings.contains(result, "<dt>assets</dt>"))
-	testing.expect(t, strings.contains(result, "<code>content</code>"))
-	testing.expect(t, strings.contains(result, "<code>assets</code>"))
-	testing.expect(t, strings.contains(result, "</dl>"))
-}
-
@@ -30,4 +30,3 @@ test_emoji_skips_invalid_shortcodes :: proc(t: ^testing.T) {
 	testing.expect_value(t, expand_emoji(":Smile:"), ":Smile:")
 	testing.expect_value(t, expand_emoji(": not real :"), ": not real :")
 }
-
@@ -1,5 +1,7 @@
 package markdown

+import cm "vendor:commonmark"
+
 import "core:fmt"
 import "core:strings"

@@ -169,27 +171,27 @@ inject_notes :: proc(html: string, sn_defs, mn_defs: map[string]string) -> strin
 		}

 		// Render definition through cmark for markdown support
-		def_html := render_inline_md(def_text)
+		def_html := cm.markdown_to_html_from_string(def_text, {.Unsafe})
+		def_html = strip_p_tags(def_html)

 		note: string
 		defer delete(note)
 		if is_margin {
-			fmt.sbprintf(
-				&parts,
+			note = fmt.aprintf(
 				`<label for="mn-%s" class="margin-toggle"></label><input type="checkbox" id="mn-%s" class="margin-toggle"><span class="marginnote">%s</span>`,
 				id,
 				id,
 				def_html,
 			)
 		} else {
-			fmt.sbprintf(
-				&parts,
+			note = fmt.aprintf(
 				`<label for="fn-%s" class="margin-toggle sidenote-number"></label><input type="checkbox" id="fn-%s" class="margin-toggle"><span class="sidenote">%s</span>`,
 				id,
 				id,
 				def_html,
 			)
 		}
+		strings.write_string(&parts, note)

 		remaining = remaining[ref_end:]
 	}
@@ -197,98 +199,14 @@ inject_notes :: proc(html: string, sn_defs, mn_defs: map[string]string) -> strin
 	return strings.to_string(parts)
 }

-// inject_footnotes finds [^id] and [*id] references in rendered HTML, numbers
-// them sequentially by order of appearance, and replaces them with <sup> links.
-// Appends a <section class="footnotes"><ol> at the end with definitions.
-// Both sidenote ([^id]) and marginnote ([*id]) references are treated equally.
-inject_footnotes :: proc(html: string, sn_defs, mn_defs: map[string]string) -> string {
-	if len(sn_defs) == 0 && len(mn_defs) == 0 {
-		return html
+// strip_p_tags removes surrounding <p></p> if the HTML is a single paragraph.
+strip_p_tags :: proc(html: string) -> string {
+	s := html
+	if len(s) > 0 && s[len(s) - 1] == '\n' {
+		s = s[:len(s) - 1]
 	}
-
-	parts: strings.Builder
-	strings.builder_init_len(&parts, 0)
-	defer strings.builder_destroy(&parts)
-
-	number_of: map[string]int = make(map[string]int, allocator = context.temp_allocator)
-	ordered_ids: [dynamic]string = make([dynamic]string, 0, allocator = context.temp_allocator)
-	next_num := 1
-
-	remaining := html
-
-	for {
-		sn_pos := strings.index(remaining, "[^")
-		mn_pos := strings.index(remaining, "[*")
-
-		is_margin := mn_pos >= 0 && (sn_pos < 0 || mn_pos < sn_pos)
-		pos := sn_pos
-		if is_margin {
-			pos = mn_pos
-		}
-		if pos < 0 {
-			strings.write_string(&parts, remaining)
-			break
-		}
-
-		strings.write_string(&parts, remaining[:pos])
-
-		close := strings.index(remaining[pos + 2:], "]")
-		if close < 0 {
-			strings.write_string(&parts, remaining[pos:])
-			break
-		}
-
-		id := remaining[pos + 2:pos + 2 + close]
-		ref_end := pos + 2 + close + 1
-
-		defs := sn_defs
-		if is_margin {
-			defs = mn_defs
-		}
-		def_text, found := defs[id]
-		if !found {
-			strings.write_string(&parts, remaining[pos:ref_end])
-			remaining = remaining[ref_end:]
-			continue
-		}
-
-		num, seen := number_of[id]
-		if !seen {
-			num = next_num
-			number_of[id] = num
-			next_num += 1
-			append(&ordered_ids, id)
-		}
-
-		fmt.sbprintf(&parts, `<sup><a href="#fn-%d" id="fnref-%d">%d</a></sup>`, num, num, num)
-
-		remaining = remaining[ref_end:]
+	if strings.has_prefix(s, "<p>") && strings.has_suffix(s, "</p>") {
+		return s[3:len(s) - 4]
 	}
-
-	if len(ordered_ids) > 0 {
-		strings.write_string(&parts, "\n<section class=\"footnotes\">\n<hr>\n<ol>\n")
-		for id in ordered_ids {
-			def_text, ok := sn_defs[id]
-			if !ok {
-				def_text, ok = mn_defs[id]
-			}
-			if !ok do continue
-
-			def_html := render_inline_md(def_text)
-			num := number_of[id]
-
-			fmt.sbprintf(
-				&parts,
-				`<li id="fn-%d">%s <a href="#fnref-%d" class="footnote-backref">↩︎</a></li>` +
-				"\n",
-				num,
-				def_html,
-				num,
-			)
-		}
-		strings.write_string(&parts, "</ol>\n</section>")
-	}
-
-	return strings.to_string(parts)
+	return s
 }
-
@@ -119,102 +119,3 @@ test_inject_notes_missing_ref :: proc(t: ^testing.T) {
 	testing.expect(t, strings.contains(out, "[^missing]"))
 	testing.expect(t, strings.contains(out, "[*missing]"))
 }
-
-@(test)
-test_inject_footnotes_basic :: proc(t: ^testing.T) {
-	html := "Text[^a] end."
-	sn := map[string]string {
-		"a" = "a footnote",
-	}
-	defer delete_map(sn)
-	mn := make(map[string]string)
-	defer delete_map(mn)
-
-	out := inject_footnotes(html, sn, mn)
-
-	testing.expect(t, strings.contains(out, `<sup><a href="#fn-1" id="fnref-1">1</a></sup>`))
-	testing.expect(t, strings.contains(out, `<li id="fn-1">a footnote`))
-	testing.expect(t, strings.contains(out, `class="footnote-backref"`))
-	testing.expect(t, strings.contains(out, `<section class="footnotes">`))
-	testing.expect(t, strings.contains(out, "</section>"))
-}
-
-@(test)
-test_inject_footnotes_numbered_by_appearance :: proc(t: ^testing.T) {
-	html := "Second[^b] then first[^a]."
-	sn := map[string]string {
-		"a" = "def a",
-		"b" = "def b",
-	}
-	defer delete_map(sn)
-	mn := make(map[string]string)
-	defer delete_map(mn)
-
-	out := inject_footnotes(html, sn, mn)
-
-	// b appears first in the text → 1, a → 2
-	testing.expect(t, strings.contains(out, `id="fnref-1">1</a></sup>`))
-	testing.expect(t, strings.contains(out, `id="fnref-2">2</a></sup>`))
-	testing.expect(t, strings.contains(out, `<li id="fn-1">def b`))
-	testing.expect(t, strings.contains(out, `<li id="fn-2">def a`))
-}
-
-@(test)
-test_inject_footnotes_marginnote_treated_same :: proc(t: ^testing.T) {
-	html := "Sidenote[^a] and marginnote[*b]."
-	sn := map[string]string {
-		"a" = "sn def",
-	}
-	defer delete_map(sn)
-	mn := map[string]string {
-		"b" = "mn def",
-	}
-	defer delete_map(mn)
-
-	out := inject_footnotes(html, sn, mn)
-
-	// Both get numbered as regular footnotes
-	testing.expect(t, strings.contains(out, `id="fnref-1">1</a></sup>`))
-	testing.expect(t, strings.contains(out, `id="fnref-2">2</a></sup>`))
-	testing.expect(t, strings.contains(out, `<li id="fn-1">sn def`))
-	testing.expect(t, strings.contains(out, `<li id="fn-2">mn def`))
-}
-
-@(test)
-test_inject_footnotes_no_defs :: proc(t: ^testing.T) {
-	html := "No notes here."
-	sn := make(map[string]string)
-	mn := make(map[string]string)
-	testing.expect(t, inject_footnotes(html, sn, mn) == html)
-}
-
-@(test)
-test_inject_footnotes_missing_def :: proc(t: ^testing.T) {
-	html := "Ref[^missing] end."
-	sn := map[string]string {
-		"other" = "x",
-	}
-	defer delete_map(sn)
-	mn := make(map[string]string)
-	defer delete_map(mn)
-
-	out := inject_footnotes(html, sn, mn)
-
-	testing.expect(t, strings.contains(out, "[^missing]"))
-	testing.expect(t, !strings.contains(out, "<section"))
-}
-
-@(test)
-test_inject_footnotes_inline_markdown :: proc(t: ^testing.T) {
-	html := "Text[^a] end."
-	sn := map[string]string {
-		"a" = "see [link](http://example.com) here",
-	}
-	defer delete_map(sn)
-	mn := make(map[string]string)
-	defer delete_map(mn)
-
-	out := inject_footnotes(html, sn, mn)
-
-	testing.expect(t, strings.contains(out, `<a href="http://example.com">link</a>`))
-}
@@ -90,12 +90,9 @@ write_escaped :: proc(b: ^strings.Builder, s: string) {
 	}
 }

-// unescape_html returns a caller-owned string (free it with delete). When the
-// input has no entities it returns a clone; otherwise it returns the builder's
-// buffer. It must NOT destroy the builder — to_string aliases that buffer, so the
-// returned string owns it.
 unescape_html :: proc(s: string) -> string {
 	sb := strings.builder_make()
+	defer strings.builder_destroy(&sb)

 	start := 0
 	for i in 0 ..< len(s) {
@@ -122,28 +119,22 @@ unescape_html :: proc(s: string) -> string {
 		strings.write_string(&sb, replacement)
 		start = i + semi + 1
 	}
-	if start == 0 do return strings.clone(s)
+	if start == 0 do return s
 	if start < len(s) do strings.write_string(&sb, s[start:])
 	return strings.to_string(sb)
 }

 highlight_block :: proc(code: string, lang: string, file_path: string) -> string {
-	g, ok := ts.grammar(lang)
-	if !ok {
+	gc := ts.load_grammar(lang)
+	if gc == nil {
 		return code
 	}
-	parser := ts.open_parser(g)
-	if parser == nil {
-		return code
-	}
-	defer ts.parser_delete(parser)

 	raw_code := unescape_html(code)
-	defer delete(raw_code)
 	raw_c := strings.clone_to_cstring(raw_code)
 	defer delete(raw_c)

-	tree := ts.parser_parse_string(parser, nil, raw_c, u32(len(raw_code)))
+	tree := ts.parser_parse_string(gc.parser, nil, raw_c, u32(len(raw_code)))
 	if tree == nil {
 		return code
 	}
@@ -165,13 +156,12 @@ highlight_block :: proc(code: string, lang: string, file_path: string) -> string
 		}
 	}

-	if g.query == nil {
+	cursor := gc.cursor
+	if cursor == nil {
 		return code
 	}
-	cursor := ts.query_cursor_new()
-	defer ts.query_cursor_delete(cursor)

-	ts.query_cursor_exec(cursor, g.query, root)
+	ts.query_cursor_exec(cursor, gc.query, root)

 	captures := make([dynamic]Capture, 0, 64, context.temp_allocator)
 	defer delete(captures)
@@ -184,7 +174,7 @@ highlight_block :: proc(code: string, lang: string, file_path: string) -> string
 		}
 		cap := match.captures[capture_idx]
 		name_len: u32
-		name_c := ts.query_capture_name_for_id(g.query, cap.index, &name_len)
+		name_c := ts.query_capture_name_for_id(gc.query, cap.index, &name_len)
 		if name_c == nil {
 			continue
 		}
@@ -1,103 +0,0 @@
-#+test
-package markdown
-
-import ts "../treesitter"
-
-import "core:mem"
-import "core:strings"
-import "core:testing"
-
-// highlight_block returns the input string unchanged (borrowed) when it can't
-// highlight — no grammar, no query, or zero captures — and a freshly-allocated
-// string when it does. Free only the latter; identity (raw_data) tells them
-// apart regardless of content.
-@(private = "file")
-free_highlight :: proc(out, code: string) {
-	if raw_data(out) != raw_data(code) {
-		delete(out)
-	}
-}
-
-// A builtin grammar (css) highlights into hl- spans, preserving the source text.
-@(test)
-test_highlight_block_css :: proc(t: ^testing.T) {
-	code := "a{color:red}"
-	out := highlight_block(code, "css", "test")
-	defer free_highlight(out, code)
-
-	testing.expect(t, strings.contains(out, "<span class=\"hl-"), "css should be highlighted")
-	testing.expect(t, strings.contains(out, "</span>"))
-	testing.expect(t, strings.contains(out, "color"))
-	testing.expect(t, strings.contains(out, "red"))
-}
-
-// The other builtin grammar (html) highlights, and the source's own '<'/'>' are
-// HTML-escaped in the output (so they can't be confused with the wrapping spans).
-@(test)
-test_highlight_block_html_escapes_source :: proc(t: ^testing.T) {
-	code := "<p>hi</p>"
-	out := highlight_block(code, "html", "test")
-	defer free_highlight(out, code)
-
-	testing.expect(t, strings.contains(out, "<span class=\"hl-"), "html should be highlighted")
-	testing.expect(t, strings.contains(out, "hi"))
-	// The literal angle brackets of the source are escaped in the output.
-	testing.expect(t, strings.contains(out, "&lt;"))
-	testing.expect(t, strings.contains(out, "&gt;"))
-}
-
-// A language with no available grammar returns the input verbatim (best-effort:
-// highlighting is optional). The returned string must be the same buffer, so
-// free_highlight leaves it alone.
-@(test)
-test_highlight_block_unknown_lang_passthrough :: proc(t: ^testing.T) {
-	code := "fn main() {}"
-	out := highlight_block(code, "no-such-language", "test")
-	defer free_highlight(out, code)
-
-	testing.expect_value(t, out, code)
-	testing.expect(t, raw_data(out) == raw_data(code), "passthrough must return the input buffer")
-}
-
-// End-to-end: highlight_code finds a fenced code block in rendered HTML and
-// rewrites its body with highlight spans while preserving the <pre><code> shell.
-// Run under a scratch arena (as production does) so the intermediate allocations
-// highlight_code doesn't individually free are reclaimed in one shot.
-@(test)
-test_highlight_code_wraps_css_block :: proc(t: ^testing.T) {
-	arena: mem.Dynamic_Arena
-	mem.dynamic_arena_init(&arena)
-	defer mem.dynamic_arena_destroy(&arena)
-	context.allocator = mem.dynamic_arena_allocator(&arena)
-
-	html := `<pre><code class="language-css">a{color:red}</code></pre>`
-	out := highlight_code(html, "test")
-
-	testing.expect(t, strings.contains(out, `<pre><code class="language-css">`))
-	testing.expect(t, strings.contains(out, "</code></pre>"))
-	testing.expect(
-		t,
-		strings.contains(out, "<span class=\"hl-"),
-		"block body should be highlighted",
-	)
-}
-
-// unescape_html must return a caller-owned allocation the caller frees exactly
-// once. The buggy version frees its builder internally (defer builder_destroy)
-// and returns the freed alias, so the caller's delete becomes a double-free that
-// the test runner's tracking allocator flags.
-@(test)
-test_unescape_html_entities_owned :: proc(t: ^testing.T) {
-	got := unescape_html("&lt;a&gt;")
-	defer delete(got)
-	testing.expect_value(t, got, "<a>")
-}
-
-// Same contract on the no-entity path: the buggy version returns the borrowed
-// input, so freeing it is a bad free (pointer not from the allocator).
-@(test)
-test_unescape_html_passthrough_owned :: proc(t: ^testing.T) {
-	got := unescape_html("plain text")
-	defer delete(got)
-	testing.expect_value(t, got, "plain text")
-}
@@ -3,11 +3,8 @@ package markdown
 import cm "vendor:commonmark"

 import "core:encoding/json"
-import "core:log"
 import "core:strings"

-import diags "../diagnostics"
-
 Extension :: enum {
 	Emoji,
 	Sidenotes,
@@ -15,43 +12,22 @@ Extension :: enum {
 	Highlight,
 	Sections,
 	HeadingIDs,
-	DefLists,
-	Footnotes,
 }

-DEFAULT_EXTENSIONS :: bit_set[Extension]{.Emoji, .Sidenotes, .Alerts, .HeadingIDs, .DefLists}
+DEFAULT_EXTENSIONS :: bit_set[Extension]{.Emoji, .Sidenotes, .Alerts, .HeadingIDs}

-EXTENSION_NAMES :: []string{
-	"emoji", "sidenotes", "alerts", "highlight", "sections",
-	"heading_ids", "deflists", "footnotes",
-}
-
-// Caller is responsible for freeing string
-process :: proc(
-	body: string,
-	ext: bit_set[Extension],
-	file_path: string,
-	allocator := context.allocator,
-) -> string {
+process :: proc(body: string, ext: bit_set[Extension], file_path: string) -> string {
 	side_notes := make(map[string]string)
 	margin_notes := make(map[string]string)
 	clean_body := body
-	if .Sidenotes in ext || .Footnotes in ext {
+	if .Sidenotes in ext {
 		clean_body, side_notes, margin_notes = strip_definitions(body)
 	}
-	if .DefLists in ext {
-		clean_body = convert_deflists(clean_body, allocator)
-	}
-	original_html := cm.markdown_to_html_from_string(clean_body, {.Unsafe})
-	html := strings.clone(original_html, allocator)
-	cm.free_string(original_html)
-
+	html := cm.markdown_to_html_from_string(clean_body, {.Unsafe})
 	if .Emoji in ext {
 		html = expand_emoji(html)
 	}
-	if .Footnotes in ext {
-		html = inject_footnotes(html, side_notes, margin_notes)
-	} else if .Sidenotes in ext {
+	if .Sidenotes in ext {
 		html = inject_notes(html, side_notes, margin_notes)
 	}
 	if .Alerts in ext {
@@ -73,19 +49,20 @@ process :: proc(
 parse_extension_list :: proc(s: string) -> (result: bit_set[Extension]) {
 	for part in strings.split(s, ",", allocator = context.temp_allocator) {
 		name := strings.to_lower(strings.trim_space(part), allocator = context.temp_allocator)
-		e, ok := extension_from_name(name)
-		if !ok {
-			if name != "" {
-				suggestion := diags.suggest_correction(EXTENSION_NAMES, name)
-				if suggestion != "" {
-					log.warnf("unknown extension '%s' (did you mean '%s'?)", name, suggestion)
-				} else {
-					log.warnf("unknown extension '%s'", name)
-				}
-			}
-			continue
+		switch name {
+		case "emoji":
+			result += {.Emoji}
+		case "sidenotes":
+			result += {.Sidenotes}
+		case "alerts":
+			result += {.Alerts}
+		case "highlight":
+			result += {.Highlight}
+		case "sections":
+			result += {.Sections}
+		case "heading_ids":
+			result += {.HeadingIDs}
 		}
-		result += {e}
 	}
 	return result
 }
@@ -93,60 +70,21 @@ parse_extension_list :: proc(s: string) -> (result: bit_set[Extension]) {
 // Given a map[Extension]bool, apply it to ext.
 apply_extension_config :: proc(ext: ^bit_set[Extension], config: json.Object) {
 	for name, val in config {
+		// TODO: Silently discards invalid values.
 		enabled := val.(json.Boolean) or_continue
-
-		e, ok := extension_from_name(name)
-		if !ok {
-			lower := strings.to_lower(name, context.temp_allocator)
-			suggestion := diags.suggest_correction(EXTENSION_NAMES, lower)
-			if suggestion != "" {
-				log.warnf("unknown extension '%s' (did you mean '%s'?)", name, suggestion)
-			} else {
-				log.warnf("unknown extension '%s'", name)
-			}
-			continue
-		}
-
-		if enabled {
-			ext^ += {e}
-		} else {
-			ext^ -= {e}
+		switch name {
+		case "emoji":
+			if enabled {ext^ += {.Emoji}} else {ext^ -= {.Emoji}}
+		case "sidenotes":
+			if enabled {ext^ += {.Sidenotes}} else {ext^ -= {.Sidenotes}}
+		case "alerts":
+			if enabled {ext^ += {.Alerts}} else {ext^ -= {.Alerts}}
+		case "highlight":
+			if enabled {ext^ += {.Highlight}} else {ext^ -= {.Highlight}}
+		case "sections":
+			if enabled {ext^ += {.Sections}} else {ext^ -= {.Sections}}
+		case "heading_ids":
+			if enabled {ext^ += {.HeadingIDs}} else {ext^ -= {.HeadingIDs}}
 		}
 	}
 }
-
-extension_from_name :: proc(name: string) -> (e: Extension, ok: bool) {
-	switch name {
-	case "emoji":
-		e = .Emoji
-		ok = true
-	case "sidenotes":
-		e = .Sidenotes
-	case "alerts":
-		e = .Alerts
-	case "highlight":
-		e = .Highlight
-	case "sections":
-		e = .Sections
-	case "heading_ids":
-		e = .HeadingIDs
-	case "deflists":
-		e = .DefLists
-	case "footnotes":
-		e = .Footnotes
-	case:
-	// Do nothing
-	}
-
-	return e, ok || e != .Emoji
-}
-
-// resolve_extension_conflicts resolves mutually exclusive extensions.
-// Footnotes and Sidenotes share the same [^id] syntax but render differently;
-// if both are enabled (e.g. from defaults + CLI), Footnotes wins.
-resolve_extension_conflicts :: proc(ext: ^bit_set[Extension]) {
-	if .Footnotes in ext^ && .Sidenotes in ext^ {
-		ext^ -= {.Sidenotes}
-	}
-}
-
@@ -1,2 +0,0 @@
-Term
-: Definition content
@@ -1,7 +0,0 @@
-Markdown
-: A lightweight markup language
-: Created by John Gruber in 2004
-
-HTML
-: A standard markup language used to create web pages
-: Used to create web pages
@@ -1,5 +0,0 @@
-Term
-: This is the first line of definition content
-   This is the second line, needs at least one space indentation
-  
-   This is a new paragraph, needs at least one space indentation and a preceding blank line
@@ -1,7 +0,0 @@
-Markdown Syntax
-: **Markdown** supports multiple text formatting:
-  - *Italic* and **Bold**
-  - [Link](https://www.markdownlang.com)
-  - `Inline code`
-  - > Block quote
-  - Lists and other elements
@@ -1,6 +0,0 @@
-Outer Term
-: Outer Definition
-
-   Inner Term
-  : Inner Definition
-  : Another Inner Definition
@@ -1 +0,0 @@
-Term: Definition
@@ -1,2 +0,0 @@
-Term
-  : Definition
@@ -1,11 +0,0 @@
-API
-: An application programming interface, a set of rules that allow different software applications to communicate with each other.
-
-Framework
-: A software library that provides a standardized structure for application development.
-
-Git
-: A distributed version control system used to track changes in a project development process.
-
-IDE
-: An integrated development environment, a software application that includes a code editor and various development tools.
@@ -1,14 +0,0 @@
-user_id
-: **Required** - The unique identifier for the user.
-: Type: `integer`
-
-name
-: **Optional** - The display name for the user.
-: Type: `string`
-: Default: `null`
-
-status
-: **Optional** - The user status.
-: Type: `string`
-: Allowed values: `active`, `inactive`, `suspended`
-: Default: `active`
@@ -1 +0,0 @@
-<dl><dt>Term</dt><dd>Definition content</dd></dl>
@@ -1 +0,0 @@
-<dl><dt>Markdown</dt><dd>A lightweight markup language</dd><dd>Created by John Gruber in 2004</dd><dt>HTML</dt><dd>A standard markup language used to create web pages</dd><dd>Used to create web pages</dd></dl>
@@ -1,3 +0,0 @@
-<dl><dt>Term</dt><dd>This is the first line of definition content</dd></dl>   This is the second line, needs at least one space indentation
-  
-   This is a new paragraph, needs at least one space indentation and a preceding blank line
@@ -1,5 +0,0 @@
-<dl><dt>Markdown Syntax</dt><dd><strong>Markdown</strong> supports multiple text formatting:</dd></dl>  - *Italic* and **Bold**
-  - [Link](https://www.markdownlang.com)
-  - `Inline code`
-  - > Block quote
-  - Lists and other elements
@@ -1 +0,0 @@
-<dl><dt>Outer Term</dt><dd>Outer Definition</dd><dt>Inner Term</dt><dd>Inner Definition</dd><dd>Another Inner Definition</dd></dl>
@@ -1 +0,0 @@
-Term: Definition
@@ -1 +0,0 @@
-<dl><dt>Term</dt><dd>Definition</dd></dl>
@@ -1 +0,0 @@
-<dl><dt>API</dt><dd>An application programming interface, a set of rules that allow different software applications to communicate with each other.</dd><dt>Framework</dt><dd>A software library that provides a standardized structure for application development.</dd><dt>Git</dt><dd>A distributed version control system used to track changes in a project development process.</dd><dt>IDE</dt><dd>An integrated development environment, a software application that includes a code editor and various development tools.</dd></dl>
@@ -1 +0,0 @@
-<dl><dt>user_id</dt><dd><strong>Required</strong> - The unique identifier for the user.</dd><dd>Type: <code>integer</code></dd><dt>name</dt><dd><strong>Optional</strong> - The display name for the user.</dd><dd>Type: <code>string</code></dd><dd>Default: <code>null</code></dd><dt>status</dt><dd><strong>Optional</strong> - The user status.</dd><dd>Type: <code>string</code></dd><dd>Allowed values: <code>active</code>, <code>inactive</code>, <code>suspended</code></dd><dd>Default: <code>active</code></dd></dl>
@@ -1,60 +0,0 @@
-# 3-way behavior comparison: ours vs Pandoc vs Hugo
-
-Captured for later review (e.g. if we ever pursue "Option B" — full CommonMark
-block semantics inside definitions). Compares `convert_deflists` output (Option A,
-inline/flat) against Pandoc (`-f markdown -t html5`) and Hugo/Goldmark for the
-four fixtures where behavior is interesting. Raw renderings live in `pandoc/` and
-`hugo/`; ours is in `../expected/`. Whitespace/formatting differences ignored
-below — only structural differences are described.
-
-## 03-multiline-def — definition with a continuation line + a second paragraph
-*How much multi-line content belongs to the `<dd>`?*
- **Ours:** only the **first line** goes in the `<dd>`. The continuation line and
-  the second paragraph fall out of the `<dl>` as **raw, unrendered markdown**.
- **Pandoc:** folds the continuation line **into** the `<dd>` (lazy continuation),
-  then **ejects** the second paragraph as a sibling `<p>` after `</dl>`.
- **Hugo:** keeps **everything inside** the `<dd>` — continuation line as text,
-  second paragraph as a nested `<p>`.
- All three differ. We capture the least, Pandoc the middle, Hugo the most.
-
-## 04-markdown-in-def — definition containing a bullet list (with a blockquote)
-*What happens to the nested list?*
- **Ours:** intro line becomes the `<dd>`; the **entire bullet list falls out as
-  raw markdown** (unrendered `- …` lines after `</dl>`).
- **Pandoc:** bullets **flattened into the `<dd>`** as inline text — no `<ul>`, the
-  `-` become literal dashes.
- **Hugo:** bullets become a **real nested `<ul>`/`<li>`** (blockquote and all)
-  inside the `<dd>`.
- All three differ; same shape as 03. We drop it, Pandoc absorbs-but-flattens,
-  Hugo builds true structure.
-
-## 05-nested — a definition list nested under another definition
-*Nest or flatten?*
- **Ours:** **flatten** — one flat `<dl>`; "Inner Term" is just another `<dt>`.
- **Pandoc:** **flatten** — structurally **identical to ours**.
- **Hugo:** **nests** a real inner `<dl>` inside the outer `<dd>`.
- Ours == Pandoc; Hugo is the odd one out.
-
-## 07-extra-space — term then a 2-space-indented `: Definition`
-*Is a 2-space colon a definition marker?*
- **Ours:** **yes** → clean `<dl>`.
- **Pandoc:** **yes** → structurally **identical to ours**.
- **Hugo:** **no** → rejects it, emits a plain `<p>`.
- Ours == Pandoc; Hugo is the odd one out.
-
-## Pattern
-
-Two clean groups:
-
- **Marker / flattening decisions (05, 07):** we land **exactly on Pandoc**. These
-  are the calls our CommonMark-marker + flat-inline design makes cleanly; Pandoc
-  agrees, Hugo diverges (nests / stricter marker).
- **Block-content-in-a-definition (03, 04):** we diverge from **both** tools, doing
-  **less** — we emit the `<dl>` for the term/first line and let the rest fall
-  through as raw markdown, rather than absorbing it into the `<dd>` (flattened like
-  Pandoc, or structured like Hugo).
-
-In short: our behavior is **"Pandoc, minus the ability to pull block content into
-a `<dd>`."** Single inline-line definitions (01, 02, 05, 07, 08, 09) match a real
-engine; genuinely multi-block definitions (03, 04) stop at the first line and pass
-the remainder through. Closing that gap is what Option B would entail.
@@ -1,44 +0,0 @@
-# Definition-list test fixtures
-
-Inputs (`../NN-name.md`) are the definition-list examples from
-<https://www.markdownlang.com/extended/definition-lists.html>.
-
-`pandoc/` and `hugo/` hold each input rendered by Pandoc (`-f markdown -t html5`)
-and Hugo (Goldmark with `markup.goldmark.extensions.definitionList = true`).
-They are retained as reference for a possible future "Option B" (full CommonMark
-block semantics inside definitions). They are NOT the test oracle for the current
-implementation.
-
-`../expected/NN.html` is the oracle for `convert_deflists`: our inline fragment
-format (no inter-tag newlines, no trailing newline), derived from the trusted
-tool per the decisions below.
-
-## Decisions (Option A — simple / inline)
-
-Marker whitespace follows CommonMark: `:` with 0–3 spaces of indent is a
-definition; 4+ (tab = 4) is a code block. Definition bodies are inline/flat.
-
-Every fixture has a concrete exact-match oracle in `../expected/NN.html`.
-
-| Fixture | Pandoc | Hugo | Decision | expected/ source |
-|---------|--------|------|----------|------------------|
-| 01-basic | `<dl>` | `<dl>` | support | hugo (agreed) |
-| 02-multiple-defs | multi `<dd>` | multi `<dd>` | support | hugo (agreed) |
-| 06-compact (`Term: Definition`) | `<p>` | `<p>` | NOT a deflist → passthrough | raw input |
-| 07-extra-space (2-space `:`) | `<dl>` | `<p>` | support (accept, CommonMark ≤3-space rule) | hand-written |
-| 08-glossary | `<dl>` | `<dl>` | support | hugo (agreed) |
-| 09-api-params | multi `<dd>` | multi `<dd>` | support | hugo (agreed) |
-| 05-nested | flattens | nested `<dl>` | flatten (matches Pandoc) | snapshot (== pandoc) |
-| 03-multiline-def | ejects 2nd para | nests 2nd para in `<dd>` | degrade | snapshot (current output) |
-| 04-markdown-in-def | flattens list | nested `<ul>`/`<blockquote>` | degrade | snapshot (current output) |
-
-A detailed 3-way behavior comparison (ours vs Pandoc vs Hugo) for the diverging
-fixtures is in `COMPARISON.md`, kept for Option-B review.
-
-03/04/05 require block-level content inside `<dd>` (multi-paragraph, nested
-lists/blockquotes, nested `<dl>`), which the inline preprocessor does not
-produce. 05 flattens exactly as Pandoc does, so its snapshot equals Pandoc's
-output. 03/04 emit a `<dl>` for the term and first line, then let the remaining
-block content fall through as markdown; their `expected/` files snapshot that
-current degraded behavior as a regression guard, not an endorsed rendering.
-Revisit via Option B (full CommonMark block semantics) if real content needs it.
@@ -1,4 +0,0 @@
-<dl>
-<dt>Term</dt>
-<dd>Definition content</dd>
-</dl>
@@ -1,8 +0,0 @@
-<dl>
-<dt>Markdown</dt>
-<dd>A lightweight markup language</dd>
-<dd>Created by John Gruber in 2004</dd>
-<dt>HTML</dt>
-<dd>A standard markup language used to create web pages</dd>
-<dd>Used to create web pages</dd>
-</dl>
@@ -1,7 +0,0 @@
-<dl>
-<dt>Term</dt>
-<dd>This is the first line of definition content
-This is the second line, needs at least one space indentation
-<p>This is a new paragraph, needs at least one space indentation and a preceding blank line</p>
-</dd>
-</dl>
@@ -1,16 +0,0 @@
-<dl>
-<dt>Markdown Syntax</dt>
-<dd><strong>Markdown</strong> supports multiple text formatting:
-<ul>
-<li><em>Italic</em> and <strong>Bold</strong></li>
-<li><a href="https://www.markdownlang.com">Link</a></li>
-<li><code>Inline code</code></li>
-<li>
-<blockquote>
-<p>Block quote</p>
-</blockquote>
-</li>
-<li>Lists and other elements</li>
-</ul>
-</dd>
-</dl>
@@ -1,10 +0,0 @@
-<dl>
-<dt>Outer Term</dt>
-<dd>Outer Definition
-<dl>
-<dt>Inner Term</dt>
-<dd>Inner Definition</dd>
-<dd>Another Inner Definition</dd>
-</dl>
-</dd>
-</dl>
@@ -1 +0,0 @@
-<p>Term: Definition</p>
@@ -1,2 +0,0 @@
-<p>Term
-: Definition</p>
@@ -1,10 +0,0 @@
-<dl>
-<dt>API</dt>
-<dd>An application programming interface, a set of rules that allow different software applications to communicate with each other.</dd>
-<dt>Framework</dt>
-<dd>A software library that provides a standardized structure for application development.</dd>
-<dt>Git</dt>
-<dd>A distributed version control system used to track changes in a project development process.</dd>
-<dt>IDE</dt>
-<dd>An integrated development environment, a software application that includes a code editor and various development tools.</dd>
-</dl>
@@ -1,14 +0,0 @@
-<dl>
-<dt>user_id</dt>
-<dd><strong>Required</strong> - The unique identifier for the user.</dd>
-<dd>Type: <code>integer</code></dd>
-<dt>name</dt>
-<dd><strong>Optional</strong> - The display name for the user.</dd>
-<dd>Type: <code>string</code></dd>
-<dd>Default: <code>null</code></dd>
-<dt>status</dt>
-<dd><strong>Optional</strong> - The user status.</dd>
-<dd>Type: <code>string</code></dd>
-<dd>Allowed values: <code>active</code>, <code>inactive</code>, <code>suspended</code></dd>
-<dd>Default: <code>active</code></dd>
-</dl>
@@ -1,6 +0,0 @@
-<dl>
-<dt>Term</dt>
-<dd>
-Definition content
-</dd>
-</dl>
@@ -1,16 +0,0 @@
-<dl>
-<dt>Markdown</dt>
-<dd>
-A lightweight markup language
-</dd>
-<dd>
-Created by John Gruber in 2004
-</dd>
-<dt>HTML</dt>
-<dd>
-A standard markup language used to create web pages
-</dd>
-<dd>
-Used to create web pages
-</dd>
-</dl>
@@ -1,9 +0,0 @@
-<dl>
-<dt>Term</dt>
-<dd>
-This is the first line of definition content This is the second line,
-needs at least one space indentation
-</dd>
-</dl>
-<p>This is a new paragraph, needs at least one space indentation and a
-preceding blank line</p>
@@ -1,9 +0,0 @@
-<dl>
-<dt>Markdown Syntax</dt>
-<dd>
-<strong>Markdown</strong> supports multiple text formatting: -
-<em>Italic</em> and <strong>Bold</strong> - <a
-href="https://www.markdownlang.com">Link</a> - <code>Inline code</code>
- &gt; Block quote - Lists and other elements
-</dd>
-</dl>
@@ -1,13 +0,0 @@
-<dl>
-<dt>Outer Term</dt>
-<dd>
-Outer Definition
-</dd>
-<dt>Inner Term</dt>
-<dd>
-Inner Definition
-</dd>
-<dd>
-Another Inner Definition
-</dd>
-</dl>
@@ -1 +0,0 @@
-<p>Term: Definition</p>
@@ -1,6 +0,0 @@
-<dl>
-<dt>Term</dt>
-<dd>
-Definition
-</dd>
-</dl>
@@ -1,22 +0,0 @@
-<dl>
-<dt>API</dt>
-<dd>
-An application programming interface, a set of rules that allow
-different software applications to communicate with each other.
-</dd>
-<dt>Framework</dt>
-<dd>
-A software library that provides a standardized structure for
-application development.
-</dd>
-<dt>Git</dt>
-<dd>
-A distributed version control system used to track changes in a project
-development process.
-</dd>
-<dt>IDE</dt>
-<dd>
-An integrated development environment, a software application that
-includes a code editor and various development tools.
-</dd>
-</dl>
@@ -1,33 +0,0 @@
-<dl>
-<dt>user_id</dt>
-<dd>
-<strong>Required</strong> - The unique identifier for the user.
-</dd>
-<dd>
-Type: <code>integer</code>
-</dd>
-<dt>name</dt>
-<dd>
-<strong>Optional</strong> - The display name for the user.
-</dd>
-<dd>
-Type: <code>string</code>
-</dd>
-<dd>
-Default: <code>null</code>
-</dd>
-<dt>status</dt>
-<dd>
-<strong>Optional</strong> - The user status.
-</dd>
-<dd>
-Type: <code>string</code>
-</dd>
-<dd>
-Allowed values: <code>active</code>, <code>inactive</code>,
-<code>suspended</code>
-</dd>
-<dd>
-Default: <code>active</code>
-</dd>
-</dl>
@@ -1,179 +0,0 @@
-package markdown
-
-import "core:log"
-import "core:strings"
-
-// generate_toc scans rendered HTML for <h1>-<h6> tags with id attributes and
-// builds a nested <ul> table of contents. Returns "" if no headings with IDs
-// are found. Must be called after inject_heading_ids.
-generate_toc :: proc(html: string, allocator := context.allocator) -> string {
-	b: strings.Builder
-	strings.builder_init(&b, allocator)
-
-	current_level := 0
-	min_level := 7
-	pos := 0
-
-	for {
-		idx, level, id, text, next_pos := next_heading(html, pos)
-		if level == 0 {
-			break
-		}
-		pos = next_pos
-
-		if level < min_level {
-			min_level = level
-		}
-
-		if current_level == 0 {
-			current_level = level
-			strings.write_string(&b, "<ul>\n")
-		} else if level > current_level {
-			if level - current_level > 1 {
-				log.warnf(
-					"heading level skips from h%d to h%d (\"%s\"); " +
-					"table of contents inserts empty levels to keep valid nesting",
-					current_level,
-					level,
-					text,
-				)
-			}
-			// The first step down nests inside the current, still-open <li>.
-			// Each further skipped level needs its own <li> to host the deeper
-			// <ul>, since a <ul> may only contain <li> — otherwise a jump of
-			// more than one level emits stacked, parentless <ul> tags.
-			strings.write_string(&b, "<ul>\n")
-			current_level += 1
-			for current_level < level {
-				strings.write_string(&b, "<li>\n<ul>\n")
-				current_level += 1
-			}
-		} else if level < current_level {
-			strings.write_string(&b, "</li>\n")
-			for current_level > level {
-				strings.write_string(&b, "</ul>\n</li>\n")
-				current_level -= 1
-			}
-		} else {
-			strings.write_string(&b, "</li>\n")
-		}
-
-		strings.write_string(&b, `<li><a href="#`)
-		strings.write_string(&b, id)
-		strings.write_string(&b, `">`)
-		strings.write_string(&b, text)
-		strings.write_string(&b, `</a>`)
-	}
-
-	if current_level == 0 {
-		return ""
-	}
-
-	strings.write_string(&b, "</li>\n")
-	for current_level > min_level {
-		strings.write_string(&b, "</ul>\n</li>\n")
-		current_level -= 1
-	}
-	strings.write_string(&b, "</ul>\n")
-
-	return strings.to_string(b)
-}
-
-// next_heading finds the next <hN> tag with an id attribute starting from pos.
-// Returns level=0 if none found.
-next_heading :: proc(
-	html: string,
-	start: int,
-) -> (
-	idx: int,
-	level: int,
-	id: string,
-	text: string,
-	next_pos: int,
-) {
-	i := start
-	for i + 3 < len(html) {
-		if html[i] == '<' && html[i + 1] == 'h' {
-			d := html[i + 2]
-			if d >= '1' && d <= '6' {
-				level = int(d - '0')
-				idx = i
-				break
-			}
-		}
-		i += 1
-	}
-
-	if level == 0 {
-		return 0, 0, "", "", len(html)
-	}
-
-	// Find end of opening tag
-	tag_end := strings.index_byte(html[idx:], '>')
-	if tag_end < 0 {
-		return 0, 0, "", "", len(html)
-	}
-	tag_end += idx
-
-	// Find id="..." within the tag
-	tag := html[idx:tag_end + 1]
-	id_pos := strings.index(tag, `id="`)
-	if id_pos < 0 {
-		// No id — skip this heading, continue searching
-		return next_heading(html, tag_end + 1)
-	}
-
-	id_start := idx + id_pos + 4
-	id_end_rel := strings.index_byte(html[id_start:], '"')
-	if id_end_rel < 0 {
-		return 0, 0, "", "", len(html)
-	}
-	id = html[id_start:id_start + id_end_rel]
-
-	// Text between > and </hN>
-	text_start := tag_end + 1
-	close_idx := strings.index(html[text_start:], "</h")
-	if close_idx < 0 {
-		return 0, 0, "", "", len(html)
-	}
-	text_end := text_start + close_idx
-	text = strip_tags(html[text_start:text_end])
-
-	// Find end of closing tag
-	next_pos = text_end + close_tag_len(html, text_end)
-
-	return idx, level, id, text, next_pos
-}
-
-// close_tag_len returns the length of the </hN> tag at pos.
-close_tag_len :: proc(html: string, pos: int) -> int {
-	if pos + 4 > len(html) {
-		return 4
-	}
-	end := strings.index_byte(html[pos:], '>')
-	if end < 0 {
-		return 4
-	}
-	return end + 1
-}
-
-// strip_tags removes HTML tags from a string, leaving only text content.
-strip_tags :: proc(s: string) -> string {
-	b: strings.Builder
-	strings.builder_init(&b, context.temp_allocator)
-
-	i := 0
-	for i < len(s) {
-		if s[i] == '<' {
-			end := strings.index_byte(s[i:], '>')
-			if end >= 0 {
-				i += end + 1
-				continue
-			}
-		}
-		strings.write_byte(&b, s[i])
-		i += 1
-	}
-	return strings.to_string(b)
-}
-
@@ -1,35 +0,0 @@
-#+test
-package markdown
-
-import "core:strings"
-import "core:testing"
-
-@(test)
-test_toc_sequential_nesting :: proc(t: ^testing.T) {
-	html := `<h1 id="a">A</h1><h2 id="b">B</h2><h1 id="c">C</h1>`
-	result := generate_toc(html, context.temp_allocator)
-	testing.expect(t, strings.contains(result, `<a href="#a">A</a>`))
-	testing.expect(t, strings.contains(result, `<a href="#b">B</a>`))
-	testing.expect(t, strings.contains(result, `<a href="#c">C</a>`))
-	// A well-formed list never places a <ul> directly inside a <ul>.
-	testing.expect(t, !strings.contains(result, "<ul>\n<ul>"))
-}
-
-// Regression: a heading level that jumps by more than one (h1 -> h3) must not
-// emit stacked, parentless <ul> tags. Each skipped level is wrapped in an <li>
-// so the list stays valid.
-@(test)
-test_toc_level_jump_stays_valid :: proc(t: ^testing.T) {
-	html := `<h1 id="a">A</h1><h3 id="b">B</h3>`
-	result := generate_toc(html, context.temp_allocator)
-	testing.expect(t, !strings.contains(result, "<ul>\n<ul>"))
-	// The intermediate level is hosted by an <li>.
-	testing.expect(t, strings.contains(result, "<li>\n<ul>"))
-	testing.expect(t, strings.contains(result, `<a href="#b">B</a>`))
-}
-
-@(test)
-test_toc_empty_when_no_headings :: proc(t: ^testing.T) {
-	result := generate_toc("<p>no headings here</p>", context.temp_allocator)
-	testing.expect_value(t, result, "")
-}
@@ -4,6 +4,7 @@ import "core:encoding/json"
 import "core:fmt"
 import "core:log"
 import "core:mem"
+import "core:os"
 import "core:strings"

 DEFAULT_WEIGHT :: 10
@@ -111,32 +112,42 @@ parse_page_menus :: proc(
 	return result
 }

-// build_menus populates site.menus by merging every source into one menu set:
-//   1. Auto-menus (sections + root-level pages)
-//   2. Config menus (thor.json "menus" key), already parsed into site.menus
-//   3. Page frontmatter menus
-// All three are combined per menu name and sorted by weight, then name.
+// build_menus populates site.menus:
+//   1. Config menus (thor.json "menus" key present) — exclusive, preserves array order
+//   2. Auto-menus (sections + root-level pages) + page frontmatter menus — merged, sorted
 //
-// If "menus" is present but empty ({}) it means explicit opt-out: no menus at
-// all, and page frontmatter menus are ignored.
+// If "menus" is present but empty ({}) it means explicit opt-out: no menus.
+// Config menus cannot be mixed with page frontmatter menus (error).
 build_menus :: proc(site: ^Site) {
-	// Explicit opt-out: "menus": {} present but empty → no menus; page
-	// frontmatter menus are ignored (with a warning).
-	// TODO(TODOS.md): silently discarding frontmatter here is harsh; revisit.
-	if site.menus != nil && len(site.menus) == 0 {
+	if site.menus != nil {
+		has_menus := false
 		for page in site.pages {
 			if len(page.menus) > 0 {
-				log.warnf(
-					"menus: config has empty menus but pages have frontmatter menu entries; ignoring page menus",
-				)
+				has_menus = true
 				break
 			}
 		}
+
+		// Already populated from config in site_apply_config
+		if len(site.menus) == 0 {
+			// Explicit opt-out ("menus": {})
+			if has_menus {
+				log.warnf(
+					"menus: config has empty menus but pages have frontmatter menu entries; ignoring page menus",
+				)
+			}
+			return
+		}
+		// Config menus active
+		if has_menus {
+			log.fatalf("menus: cannot mix config menus with frontmatter menus")
+			os.exit(1)
+		}
+		warn_all_duplicate_weights(site)
 		return
 	}

-	// Merge all sources: auto section/root menus, config menus (already in
-	// site.menus if present), then page frontmatter menus.
+	// No config menus — auto-generate, then merge page menus on top
 	collect_auto_menus(site)
 	merge_page_menus(site)
 	warn_all_duplicate_weights(site)
@@ -242,17 +253,8 @@ collect_auto_menus :: proc(site: ^Site) {
 		return
 	}

-	if site.menus == nil {
-		site.menus = make(map[string][]Menu_Entry, alloc)
-	}
-	// Merge into any existing "main" (e.g. config menus) rather than clobbering.
-	if existing, ok := site.menus["main"]; ok {
-		merged := make([dynamic]Menu_Entry, 0, len(existing) + len(entries), alloc)
-		append(&merged, ..existing)
-		append(&merged, ..entries[:])
-		entries = merged
-	}
 	sort_menu_entries(entries[:])
+	site.menus = make(map[string][]Menu_Entry, alloc)
 	site.menus["main"] = entries[:]
 }

@@ -100,6 +100,7 @@ test_menus_nil_input :: proc(t: ^testing.T) {

@(test)
 test_menus_invalid_type :: proc(t: ^testing.T) {
+	context.logger = log.nil_logger()
 	arena: mem.Dynamic_Arena
 	mem.dynamic_arena_init(&arena)
 	defer mem.dynamic_arena_destroy(&arena)
@@ -112,6 +113,7 @@ test_menus_invalid_type :: proc(t: ^testing.T) {

@(test)
 test_menus_array_non_string :: proc(t: ^testing.T) {
+	context.logger = log.nil_logger()
 	arena: mem.Dynamic_Arena
 	mem.dynamic_arena_init(&arena)
 	defer mem.dynamic_arena_destroy(&arena)
@@ -128,6 +130,7 @@ test_menus_array_non_string :: proc(t: ^testing.T) {

@(test)
 test_menus_object_non_object_value :: proc(t: ^testing.T) {
+	context.logger = log.nil_logger()
 	arena: mem.Dynamic_Arena
 	mem.dynamic_arena_init(&arena)
 	defer mem.dynamic_arena_destroy(&arena)
@@ -143,6 +146,7 @@ test_menus_object_non_object_value :: proc(t: ^testing.T) {

@(test)
 test_menus_non_numeric_weight :: proc(t: ^testing.T) {
+	context.logger = log.nil_logger()
 	arena: mem.Dynamic_Arena
 	mem.dynamic_arena_init(&arena)
 	defer mem.dynamic_arena_destroy(&arena)
@@ -157,6 +161,7 @@ test_menus_non_numeric_weight :: proc(t: ^testing.T) {

@(test)
 test_menus_empty_title :: proc(t: ^testing.T) {
+	context.logger = log.nil_logger()
 	arena: mem.Dynamic_Arena
 	mem.dynamic_arena_init(&arena)
 	defer mem.dynamic_arena_destroy(&arena)
@@ -421,71 +426,6 @@ test_auto_menus_no_duplicate_with_frontmatter :: proc(t: ^testing.T) {
 	testing.expect_value(t, main[0].name, "Ideas")
 }

-// --- config + frontmatter + auto merge tests ---
-
-@(test)
-test_build_menus_config_merges_frontmatter :: proc(t: ^testing.T) {
-	site: Site
-	mem.dynamic_arena_init(&site.arena)
-	defer mem.dynamic_arena_destroy(&site.arena)
-	context.allocator = site_allocator(&site)
-
-	// Simulate config menus as site_apply_config would populate them.
-	site.menus = make(map[string][]Menu_Entry, site_allocator(&site))
-	cfg := make([dynamic]Menu_Entry, 0, 1, site_allocator(&site))
-	append(&cfg, Menu_Entry{name = "GitHub", url = "https://example.com", weight = 20})
-	site.menus["main"] = cfg[:]
-
-	// A root-level page opting into "main" via frontmatter.
-	page := make_page("Ideas", "/ideas/")
-	page.menus = parse_page_menus(parse_raw(`"main"`), page, site_allocator(&site))
-
-	site.pages = make(#soa[dynamic]Page, 0, 1, site_allocator(&site))
-	append(&site.pages, page)
-
-	build_menus(&site)
-
-	main, ok := site.menus["main"]
-	testing.expect(t, ok)
-	// GitHub (config, weight 20) + Ideas (frontmatter, nil → DEFAULT_WEIGHT 10).
-	testing.expect(t, len(main) == 2, "expected config + frontmatter merged")
-	testing.expect_value(t, main[0].name, "Ideas")
-	testing.expect_value(t, main[1].name, "GitHub")
-}
-
-@(test)
-test_build_menus_config_merges_auto :: proc(t: ^testing.T) {
-	site: Site
-	mem.dynamic_arena_init(&site.arena)
-	defer mem.dynamic_arena_destroy(&site.arena)
-	context.allocator = site_allocator(&site)
-
-	// Config menu in a different bucket ("footer"); auto-gen owns "main".
-	site.menus = make(map[string][]Menu_Entry, site_allocator(&site))
-	cfg := make([dynamic]Menu_Entry, 0, 1, site_allocator(&site))
-	append(&cfg, Menu_Entry{name = "GitHub", url = "https://example.com", weight = 20})
-	site.menus["footer"] = cfg[:]
-
-	// A section (non-index) page triggers an auto "main" section entry.
-	page := make_page("Post", "/blog/post/")
-	page.section = "blog"
-
-	site.pages = make(#soa[dynamic]Page, 0, 1, site_allocator(&site))
-	append(&site.pages, page)
-
-	build_menus(&site)
-
-	footer, ok_f := site.menus["footer"]
-	testing.expect(t, ok_f, "config footer menu should be preserved")
-	testing.expect(t, len(footer) == 1)
-	testing.expect_value(t, footer[0].name, "GitHub")
-
-	main, ok_m := site.menus["main"]
-	testing.expect(t, ok_m, "auto-generated main menu should exist alongside config")
-	testing.expect(t, len(main) == 1)
-	testing.expect_value(t, main[0].name, "Blog")
-}
-
 // --- warn_duplicate_weights tests ---

@(test)
@@ -12,31 +12,18 @@ Range :: struct {
 	end:   u32,
 }

-// minify_html collapses insignificant whitespace, strips comments, and minifies
-// inline <style> bodies (via minify_css). The caller owns html_parser and
-// css_parser and passes them in; both may be reused across calls. On success it
-// returns (owned string, true) and the caller must free the string. On failure
-// (nil html_parser, parse failure, or syntax errors) it returns ("", false) and
-// the caller should keep the original source unchanged.
-minify_html :: proc(
-	html_parser: ts.Parser,
-	css_parser: ts.Parser,
-	source: string,
-) -> (
-	minified: string,
-	ok: bool,
-) {
-	if html_parser == nil {
-		return minified, ok
+minify_html :: proc(source: string) -> string {
+	gc := ts.ensure_parser("html")
+	if gc == nil {
+		return source
 	}

 	source_c := strings.clone_to_cstring(source)
 	defer delete(source_c)

-	tree := ts.parser_parse_string(html_parser, nil, source_c, u32(len(source)))
+	tree := ts.parser_parse_string(gc.parser, nil, source_c, u32(len(source)))
 	if tree == nil {
-		ok = false
-		return minified, ok
+		return source
 	}
 	defer ts.tree_delete(tree)

@@ -44,24 +31,20 @@ minify_html :: proc(

 	if ts.node_has_error(root) {
 		log.warnf("minify: HTML parse errors, skipping minification")
-		ok = false
-		return minified, ok
+		return source
 	}

 	comments: [dynamic]Range
 	defer delete(comments)
 	preserves: [dynamic]Range
 	defer delete(preserves)
-	styles: [dynamic]Range
-	defer delete(styles)

-	collect_html_ranges(root, source, &comments, &preserves, &styles)
+	collect_html_ranges(root, source, &comments, &preserves)

 	sb := strings.builder_make()

 	ci := 0
 	pi := 0
-	si := 0
 	i := 0
 	last_written: u8 = 0

@@ -78,27 +61,6 @@ minify_html :: proc(
 			continue
 		}

-		if si < len(styles) && u32(i) >= styles[si].start {
-			s := styles[si]
-			seg := source[i:s.end]
-			if minified, mok := minify_css(css_parser, seg); mok {
-				strings.write_string(&sb, minified)
-				if len(minified) > 0 {
-					last_written = minified[len(minified) - 1]
-				}
-				delete(minified)
-			} else {
-				// CSS minify skipped/failed: keep the style body verbatim.
-				strings.write_string(&sb, seg)
-				if len(seg) > 0 {
-					last_written = seg[len(seg) - 1]
-				}
-			}
-			i = int(s.end)
-			si += 1
-			continue
-		}
-
 		if ci < len(comments) && u32(i) >= comments[ci].start {
 			i = int(comments[ci].end)
 			ci += 1
@@ -131,10 +93,7 @@ minify_html :: proc(
 		}
 	}

-
-	minified, ok = strings.to_string(sb), true
-
-	return minified, ok
+	return strings.to_string(sb)
 }

 collect_html_ranges :: proc(
@@ -142,7 +101,6 @@ collect_html_ranges :: proc(
 	source: string,
 	comments: ^[dynamic]Range,
 	preserves: ^[dynamic]Range,
-	styles: ^[dynamic]Range,
 ) {
 	child_count := ts.node_named_child_count(node)
 	for i in 0 ..< child_count {
@@ -154,14 +112,7 @@ collect_html_ranges :: proc(
 				comments,
 				Range{start = ts.node_start_byte(child), end = ts.node_end_byte(child)},
 			)
-		} else if type_str == "style_element" {
-			// Preserve the <style> tags but minify the CSS body (the raw_text
-			// child) via the CSS-aware pass.
-			raw, ok := style_raw_text(child)
-			if ok {
-				append(styles, Range{start = ts.node_start_byte(raw), end = ts.node_end_byte(raw)})
-			}
-		} else if type_str == "script_element" {
+		} else if type_str == "script_element" || type_str == "style_element" {
 			append(
 				preserves,
 				Range{start = ts.node_start_byte(child), end = ts.node_end_byte(child)},
@@ -174,27 +125,14 @@ collect_html_ranges :: proc(
 					Range{start = ts.node_start_byte(child), end = ts.node_end_byte(child)},
 				)
 			} else {
-				collect_html_ranges(child, source, comments, preserves, styles)
+				collect_html_ranges(child, source, comments, preserves)
 			}
 		} else {
-			collect_html_ranges(child, source, comments, preserves, styles)
+			collect_html_ranges(child, source, comments, preserves)
 		}
 	}
 }

-// style_raw_text returns the raw_text (CSS body) child of a style_element.
-// ok is false for an empty <style></style>, which has no raw_text child.
-style_raw_text :: proc(style_element: ts.Node) -> (raw: ts.Node, ok: bool) {
-	child_count := ts.node_named_child_count(style_element)
-	for i in 0 ..< child_count {
-		child := ts.node_named_child(style_element, u32(i))
-		if string(ts.node_type(child)) == "raw_text" {
-			return child, true
-		}
-	}
-	return {}, false
-}
-
 html_tag_name :: proc(element: ts.Node, source: string) -> string {
 	child_count := ts.node_named_child_count(element)
 	for i in 0 ..< child_count {
@@ -230,23 +168,18 @@ is_css_delim :: proc(c: u8) -> bool {
 	return false
 }

-// minify_css collapses insignificant whitespace and strips comments. The caller
-// owns `parser` and passes it in; it may be reused across calls. On success it
-// returns (owned string, true) and the caller must free the string. On failure
-// (nil parser, parse failure, or syntax errors) it returns ("", false) and the
-// caller should keep the original source unchanged.
-minify_css :: proc(parser: ts.Parser, source: string) -> (minified: string, ok: bool) {
-	if parser == nil {
-		return minified, ok
+minify_css :: proc(source: string) -> string {
+	gc := ts.ensure_parser("css")
+	if gc == nil {
+		return source
 	}

 	source_c := strings.clone_to_cstring(source)
 	defer delete(source_c)

-	tree := ts.parser_parse_string(parser, nil, source_c, u32(len(source)))
+	tree := ts.parser_parse_string(gc.parser, nil, source_c, u32(len(source)))
 	if tree == nil {
-		ok = false
-		return minified, ok
+		return source
 	}
 	defer ts.tree_delete(tree)

@@ -254,8 +187,7 @@ minify_css :: proc(parser: ts.Parser, source: string) -> (minified: string, ok:

 	if ts.node_has_error(root) {
 		log.warnf("minify: CSS parse errors, skipping minification")
-		ok = false
-		return minified, ok
+		return source
 	}

 	comments: [dynamic]Range
@@ -292,14 +224,7 @@ minify_css :: proc(parser: ts.Parser, source: string) -> (minified: string, ok:
 				next = source[j]
 			}

-			// Emit a single separating space, but never a leading space
-			// (last_written == 0), a doubled space (last_written == ' ',
-			// which happens when a stripped comment sits between two
-			// whitespace runs), or one adjacent to a delimiter.
-			if last_written != 0 &&
-			   last_written != ' ' &&
-			   !is_css_delim(last_written) &&
-			   !is_css_delim(next) {
+			if !is_css_delim(last_written) && !is_css_delim(next) {
 				strings.write_byte(&sb, ' ')
 				last_written = ' '
 			}
@@ -311,8 +236,7 @@ minify_css :: proc(parser: ts.Parser, source: string) -> (minified: string, ok:
 		}
 	}

-	minified, ok = strings.to_string(sb), true
-	return minified, ok
+	return strings.to_string(sb)
 }

 collect_css_comments :: proc(node: ts.Node, comments: ^[dynamic]Range) {
@@ -1,98 +0,0 @@
-#+test
-package main
-
-import ts "treesitter"
-
-import "core:strings"
-import "core:testing"
-
-// The treesitter registry is thread-safe and self-initializing (grammar() lazily
-// binds it to the heap), and parsers are caller-owned. Each test opens its own
-// parser(s) via ts.open_parser, passes them to minify_*, and frees the returned
-// string — so the multi-threaded test runner needs no shared state or locking.
-
-@(private = "file")
-open_parsers :: proc(t: ^testing.T) -> (html_parser, css_parser: ts.Parser) {
-	hg, hok := ts.grammar("html")
-	cg, cok := ts.grammar("css")
-	testing.expect(t, hok, "html grammar should load")
-	testing.expect(t, cok, "css grammar should load")
-	return ts.open_parser(hg), ts.open_parser(cg)
-}
-
-@(private = "file")
-expect_css :: proc(t: ^testing.T, parser: ts.Parser, input, want: string) {
-	got, ok := minify_css(parser, input)
-	testing.expect(t, ok)
-	defer delete(got)
-	testing.expect_value(t, got, want)
-}
-
-// --- minify_html: inline <style> bodies get CSS-minified ---
-
-@(test)
-test_minify_html_inline_style :: proc(t: ^testing.T) {
-	hp, cp := open_parsers(t)
-	defer ts.parser_delete(hp)
-	defer ts.parser_delete(cp)
-
-	input := `<html><head><style>
-body {
-  color:  red;
-}
-</style></head><body><p>hi</p></body></html>`
-	result, ok := minify_html(hp, cp, input)
-	testing.expect(t, ok)
-	defer delete(result)
-
-	// The CSS body is collapsed by minify_css (delimiters lose surrounding
-	// whitespace)...
-	testing.expect(t, strings.contains(result, "body{color:red;}"))
-	// ...while the <style> tags themselves survive.
-	testing.expect(t, strings.contains(result, "<style>"))
-	testing.expect(t, strings.contains(result, "</style>"))
-}
-
-@(test)
-test_minify_html_empty_style :: proc(t: ^testing.T) {
-	hp, cp := open_parsers(t)
-	defer ts.parser_delete(hp)
-	defer ts.parser_delete(cp)
-
-	// An empty <style></style> has no raw_text child; must not crash.
-	input := `<html><head><style></style></head><body></body></html>`
-	result, ok := minify_html(hp, cp, input)
-	testing.expect(t, ok)
-	defer delete(result)
-	testing.expect(t, strings.contains(result, "<style></style>"))
-}
-
-// --- minify_css: whitespace around stripped comments / leading edge ---
-
-@(test)
-test_minify_css_leading_and_comment :: proc(t: ^testing.T) {
-	cg, cok := ts.grammar("css")
-	testing.expect(t, cok, "css grammar should load")
-	cp := ts.open_parser(cg)
-	defer ts.parser_delete(cp)
-
-	// Leading whitespace and a comment between two whitespace runs must not
-	// leave a leading space or a doubled space.
-	expect_css(t, cp, "\n/* c */\na {}", "a{}")
-	expect_css(t, cp, "  a {}", "a{}")
-	expect_css(t, cp, "a {}\n/* trailing */", "a{}")
-}
-
-@(test)
-test_minify_html_script_preserved :: proc(t: ^testing.T) {
-	hp, cp := open_parsers(t)
-	defer ts.parser_delete(hp)
-	defer ts.parser_delete(cp)
-
-	// <script> stays byte-for-byte (whitespace-significant JS).
-	input := "<html><body><script>\nconst s = \"a    b\";\n</script></body></html>"
-	result, ok := minify_html(hp, cp, input)
-	testing.expect(t, ok)
-	defer delete(result)
-	testing.expect(t, strings.contains(result, "const s = \"a    b\";"))
-}
@@ -4,4 +4,3 @@
 - [ ] Simplify `trim_standalone_whitespace` — four temp arrays (`li_buf`, `ri_buf`, `left_done`, `right_done`) could potentially be reduced to two by merging the done-tracking into the index buffers using sentinel values (e.g. `-2` = already trimmed).
 - [ ] Adjacent standalone tags don't capture indent for the second tag (e.g. `  {{>a}}\n  {{>b}}` — `{{>b}}` gets `indent=""` because its left text was consumed by `{{>a}}`'s right-trim). Blocks recover via `deindent_blocks`, but partials/parents have no fallback.
 - [ ] lazy load partials ? (with mutex)
- [ ] Support changing delimiters (Set Delimiter tag `{{=<% %>=}}`). Delimiters `{{`/`}}`/`{{{`/`}}}` are hardcoded in [tokenizer.odin](./tokenizer.odin); needs an `=` sigil case, mutable open/close delimiters threaded through `tokenize`, and wiring up `spec/specs/delimiters.json` in `spec_test.odin` (fixture present but not run).
@@ -137,22 +137,6 @@ resolve_name :: proc(name: string, ctx: []any) -> any {
 	return result
 }

-collect_map_keys :: proc(container: any, allocator := context.temp_allocator) -> []string {
-	val, info := base_value(container)
-	if info == nil do return nil
-	if _, ok := info.variant.(runtime.Type_Info_Map); !ok {
-		return nil
-	}
-	out := make([dynamic]string, 0, 4, allocator)
-	it := 0
-	for {
-		key, _ := reflect.iterate_map(val, &it) or_break
-		key_str := key.(string) or_continue
-		append(&out, key_str)
-	}
-	return out[:]
-}
-
 // is_truthy checks mustache truthiness.
 is_truthy :: proc(a: any) -> bool {
 	if a == nil {
@@ -175,6 +159,32 @@ is_truthy :: proc(a: any) -> bool {
 	}
 }

+call_interp_lambda :: proc(val: any) -> (result: string, ok: bool) {
+	switch v in val {
+	case proc() -> string:
+		return v(), true
+	case proc() -> int:
+		return fmt.tprintf("%d", v()), true
+	case proc() -> bool:
+		return "true" if v() else "false", true
+	case:
+		return "", false
+	}
+}
+
+call_section_lambda :: proc(val: any, text: string) -> (result: string, ok: bool) {
+	switch v in val {
+	case proc(_: string) -> string:
+		return v(text), true
+	case proc(_: string) -> int:
+		return fmt.tprintf("%d", v(text)), true
+	case proc(_: string) -> bool:
+		return "true" if v(text) else "false", true
+	case:
+		return "", false
+	}
+}
+
 // list_info returns element type info, count, and data pointer for a list value.
 // Returns elem_info=nil if the value is not a list.
 list_info :: proc(a: any) -> (elem_info: ^runtime.Type_Info, count: int, data: rawptr) {
@@ -292,4 +302,3 @@ write_value :: proc(b: ^strings.Builder, a: any, escape: bool) {
 		strings.write_string(b, s[start:])
 	}
 }
-
@@ -1,149 +0,0 @@
-package mustache
-
-import "core:os"
-import "core:path/filepath"
-import "core:strings"
-
-Diag_Case :: struct {
-	name:     string,
-	enable:   bool,
-	should:   string, // "error" | "warn" | "pass"
-	expected: string, // exact formatted diagnostic output
-	input:    string,
-}
-
-load_diag_cases :: proc(path: string, allocator := context.temp_allocator) -> []Diag_Case {
-	dirs := [?]string{#directory, path}
-	full_path, _ := filepath.join(dirs[:], allocator)
-	raw_data, err := os.read_entire_file(full_path, allocator)
-	if err != nil {
-		return {}
-	}
-
-	content := string(raw_data)
-	lines := strings.split(content, "\n", allocator)
-
-	cases := make([dynamic]Diag_Case, 0, 16, allocator)
-
-	current: Diag_Case
-	has_current := false
-
-	// Multiline block state
-	in_block := false
-	block_key: string
-	block_indent := -1
-	block_sb: strings.Builder
-
-	for line in lines {
-		raw := strings.trim_right(line, "\r")
-
-		if in_block {
-			// Check if this line belongs to the multiline block
-			is_indented := len(raw) == 0 || raw[0] == ' ' || raw[0] == '\t'
-			if is_indented {
-				if block_indent < 0 && len(raw) > 0 {
-					// Determine indentation from first content line
-					block_indent = 0
-					for block_indent < len(raw) && raw[block_indent] == ' ' {
-						block_indent += 1
-					}
-				}
-				// Strip block indentation
-				stripped := raw
-				if block_indent > 0 && len(stripped) >= block_indent {
-					stripped = stripped[block_indent:]
-				}
-				if strings.builder_len(block_sb) > 0 {
-					strings.write_string(&block_sb, "\n")
-				}
-				strings.write_string(&block_sb, stripped)
-				continue
-			} else {
-				// End of multiline block
-				in_block = false
-				strings.write_string(&block_sb, "\n") // YAML | trailing newline
-				value := strings.clone(strings.to_string(block_sb), allocator)
-				strings.builder_destroy(&block_sb)
-				set_field(&current, block_key, value)
-				block_indent = -1
-				// Fall through to process this line
-			}
-		}
-
-		trimmed := strings.trim_space(raw)
-
-		// Entry separator
-		if trimmed == "---" {
-			if has_current {
-				append(&cases, current)
-				current = {}
-				has_current = false
-			}
-			continue
-		}
-
-		// Skip blank lines outside blocks
-		if trimmed == "" {
-			continue
-		}
-
-		// Multiline block start: "key: |"
-		if strings.has_suffix(trimmed, ": |") {
-			colon := strings.last_index(trimmed, ":")
-			if colon > 0 {
-				block_key = strings.trim_space(trimmed[:colon])
-				strings.builder_init(&block_sb, allocator)
-				in_block = true
-				has_current = true
-				continue
-			}
-		}
-
-		// Simple key: value
-		colon := strings.index(trimmed, ":")
-		if colon > 0 {
-			key := strings.trim_space(trimmed[:colon])
-			value := strings.trim_space(trimmed[colon + 1:])
-			// Strip matching outer quotes
-			if len(value) >= 2 {
-				if (value[0] == '"' && value[len(value) - 1] == '"') ||
-				   (value[0] == '\'' && value[len(value) - 1] == '\'') {
-					value = value[1:len(value) - 1]
-				}
-			}
-			set_field(&current, key, value)
-			has_current = true
-		}
-	}
-
-	// Flush trailing multiline block
-	if in_block {
-		strings.write_string(&block_sb, "\n") // YAML | trailing newline
-		value := strings.clone(strings.to_string(block_sb), allocator)
-		strings.builder_destroy(&block_sb)
-		set_field(&current, block_key, value)
-	}
-
-	// Flush last entry
-	if has_current {
-		append(&cases, current)
-	}
-
-	return cases[:]
-}
-
-set_field :: proc(c: ^Diag_Case, key, value: string) {
-	switch key {
-	case "name":
-		c.name = value
-	case "enable":
-		c.enable = value == "true"
-	case "should":
-		c.should = value
-	case "expected":
-		c.expected = value
-	case "input":
-		c.input = value
-	}
-}
-
@@ -1,6 +1,61 @@
 package mustache

-import diags "../diagnostics"
+import "core:fmt"
+import "core:os"
+import "core:strings"
+import "core:terminal/ansi"
+import "core:unicode/utf8"
+
+// line_col returns the 1-indexed line and column for a byte offset in source.
+// Newlines ('\n') separate lines; '\r' is treated as part of '\r\n'. Column is
+// counted in bytes from the start of the line.
+line_col :: proc(source: string, pos_in: int) -> (line: int, col: int) {
+	pos := pos_in
+	if pos < 0 {
+		return 1, 1
+	}
+	if pos > len(source) {
+		pos = len(source)
+	}
+	line = 1
+	col = 1
+	for i := 0; i < pos; i += 1 {
+		if source[i] == '\n' {
+			line += 1
+			col = 1
+		} else {
+			col += 1
+		}
+	}
+	return
+}
+
+// line_text returns the Nth (1-indexed) line of source, without the trailing
+// newline. Returns "" if line is out of range.
+line_text :: proc(source: string, line: int) -> string {
+	if line < 1 {
+		return ""
+	}
+	current := 1
+	start := 0
+	for i := 0; i < len(source); i += 1 {
+		if current == line {
+			end := i
+			for end < len(source) && source[end] != '\n' {
+				end += 1
+			}
+			return source[start:end]
+		}
+		if source[i] == '\n' {
+			current += 1
+			start = i + 1
+		}
+	}
+	if current == line {
+		return source[start:]
+	}
+	return ""
+}

 // context_extent returns the byte offset of the start of the line containing
 // pos, plus the byte offsets of the start and end of the mustache tag at pos.
@@ -60,11 +115,68 @@ context_extent :: proc(
 	return line_start, pos, pos + 1
 }

-// format_tag_error is a mustache-aware wrapper around diags.format_error.
-// It uses context_extent to compute the caret underline for `{{ }}` tags,
-// falling back to a single-rune underline when pos is not inside a tag.
-// When `span > 0`, the caret extends from pos to pos + span.
-format_tag_error :: proc(
+// should_colorize returns true if stderr is a TTY and color output is wanted.
+should_colorize :: proc() -> bool {
+	return os.is_tty(os.stderr)
+}
+
+// count_lines returns the number of '\n'-separated lines in source.
+// A trailing newline does not add an extra line.
+count_lines :: proc(source: string) -> int {
+	if len(source) == 0 {
+		return 1
+	}
+	n := 1
+	for c in source {
+		if c == '\n' {
+			n += 1
+		}
+	}
+	// Drop phantom last line if source ends with '\n'.
+	if len(source) > 0 && source[len(source) - 1] == '\n' {
+		n -= 1
+	}
+	return n
+}
+
+// digit_count returns the number of decimal digits in n (min 1).
+digit_count :: proc(n: int) -> int {
+	if n <= 0 {
+		return 1
+	}
+	c := 0
+	x := n
+	for x > 0 {
+		c += 1
+		x /= 10
+	}
+	return c
+}
+
+// display_width returns the number of terminal cells `s` occupies.
+// For ASCII this is byte length; for UTF-8 we count runes (combining
+// marks and wide CJK chars are still approximate).
+display_width :: proc(s: string) -> int {
+	return utf8.rune_count_in_string(s)
+}
+
+// format_error produces a rust-style multi-line diagnostic string.
+//
+//   <msg>
+//    --> <path>:<line>:<col>
+//     |
+//  N | <source line N-2>
+//  N | <source line N-1>
+//  N | <source line N — the error line>
+//     |     ^^^^^^^^^^^ <hint>
+//  N | <source line N+1>
+//  N | <source line N+2>
+//     |
+//
+// `context_before`/`context_after` lines of context are shown around the
+// error line. Line numbers are right-aligned to the width of the largest
+// line number shown.
+format_error :: proc(
 	path: string,
 	source: string,
 	pos: int,
@@ -75,25 +187,135 @@ format_tag_error :: proc(
 	colorize: bool = false,
 	span: int = 0,
 ) -> string {
-	caret_start := pos
-	caret_end := pos + 1
-	if span > 0 {
-		caret_end = pos + span
-	} else {
-		_, ts, te := context_extent(source, pos)
-		caret_start = ts
-		caret_end = te
+	line, col := line_col(source, pos)
+	total_lines := count_lines(source)
+
+	start_line := line - context_before
+	if start_line < 1 {
+		start_line = 1
 	}
-	return diags.format_error(
-		path, source, pos, msg, hint,
-		context_before, context_after,
-		colorize,
-		caret_start, caret_end,
-	)
+	end_line := line + context_after
+	if end_line > total_lines {
+		end_line = total_lines
+	}
+
+	// Width of the line-number column (right-align).
+	width := digit_count(end_line)
+	if width < 1 {
+		width = 1
+	}
+
+	sb := strings.builder_make(context.temp_allocator)
+	defer strings.builder_destroy(&sb)
+
+	color := colorize
+	red, faint, reset := "", "", ""
+	if color {
+		red = ansi.CSI + ansi.FG_RED + ansi.SGR
+		faint = ansi.CSI + ansi.FAINT + ansi.SGR
+		reset = ansi.CSI + ansi.RESET + ansi.SGR
+	}
+
+	// Header line: message.
+	strings.write_string(&sb, msg)
+	strings.write_byte(&sb, '\n')
+
+	// Location line: "  --> path:line:col" (width spaces + arrow).
+	strings.write_string(&sb, faint)
+	for _ in 0 ..< width {
+		strings.write_byte(&sb, ' ')
+	}
+	strings.write_string(&sb, "--> ")
+	strings.write_string(&sb, reset)
+	strings.write_string(&sb, fmt.tprintf("%s:%d:%d\n", path, line, col))
+
+	// Top gutter line.
+	write_gutter(&sb, width, faint, reset)
+
+	// Caret extent for the error line.
+	line_start, token_start, token_end := context_extent(source, pos)
+	if span > 0 {
+		token_start = pos
+		token_end = pos + span
+	}
+	caret_start_col := token_start - line_start + 1
+	caret_end_col := token_end - line_start + 1
+	if caret_end_col <= caret_start_col {
+		caret_end_col = caret_start_col + 1
+	}
+
+	// Context lines.
+	for n in start_line ..= end_line {
+		// Line number (right-aligned, faint).
+		num_str := fmt.tprintf("%d", n)
+		strings.write_string(&sb, faint)
+		for _ in 0 ..< width - len(num_str) {
+			strings.write_byte(&sb, ' ')
+		}
+		strings.write_string(&sb, num_str)
+		strings.write_string(&sb, " | ")
+		strings.write_string(&sb, reset)
+
+		strings.write_string(&sb, line_text(source, n))
+		strings.write_byte(&sb, '\n')
+
+		// After the error line, emit the caret row.
+		if n == line {
+			strings.write_string(&sb, faint)
+			for _ in 0 ..< width + 1 {
+				strings.write_byte(&sb, ' ')
+			}
+			strings.write_string(&sb, "| ")
+			strings.write_string(&sb, reset)
+
+			for _ in 1 ..< caret_start_col {
+				strings.write_byte(&sb, ' ')
+			}
+			if color {
+				strings.write_string(&sb, red)
+			}
+			for _ in 0 ..< caret_end_col - caret_start_col {
+				strings.write_byte(&sb, '^')
+			}
+			if color {
+				strings.write_string(&sb, reset)
+			}
+			if hint != "" {
+				strings.write_byte(&sb, ' ')
+				if color {
+					strings.write_string(&sb, faint)
+				}
+				strings.write_string(&sb, hint)
+				if color {
+					strings.write_string(&sb, reset)
+				}
+			}
+			strings.write_byte(&sb, '\n')
+		}
+	}
+
+	// Trailing gutter line for visual closure.
+	write_gutter(&sb, width, faint, reset)
+
+	return strings.to_string(sb)
+}
+
+// write_gutter emits a faint pipe-only gutter line: `<width+1 spaces> |`.
+write_gutter :: proc(sb: ^strings.Builder, width: int, faint: string, reset: string) {
+	strings.write_string(sb, faint)
+	for _ in 0 ..< width + 1 {
+		strings.write_byte(sb, ' ')
+	}
+	strings.write_string(sb, "|")
+	strings.write_string(sb, reset)
+	strings.write_byte(sb, '\n')
 }

 // format_render_error produces a diagnostic for an Error value using the
 // template's path and source for context. Returns "" for nil errors.
+// If the error carries its own source/path (from tag_error), those are used
+// instead of the passed-in template — this ensures errors inside partials
+// point at the correct file.
 format_render_error :: proc(err: Error, tmpl: Template, colorize: bool = false) -> string {
 	if err == nil {
 		return ""
@@ -104,22 +326,5 @@ format_render_error :: proc(err: Error, tmpl: Template, colorize: bool = false)
 	if path == "" {
 		path = "<input>"
 	}
-
-	caret_start := b.pos
-	caret_end := b.pos + 1
-	if b.span > 0 {
-		caret_end = b.pos + b.span
-	} else {
-		_, ts, te := context_extent(source, b.pos)
-		caret_start = ts
-		caret_end = te
-	}
-
-	return diags.format_error(
-		path, source, b.pos, b.msg,
-		hint = b.hint,
-		colorize = colorize,
-		caret_start = caret_start,
-		caret_end = caret_end,
-	)
+	return format_error(path, source, b.pos, b.msg, hint = b.hint, colorize = colorize, span = b.span)
 }
@@ -1,22 +1,100 @@
 #+test
 package mustache

-import "core:encoding/json"
 import "core:fmt"
-import "core:log"
-import "core:os"
-import "core:path/filepath"
 import "core:strings"
 import "core:testing"

 // ---------------------------------------------------------------------------
-// format_tag_error — golden output tests
+// line_col / line_text / count_lines / digit_count — primitive helpers
+// ---------------------------------------------------------------------------
+
+@(test)
+test_line_col_basic :: proc(t: ^testing.T) {
+	src := "abc\ndef\nghi"
+	cases := [?]struct {
+		pos:  int,
+		line: int,
+		col:  int,
+	}{{0, 1, 1}, {2, 1, 3}, {3, 1, 4}, {4, 2, 1}, {6, 2, 3}, {7, 2, 4}, {8, 3, 1}}
+	for c in cases {
+		l, col := line_col(src, c.pos)
+		testing.expect(t, l == c.line, fmt.tprintf("pos %d: line %d, want %d", c.pos, l, c.line))
+		testing.expect(t, col == c.col, fmt.tprintf("pos %d: col %d, want %d", c.pos, col, c.col))
+	}
+}
+
+@(test)
+test_line_col_empty :: proc(t: ^testing.T) {
+	l, col := line_col("", 0)
+	testing.expect_value(t, l, 1)
+	testing.expect_value(t, col, 1)
+}
+
+@(test)
+test_line_col_negative :: proc(t: ^testing.T) {
+	l, col := line_col("abc", -1)
+	testing.expect_value(t, l, 1)
+	testing.expect_value(t, col, 1)
+}
+
+@(test)
+test_line_col_past_end :: proc(t: ^testing.T) {
+	l, col := line_col("abc", 100)
+	testing.expect_value(t, l, 1)
+	testing.expect_value(t, col, 4)
+}
+
+@(test)
+test_line_text_first :: proc(t: ^testing.T) {
+	src := "first\nsecond\nthird"
+	testing.expect_value(t, line_text(src, 1), "first")
+	testing.expect_value(t, line_text(src, 2), "second")
+	testing.expect_value(t, line_text(src, 3), "third")
+}
+
+@(test)
+test_line_text_trailing_newline :: proc(t: ^testing.T) {
+	src := "first\nsecond\n"
+	testing.expect_value(t, line_text(src, 1), "first")
+	testing.expect_value(t, line_text(src, 2), "second")
+	testing.expect_value(t, line_text(src, 3), "")
+}
+
+@(test)
+test_line_text_out_of_range :: proc(t: ^testing.T) {
+	testing.expect_value(t, line_text("abc", 5), "")
+	testing.expect_value(t, line_text("abc", 0), "")
+}
+
+@(test)
+test_count_lines :: proc(t: ^testing.T) {
+	testing.expect_value(t, count_lines(""), 1)
+	testing.expect_value(t, count_lines("abc"), 1)
+	testing.expect_value(t, count_lines("a\nb"), 2)
+	testing.expect_value(t, count_lines("a\nb\n"), 2)
+	testing.expect_value(t, count_lines("a\nb\nc"), 3)
+}
+
+@(test)
+test_digit_count :: proc(t: ^testing.T) {
+	testing.expect_value(t, digit_count(0), 1)
+	testing.expect_value(t, digit_count(1), 1)
+	testing.expect_value(t, digit_count(9), 1)
+	testing.expect_value(t, digit_count(10), 2)
+	testing.expect_value(t, digit_count(99), 2)
+	testing.expect_value(t, digit_count(100), 3)
+	testing.expect_value(t, digit_count(-5), 1)
+}
+
+// ---------------------------------------------------------------------------
+// format_error — golden output tests
 // ---------------------------------------------------------------------------

@(test)
 test_format_error_basic :: proc(t: ^testing.T) {
 	src := "line 1\nline 2\n{{bad}}\nline 4\nline 5"
-	out := format_tag_error("p.html", src, 14, "unknown key 'bad'", "", colorize = false)
+	out := format_error("p.html", src, 14, "unknown key 'bad'", "", colorize = false)
 	expected := `unknown key 'bad'
 --> p.html:3:1
  |
@@ -34,7 +112,7 @@ test_format_error_basic :: proc(t: ^testing.T) {
@(test)
 test_format_error_with_hint :: proc(t: ^testing.T) {
 	src := "line 1\nline 2\n{{titel}}\nline 4\nline 5"
-	out := format_tag_error(
+	out := format_error(
 		"post.html",
 		src,
 		14,
@@ -59,7 +137,7 @@ test_format_error_with_hint :: proc(t: ^testing.T) {
@(test)
 test_format_error_no_hint_omits_trailing_space :: proc(t: ^testing.T) {
 	src := "{{bad}}"
-	out := format_tag_error("p.html", src, 0, "msg", "", colorize = false)
+	out := format_error("p.html", src, 0, "msg", "", colorize = false)
 	// Caret line ends immediately after the carets — no trailing space.
 	testing.expect(t, strings.contains(out, "^^^^^^^\n"), out)
 	testing.expect(t, !strings.contains(out, "^^^^^^^ \n"), out)
@@ -72,7 +150,7 @@ test_format_error_no_hint_omits_trailing_space :: proc(t: ^testing.T) {
@(test)
 test_format_error_first_line_only_after_context :: proc(t: ^testing.T) {
 	src := "{{bad}}\nline 2\nline 3\nline 4\nline 5"
-	out := format_tag_error("p.html", src, 0, "msg", "", colorize = false)
+	out := format_error("p.html", src, 0, "msg", "", colorize = false)
 	expected := `msg
 --> p.html:1:1
  |
@@ -88,7 +166,7 @@ test_format_error_first_line_only_after_context :: proc(t: ^testing.T) {
@(test)
 test_format_error_last_line_only_before_context :: proc(t: ^testing.T) {
 	src := "line 1\nline 2\nline 3\nline 4\n{{bad}}"
-	out := format_tag_error("p.html", src, 28, "msg", "", colorize = false)
+	out := format_error("p.html", src, 28, "msg", "", colorize = false)
 	expected := `msg
 --> p.html:5:1
  |
@@ -104,7 +182,7 @@ test_format_error_last_line_only_before_context :: proc(t: ^testing.T) {
@(test)
 test_format_error_short_source_clamped :: proc(t: ^testing.T) {
 	src := "x\n{{bad}}\ny"
-	out := format_tag_error("p.html", src, 2, "msg", "", colorize = false)
+	out := format_error("p.html", src, 2, "msg", "", colorize = false)
 	expected := `msg
 --> p.html:2:1
  |
@@ -120,7 +198,7 @@ test_format_error_short_source_clamped :: proc(t: ^testing.T) {
@(test)
 test_format_error_single_line_source :: proc(t: ^testing.T) {
 	src := "{{bad}}"
-	out := format_tag_error("p.html", src, 0, "msg", "", colorize = false)
+	out := format_error("p.html", src, 0, "msg", "", colorize = false)
 	expected := `msg
 --> p.html:1:1
  |
@@ -136,7 +214,7 @@ test_format_error_two_digit_line_numbers :: proc(t: ^testing.T) {
 	// 12-line source; error on line 9. end_line=11 → width=2.
 	src := "l01\nl02\nl03\nl04\nl05\nl06\nl07\nl08\n{{bad}}\nl10\nl11\nl12"
 	// Position of `{{bad}}`: 8 lines of "l0N\n" = 8*4 = 32 bytes.
-	out := format_tag_error("p.html", src, 32, "msg", "", colorize = false)
+	out := format_error("p.html", src, 32, "msg", "", colorize = false)
 	testing.expect(t, strings.contains(out, "  --> p.html:9:1\n"), out)
 	testing.expect(t, strings.contains(out, "   |\n"), out)
 	testing.expect(t, strings.contains(out, " 7 | l07\n"), out)
@@ -163,7 +241,7 @@ test_format_error_three_digit_line_numbers :: proc(t: ^testing.T) {
 		pos += len(parts[i - 1]) + 1
 	}

-	out := format_tag_error("p.html", src, pos, "msg", "", colorize = false)
+	out := format_error("p.html", src, pos, "msg", "", colorize = false)
 	testing.expect(t, strings.contains(out, "   --> p.html:100:1\n"), out)
 	testing.expect(t, strings.contains(out, "    |\n"), out)
 	testing.expect(t, strings.contains(out, " 98 | l098\n"), out)
@@ -178,7 +256,7 @@ test_format_error_three_digit_line_numbers :: proc(t: ^testing.T) {
@(test)
 test_caret_at_column_1 :: proc(t: ^testing.T) {
 	src := "{{bad}} at start"
-	out := format_tag_error("p.html", src, 0, "msg", "", colorize = false)
+	out := format_error("p.html", src, 0, "msg", "", colorize = false)
 	// Caret line should start with "^" right after "| " (no leading spaces).
 	testing.expect(t, strings.contains(out, "  | ^^^^^^^\n"), out)
 }
@@ -187,7 +265,7 @@ test_caret_at_column_1 :: proc(t: ^testing.T) {
 test_caret_at_column_N :: proc(t: ^testing.T) {
 	src := "    {{bad}}"
 	// pos=4 is the first '{'. Line 1, col 5.
-	out := format_tag_error("p.html", src, 4, "msg", "", colorize = false)
+	out := format_error("p.html", src, 4, "msg", "", colorize = false)
 	// 4 leading spaces, then 7 carets.
 	testing.expect(t, strings.contains(out, "  |     ^^^^^^^\n"), out)
 }
@@ -195,7 +273,7 @@ test_caret_at_column_N :: proc(t: ^testing.T) {
@(test)
 test_caret_width_matches_token :: proc(t: ^testing.T) {
 	src := "{{x}}"
-	out := format_tag_error("p.html", src, 0, "msg", "", colorize = false)
+	out := format_error("p.html", src, 0, "msg", "", colorize = false)
 	// {{x}} is 5 chars wide.
 	testing.expect(t, strings.contains(out, "  | ^^^^^\n"), out)
 }
@@ -207,7 +285,7 @@ test_caret_width_matches_token :: proc(t: ^testing.T) {
@(test)
 test_context_before_zero :: proc(t: ^testing.T) {
 	src := "l1\nl2\nl3\n{{bad}}\nl5\nl6"
-	out := format_tag_error(
+	out := format_error(
 		"p.html",
 		src,
 		9,
@@ -231,7 +309,7 @@ test_context_before_zero :: proc(t: ^testing.T) {
@(test)
 test_context_after_zero :: proc(t: ^testing.T) {
 	src := "l1\nl2\nl3\n{{bad}}\nl5\nl6"
-	out := format_tag_error(
+	out := format_error(
 		"p.html",
 		src,
 		9,
@@ -255,7 +333,7 @@ test_context_after_zero :: proc(t: ^testing.T) {
@(test)
 test_context_both_zero :: proc(t: ^testing.T) {
 	src := "l1\nl2\nl3\n{{bad}}\nl5\nl6"
-	out := format_tag_error(
+	out := format_error(
 		"p.html",
 		src,
 		9,
@@ -282,7 +360,7 @@ test_context_both_zero :: proc(t: ^testing.T) {
@(test)
 test_gutter_pipes_align_with_source_pipe :: proc(t: ^testing.T) {
 	src := "l1\n{{bad}}\nl3"
-	out := format_tag_error("p.html", src, 3, "msg", "", colorize = false)
+	out := format_error("p.html", src, 3, "msg", "", colorize = false)
 	// All "|" characters should appear at the same column.
 	// For width=1: source line is "N | ...", so "|" at col 2.
 	// Empty gutter is "  |" (width+1 spaces + "|"), so "|" at col 2.
@@ -304,7 +382,7 @@ test_gutter_pipes_align_with_source_pipe :: proc(t: ^testing.T) {
@(test)
 test_arrow_points_at_pipe :: proc(t: ^testing.T) {
 	src := "{{bad}}"
-	out := format_tag_error("p.html", src, 0, "msg", "", colorize = false)
+	out := format_error("p.html", src, 0, "msg", "", colorize = false)
 	// For width=1: arrow line is " --> ..." so ">" at col 3.
 	// Pipe lines are "  |" so "|" at col 2.
 	lines := strings.split(out, "\n", context.temp_allocator)
@@ -346,7 +424,7 @@ test_format_render_error_dispatch :: proc(t: ^testing.T) {
 	}

 	b := body(parse_err)
-	out := format_tag_error("test.html", src, b.pos, b.msg, colorize = false)
+	out := format_error("test.html", src, b.pos, b.msg, colorize = false)
 	testing.expect(t, strings.contains(out, "unclosed section"), out)
 	testing.expect(t, strings.contains(out, "test.html:"), out)
 }
@@ -479,268 +557,3 @@ test_parse_error_pipe_parse_in_inverted_keeps_double_braces :: proc(t: ^testing.
 		fmt.tprintf("msg should contain literal '{{^', got %q", b.msg),
 	)
 }
-
-// ---------------------------------------------------------------------------
-// Data-driven diagnostic cases (from DIAGNOSTIC_TESTS.yaml)
-//
-// write_diagnostic_tests generates diagnostic_test_cases_test.odin from the
-// enabled cases in DIAGNOSTIC_TESTS.yaml. On the first run (or when the YAML
-// changes), it regenerates the file and fails. On the second run, each case
-// runs as an individual test via run_diag_case.
-// ---------------------------------------------------------------------------
-
-Diag_Context :: struct {
-	items:  [5]string,
-	word:   string,
-	params: json.Object,
-	now:    string,
-}
-
-diag_ctx :: proc() -> Diag_Context {
-	params_raw := `{ "author": { "name": "foo bar"}}`
-	params_val, err := json.parse(params_raw, allocator = context.temp_allocator)
-	assert(err == nil)
-	params, ok := params_val.(json.Object)
-	assert(ok)
-	return Diag_Context {
-		items = {"a", "b", "c", "d", "e"},
-		word = "Hello, World!",
-		params = params,
-		now = "2026-08-10T16:21:55-04:00",
-	}
-}
-
-diag_partials :: proc() -> map[string]Template {
-	dirs := [?]string{#directory, "fixtures/layouts/partials"}
-	partials_dir, _ := filepath.join(dirs[:], context.temp_allocator)
-	partials := load_partials(partials_dir, allocator = context.temp_allocator)
-	// #directory expands to an absolute, build-environment-specific path (e.g.
-	// /build/... inside the nix sandbox vs a local checkout). Strip it so
-	// partial diagnostics reference a stable path across environments, keeping
-	// the golden expected output deterministic.
-	for key, tmpl in partials {
-		t := tmpl
-		rel := strings.trim_prefix(t.path, #directory)
-		t.path = strings.trim_left(rel, "/")
-		partials[key] = t
-	}
-	return partials
-}
-
-diag_slug :: proc(name: string) -> string {
-	sb: strings.Builder
-	strings.builder_init(&sb, context.temp_allocator)
-	prev_underscore := false
-	for r in name {
-		if (r >= 'a' && r <= 'z') || (r >= '0' && r <= '9') {
-			strings.write_rune(&sb, r)
-			prev_underscore = false
-		} else if r >= 'A' && r <= 'Z' {
-			strings.write_rune(&sb, r + 32)
-			prev_underscore = false
-		} else {
-			if !prev_underscore && strings.builder_len(sb) > 0 {
-				strings.write_rune(&sb, '_')
-				prev_underscore = true
-			}
-		}
-	}
-	s := strings.to_string(sb)
-	if len(s) > 0 && s[len(s) - 1] == '_' {
-		s = s[:len(s) - 1]
-	}
-	return s
-}
-
-render_silenced :: proc(
-	tmpl: Template,
-	data: any,
-	partials: map[string]Template = nil,
-	allocator := context.allocator,
-	warnings: ^[dynamic]Error = nil,
-) -> (string, Error) {
-	saved := context.logger
-	context.logger = log.nil_logger()
-	result, err := render(tmpl, data, partials, allocator, warnings)
-	context.logger = saved
-	return result, err
-}
-
-run_diag_case :: proc(t: ^testing.T, name: string) {
-	cases := load_diag_cases("../DIAGNOSTIC_TESTS.yaml")
-	ctx := diag_ctx()
-	partials := diag_partials()
-
-	for c in cases {
-		if c.name != name do continue
-
-		switch c.should {
-		case "error":
-			tmpl, perr := parse(c.input, "<test>", allocator = context.temp_allocator)
-			if perr != nil {
-				formatted := format_render_error(perr, tmpl, colorize = false)
-				testing.expect_value(t, formatted, c.expected)
-				return
-			}
-
-			_, rerr := render_silenced(tmpl, ctx, partials, context.temp_allocator)
-			testing.expect(
-				t,
-				rerr != nil,
-				fmt.tprintf("[%s] should error but render succeeded", c.name),
-			)
-			if rerr == nil do return
-
-			formatted := format_render_error(rerr, tmpl, colorize = false)
-			testing.expect_value(t, formatted, c.expected)
-
-		case "pass":
-			tmpl, perr := parse(c.input, "<test>", allocator = context.temp_allocator)
-			testing.expect(
-				t,
-				perr == nil,
-				fmt.tprintf(
-					"[%s] should parse, got: %s",
-					c.name,
-					perr != nil ? format_render_error(perr, tmpl, colorize = false) : "",
-				),
-			)
-			if perr != nil do return
-
-			_, rerr := render_silenced(tmpl, ctx, partials, context.temp_allocator)
-			testing.expect(
-				t,
-				rerr == nil,
-				fmt.tprintf(
-					"[%s] should not error, got: %s",
-					c.name,
-					rerr != nil ? format_render_error(rerr, tmpl, colorize = false) : "",
-				),
-			)
-
-		case "ok":
-			tmpl, perr := parse(c.input, "<test>", allocator = context.temp_allocator)
-			testing.expect(
-				t,
-				perr == nil,
-				fmt.tprintf(
-					"[%s] should parse, got: %s",
-					c.name,
-					perr != nil ? format_render_error(perr, tmpl, colorize = false) : "",
-				),
-			)
-			if perr != nil do return
-
-			result, rerr := render_silenced(tmpl, ctx, partials, context.temp_allocator)
-			testing.expect(
-				t,
-				rerr == nil,
-				fmt.tprintf(
-					"[%s] should not error, got: %s",
-					c.name,
-					rerr != nil ? format_render_error(rerr, tmpl, colorize = false) : "",
-				),
-			)
-			if rerr != nil do return
-
-			testing.expect_value(t, result, c.expected)
-
-		case "warn":
-			tmpl, perr := parse(c.input, "<test>", allocator = context.temp_allocator)
-			testing.expect(
-				t,
-				perr == nil,
-				fmt.tprintf(
-					"[%s] should parse, got: %s",
-					c.name,
-					perr != nil ? format_render_error(perr, tmpl, colorize = false) : "",
-				),
-			)
-			if perr != nil do return
-
-			warnings := make([dynamic]Error, 0, 4, context.temp_allocator)
-			_, rerr := render_silenced(tmpl, ctx, partials, context.temp_allocator, &warnings)
-			testing.expect(
-				t,
-				rerr == nil,
-				fmt.tprintf(
-					"[%s] should not error, got: %s",
-					c.name,
-					rerr != nil ? format_render_error(rerr, tmpl, colorize = false) : "",
-				),
-			)
-			if rerr != nil do return
-
-			testing.expect(
-				t,
-				len(warnings) > 0,
-				fmt.tprintf("[%s] should produce a warning", c.name),
-			)
-			if len(warnings) == 0 do return
-
-			formatted := format_render_error(warnings[0], tmpl, colorize = false)
-			testing.expect_value(t, formatted, c.expected)
-
-		case:
-			testing.expectf(t, false, "[%s] unknown should value: '%s'", c.name, c.should)
-		}
-		return
-	}
-
-	testing.expectf(t, false, "case '%s' not found in DIAGNOSTIC_TESTS.yaml", name)
-}
-
-@(test)
-write_diagnostic_tests :: proc(t: ^testing.T) {
-	cases := load_diag_cases("../DIAGNOSTIC_TESTS.yaml")
-	testing.expect(t, len(cases) > 0, "should load diagnostic cases")
-
-	sb: strings.Builder
-	strings.builder_init(&sb, context.temp_allocator)
-	strings.write_string(&sb, "#+test\n")
-	strings.write_string(&sb, "package mustache\n\n")
-	strings.write_string(&sb, "import \"core:testing\"\n\n")
-
-	count := 0
-	for c in cases {
-		if !c.enable do continue
-		if c.should == "" {
-			testing.expectf(t, false, "[%s] has enable: true but no should value", c.name)
-			continue
-		}
-		count += 1
-		slug := diag_slug(c.name)
-		strings.write_string(&sb, "@(test)\n")
-		strings.write_string(&sb, "test_diag_")
-		strings.write_string(&sb, slug)
-		strings.write_string(&sb, " :: proc(t: ^testing.T) {\n")
-		strings.write_string(&sb, "\trun_diag_case(t, \"")
-		strings.write_string(&sb, c.name)
-		strings.write_string(&sb, "\")\n}\n\n")
-	}
-
-	testing.expect(t, count > 0, "at least one case should be enabled")
-
-	generated := strings.to_string(sb)
-
-	dirs := [?]string{#directory, "diagnostic_test_cases_test.odin"}
-	gen_path, _ := filepath.join(dirs[:], context.temp_allocator)
-
-	existing_data, _ := os.read_entire_file_from_path(gen_path, context.temp_allocator)
-	existing := string(existing_data)
-
-	if existing == generated do return
-
-	werr := os.write_entire_file(gen_path, transmute([]byte)generated)
-	if werr != nil {
-		testing.expectf(t, false, "failed to write %s: %v", gen_path, werr)
-		return
-	}
-	testing.expectf(
-		t,
-		false,
-		"diagnostic_test_cases_test.odin regenerated (%d cases), re-run tests to see results",
-		count,
-	)
-}
-
@@ -1,45 +0,0 @@
-#+test
-package mustache
-
-import "core:testing"
-
-@(test)
-test_diag_empty_tag :: proc(t: ^testing.T) {
-	run_diag_case(t, "empty tag")
-}
-
-@(test)
-test_diag_empty_raw_tag :: proc(t: ^testing.T) {
-	run_diag_case(t, "empty raw tag")
-}
-
-@(test)
-test_diag_missing_pipe_arguement :: proc(t: ^testing.T) {
-	run_diag_case(t, "Missing pipe arguement")
-}
-
-@(test)
-test_diag_first_last_with_zero_argument :: proc(t: ^testing.T) {
-	run_diag_case(t, "first/last with zero argument")
-}
-
-@(test)
-test_diag_first_last_with_negative_argument :: proc(t: ^testing.T) {
-	run_diag_case(t, "first/last with negative argument")
-}
-
-@(test)
-test_diag_first_succeeds_on_params_author_name :: proc(t: ^testing.T) {
-	run_diag_case(t, "first succeeds on params.author.name")
-}
-
-@(test)
-test_diag_first_last_in_partial :: proc(t: ^testing.T) {
-	run_diag_case(t, "first/last in partial")
-}
-
-@(test)
-test_diag_pipes_are_ignored_in_comments :: proc(t: ^testing.T) {
-	run_diag_case(t, "Pipes are ignored in comments")
-}
-
@@ -1,12 +0,0 @@
-<footer>
-  <menu>
-    <a class="goto-top opacity-0" href="#">{{> icons/chevron_up}}</a>
-    {{#params.social}}
-    <a href="{{url}}" target="_blank" rel="noopener noreferrer me" title="{{name}}">{{>* icon}}</a>
-    {{/params.social}}
-  </menu>
-  <p>Proudly built with <a href="https://github.com/sbrow/thor/">Thor</a> and <a
-      href="https://edwardtufte.github.io/tufte-css/">Tufte CSS</a>
-  </p>
-  <p><small>&copy;</small> {{! now | format "2006" }} {{params.author.name | first -3 }}</p>
-</footer>
@@ -129,12 +129,12 @@ match_token :: proc(b: ^strings.Builder, dt: Date_Components, s: string) -> int
 	if strings.has_prefix(
 		s,
 		"January",
-	) {fmt.sbprintf(b, "%s", time.Month(dt.month)); return 7}
+	) {strings.write_string(b, fmt.tprintf("%s", time.Month(dt.month))); return 7}
 	if strings.has_prefix(s, "Monday") {emit_weekday(b, dt, full = true); return 6}
 	if strings.has_prefix(
 		s,
 		"2006",
-	) {fmt.sbprintf(b, "%04d", dt.year); return 4}
+	) {strings.write_string(b, fmt.tprintf("%04d", dt.year)); return 4}
 	if strings.has_prefix(s, "MST") {
 		abbr := dt.tz_abbr
 		if len(abbr) == 0 do abbr = "UTC"
@@ -146,37 +146,37 @@ match_token :: proc(b: ^strings.Builder, dt: Date_Components, s: string) -> int
 	if strings.has_prefix(
 		s,
 		"06",
-	) {fmt.sbprintf(b, "%02d", dt.year % 100); return 2}
-	if strings.has_prefix(s, "02") {fmt.sbprintf(b, "%02d", dt.day); return 2}
+	) {strings.write_string(b, fmt.tprintf("%02d", dt.year % 100)); return 2}
+	if strings.has_prefix(s, "02") {strings.write_string(b, fmt.tprintf("%02d", dt.day)); return 2}
 	if strings.has_prefix(
 		s,
 		"15",
-	) {fmt.sbprintf(b, "%02d", dt.hour); return 2}
+	) {strings.write_string(b, fmt.tprintf("%02d", dt.hour)); return 2}
 	if strings.has_prefix(
 		s,
 		"04",
-	) {fmt.sbprintf(b, "%02d", dt.minute); return 2}
+	) {strings.write_string(b, fmt.tprintf("%02d", dt.minute)); return 2}
 	if strings.has_prefix(
 		s,
 		"05",
-	) {fmt.sbprintf(b, "%02d", dt.second); return 2}
+	) {strings.write_string(b, fmt.tprintf("%02d", dt.second)); return 2}
 	if strings.has_prefix(
 		s,
 		"01",
-	) {fmt.sbprintf(b, "%02d", dt.month); return 2}
+	) {strings.write_string(b, fmt.tprintf("%02d", dt.month)); return 2}
 	if strings.has_prefix(s, "03") {emit_hour_12(b, dt, pad = true); return 2}
 	if strings.has_prefix(s, "PM") {emit_am_pm(b, dt); return 2}
 	if strings.has_prefix(s, "pm") {emit_am_pm_lower(b, dt); return 2}
 	if len(s) >= 1 {
 		switch s[0] {
 		case '2':
-			fmt.sbprintf(b, "%d", dt.day); return 1
+			strings.write_string(b, fmt.tprintf("%d", dt.day)); return 1
 		case '1':
-			fmt.sbprintf(b, "%d", dt.month); return 1
+			strings.write_string(b, fmt.tprintf("%d", dt.month)); return 1
 		case '4':
-			fmt.sbprintf(b, "%d", dt.minute); return 1
+			strings.write_string(b, fmt.tprintf("%d", dt.minute)); return 1
 		case '5':
-			fmt.sbprintf(b, "%d", dt.second); return 1
+			strings.write_string(b, fmt.tprintf("%d", dt.second)); return 1
 		case '3':
 			emit_hour_12(b, dt, pad = false); return 1
 		case:
@@ -0,0 +1,154 @@
+#+test
+package mustache
+
+import "core:fmt"
+import "core:testing"
+
+/*
+
+// --- Spec test 1: Interpolation ---
+// A lambda's return value should be interpolated.
+
+Interp_Data :: struct {
+	lambda: proc() -> string,
+	planet: string,
+}
+
+Interp_Data_Int :: struct {
+	lambda: proc() -> int,
+	planet: string,
+}
+
+@(test)
+test_lambda_interpolation :: proc(t: ^testing.T) {
+	data := Interp_Data {
+		lambda = proc() -> string {return "world"},
+	}
+	tpl, _ := parse("Hello, {{lambda}}!", "<test>", context.temp_allocator)
+	result, _ := render(tpl, data, {}, context.temp_allocator)
+	testing.expect_value(t, result, "Hello, world!")
+}
+
+// --- Spec test 2: Interpolation - Expansion ---
+// A lambda's return value should be parsed.
+
+@(test)
+test_lambda_interpolation_expansion :: proc(t: ^testing.T) {
+	data := Interp_Data {
+		lambda = proc() -> string {return "{{planet}}"},
+		planet = "world",
+	}
+	tpl, _ := parse("Hello, {{lambda}}!", "<test>", context.temp_allocator)
+	result, _ := render(tpl, data, {}, context.temp_allocator)
+	testing.expect_value(t, result, "Hello, world!")
+}
+
+// --- Spec test 4: Interpolation - Multiple Calls ---
+// Interpolated lambdas should not be cached.
+
+counter_lambda :: proc() -> int {
+	@(static) call_count := 0
+	call_count += 1
+	return call_count
+}
+
+@(test)
+test_lambda_interpolation_multiple_calls :: proc(t: ^testing.T) {
+	data := Interp_Data_Int {
+		lambda = counter_lambda,
+	}
+	tpl, _ := parse("{{lambda}} == {{{lambda}}} == {{lambda}}", "<test>", context.temp_allocator)
+	result, _ := render(tpl, data, {}, context.temp_allocator)
+	testing.expect_value(t, result, "1 == 2 == 3")
+}
+
+// --- Spec test 5: Escaping ---
+// Lambda results should be appropriately escaped.
+
+@(test)
+test_lambda_escaping :: proc(t: ^testing.T) {
+	data := Interp_Data {
+		lambda = proc() -> string {return ">"},
+	}
+	tpl, _ := parse("<{{lambda}}{{{lambda}}}", "<test>", context.temp_allocator)
+	result, _ := render(tpl, data, {}, context.temp_allocator)
+	testing.expect_value(t, result, "<&gt;>")
+}
+
+// --- Spec test 6: Section ---
+// Lambdas used for sections should receive the raw section string.
+
+Section_Data :: struct {
+	lambda: proc(_: string) -> string,
+	x:      string,
+	planet: string,
+}
+
+@(test)
+test_lambda_section :: proc(t: ^testing.T) {
+	data := Section_Data {
+		lambda = proc(text: string) -> string {
+			if text == "{{x}}" {return "yes"} else {return "no"}
+		},
+		x = "Error!",
+	}
+	tpl, _ := parse("<{{#lambda}}{{x}}{{/lambda}}>", "<test>", context.temp_allocator)
+	result, _ := render(tpl, data, {}, context.temp_allocator)
+	testing.expect_value(t, result, "<yes>")
+}
+
+// --- Spec test 7: Section - Expansion ---
+// Lambdas used for sections should have their results parsed.
+
+@(test)
+test_lambda_section_expansion :: proc(t: ^testing.T) {
+	data := Section_Data {
+		lambda = proc(text: string) -> string {
+			return fmt.tprintf("%s{{{{planet}}}}%s", text, text)
+		},
+		planet = "Earth",
+	}
+	tpl, _ := parse("<{{#lambda}}-{{/lambda}}>", "<test>", context.temp_allocator)
+	result, _ := render(tpl, data, {}, context.temp_allocator)
+	testing.expect_value(t, result, "<-Earth->")
+}
+
+// --- Spec test 9: Section - Multiple Calls ---
+// Lambdas used for sections should not be cached.
+
+@(test)
+test_lambda_section_multiple_calls :: proc(t: ^testing.T) {
+	data := Section_Data {
+		lambda = proc(text: string) -> string {
+			return fmt.tprintf("__%s__", text)
+		},
+	}
+	tpl, _ := parse(
+		"{{#lambda}}FILE{{/lambda}} != {{#lambda}}LINE{{/lambda}}",
+		"<test>",
+		context.temp_allocator,
+	)
+	result, _ := render(tpl, data, {}, context.temp_allocator)
+	testing.expect_value(t, result, "__FILE__ != __LINE__")
+}
+
+// --- Spec test 10: Inverted Section ---
+// Lambdas used for inverted sections should be considered truthy.
+
+Inverted_Data :: struct {
+	lambda: proc(_: string) -> bool,
+	static: string,
+}
+
+@(test)
+test_lambda_inverted_section :: proc(t: ^testing.T) {
+	data := Inverted_Data {
+		lambda = proc(text: string) -> bool {return false},
+		static = "static",
+	}
+	tpl, _ := parse("<{{^lambda}}{{static}}{{/lambda}}>", "<test>", context.temp_allocator)
+	result, _ := render(tpl, data, {}, context.temp_allocator)
+	testing.expect_value(t, result, "<>")
+}
+
+*/
@@ -5,8 +5,6 @@ import "core:fmt"
 import "core:log"
 import "core:strings"

-import diags "../diagnostics"
-
 // A hypothetical maximum context depth. Trying to pass more than this many items
 // to render(tpl, data), or nesting templates further than this depth would be
 // an error.
@@ -25,7 +23,6 @@ Context_Stack :: [dynamic]any
 Error_Kind :: enum {
 	Syntax, // parse-time: malformed template
 	Data, // render-time: template fine, data wrong (e.g. filter misuse)
-	Warning, // non-fatal: operation succeeded but user should know
 }

 Error_Body :: struct {
@@ -60,13 +57,13 @@ tag_error :: proc(err: Error, tmpl: Template) -> Error {
 	if err == nil do return nil
 	b := body(err)
 	return Error_Body {
-		msg = b.msg,
-		pos = b.pos,
-		span = b.span,
-		hint = b.hint,
-		kind = b.kind,
+		msg    = b.msg,
+		pos    = b.pos,
+		span   = b.span,
+		hint   = b.hint,
+		kind   = b.kind,
 		source = tmpl.source,
-		path = tmpl.path,
+		path   = tmpl.path,
 	}
 }

@@ -156,22 +153,18 @@ parse :: proc(
 	tmpl: Template,
 	err: Error,
 ) {
-	tmpl.source = source
-	tmpl.path = path
-
 	tokens, terr := tokenize(source, tokens_allocator)
 	if terr != nil {
-		return tmpl, terr
+		return {}, terr
 	}

 	tmpl.nodes, err = parse_tokens(tokens[:], source, allocator)
 	if err != nil {
 		delete(tmpl.nodes)
-		e := body(err)
-		e.source = source
-		e.path = path
-		return {}, e
+		return {}, err
 	}
+	tmpl.source = source
+	tmpl.path = path
 	deindent_blocks(tmpl.nodes[:], allocator)
 	return tmpl, nil
 }
@@ -181,8 +174,6 @@ render :: proc(
 	data: any,
 	partials: map[string]Template = nil,
 	allocator := context.allocator,
-	warnings: ^[dynamic]Error = nil,
-	tools: ^Tool_Registry = nil,
 ) -> (
 	result: string,
 	err: Error,
@@ -208,15 +199,7 @@ render :: proc(
 	}

 	all_nodes := tmpl.nodes[:]
-	err = render_nodes(
-		tmpl,
-		all_nodes,
-		&ctx,
-		partials,
-		&builder,
-		warnings = warnings,
-		tools = tools,
-	)
+	err = render_nodes(tmpl, all_nodes, &ctx, partials, &builder)
 	if err != nil {
 		return result, err
 	}
@@ -280,22 +263,9 @@ parse_section :: proc(
 			pos^ += 1

 		case .Variable:
-			if len(strings.trim_space(tok.value)) == 0 {
-				return Error_Body {
-					msg = "Empty tag",
-					pos = tok.pos,
-					kind = .Syntax,
-					hint = "tag must contain a key",
-				}
-			}
 			idx := len(nodes)
 			append(nodes, Node{kind = .Variable})
-			pipe_key, perr := parse_pipeline(
-				tok.value,
-				&nodes[idx].filters,
-				tok.pos,
-				tag_content_base(source, tok.pos),
-			)
+			pipe_key, perr := parse_pipeline(tok.value, &nodes[idx].filters, tok.pos, tag_content_base(source, tok.pos))
 			if perr != nil {
 				return Error_Body {
 					msg = fmt.tprintf("pipe parse error in '{{{{%s}}}}': %v", tok.value, perr),
@@ -307,22 +277,9 @@ parse_section :: proc(
 			pos^ += 1

 		case .Unescaped:
-			if len(strings.trim_space(tok.value)) == 0 {
-				return Error_Body {
-					msg = "Empty tag",
-					pos = tok.pos,
-					kind = .Syntax,
-					hint = "tag must contain a key",
-				}
-			}
 			idx := len(nodes)
 			append(nodes, Node{kind = .Unescaped})
-			pipe_key, perr := parse_pipeline(
-				tok.value,
-				&nodes[idx].filters,
-				tok.pos,
-				tag_content_base(source, tok.pos),
-			)
+			pipe_key, perr := parse_pipeline(tok.value, &nodes[idx].filters, tok.pos, tag_content_base(source, tok.pos))
 			if perr != nil {
 				return Error_Body {
 					msg = fmt.tprintf("pipe parse error in '{{{{&%s}}}}': %v", tok.value, perr),
@@ -342,12 +299,7 @@ parse_section :: proc(
 			content_start := 0
 			if pos^ < len(tokens) {content_start = tokens[pos^].pos}
 			append(nodes, Node{kind = .Section, pos = tok.pos})
-			pipe_key, perr := parse_pipeline(
-				tok.value,
-				&nodes[idx].filters,
-				tok.pos,
-				tag_content_base(source, tok.pos),
-			)
+			pipe_key, perr := parse_pipeline(tok.value, &nodes[idx].filters, tok.pos, tag_content_base(source, tok.pos))
 			if perr != nil {
 				return Error_Body {
 					msg = fmt.tprintf("pipe parse error in '{{{{#%s}}}}': %v", tok.value, perr),
@@ -368,12 +320,7 @@ parse_section :: proc(
 			content_start := 0
 			if pos^ < len(tokens) {content_start = tokens[pos^].pos}
 			append(nodes, Node{kind = .Inverted, pos = tok.pos})
-			pipe_key, perr := parse_pipeline(
-				tok.value,
-				&nodes[idx].filters,
-				tok.pos,
-				tag_content_base(source, tok.pos),
-			)
+			pipe_key, perr := parse_pipeline(tok.value, &nodes[idx].filters, tok.pos, tag_content_base(source, tok.pos))
 			if perr != nil {
 				return Error_Body {
 					msg = fmt.tprintf("pipe parse error in '{{{{^%s}}}}': %v", tok.value, perr),
@@ -596,8 +543,6 @@ render_template :: proc(
 	b: ^strings.Builder,
 	blocks: map[string]Block_Override,
 	indent: string,
-	warnings: ^[dynamic]Error = nil,
-	tools: ^Tool_Registry = nil,
 ) -> Error {
 	if len(indent) > 0 {
 		state := Indent_State {
@@ -605,9 +550,9 @@ render_template :: proc(
 			at_line_start = false,
 		}
 		strings.write_string(b, indent) // first line always gets indent
-		return render_nodes(pt, pt.nodes[:], ctx, partials, b, blocks, &state, warnings, tools)
+		return render_nodes(pt, pt.nodes[:], ctx, partials, b, blocks, &state)
 	}
-	return render_nodes(pt, pt.nodes[:], ctx, partials, b, blocks, nil, warnings, tools)
+	return render_nodes(pt, pt.nodes[:], ctx, partials, b, blocks, nil)
 }

 write_indented :: proc(
@@ -651,8 +596,6 @@ render_nodes :: proc(
 	b: ^strings.Builder,
 	blocks: map[string]Block_Override = nil,
 	indent_state: ^Indent_State = nil,
-	warnings: ^[dynamic]Error = nil,
-	tools: ^Tool_Registry = nil,
 ) -> Error {
 	i := 0
 	for i < len(nodes) {
@@ -676,21 +619,36 @@ render_nodes :: proc(
 				warn_unknown_key(current, ctx[:], node)
 			}
 			if len(node.filters) > 0 {
-				transformed, perr := apply_pipeline(
-					val,
-					node.filters[:],
-					node.pos,
-					ctx[:],
-					current,
-					warnings,
-					tools,
-				)
+				transformed, perr := apply_pipeline(val, node.filters[:], node.pos, ctx[:])
 				if perr != nil {
 					return tag_error(perr, current)
 				}
 				val = transformed
 			}
-			write_value(b, val, escape = true)
+			if result_str, ok := call_interp_lambda(val); ok {
+				sub_tpl, perr := parse(
+					result_str,
+					fmt.tprintf("<lambda output from '%s'>", node.key),
+					context.temp_allocator,
+					context.temp_allocator,
+				)
+				if perr == nil {
+					temp: strings.Builder
+					strings.builder_init(&temp, context.temp_allocator)
+					render_nodes(
+						sub_tpl,
+						sub_tpl.nodes[:],
+						ctx,
+						partials,
+						&temp,
+						blocks,
+						nil,
+					) or_return
+					write_value(b, strings.to_string(temp), escape = true)
+				}
+			} else {
+				write_value(b, val, escape = true)
+			}
 			i += 1

 		case .Unescaped:
@@ -703,21 +661,36 @@ render_nodes :: proc(
 				warn_unknown_key(current, ctx[:], node)
 			}
 			if len(node.filters) > 0 {
-				transformed, perr := apply_pipeline(
-					val,
-					node.filters[:],
-					node.pos,
-					ctx[:],
-					current,
-					warnings,
-					tools,
-				)
+				transformed, perr := apply_pipeline(val, node.filters[:], node.pos, ctx[:])
 				if perr != nil {
 					return tag_error(perr, current)
 				}
 				val = transformed
 			}
-			write_value(b, val, escape = false)
+			if result_str, ok := call_interp_lambda(val); ok {
+				sub_tpl, perr := parse(
+					result_str,
+					fmt.tprintf("<lambda output from '%s'>", node.key),
+					context.temp_allocator,
+					context.temp_allocator,
+				)
+				if perr == nil {
+					temp: strings.Builder
+					strings.builder_init(&temp, context.temp_allocator)
+					render_nodes(
+						sub_tpl,
+						sub_tpl.nodes[:],
+						ctx,
+						partials,
+						&temp,
+						blocks,
+						nil,
+					) or_return
+					write_value(b, strings.to_string(temp), escape = false)
+				}
+			} else {
+				write_value(b, val, escape = false)
+			}
 			i += 1

 		case .Section:
@@ -726,21 +699,31 @@ render_nodes :: proc(
 				warn_unknown_key(current, ctx[:], node)
 			}
 			if len(node.filters) > 0 {
-				transformed, perr := apply_pipeline(
-					val,
-					node.filters[:],
-					node.pos,
-					ctx[:],
-					current,
-					warnings,
-					tools,
-				)
+				transformed, perr := apply_pipeline(val, node.filters[:], node.pos, ctx[:])
 				if perr != nil {
 					return tag_error(perr, current)
 				}
 				val = transformed
 			}
-			if is_truthy(val) {
+			if result_str, ok := call_section_lambda(val, node.content); ok {
+				sub_tpl, perr := parse(
+					result_str,
+					fmt.tprintf("<lambda output from '%s'>", node.key),
+					context.temp_allocator,
+					context.temp_allocator,
+				)
+				if perr == nil {
+					render_nodes(
+						sub_tpl,
+						sub_tpl.nodes[:],
+						ctx,
+						partials,
+						b,
+						blocks,
+						nil,
+					) or_return
+				}
+			} else if is_truthy(val) {
 				children := node.children
 				elem_info, count, data := list_info(val)
 				if elem_info != nil {
@@ -756,8 +739,6 @@ render_nodes :: proc(
 							b,
 							blocks,
 							indent_state,
-							warnings,
-							tools,
 						) or_return
 					}
 				} else {
@@ -771,7 +752,6 @@ render_nodes :: proc(
 						b,
 						blocks,
 						indent_state,
-						warnings,
 					) or_return
 				}
 			}
@@ -783,15 +763,7 @@ render_nodes :: proc(
 				warn_unknown_key(current, ctx[:], node)
 			}
 			if len(node.filters) > 0 {
-				transformed, perr := apply_pipeline(
-					val,
-					node.filters[:],
-					node.pos,
-					ctx[:],
-					current,
-					warnings,
-					tools,
-				)
+				transformed, perr := apply_pipeline(val, node.filters[:], node.pos, ctx[:])
 				if perr != nil {
 					return tag_error(perr, current)
 				}
@@ -806,8 +778,6 @@ render_nodes :: proc(
 					b,
 					blocks,
 					indent_state,
-					warnings,
-					tools,
 				) or_return
 			}
 			i += 1 + len(node.children)
@@ -822,7 +792,7 @@ render_nodes :: proc(
 			if !found {
 				warn_missing_partial(current, partials, node, name)
 			} else {
-				render_template(pt, ctx, partials, b, nil, node.indent, warnings, tools) or_return
+				render_template(pt, ctx, partials, b, nil, node.indent) or_return
 				if indent_state != nil {
 					indent_state.at_line_start = false
 				}
@@ -857,8 +827,6 @@ render_nodes :: proc(
 					&temp,
 					content_blocks,
 					nil,
-					warnings,
-					tools,
 				) or_return
 				at_ls := true
 				write_indented(b, node.indent, strings.to_string(temp), &at_ls)
@@ -871,8 +839,6 @@ render_nodes :: proc(
 					b,
 					content_blocks,
 					indent_state,
-					warnings,
-					tools,
 				) or_return
 			}
 			i += 1 + len(node.children)
@@ -885,16 +851,7 @@ render_nodes :: proc(
 				warn_missing_partial(current, partials, node, node.key)
 			} else {
 				warn_unmatched_block_overrides(current, pt, parent_children)
-				render_template(
-					pt,
-					ctx,
-					partials,
-					b,
-					merged,
-					node.indent,
-					warnings,
-					tools,
-				) or_return
+				render_template(pt, ctx, partials, b, merged, node.indent) or_return
 				if indent_state != nil {
 					indent_state.at_line_start = false
 				}
@@ -949,7 +906,7 @@ warn_unknown_key :: proc(current: Template, ctx: []any, node: Node) {

 	hint := ""
 	if len(available) > 0 {
-		suggestion := diags.suggest_correction(available, missing)
+		suggestion := suggest_correction(available, missing)
 		if suggestion != "" {
 			hint = fmt.tprintf("did you mean '%s'?", suggestion)
 		}
@@ -959,14 +916,7 @@ warn_unknown_key :: proc(current: Template, ctx: []any, node: Node) {
 	if path == "" {
 		path = "<input>"
 	}
-	diag := format_tag_error(
-		path,
-		current.source,
-		node.pos,
-		msg,
-		hint,
-		colorize = diags.should_colorize(),
-	)
+	diag := format_error(path, current.source, node.pos, msg, hint, colorize = should_colorize())
 	log.warnf("%s", diag)
 }

@@ -980,7 +930,7 @@ warn_missing_partial :: proc(
 ) {
 	hint := ""
 	available := collect_partial_names(partials)
-	suggestion := diags.suggest_correction(available, name)
+	suggestion := suggest_correction(available, name)
 	if suggestion != "" {
 		hint = fmt.tprintf("did you mean '%s'?", suggestion)
 	}
@@ -989,14 +939,7 @@ warn_missing_partial :: proc(
 	if path == "" {
 		path = "<input>"
 	}
-	diag := format_tag_error(
-		path,
-		current.source,
-		node.pos,
-		msg,
-		hint,
-		colorize = diags.should_colorize(),
-	)
+	diag := format_error(path, current.source, node.pos, msg, hint, colorize = should_colorize())
 	log.warnf("%s", diag)
 }

@@ -1032,7 +975,7 @@ warn_unmatched_block_overrides :: proc(
 		}

 		hint := ""
-		suggestion := diags.suggest_correction(available, child.key)
+		suggestion := suggest_correction(available, child.key)
 		if suggestion != "" {
 			hint = fmt.tprintf("did you mean '%s'?", suggestion)
 		}
@@ -1045,13 +988,13 @@ warn_unmatched_block_overrides :: proc(
 		if path == "" {
 			path = "<input>"
 		}
-		diag := format_tag_error(
+		diag := format_error(
 			path,
 			current.source,
 			child.pos,
 			msg,
 			hint,
-			colorize = diags.should_colorize(),
+			colorize = should_colorize(),
 		)
 		log.warnf("%s", diag)
 	}
@@ -1075,13 +1018,6 @@ warn_context_depth :: proc(current: Template, node: Node) {
 	if path == "" {
 		path = "<input>"
 	}
-	diag := format_tag_error(
-		path,
-		current.source,
-		node.pos,
-		msg,
-		"",
-		colorize = diags.should_colorize(),
-	)
+	diag := format_error(path, current.source, node.pos, msg, "", colorize = should_colorize())
 	log.warnf("%s", diag)
 }
@@ -1,6 +1,7 @@
 #+test
 package mustache

+import "core:log"
 import "core:mem"
 import "core:strings"
 import "core:testing"
@@ -127,9 +128,10 @@ leak_repeated_render :: proc(t: ^testing.T) {

 // A deeply nested map pushes the context stack past MAX_CONTEXT_DEPTH (16).
 // The depth warning must be non-fatal: rendering still succeeds.
-/*
@(test)
 test_context_depth_warns :: proc(t: ^testing.T) {
+	context.logger = log.nil_logger()
+
 	AMT :: MAX_CONTEXT_DEPTH + 2

 	// Build AMT nested {x: {...}} levels; the innermost holds `leaf`.
@@ -155,9 +157,7 @@ test_context_depth_warns :: proc(t: ^testing.T) {
 		return
 	}

-	result, rerr := render_silenced(tmpl, data, allocator = context.temp_allocator)
+	result, rerr := render(tmpl, data, allocator = context.temp_allocator)
 	testing.expect(t, rerr == nil, "depth warning must be non-fatal")
 	testing.expect_value(t, result, "found")
 }
-*/
-
@@ -1,48 +0,0 @@
-package mustache
-
-import "core:fmt"
-import "core:log"
-import "core:os"
-import "core:strings"
-
-load_partials :: proc(
-	dir: string,
-	base: string = "",
-	allocator := context.allocator,
-) -> map[string]Template {
-	partials := make(map[string]Template, allocator)
-
-	entries, err := os.read_all_directory_by_path(dir, context.temp_allocator)
-	if err != nil {
-		log.warnf("cannot read partials dir %s: %v", dir, err)
-		return partials
-	}
-	defer os.file_info_slice_delete(entries, context.temp_allocator)
-
-	for entry in entries {
-		if entry.type == .Regular {
-			if !strings.has_suffix(entry.name, ".html") do continue
-			stripped := entry.name[:len(entry.name) - len(".html")]
-			key := fmt.tprintf("%s%s", base, stripped)
-			data, ferr := os.read_entire_file_from_path(entry.fullpath, allocator)
-			if ferr != nil do continue
-			source := string(data)
-			path := strings.clone(entry.fullpath, allocator)
-			tmpl, perr := parse(source, path, allocator = allocator)
-			if perr != nil {
-				b := body(perr)
-				log.warnf("failed to parse partial %s: %s", entry.fullpath, b.msg)
-				continue
-			}
-			partials[key] = tmpl
-		} else if entry.type == .Directory {
-			sub_base := fmt.tprintf("%s%s/", base, entry.name)
-			sub := load_partials(entry.fullpath, sub_base, allocator)
-			for k, v in sub {
-				partials[k] = v
-			}
-		}
-	}
-
-	return partials
-}
@@ -4,10 +4,8 @@ import "base:runtime"
 import "core:fmt"
 import "core:log"
 import "core:reflect"
-import "core:strconv"
 import "core:strings"
-
-import diags "../diagnostics"
+import "core:time"
 import "core:time/datetime"

 // MAX_PIPES was chosen arbitrarily. It holds no performance or logical
@@ -22,10 +20,6 @@ DEFAULT_DATE_FORMAT :: "2 Jan 2006"
 Pipe_Op :: enum {
 	Format,
 	Group_By,
-	First,
-	Last,
-	Rel_Url,
-	Tool,
 }

 // Pipe op names are derived from the enum via reflection (lowercased).
@@ -62,10 +56,9 @@ pipe_op_candidates :: proc(allocator := context.temp_allocator) -> []string {
 }

 Pipe_Filter :: struct {
-	op:      string,
-	args:    [dynamic; MAX_PIPE_ARGS]string,
-	op_pos:  int,
-	end_pos: int,
+	op:     string,
+	args:   [dynamic; MAX_PIPE_ARGS]string,
+	op_pos: int,
 }

 Group :: struct {
@@ -138,7 +131,11 @@ parse_pipeline :: proc(
 	pipe_count := strings.count(content, "|")
 	if pipe_count > MAX_PIPES {
 		return "", Error_Body {
-			msg = fmt.tprintf("pipe expression has %d filters, max is %d", pipe_count, MAX_PIPES),
+			msg = fmt.tprintf(
+				"pipe expression has %d filters, max is %d",
+				pipe_count,
+				MAX_PIPES,
+			),
 			pos = pos,
 			kind = .Syntax,
 		}
@@ -200,9 +197,8 @@ parse_pipeline :: proc(
 		}

 		filter := Pipe_Filter {
-			op      = tokens[0],
-			op_pos  = content_base + op_offset_in_content,
-			end_pos = content_base + op_offset_in_content + len(seg),
+			op     = tokens[0],
+			op_pos = content_base + op_offset_in_content,
 		}
 		for j in 1 ..< len(tokens) {
 			append(&filter.args, tokens[j])
@@ -224,28 +220,14 @@ apply_pipeline :: proc(
 	filters: []Pipe_Filter,
 	pos: int,
 	ctx: []any,
-	tmpl: Template,
-	warnings: ^[dynamic]Error = nil,
-	tools: ^Tool_Registry = nil,
 ) -> (
 	current: any,
 	err: Error,
 ) {
 	current = value
 	for &filter in filters {
-		result, ferr := apply_filter(current, &filter, pos, ctx, tmpl, tools)
-		if ferr != nil {
-			b := body(ferr)
-			if b.kind != .Warning {
-				return result, ferr
-			}
-			if warnings != nil {
-				append(warnings, ferr)
-			} else {
-				log.warnf("%s", format_render_error(ferr, tmpl))
-			}
-		}
-		current = result
+		current = apply_filter(current, &filter, pos, ctx) or_return
+		log.debugf("applied: filter=%v before=%s after=%s pos=%d", filter, value, current, pos)
 	}
 	return
 }
@@ -274,27 +256,17 @@ resolve_format_string :: proc(name: string, ctx: []any, pos: int) -> (string, Er
 	return str, nil
 }

-apply_filter :: proc(
-	value: any,
-	filter: ^Pipe_Filter,
-	pos: int,
-	ctx: []any,
-	tmpl: Template,
-	tools: ^Tool_Registry = nil,
-) -> (
-	any,
-	Error,
-) {
+apply_filter :: proc(value: any, filter: ^Pipe_Filter, pos: int, ctx: []any) -> (any, Error) {
 	op, ok := pipe_op_from_string(filter.op)
 	if !ok {
 		hint := ""
-		suggestion := diags.suggest_correction(pipe_op_candidates(), filter.op)
+		suggestion := suggest_correction(pipe_op_candidates(), filter.op)
 		if suggestion != "" {
 			hint = fmt.tprintf("did you mean '%s'?", suggestion)
 		}
 		return nil, Error_Body {
-			msg = fmt.tprintf("unknown pipe op '%s'", filter.op),
-			pos = filter.op_pos,
+			msg  = fmt.tprintf("unknown pipe op '%s'", filter.op),
+			pos  = filter.op_pos,
 			span = len(filter.op),
 			kind = .Data,
 			hint = hint,
@@ -303,8 +275,7 @@ apply_filter :: proc(

 	switch op {
 	case .Group_By:
-		span := filter.end_pos - filter.op_pos
-		return apply_group_by(value, filter.args[:], filter.op_pos, span)
+		return apply_group_by(value, filter.args[:], pos)
 	case .Format:
 		str, ok := reflect.as_string(value)
 		if !ok {
@@ -343,89 +314,10 @@ apply_filter :: proc(
 		} else {
 			return any{new_clone(str2, context.temp_allocator), typeid_of(string)}, nil
 		}
-	case .First:
-		span := filter.end_pos - filter.op_pos
-		return take(value, filter.args[:], true, tmpl, filter.op_pos, span)
-	case .Last:
-		span := filter.end_pos - filter.op_pos
-		return take(value, filter.args[:], false, tmpl, filter.op_pos, span)
-	case .Rel_Url:
-		bv, _ := base_value(value)
-		str, ok := reflect.as_string(bv)
-		if !ok {
-			return value, Error_Body {
-				msg = "rel_url may only be used on strings",
-				pos = pos,
-				kind = .Data,
-			}
-		}
-		result := rel_url(str, ctx)
-		return any{new_clone(result, context.temp_allocator), typeid_of(string)}, nil
-	case .Tool:
-		if len(filter.args) != 1 {
-			return value, Error_Body {
-				msg = fmt.tprintf(
-					"tool expects 1 argument (the tool name), got %d",
-					len(filter.args),
-				),
-				pos = filter.op_pos,
-				span = filter.end_pos - filter.op_pos,
-				kind = .Data,
-			}
-		}
-		bv, _ := base_value(value)
-		str, ok := reflect.as_string(bv)
-		if !ok {
-			return value, Error_Body {
-				msg = "tool may only be used on strings",
-				pos = pos,
-				kind = .Data,
-			}
-		}
-		result, terr := run_tool(tools, filter.args[0], str, filter.op_pos)
-		if terr != nil {
-			return value, terr
-		}
-		return any{new_clone(result, context.temp_allocator), typeid_of(string)}, nil
 	}
 	return {}, nil
 }

-// rel_url prefixes a root-relative path (e.g. "/faq/") with the site's
-// base_path so links resolve correctly under a subpath deployment. Values that
-// are already absolute — a scheme ("http://"), protocol-relative ("//"), an
-// in-page fragment ("#"), or a "mailto:" — are returned unchanged, which keeps
-// external menu and footer links intact. base_path is read from the context
-// stack the same way `format` reads date_format/timezone.
-rel_url :: proc(path: string, ctx: []any) -> string {
-	if path == "" {
-		return path
-	}
-	if strings.has_prefix(path, "http://") ||
-	   strings.has_prefix(path, "https://") ||
-	   strings.has_prefix(path, "//") ||
-	   strings.has_prefix(path, "#") ||
-	   strings.has_prefix(path, "mailto:") {
-		return path
-	}
-
-	base := ""
-	if raw := resolve_name("base_path", ctx); raw != nil {
-		if s, ok := reflect.as_string(raw); ok {
-			base = s
-		}
-	}
-	if base == "" {
-		return path
-	}
-
-	// Only prefix root-relative paths; leave document-relative ones alone.
-	if path[0] == '/' {
-		return fmt.tprintf("%s%s", base, path)
-	}
-	return path
-}
-
 apply_format :: proc(
 	iso: string,
 	args: []string,
@@ -463,40 +355,19 @@ apply_format :: proc(
 }

 // Groups preserve first-appearance order from the input list.
-apply_group_by :: proc(
-	value: any,
-	args: []string,
-	pos: int,
-	span: int,
-) -> (
-	result: any,
-	err: Error,
-) {
+apply_group_by :: proc(value: any, args: []string, pos: int) -> (result: any, err: Error) {
 	if len(args) != 1 {
-		hint := ""
-		if len(args) == 0 {
-			hint = "missing argument"
-		} else {
-			hint = "too many arguments"
-		}
 		return nil, Error_Body {
 			msg = fmt.tprintf("group_by expects 1 argument, got %d", len(args)),
 			pos = pos,
-			span = span,
 			kind = .Data,
-			hint = hint,
 		}
 	}
 	field := args[0]

 	elem_info, count, data := list_info(value)
 	if elem_info == nil {
-		return nil, Error_Body {
-			msg = "group_by expects a list",
-			pos = pos,
-			span = span,
-			kind = .Data,
-		}
+		return nil, Error_Body{msg = "group_by expects a list", pos = pos, kind = .Data}
 	}

 	groups := make([dynamic]Group, 0, 8, context.temp_allocator)
@@ -538,112 +409,3 @@ apply_group_by :: proc(

 	return groups, nil
 }
-
-take :: proc(
-	value: any,
-	args: []string,
-	from_start: bool,
-	tmpl: Template,
-	op_pos: int,
-	span: int,
-) -> (
-	any,
-	Error,
-) {
-	unwrapped, _ := base_value(value)
-	elem_info, count, data := list_info(unwrapped)
-	is_list := elem_info != nil
-	from_start := from_start
-	warning: Error = nil
-
-	str: string
-	if !is_list {
-		s, is_str := reflect.as_string(unwrapped)
-		if !is_str {
-			return nil, Error_Body {
-				msg = "requires a list or string",
-				pos = op_pos,
-				span = span,
-				kind = .Data,
-				source = tmpl.source,
-				path = tmpl.path,
-			}
-		}
-		str = s
-		count = len(str)
-	}
-
-	n := 1
-	if !is_list && len(args) == 0 {
-		warning = Error_Body {
-			msg    = "defaulting to n=1 for string",
-			pos    = op_pos,
-			span   = span,
-			kind   = .Warning,
-			source = tmpl.source,
-			path   = tmpl.path,
-			hint   = "consider specifying n explicitly",
-		}
-	}
-	if len(args) > 0 {
-		parsed, ok := strconv.parse_int(args[0])
-		if !ok {
-			return nil, Error_Body {
-				msg = fmt.tprintf("invalid argument '%s'", args[0]),
-				pos = op_pos,
-				span = span,
-				kind = .Data,
-				source = tmpl.source,
-				path = tmpl.path,
-			}
-		}
-		n = parsed
-	}
-
-	if n == 0 {
-		return nil, Error_Body {
-			msg = "argument must not be zero",
-			pos = op_pos,
-			span = span,
-			kind = .Data,
-			source = tmpl.source,
-			path = tmpl.path,
-		}
-	}
-
-	if n < 0 {
-		op_name := from_start ? "first" : "last"
-		alt_name := from_start ? "last" : "first"
-		warning = Error_Body {
-			msg    = fmt.tprintf("%s %d is equivalent to %s %d", op_name, n, alt_name, 0 - n),
-			pos    = op_pos,
-			span   = span,
-			kind   = .Warning,
-			source = tmpl.source,
-			path   = tmpl.path,
-			hint   = "consider using the clearer form",
-		}
-		from_start = !from_start
-		n = 0 - n
-	}
-
-	start: int
-	end: int
-	if from_start {
-		start = 0
-		end = min(count, n)
-	} else {
-		start = max(0, count - n)
-		end = count
-	}
-
-	if is_list {
-		result := make([dynamic]any, 0, max(0, end - start), context.temp_allocator)
-		for j in start ..< end {
-			append(&result, extract_list_element(elem_info, data, j))
-		}
-		return any{new_clone(result, context.temp_allocator), typeid_of([dynamic]any)}, warning
-	} else {
-		return any{new_clone(str[start:end], context.temp_allocator), typeid_of(string)}, warning
-	}
-}
@@ -1,4 +1,3 @@
-#+feature dynamic-literals
 #+test
 package mustache

@@ -568,11 +567,8 @@ test_format_mst_no_offset_no_timezone :: proc(t: ^testing.T) {

@(test)
 test_unknown_pipe_op_suggestion :: proc(t: ^testing.T) {
-	filter := Pipe_Filter {
-		op     = "formats",
-		op_pos = 0,
-	}
-	_, err := apply_filter("2026-01-15", &filter, 0, nil, {})
+	filter := Pipe_Filter{op = "formats", op_pos = 0}
+	_, err := apply_filter("2026-01-15", &filter, 0, nil)
 	testing.expect(t, err != nil, "should error on unknown op")
 	b := body(err)
 	testing.expect(t, strings.contains(b.hint, "format"), "hint should suggest 'format'")
@@ -581,146 +577,9 @@ test_unknown_pipe_op_suggestion :: proc(t: ^testing.T) {

@(test)
 test_unknown_pipe_op_no_suggestion :: proc(t: ^testing.T) {
-	filter := Pipe_Filter {
-		op     = "xyz",
-		op_pos = 0,
-	}
-	_, err := apply_filter("2026-01-15", &filter, 0, nil, {})
+	filter := Pipe_Filter{op = "xyz", op_pos = 0}
+	_, err := apply_filter("2026-01-15", &filter, 0, nil)
 	testing.expect(t, err != nil, "should error on unknown op")
 	b := body(err)
 	testing.expect(t, b.hint == "", "no suggestion expected for 'xyz'")
 }
-
-// Group 1: Lists
-
-First_Last_List_Data: map[string][5]string = {
-	"items" = [5]string{"foo", "bar", "bat", "qux", "quz"},
-}
-
-@(test)
-test_first_list_no_args_equals_first_one :: proc(t: ^testing.T) {
-	tpl_a, _ := parse(
-		"{{#items | first}}<li>{{.}}</li>{{/items}}",
-		"<test>",
-		allocator = context.temp_allocator,
-	)
-	result_a, _ := render(tpl_a, First_Last_List_Data, {}, context.temp_allocator)
-
-	tpl_b, _ := parse(
-		"{{#items | first 1}}<li>{{.}}</li>{{/items}}",
-		"<test>",
-		allocator = context.temp_allocator,
-	)
-	result_b, _ := render(tpl_b, First_Last_List_Data, {}, context.temp_allocator)
-
-	testing.expect_value(t, result_a, "<li>foo</li>")
-	testing.expect_value(t, result_a, result_b)
-}
-
-@(test)
-test_first_list_n :: proc(t: ^testing.T) {
-	data := First_Last_List_Data
-	tpl, _ := parse(
-		"{{#items | first 3}}<li>{{.}}</li>{{/items}}",
-		"<test>",
-		allocator = context.temp_allocator,
-	)
-	result, _ := render(tpl, data, {}, context.temp_allocator)
-	testing.expect_value(t, result, "<li>foo</li><li>bar</li><li>bat</li>")
-}
-
-@(test)
-test_first_list_negative_equals_last :: proc(t: ^testing.T) {
-	data := First_Last_List_Data
-	tpl, _ := parse(
-		"{{#items | first -2}}<li>{{.}}</li>{{/items}}",
-		"<test>",
-		allocator = context.temp_allocator,
-	)
-	result, _ := render(tpl, data, {}, context.temp_allocator)
-	testing.expect_value(t, result, "<li>qux</li><li>quz</li>")
-}
-
-@(test)
-test_last_list_no_args_equals_last_one :: proc(t: ^testing.T) {
-	data := First_Last_List_Data
-	tpl_a, _ := parse(
-		"{{#items | last}}<li>{{.}}</li>{{/items}}",
-		"<test>",
-		allocator = context.temp_allocator,
-	)
-	result_a, _ := render(tpl_a, data, {}, context.temp_allocator)
-
-	tpl_b, _ := parse(
-		"{{#items | last 1}}<li>{{.}}</li>{{/items}}",
-		"<test>",
-		allocator = context.temp_allocator,
-	)
-	result_b, _ := render(tpl_b, data, {}, context.temp_allocator)
-
-	testing.expect_value(t, result_a, "<li>quz</li>")
-	testing.expect_value(t, result_b, "<li>quz</li>")
-}
-
-@(test)
-test_last_list_n :: proc(t: ^testing.T) {
-	data := First_Last_List_Data
-	tpl, _ := parse(
-		"{{#items | last 3}}<li>{{.}}</li>{{/items}}",
-		"<test>",
-		allocator = context.temp_allocator,
-	)
-	result, _ := render(tpl, data, {}, context.temp_allocator)
-	testing.expect_value(t, result, "<li>bat</li><li>qux</li><li>quz</li>")
-}
-
-@(test)
-test_last_list_negative_equals_first :: proc(t: ^testing.T) {
-	data := First_Last_List_Data
-	tpl, _ := parse(
-		"{{#items | last -2}}<li>{{.}}</li>{{/items}}",
-		"<test>",
-		allocator = context.temp_allocator,
-	)
-	result, _ := render(tpl, data, {}, context.temp_allocator)
-	testing.expect_value(t, result, "<li>foo</li><li>bar</li>")
-}
-
-// Group 2: Strings
-
-First_Last_String_Data := map[string]string {
-	"word" = "Hello, World!",
-}
-
-@(test)
-test_first_string_n :: proc(t: ^testing.T) {
-	data := First_Last_String_Data
-	tpl, _ := parse("{{word | first 5}}", "<test>", allocator = context.temp_allocator)
-	result, _ := render(tpl, data, {}, context.temp_allocator)
-	testing.expect_value(t, result, "Hello")
-}
-
-@(test)
-test_last_string_n :: proc(t: ^testing.T) {
-	data := First_Last_String_Data
-	tpl, _ := parse("{{word | last 6}}", "<test>", allocator = context.temp_allocator)
-	result, _ := render(tpl, data, {}, context.temp_allocator)
-	testing.expect_value(t, result, "World!")
-}
-
-@(test)
-test_first_string_negative_equals_last :: proc(t: ^testing.T) {
-	data := First_Last_String_Data
-	tpl, _ := parse("{{word | first -6}}", "<test>", allocator = context.temp_allocator)
-	result, _ := render(tpl, data, {}, context.temp_allocator)
-	testing.expect_value(t, result, "World!")
-}
-
-@(test)
-test_last_string_negative_equals_first :: proc(t: ^testing.T) {
-	data := First_Last_String_Data
-	tpl, _ := parse("{{word | last -5}}", "<test>", allocator = context.temp_allocator)
-	result, _ := render(tpl, data, {}, context.temp_allocator)
-	testing.expect_value(t, result, "Hello")
-}
-
@@ -3,6 +3,7 @@ package mustache

 import "core:encoding/json"
 import "core:fmt"
+import "core:log"
 import "core:os"
 import "core:path/filepath"
 import "core:testing"
@@ -54,6 +55,10 @@ run_one_test :: proc(
 	test: json.Object,
 	name, template_src, expected: string,
 ) -> bool {
+	// Spec tests deliberately exercise missing keys, missing partials, etc.
+	// Silence the warnings to keep test output readable.
+	context.logger = log.nil_logger()
+
 	partials := make_map(map[string]Template, context.temp_allocator)

 	if "partials" in test {
@@ -79,7 +84,7 @@ run_one_test :: proc(
 	}
 	defer delete(tmpl.nodes)

-	result, rerr := render_silenced(tmpl, test["data"], partials)
+	result, rerr := render(tmpl, test["data"], partials)
 	if rerr != nil {
 		testing.expectf(t, false, "[%s] render error", name)
 		return false
--- a/Show More
+++ b/Show More
Author	SHA1	Message	Date
Spencer Brower	18733f31d1	feat: Errors are now tracked, so duplicate errors don't report more than once.	2026-07-31 10:15:33 -04:00
Spencer Brower	a4db4c2c58	chore: Updated TODOS.md.	2026-07-31 09:33:15 -04:00
Spencer Brower	9e93c012b6	feat: help for misspelled pipes is now shown to the user.	2026-07-31 09:25:38 -04:00
Spencer Brower	0d09db8d26	refactor: Filters are now read as an enum.	2026-07-31 09:24:01 -04:00
Spencer Brower	4948eb05a9	chore: Removed old plan file.	2026-07-31 08:57:36 -04:00
Spencer Brower	e71c2a94e5	fix: pipe Typos diagnostic now highlights only the broken pipe name.	2026-07-30 18:34:49 -04:00
Spencer Brower	ef7327cc52	fix: typos in pipe arguments (i.e. format) now show a proper diagnostic.	2026-07-30 18:22:53 -04:00
Spencer Brower	7b67201791	chore: Added diagnostics plan.	2026-07-30 14:55:09 -04:00
Spencer Brower	0307ab647f	fix: Fixed a bug with the weight checking.	2026-07-30 14:34:27 -04:00
Spencer Brower	9e19b0dc2c	build(nix): Updated prettier config.	2026-07-30 13:51:20 -04:00
Spencer Brower	6d97a23568	chore: Removed completed todos.	2026-07-30 13:20:58 -04:00
Spencer Brower	89c2ffbf78	style: Cleaned up `#partial` switches.	2026-07-30 11:12:31 -04:00
Spencer Brower	4659aa8dc2	feat: Added duplicate weights warning.	2026-07-30 11:07:31 -04:00
Spencer Brower	d03e3469b3	refactor: weights are now wrapped in a Maybe.	2026-07-30 09:58:53 -04:00
Spencer Brower	df14d76a85	feat: Added `weight` field to Page.	2026-07-29 15:20:35 -04:00
Spencer Brower	c48afafd8a	feat: Menu items get sorted by weight.	2026-07-29 15:15:02 -04:00
Spencer Brower	92d7b7525a	refactor: `Page.menu: string` is now `Page.menus: map[string]Menu_Entry	2026-07-29 15:15:02 -04:00
Spencer Brower	b54eb76495	nix: Fixed broken package build.	2026-07-29 15:15:02 -04:00
Spencer Brower	137b691a71	chore: Added PROBLEMS.md	2026-07-29 15:15:02 -04:00
Spencer Brower	06e411a329	chore: Updated TODOS.md	2026-07-29 15:15:02 -04:00
Spencer Brower	2299dcd652	style: Ran `typos`.	2026-07-29 15:15:01 -04:00
Spencer Brower	40bd2ad810	style: Ran nixpkgs-fmt.	2026-07-29 15:15:01 -04:00
Spencer Brower	eabb8a1430	style: ran deadnix.	2026-07-29 15:15:01 -04:00
Spencer Brower	c634e2e718	style: Ran `odinfmt`.	2026-07-29 15:15:01 -04:00
Spencer Brower	f839ec0f04	build(nix): Updated all flake inputs.	2026-07-29 15:15:01 -04:00
Spencer Brower	df0ed5a8f1	feat: Added menu configuration.	2026-07-29 15:15:01 -04:00
Spencer Brower	fb32ce90fc	chore: Updated `TODOS.md`	2026-07-28 17:03:59 -04:00
Spencer Brower	0abac89d88	feat: Warns the user if they exceed 16 context frames.	2026-07-28 16:58:46 -04:00
Spencer Brower	1824bae26c	docs: Updated README.md	2026-07-28 13:14:30 -04:00
Spencer Brower	5627215d6f	feat: Added `site` to the template context stack.	2026-07-28 13:07:14 -04:00
Spencer Brower	9d1954d70a	feat: page prop now falls through (implicit access).	2026-07-28 12:04:54 -04:00
Spencer Brower	fcf349eb51	fix: Removed `content` from Template_Context. now accessible only through page.content.	2026-07-28 11:45:55 -04:00
Spencer Brower	d51172c836	feat: Added `page` to template context.	2026-07-28 11:41:21 -04:00
				`@@ -1 +0,0 @@`
				`{{#site.title}}{{.}}{{#page.title}} \| {{/page.title}}{{/site.title}}{{page.title}}`
				`@@ -1 +0,0 @@`
				`<dl><dt>Term</dt><dd>Definition content</dd></dl>`
				`@@ -1 +0,0 @@`
				`<dl><dt>Markdown</dt><dd>A lightweight markup language</dd><dd>Created by John Gruber in 2004</dd><dt>HTML</dt><dd>A standard markup language used to create web pages</dd><dd>Used to create web pages</dd></dl>`
				`@@ -1 +0,0 @@`
				`<dl><dt>Outer Term</dt><dd>Outer Definition</dd><dt>Inner Term</dt><dd>Inner Definition</dd><dd>Another Inner Definition</dd></dl>`
				`@@ -1 +0,0 @@`
				<dl><dt>API</dt><dd>An application programming interface, a set of rules that allow different software applications to communicate with each other.</dd><dt>Framework</dt><dd>A software library that provides a standardized structure for application development.</dd><dt>Git</dt><dd>A distributed version control system used to track changes in a project development process.</dd><dt>IDE</dt><dd>An integrated development environment, a software application that includes a code editor and various development tools.</dd></dl>
				`@@ -1 +0,0 @@`
				<dl><dt>user_id</dt><dd><strong>Required</strong> - The unique identifier for the user.</dd><dd>Type: <code>integer</code></dd><dt>name</dt><dd><strong>Optional</strong> - The display name for the user.</dd><dd>Type: <code>string</code></dd><dd>Default: <code>null</code></dd><dt>status</dt><dd><strong>Optional</strong> - The user status.</dd><dd>Type: <code>string</code></dd><dd>Allowed values: <code>active</code>, <code>inactive</code>, <code>suspended</code></dd><dd>Default: <code>active</code></dd></dl>