pantsdown 2.1.7 → 2.2.1

package/CLAUDE.md

@@ -0,0 +1,72 @@
+# CLAUDE.md
+
+This file provides guidance to Claude Code (claude.ai/code) when working with code in this repository.
+
+## Project Overview
+
+Pantsdown is a Markdown to HTML converter that renders markdown similar to GitHub's styling. It was built specifically for [github-preview.nvim](https://github.com/wallpants/github-preview.nvim). Based on [Marked](https://github.com/markedjs/marked).
+
+## Commands
+
+```bash
+# Type checking
+bun run typecheck
+
+# Linting
+bun run lint
+
+# Both typecheck and lint
+bun run check
+
+# Run tests (uses bun:test with happy-dom)
+bun test
+
+# Run a single test file
+bun test tests/parse.test.ts
+
+# Update test snapshots
+bun test --update-snapshots
+
+# Format code
+bun run format
+
+# Build docs
+bun run docs:build
+```
+
+## Architecture
+
+The parsing pipeline follows a classic compiler pattern:
+
+1. **Lexer** (`src/lexer.ts`) - Tokenizes markdown source into an array of tokens
+   - Uses `Tokenizer` for the actual token creation
+   - Processes block-level tokens first, then inline tokens
+   - Tracks source maps for line number references
+   - Collects footnotes separately
+
+2. **Tokenizer** (`src/tokenizer.ts`) - Creates tokens from markdown patterns
+   - Uses regex rules from `src/rules/block.ts` and `src/rules/inline.ts`
+
+3. **Parser** (`src/parser.ts`) - Converts tokens to HTML by dispatching to the Renderer
+   - Recursively processes nested tokens
+
+4. **Renderer** (`src/renderer.ts`) - Produces HTML output for each token type
+   - Uses highlight.js for syntax highlighting
+   - Uses github-slugger for heading anchors
+   - Handles special cases like mermaid diagrams and alerts
+
+Entry point: `Pantsdown` class in `src/pantsdown.ts` coordinates the pipeline.
+
+## Key Types
+
+All token types are defined in `src/types.ts`. The `Token` union type covers all possible markdown elements (headings, code blocks, lists, tables, footnotes, alerts, etc.).
+
+## Output
+
+`pantsdown.parse(markdown)` returns `{ html, javascript }`:
+- `html`: The rendered HTML string
+- `javascript`: A script for interactive features (task list checkboxes, copy buttons)
+
+## Styling
+
+CSS is in `src/css/styles.css`. Requires a parent element with classes `pantsdown light` or `pantsdown dark` (optionally with `high-contrast`).

package/package.json

@@ -2,7 +2,7 @@
   "name": "pantsdown",
   "author": "wallpants",
   "type": "module",
-  "version": "2.1.7",
+  "version": "2.2.1",
   "license": "MIT",
   "description": "Markdown to 'GitHub HTML' parser",
   "repository": {
@@ -51,21 +51,23 @@
   },
   "dependencies": {
     "github-slugger": "^2.0.0",
-    "highlight.js": "^11.11.1"
+    "highlight.js": "^11.11.1",
+    "katex": "^0.16.28"
   },
   "devDependencies": {
-    "@commitlint/config-conventional": "^19.6.0",
-    "@commitlint/cz-commitlint": "^19.6.1",
-    "@happy-dom/global-registrator": "^16.3.0",
-    "@types/bun": "^1.1.14",
-    "@typescript-eslint/eslint-plugin": "^8.19.0",
-    "@typescript-eslint/parser": "^8.19.0",
+    "@commitlint/config-conventional": "^19.8.1",
+    "@commitlint/cz-commitlint": "^19.8.1",
+    "@happy-dom/global-registrator": "^16.8.1",
+    "@types/bun": "^1.3.7",
+    "@types/katex": "^0.16.8",
+    "@typescript-eslint/eslint-plugin": "^8.54.0",
+    "@typescript-eslint/parser": "^8.54.0",
     "commitizen": "^4.3.1",
-    "commitlint": "^19.6.1",
+    "commitlint": "^19.8.1",
     "eslint": "^8.57.1",
-    "prettier": "^3.4.2",
-    "prettier-plugin-organize-imports": "^4.1.0",
-    "semantic-release": "^24.2.1",
-    "typescript": "5.6.3"
+    "prettier": "^3.8.1",
+    "prettier-plugin-organize-imports": "^4.3.0",
+    "semantic-release": "^24.2.9",
+    "typescript": "5.7.2"
   }
 }

package/src/css/styles.css

@@ -113,8 +113,9 @@
     margin: 0;
     color: var(--color-fg-default);
     background-color: var(--color-canvas-default);
-    font-family: -apple-system, BlinkMacSystemFont, "Segoe UI", "Noto Sans", Helvetica, Arial,
-        sans-serif, "Apple Color Emoji", "Segoe UI Emoji";
+    font-family:
+        -apple-system, BlinkMacSystemFont, "Segoe UI", "Noto Sans", Helvetica, Arial, sans-serif,
+        "Apple Color Emoji", "Segoe UI Emoji";
     font-size: 16px;
     line-height: 1.5;
     word-wrap: break-word;
@@ -1323,3 +1324,36 @@
 /* .pantsdown.dark .hljs-tag { */
 /*     /1* purposely ignored *1/ */
 /* } */
+
+/*
+* * * * * * * * * * * * * * * * * * * * * * * * * * * * *
+*                                                       *
+*                  KaTeX / LaTeX                        *
+*                                                       *
+* * * * * * * * * * * * * * * * * * * * * * * * * * * * *
+*/
+
+.pantsdown .katex-block {
+    display: block;
+    margin: 16px 0;
+    text-align: center;
+    overflow-x: auto;
+    overflow-y: hidden;
+}
+
+.pantsdown .katex-inline {
+    display: inline;
+}
+
+.pantsdown .katex-error {
+    color: var(--color-danger-fg);
+    background-color: var(--color-canvas-subtle);
+    padding: 0.2em 0.4em;
+    border-radius: 6px;
+}
+
+.pantsdown .katex-error code {
+    color: inherit;
+    background-color: transparent;
+    padding: 0;
+}

package/src/lexer.ts

@@ -111,6 +111,13 @@ export class Lexer {
                 continue;
             }
 
+            // latexBlock
+            if ((token = this.tokenizer.latexBlock(src))) {
+                src = src.substring(token.raw.length);
+                tokens.push(token);
+                continue;
+            }
+
             // fences
             if ((token = this.tokenizer.fences(src))) {
                 src = src.substring(token.raw.length);
@@ -169,8 +176,8 @@ export class Lexer {
                     lastToken.text += "\n" + token.raw;
                     const lastInline = this.inlineQueue[this.inlineQueue.length - 1];
                     if (lastInline) lastInline.src = lastToken.text;
-                } else if (!this.links[token.tag]) {
-                    this.links[token.tag] = {
+                } else {
+                    this.links[token.tag] ??= {
                         href: token.href,
                         title: token.title,
                     };
@@ -215,7 +222,7 @@ export class Lexer {
             if ((token = this.tokenizer.text(src))) {
                 src = src.substring(token.raw.length);
                 lastToken = tokens[tokens.length - 1];
-                if (lastToken && lastToken.type === "text") {
+                if (lastToken?.type === "text") {
                     lastToken.raw += "\n" + token.raw;
                     lastToken.text += "\n" + token.text;
                     this.inlineQueue.pop();
@@ -291,6 +298,13 @@ export class Lexer {
             }
             keepPrevChar = false;
 
+            // latexInline (before escape to handle \(...\) syntax)
+            if ((token = this.tokenizer.latexInline(src))) {
+                src = src.substring(token.raw.length);
+                tokens.push(token);
+                continue;
+            }
+
             // escape
             if ((token = this.tokenizer.escape(src))) {
                 src = src.substring(token.raw.length);
@@ -391,7 +405,7 @@ export class Lexer {
                 }
                 keepPrevChar = true;
                 lastToken = tokens[tokens.length - 1];
-                if (lastToken && lastToken.type === "text") {
+                if (lastToken?.type === "text") {
                     lastToken.raw += token.raw;
                     lastToken.text += token.text;
                 } else {

package/src/parser.ts

@@ -184,6 +184,10 @@ export class Parser {
                     out += top ? this.renderer.paragraph(body, textToken.sourceMap) : body;
                     continue;
                 }
+                case "latexBlock": {
+                    out += this.renderer.latexBlock(token.text, token.sourceMap);
+                    continue;
+                }
 
                 default: {
                     const errMsg = 'Token with "' + token.type + '" type was not found.';
@@ -253,6 +257,10 @@ export class Parser {
                     out += this.renderer.text(token.text);
                     break;
                 }
+                case "latexInline": {
+                    out += this.renderer.latexInline(token.text);
+                    break;
+                }
                 default: {
                     const errMsg = 'Token with "' + token.type + '" type was not found.';
                     throw new Error(errMsg);

package/src/renderer.ts

@@ -1,5 +1,6 @@
 import GithubSlugger from "github-slugger";
 import hljs from "highlight.js";
+import katex from "katex";
 import { type Pantsdown } from "./pantsdown.ts";
 import { inline } from "./rules/inline.ts";
 import { type HTMLAttrs, type SourceMap, type Tokens } from "./types.ts";
@@ -25,18 +26,24 @@ export class Renderer {
 
     code(code: string, infostring: string | undefined, sourceMap: SourceMap): string {
         const lang = /^\S*/.exec(infostring ?? "")?.[0];
-        code = code.replace(/\n$/, "") + "\n";
+        code = code.replace(/\n$/, "");
 
         const attrs: HTMLAttrs = [];
 
         if (lang === "mermaid") {
             attrs.push(["class", "mermaid-container mermaid"]);
-        } else {
-            const language = lang && hljs.getLanguage(lang) ? lang : "plaintext";
-            code = hljs.highlight(code, { language }).value;
-            code = `<code class="hljs language-${escape(language)}">${code}</code>`;
+            const result = `<pre style="position: relative;">` + code + "\n" + `</pre>`;
+            return injectHtmlAttributes(result, attrs, sourceMap);
         }
 
+        if (lang === "math") {
+            return this.latexBlock(code, sourceMap);
+        }
+
+        const language = lang && hljs.getLanguage(lang) ? lang : "plaintext";
+        code = hljs.highlight(code + "\n", { language }).value;
+        code = `<code class="hljs language-${escape(language)}">${code}</code>`;
+
         const result = `<pre style="position: relative;">` + code + `</pre>`;
         return injectHtmlAttributes(result, attrs, sourceMap);
     }
@@ -212,4 +219,40 @@ export class Renderer {
     text(text: string): string {
         return text;
     }
+
+    latexBlock(latex: string, sourceMap: SourceMap): string {
+        try {
+            const rendered = katex.renderToString(latex, {
+                displayMode: true,
+                throwOnError: false,
+                output: "html",
+                trust: false,
+            });
+            return injectHtmlAttributes(
+                `<div class="katex-block">${rendered}</div>\n`,
+                [],
+                sourceMap,
+            );
+        } catch {
+            return injectHtmlAttributes(
+                `<div class="katex-block katex-error"><code>${escape(latex)}</code></div>\n`,
+                [],
+                sourceMap,
+            );
+        }
+    }
+
+    latexInline(latex: string): string {
+        try {
+            const rendered = katex.renderToString(latex, {
+                displayMode: false,
+                throwOnError: false,
+                output: "html",
+                trust: false,
+            });
+            return `<span class="katex-inline">${rendered}</span>`;
+        } catch {
+            return `<span class="katex-inline katex-error"><code>${escape(latex)}</code></span>`;
+        }
+    }
 }

package/src/rules/block.ts

@@ -16,7 +16,8 @@ type BlockRuleNames =
     | "bullet"
     | "listItemStart"
     | "footnote"
-    | "paragraph";
+    | "paragraph"
+    | "latexBlock";
 
 export const label = /(?!\s*\])(?:\\.|[^\[\]\\])+/;
 
@@ -136,4 +137,5 @@ export const block: Record<BlockRuleNames, RegExp> = {
     paragraph: block_paragraph,
     footnote: /^\[\^([^\]\n]+)\]:(?:[ \t]+|[\n]*?|$)([^\n]*?(?:\n|$)(?:[\n]*?[ ]{4,}[^\n]*)*)/,
     text: /^[^\n]+/,
+    latexBlock: /^(?:\$\$([^$]*(?:\$(?!\$)[^$]*)*)\$\$|\\\[([\s\S]*?)\\\])/,
 };

package/src/rules/inline.ts

@@ -19,7 +19,8 @@ type InlineRuleNames =
     | "punctuation"
     | "blockSkip"
     | "footnoteRef"
-    | "backpedal";
+    | "backpedal"
+    | "latexInline";
 
 // list of unicode punctuation marks, plus any missing characters from CommonMark spec
 const punctuation = "\\p{P}\\p{S}";
@@ -112,7 +113,7 @@ const inline_backpedal =
 const inline_del = /^(~~?)(?=[^\s~])([\s\S]*?[^\s~])\1(?=[^~]|$)/;
 
 const inline_text =
-    /^([`~]+|[^`~])(?:(?= {2,}\n)|(?=[a-zA-Z0-9.!#$%&'*+\/=?_`{\|}~-]+@)|[\s\S]*?(?:(?=[\\<!\[`*~_]|\b_|https?:\/\/|ftp:\/\/|www\.|$)|[^ ](?= {2,}\n)|[^a-zA-Z0-9.!#$%&'*+\/=?_`{\|}~-](?=[a-zA-Z0-9.!#$%&'*+\/=?_`{\|}~-]+@)))/;
+    /^([`~]+|[^`~])(?:(?= {2,}\n)|(?=[a-zA-Z0-9.!#$%&'*+\/=?_`{\|}~-]+@)|[\s\S]*?(?:(?=[\\<!\[`*~_$]|\b_|https?:\/\/|ftp:\/\/|www\.|$)|[^ ](?= {2,}\n)|[^a-zA-Z0-9.!#$%&'*+\/=?_`{\|}~-](?=[a-zA-Z0-9.!#$%&'*+\/=?_`{\|}~-]+@)))/;
 
 const inline_url = edit(/^((?:ftp|https?):\/\/|www\.)(?:[a-zA-Z0-9\-]+\.?)+[^\s<]*|^email/, "i")
     .replace("email", extended_email)
@@ -139,4 +140,5 @@ export const inline: Omit<Record<InlineRuleNames, RegExp>, "emStrong"> & {
     blockSkip: inline_blockSkip,
     backpedal: inline_backpedal,
     footnoteRef: /^\[\^([^\]\n]+)\]/,
+    latexInline: /^(?:\$(?!\$)([^\s$](?:[^$\n]*[^\s$])?)\$(?!\$)|\\\((.+?)\\\))/,
 };

package/src/tokenizer.ts

@@ -716,6 +716,7 @@ export class Tokenizer {
 
         if (!nextChar || !prevChar || inline.punctuation.exec(prevChar)) {
             // unicode Regex counts emoji as 1 char; spread into array for proper count (used multiple times below)
+            // eslint-disable-next-line @typescript-eslint/no-misused-spread
             const lLength = [...match[0]].length - 1;
             let rDelim,
                 rLength,
@@ -736,6 +737,7 @@ export class Tokenizer {
 
                 if (!rDelim) continue; // skip single * in __abc*abc__
 
+                // eslint-disable-next-line @typescript-eslint/no-misused-spread
                 rLength = [...rDelim].length;
 
                 if (match[3] || match[4]) {
@@ -757,6 +759,7 @@ export class Tokenizer {
                 // Remove extra characters. *a*** -> *a*
                 rLength = Math.min(rLength, rLength + delimTotal + midDelimTotal);
                 // char length can be >1 for unicode characters;
+                // eslint-disable-next-line @typescript-eslint/no-misused-spread
                 const lastCharLength = [...match[0]][0]!.length;
                 const raw = src.slice(0, lLength + match.index + lastCharLength + rLength);
 
@@ -920,4 +923,33 @@ export class Tokenizer {
             text,
         };
     }
+
+    latexBlock(src: string): Tokens["LatexBlock"] | undefined {
+        const cap = block.latexBlock.exec(src);
+        if (!cap) return undefined;
+
+        // cap[1] is from $$...$$ syntax, cap[2] is from \[...\] syntax
+        const text = cap[1] ?? cap[2] ?? "";
+
+        return {
+            type: "latexBlock",
+            raw: cap[0],
+            text: text.trim(),
+            sourceMap: this.lexer.getSourceMap(cap[0]),
+        };
+    }
+
+    latexInline(src: string): Tokens["LatexInline"] | undefined {
+        const cap = inline.latexInline.exec(src);
+        if (!cap) return undefined;
+
+        // cap[1] is from $...$ syntax, cap[2] is from \(...\) syntax
+        const text = cap[1] ?? cap[2] ?? "";
+
+        return {
+            type: "latexInline",
+            raw: cap[0],
+            text,
+        };
+    }
 }

package/src/types.ts

@@ -186,6 +186,17 @@ export interface Tokens extends Record<string, BaseToken> {
         raw: string;
         items: Tokens["Footnote"][];
     };
+    LatexBlock: {
+        type: "latexBlock";
+        raw: string;
+        text: string;
+        sourceMap: SourceMap;
+    };
+    LatexInline: {
+        type: "latexInline";
+        raw: string;
+        text: string;
+    };
 }
 
 export type Token =
@@ -214,7 +225,9 @@ export type Token =
     | Tokens["Alert"]
     | Tokens["Footnote"]
     | Tokens["FootnoteRef"]
-    | Tokens["Footnotes"];
+    | Tokens["Footnotes"]
+    | Tokens["LatexBlock"]
+    | Tokens["LatexInline"];
 
 export type Links = Record<string, { href: string; title: string }>;