marked 14.0.0 → 14.1.1

package/lib/marked.d.cts

@@ -271,54 +271,6 @@ declare const inlineNormal: {
 	url: RegExp;
 };
 export type InlineKeys = keyof typeof inlineNormal;
-/**
- * exports
- */
-export declare const block: {
-	normal: {
-		blockquote: RegExp;
-		code: RegExp;
-		def: RegExp;
-		fences: RegExp;
-		heading: RegExp;
-		hr: RegExp;
-		html: RegExp;
-		lheading: RegExp;
-		list: RegExp;
-		newline: RegExp;
-		paragraph: RegExp;
-		table: RegExp;
-		text: RegExp;
-	};
-	gfm: Record<"code" | "blockquote" | "hr" | "html" | "table" | "text" | "heading" | "list" | "paragraph" | "def" | "fences" | "lheading" | "newline", RegExp>;
-	pedantic: Record<"code" | "blockquote" | "hr" | "html" | "table" | "text" | "heading" | "list" | "paragraph" | "def" | "fences" | "lheading" | "newline", RegExp>;
-};
-export declare const inline: {
-	normal: {
-		_backpedal: RegExp;
-		anyPunctuation: RegExp;
-		autolink: RegExp;
-		blockSkip: RegExp;
-		br: RegExp;
-		code: RegExp;
-		del: RegExp;
-		emStrongLDelim: RegExp;
-		emStrongRDelimAst: RegExp;
-		emStrongRDelimUnd: RegExp;
-		escape: RegExp;
-		link: RegExp;
-		nolink: RegExp;
-		punctuation: RegExp;
-		reflink: RegExp;
-		reflinkSearch: RegExp;
-		tag: RegExp;
-		text: RegExp;
-		url: RegExp;
-	};
-	gfm: Record<"link" | "code" | "url" | "br" | "del" | "text" | "escape" | "tag" | "reflink" | "autolink" | "nolink" | "_backpedal" | "anyPunctuation" | "blockSkip" | "emStrongLDelim" | "emStrongRDelimAst" | "emStrongRDelimUnd" | "punctuation" | "reflinkSearch", RegExp>;
-	breaks: Record<"link" | "code" | "url" | "br" | "del" | "text" | "escape" | "tag" | "reflink" | "autolink" | "nolink" | "_backpedal" | "anyPunctuation" | "blockSkip" | "emStrongLDelim" | "emStrongRDelimAst" | "emStrongRDelimUnd" | "punctuation" | "reflinkSearch", RegExp>;
-	pedantic: Record<"link" | "code" | "url" | "br" | "del" | "text" | "escape" | "tag" | "reflink" | "autolink" | "nolink" | "_backpedal" | "anyPunctuation" | "blockSkip" | "emStrongLDelim" | "emStrongRDelimAst" | "emStrongRDelimUnd" | "punctuation" | "reflinkSearch", RegExp>;
-};
 export interface Rules {
 	block: Record<BlockKeys, RegExp>;
 	inline: Record<InlineKeys, RegExp>;
@@ -358,6 +310,7 @@ declare class _Tokenizer {
 }
 declare class _Hooks {
 	options: MarkedOptions;
+	block: boolean | undefined;
 	constructor(options?: MarkedOptions);
 	static passThroughHooks: Set<string>;
 	/**
@@ -372,6 +325,14 @@ declare class _Hooks {
 	 * Process all tokens before walk tokens
 	 */
 	processAllTokens(tokens: Token[] | TokensList): Token[] | TokensList;
+	/**
+	 * Provide function to tokenize markdown
+	 */
+	provideLexer(): typeof _Lexer.lexInline;
+	/**
+	 * Provide function to parse tokens
+	 */
+	provideParser(): typeof _Parser.parse;
 }
 export interface TokenizerThis {
 	lexer: _Lexer;
@@ -394,7 +355,7 @@ export interface RendererExtension {
 	renderer: RendererExtensionFunction;
 }
 export type TokenizerAndRendererExtension = TokenizerExtension | RendererExtension | (TokenizerExtension & RendererExtension);
-export type HooksApi = Omit<_Hooks, "constructor" | "options">;
+export type HooksApi = Omit<_Hooks, "constructor" | "options" | "block">;
 export type HooksObject = {
 	[K in keyof HooksApi]?: (this: _Hooks, ...args: Parameters<HooksApi[K]>) => ReturnType<HooksApi[K]> | Promise<ReturnType<HooksApi[K]>>;
 };
@@ -428,6 +389,8 @@ export interface MarkedExtension {
 	 * preprocess is called to process markdown before sending it to marked.
 	 * processAllTokens is called with the TokensList before walkTokens.
 	 * postprocess is called to process html after marked has finished parsing.
+	 * provideLexer is called to provide a function to tokenize markdown.
+	 * provideParser is called to provide a function to parse tokens.
 	 */
 	hooks?: HooksObject | undefined | null;
 	/**

package/lib/marked.d.ts

@@ -271,54 +271,6 @@ declare const inlineNormal: {
 	url: RegExp;
 };
 export type InlineKeys = keyof typeof inlineNormal;
-/**
- * exports
- */
-export declare const block: {
-	normal: {
-		blockquote: RegExp;
-		code: RegExp;
-		def: RegExp;
-		fences: RegExp;
-		heading: RegExp;
-		hr: RegExp;
-		html: RegExp;
-		lheading: RegExp;
-		list: RegExp;
-		newline: RegExp;
-		paragraph: RegExp;
-		table: RegExp;
-		text: RegExp;
-	};
-	gfm: Record<"code" | "blockquote" | "hr" | "html" | "table" | "text" | "heading" | "list" | "paragraph" | "def" | "fences" | "lheading" | "newline", RegExp>;
-	pedantic: Record<"code" | "blockquote" | "hr" | "html" | "table" | "text" | "heading" | "list" | "paragraph" | "def" | "fences" | "lheading" | "newline", RegExp>;
-};
-export declare const inline: {
-	normal: {
-		_backpedal: RegExp;
-		anyPunctuation: RegExp;
-		autolink: RegExp;
-		blockSkip: RegExp;
-		br: RegExp;
-		code: RegExp;
-		del: RegExp;
-		emStrongLDelim: RegExp;
-		emStrongRDelimAst: RegExp;
-		emStrongRDelimUnd: RegExp;
-		escape: RegExp;
-		link: RegExp;
-		nolink: RegExp;
-		punctuation: RegExp;
-		reflink: RegExp;
-		reflinkSearch: RegExp;
-		tag: RegExp;
-		text: RegExp;
-		url: RegExp;
-	};
-	gfm: Record<"link" | "code" | "url" | "br" | "del" | "text" | "escape" | "tag" | "reflink" | "autolink" | "nolink" | "_backpedal" | "anyPunctuation" | "blockSkip" | "emStrongLDelim" | "emStrongRDelimAst" | "emStrongRDelimUnd" | "punctuation" | "reflinkSearch", RegExp>;
-	breaks: Record<"link" | "code" | "url" | "br" | "del" | "text" | "escape" | "tag" | "reflink" | "autolink" | "nolink" | "_backpedal" | "anyPunctuation" | "blockSkip" | "emStrongLDelim" | "emStrongRDelimAst" | "emStrongRDelimUnd" | "punctuation" | "reflinkSearch", RegExp>;
-	pedantic: Record<"link" | "code" | "url" | "br" | "del" | "text" | "escape" | "tag" | "reflink" | "autolink" | "nolink" | "_backpedal" | "anyPunctuation" | "blockSkip" | "emStrongLDelim" | "emStrongRDelimAst" | "emStrongRDelimUnd" | "punctuation" | "reflinkSearch", RegExp>;
-};
 export interface Rules {
 	block: Record<BlockKeys, RegExp>;
 	inline: Record<InlineKeys, RegExp>;
@@ -358,6 +310,7 @@ declare class _Tokenizer {
 }
 declare class _Hooks {
 	options: MarkedOptions;
+	block: boolean | undefined;
 	constructor(options?: MarkedOptions);
 	static passThroughHooks: Set<string>;
 	/**
@@ -372,6 +325,14 @@ declare class _Hooks {
 	 * Process all tokens before walk tokens
 	 */
 	processAllTokens(tokens: Token[] | TokensList): Token[] | TokensList;
+	/**
+	 * Provide function to tokenize markdown
+	 */
+	provideLexer(): typeof _Lexer.lexInline;
+	/**
+	 * Provide function to parse tokens
+	 */
+	provideParser(): typeof _Parser.parse;
 }
 export interface TokenizerThis {
 	lexer: _Lexer;
@@ -394,7 +355,7 @@ export interface RendererExtension {
 	renderer: RendererExtensionFunction;
 }
 export type TokenizerAndRendererExtension = TokenizerExtension | RendererExtension | (TokenizerExtension & RendererExtension);
-export type HooksApi = Omit<_Hooks, "constructor" | "options">;
+export type HooksApi = Omit<_Hooks, "constructor" | "options" | "block">;
 export type HooksObject = {
 	[K in keyof HooksApi]?: (this: _Hooks, ...args: Parameters<HooksApi[K]>) => ReturnType<HooksApi[K]> | Promise<ReturnType<HooksApi[K]>>;
 };
@@ -428,6 +389,8 @@ export interface MarkedExtension {
 	 * preprocess is called to process markdown before sending it to marked.
 	 * processAllTokens is called with the TokensList before walkTokens.
 	 * postprocess is called to process html after marked has finished parsing.
+	 * provideLexer is called to provide a function to tokenize markdown.
+	 * provideParser is called to provide a function to parse tokens.
 	 */
 	hooks?: HooksObject | undefined | null;
 	/**

package/lib/marked.esm.js

@@ -1,5 +1,5 @@
 /**
- * marked v14.0.0 - a markdown parser
+ * marked v14.1.1 - a markdown parser
  * Copyright (c) 2011-2024, Christopher Jeffrey. (MIT Licensed)
  * https://github.com/markedjs/marked
  */
@@ -247,7 +247,7 @@ class _Tokenizer {
     code(src) {
         const cap = this.rules.block.code.exec(src);
         if (cap) {
-            const text = cap[0].replace(/^ {1,4}/gm, '');
+            const text = cap[0].replace(/^(?: {1,4}| {0,3}\t)/gm, '');
             return {
                 type: 'code',
                 raw: cap[0],
@@ -431,7 +431,7 @@ class _Tokenizer {
                     itemContents = line.slice(indent);
                     indent += cap[1].length;
                 }
-                if (blankLine && /^ *$/.test(nextLine)) { // Items begin with at most one blank line
+                if (blankLine && /^[ \t]*$/.test(nextLine)) { // Items begin with at most one blank line
                     raw += nextLine + '\n';
                     src = src.substring(nextLine.length + 1);
                     endEarly = true;
@@ -444,10 +444,15 @@ class _Tokenizer {
                     // Check if following lines should be included in List Item
                     while (src) {
                         const rawLine = src.split('\n', 1)[0];
+                        let nextLineWithoutTabs;
                         nextLine = rawLine;
                         // Re-align to follow commonmark nesting rules
                         if (this.options.pedantic) {
                             nextLine = nextLine.replace(/^ {1,4}(?=( {4})*[^ ])/g, '  ');
+                            nextLineWithoutTabs = nextLine;
+                        }
+                        else {
+                            nextLineWithoutTabs = nextLine.replace(/\t/g, '    ');
                         }
                         // End list item if found code fences
                         if (fencesBeginRegex.test(nextLine)) {
@@ -462,11 +467,11 @@ class _Tokenizer {
                             break;
                         }
                         // Horizontal rule found
-                        if (hrRegex.test(src)) {
+                        if (hrRegex.test(nextLine)) {
                             break;
                         }
-                        if (nextLine.search(/[^ ]/) >= indent || !nextLine.trim()) { // Dedent if possible
-                            itemContents += '\n' + nextLine.slice(indent);
+                        if (nextLineWithoutTabs.search(/[^ ]/) >= indent || !nextLine.trim()) { // Dedent if possible
+                            itemContents += '\n' + nextLineWithoutTabs.slice(indent);
                         }
                         else {
                             // not enough indentation
@@ -474,7 +479,7 @@ class _Tokenizer {
                                 break;
                             }
                             // paragraph continuation unless last line was a different block level element
-                            if (line.search(/[^ ]/) >= 4) { // indented code block
+                            if (line.replace(/\t/g, '    ').search(/[^ ]/) >= 4) { // indented code block
                                 break;
                             }
                             if (fencesBeginRegex.test(line)) {
@@ -493,7 +498,7 @@ class _Tokenizer {
                         }
                         raw += rawLine + '\n';
                         src = src.substring(rawLine.length + 1);
-                        line = nextLine.slice(indent);
+                        line = nextLineWithoutTabs.slice(indent);
                     }
                 }
                 if (!list.loose) {
@@ -501,7 +506,7 @@ class _Tokenizer {
                     if (endsWithBlankLine) {
                         list.loose = true;
                     }
-                    else if (/\n *\n *$/.test(raw)) {
+                    else if (/\n[ \t]*\n[ \t]*$/.test(raw)) {
                         endsWithBlankLine = true;
                     }
                 }
@@ -963,15 +968,15 @@ class _Tokenizer {
 /**
  * Block-Level Grammar
  */
-const newline = /^(?: *(?:\n|$))+/;
-const blockCode = /^( {4}[^\n]+(?:\n(?: *(?:\n|$))*)?)+/;
+const newline = /^(?:[ \t]*(?:\n|$))+/;
+const blockCode = /^((?: {4}| {0,3}\t)[^\n]+(?:\n(?:[ \t]*(?:\n|$))*)?)+/;
 const fences = /^ {0,3}(`{3,}(?=[^`\n]*(?:\n|$))|~{3,})([^\n]*)(?:\n|$)(?:|([\s\S]*?)(?:\n|$))(?: {0,3}\1[~`]* *(?=\n|$)|$)/;
 const hr = /^ {0,3}((?:-[\t ]*){3,}|(?:_[ \t]*){3,}|(?:\*[ \t]*){3,})(?:\n+|$)/;
 const heading = /^ {0,3}(#{1,6})(?=\s|$)(.*)(?:\n+|$)/;
 const bullet = /(?:[*+-]|\d{1,9}[.)])/;
 const lheading = edit(/^(?!bull |blockCode|fences|blockquote|heading|html)((?:.|\n(?!\s*?\n|bull |blockCode|fences|blockquote|heading|html))+?)\n {0,3}(=+|-+) *(?:\n+|$)/)
     .replace(/bull/g, bullet) // lists can interrupt
-    .replace(/blockCode/g, / {4}/) // indented code blocks can interrupt
+    .replace(/blockCode/g, /(?: {4}| {0,3}\t)/) // indented code blocks can interrupt
     .replace(/fences/g, / {0,3}(?:`{3,}|~{3,})/) // fenced code blocks can interrupt
     .replace(/blockquote/g, / {0,3}>/) // blockquote can interrupt
     .replace(/heading/g, / {0,3}#{1,6}/) // ATX heading can interrupt
@@ -980,7 +985,7 @@ const lheading = edit(/^(?!bull |blockCode|fences|blockquote|heading|html)((?:.|
 const _paragraph = /^([^\n]+(?:\n(?!hr|heading|lheading|blockquote|fences|list|html|table| +\n)[^\n]+)*)/;
 const blockText = /^[^\n]+/;
 const _blockLabel = /(?!\s*\])(?:\\.|[^\[\]\\])+/;
-const def = edit(/^ {0,3}\[(label)\]: *(?:\n *)?([^<\s][^\s]*|<.*?>)(?:(?: +(?:\n *)?| *\n *)(title))? *(?:\n+|$)/)
+const def = edit(/^ {0,3}\[(label)\]: *(?:\n[ \t]*)?([^<\s][^\s]*|<.*?>)(?:(?: +(?:\n[ \t]*)?| *\n[ \t]*)(title))? *(?:\n+|$)/)
     .replace('label', _blockLabel)
     .replace('title', /(?:"(?:\\"?|[^"\\])*"|'[^'\n]*(?:\n[^'\n]+)*\n?'|\([^()]*\))/)
     .getRegex();
@@ -1000,9 +1005,9 @@ const html = edit('^ {0,3}(?:' // optional indentation
     + '|<\\?[\\s\\S]*?(?:\\?>\\n*|$)' // (3)
     + '|<![A-Z][\\s\\S]*?(?:>\\n*|$)' // (4)
     + '|<!\\[CDATA\\[[\\s\\S]*?(?:\\]\\]>\\n*|$)' // (5)
-    + '|</?(tag)(?: +|\\n|/?>)[\\s\\S]*?(?:(?:\\n *)+\\n|$)' // (6)
-    + '|<(?!script|pre|style|textarea)([a-z][\\w-]*)(?:attribute)*? */?>(?=[ \\t]*(?:\\n|$))[\\s\\S]*?(?:(?:\\n *)+\\n|$)' // (7) open tag
-    + '|</(?!script|pre|style|textarea)[a-z][\\w-]*\\s*>(?=[ \\t]*(?:\\n|$))[\\s\\S]*?(?:(?:\\n *)+\\n|$)' // (7) closing tag
+    + '|</?(tag)(?: +|\\n|/?>)[\\s\\S]*?(?:(?:\\n[ \t]*)+\\n|$)' // (6)
+    + '|<(?!script|pre|style|textarea)([a-z][\\w-]*)(?:attribute)*? */?>(?=[ \\t]*(?:\\n|$))[\\s\\S]*?(?:(?:\\n[ \t]*)+\\n|$)' // (7) open tag
+    + '|</(?!script|pre|style|textarea)[a-z][\\w-]*\\s*>(?=[ \\t]*(?:\\n|$))[\\s\\S]*?(?:(?:\\n[ \t]*)+\\n|$)' // (7) closing tag
     + ')', 'i')
     .replace('comment', _comment)
     .replace('tag', _tag)
@@ -1049,7 +1054,7 @@ const gfmTable = edit('^ *([^\\n ].*)\\n' // Header
     .replace('hr', hr)
     .replace('heading', ' {0,3}#{1,6}(?:\\s|$)')
     .replace('blockquote', ' {0,3}>')
-    .replace('code', ' {4}[^\\n]')
+    .replace('code', '(?: {4}| {0,3}\t)[^\\n]')
     .replace('fences', ' {0,3}(?:`{3,}(?=[^`\\n]*\\n)|~{3,})[^\\n]*\\n')
     .replace('list', ' {0,3}(?:[*+-]|1[.)]) ') // only lists starting from 1 can interrupt
     .replace('html', '</?(?:tag)(?: +|\\n|/?>)|<(?:script|pre|style|textarea|!--)')
@@ -1329,11 +1334,6 @@ class _Lexer {
         if (this.options.pedantic) {
             src = src.replace(/\t/g, '    ').replace(/^ +$/gm, '');
         }
-        else {
-            src = src.replace(/^( *)(\t+)/gm, (_, leading, tabs) => {
-                return leading + '    '.repeat(tabs.length);
-            });
-        }
         let token;
         let lastToken;
         let cutSrc;
@@ -2076,6 +2076,7 @@ class _Parser {
 
 class _Hooks {
     options;
+    block;
     constructor(options) {
         this.options = options || _defaults;
     }
@@ -2102,13 +2103,25 @@ class _Hooks {
     processAllTokens(tokens) {
         return tokens;
     }
+    /**
+     * Provide function to tokenize markdown
+     */
+    provideLexer() {
+        return this.block ? _Lexer.lex : _Lexer.lexInline;
+    }
+    /**
+     * Provide function to parse tokens
+     */
+    provideParser() {
+        return this.block ? _Parser.parse : _Parser.parseInline;
+    }
 }
 
 class Marked {
     defaults = _getDefaults();
     options = this.setOptions;
-    parse = this.parseMarkdown(_Lexer.lex, _Parser.parse);
-    parseInline = this.parseMarkdown(_Lexer.lexInline, _Parser.parseInline);
+    parse = this.parseMarkdown(true);
+    parseInline = this.parseMarkdown(false);
     Parser = _Parser;
     Renderer = _Renderer;
     TextRenderer = _TextRenderer;
@@ -2281,8 +2294,8 @@ class Marked {
                     if (!(prop in hooks)) {
                         throw new Error(`hook '${prop}' does not exist`);
                     }
-                    if (prop === 'options') {
-                        // ignore options property
+                    if (['options', 'block'].includes(prop)) {
+                        // ignore options and block properties
                         continue;
                     }
                     const hooksProp = prop;
@@ -2340,7 +2353,7 @@ class Marked {
     parser(tokens, options) {
         return _Parser.parse(tokens, options ?? this.defaults);
     }
-    parseMarkdown(lexer, parser) {
+    parseMarkdown(blockType) {
         // eslint-disable-next-line @typescript-eslint/no-explicit-any
         const parse = (src, options) => {
             const origOpt = { ...options };
@@ -2360,7 +2373,10 @@ class Marked {
             }
             if (opt.hooks) {
                 opt.hooks.options = opt;
+                opt.hooks.block = blockType;
             }
+            const lexer = opt.hooks ? opt.hooks.provideLexer() : (blockType ? _Lexer.lex : _Lexer.lexInline);
+            const parser = opt.hooks ? opt.hooks.provideParser() : (blockType ? _Parser.parse : _Parser.parseInline);
             if (opt.async) {
                 return Promise.resolve(opt.hooks ? opt.hooks.preprocess(src) : src)
                     .then(src => lexer(src, opt))