R2-Explorer/packages/dashboard/src/parsers/markdown.js

/***   Regex Markdown Parser by chalarangelo   ***/
// Replaces 'regex' with 'replacement' in 'str'
// Curry function, usage: replaceRegex(regexVar, replacementVar) (strVar)
const replaceRegex = (regex, replacement) => (str) =>
	str.replace(regex, replacement);
// Regular expressions for Markdown (a bit strict, but they work)
const codeBlockRegex = /((\n\t)(.*))+/g;
const inlineCodeRegex = /(`)(.*?)\1/g;
const imageRegex = /!\[([^\[]+)\]\(([^\)]+)\)/g;
const linkRegex = /\[([^\[]+)\]\(([^\)]+)\)/g;
const headingRegex = /\n(#+\s*)(.*)/g;
const boldItalicsRegex = /(\*{1,2})(.*?)\1/g;
const strikethroughRegex = /(\~\~)(.*?)\1/g;
const blockquoteRegex = /\n(>|\>)(.*)/g;
const horizontalRuleRegex = /\n((\-{3,})|(={3,}))/g;
const unorderedListRegex = /(\n\s*(\-|\+)\s.*)+/g;
const orderedListRegex = /(\n\s*([0-9]+\.)\s.*)+/g;
const paragraphRegex =
	/\n+(?!<pre>)(?!<h)(?!<ul>)(?!<blockquote)(?!<hr)(?!\t)([^\n]+)\n/g;
// Replacer functions for Markdown
const codeBlockReplacer = (fullMatch) => `\n<pre>${fullMatch}</pre>`;
const inlineCodeReplacer = (fullMatch, tagStart, tagContents) =>
	`<code>${tagContents}</code>`;
const imageReplacer = (fullMatch, tagTitle, tagURL) =>
	`<img src="${tagURL}" alt="${tagTitle}" />`;
const linkReplacer = (fullMatch, tagTitle, tagURL) =>
	`<a href="${tagURL}">${tagTitle}</a>`;
const headingReplacer = (fullMatch, tagStart, tagContents) =>
	`\n<h${tagStart.trim().length}>${tagContents}</h${tagStart.trim().length}>`;
const boldItalicsReplacer = (fullMatch, tagStart, tagContents) =>
	`<${tagStart.trim().length === 1 ? "em" : "strong"}>${tagContents}</${tagStart.trim().length === 1 ? "em" : "strong"}>`;
const strikethroughReplacer = (fullMatch, tagStart, tagContents) =>
	`<del>${tagContents}</del>`;
const blockquoteReplacer = (fullMatch, tagStart, tagContents) =>
	`\n<blockquote>${tagContents}</blockquote>`;
const horizontalRuleReplacer = (fullMatch) => "\n<hr />";
const unorderedListReplacer = (fullMatch) => {
	let items = "";
	fullMatch
		.trim()
		.split("\n")
		.forEach((item) => {
			items += `<li>${item.substring(2)}</li>`;
		});
	return `\n<ul>${items}</ul>`;
};
const orderedListReplacer = (fullMatch) => {
	let items = "";
	fullMatch
		.trim()
		.split("\n")
		.forEach((item) => {
			items += `<li>${item.substring(item.indexOf(".") + 2)}</li>`;
		});
	return `\n<ol>${items}</ol>`;
};
const paragraphReplacer = (fullMatch, tagContents) => `<p>${tagContents}</p>`;
// Rules for Markdown parsing (use in order of appearance for best results)
const replaceCodeBlocks = replaceRegex(codeBlockRegex, codeBlockReplacer);
const replaceInlineCodes = replaceRegex(inlineCodeRegex, inlineCodeReplacer);
const replaceImages = replaceRegex(imageRegex, imageReplacer);
const replaceLinks = replaceRegex(linkRegex, linkReplacer);
const replaceHeadings = replaceRegex(headingRegex, headingReplacer);
const replaceBoldItalics = replaceRegex(boldItalicsRegex, boldItalicsReplacer);
const replaceceStrikethrough = replaceRegex(
	strikethroughRegex,
	strikethroughReplacer,
);
const replaceBlockquotes = replaceRegex(blockquoteRegex, blockquoteReplacer);
const replaceHorizontalRules = replaceRegex(
	horizontalRuleRegex,
	horizontalRuleReplacer,
);
const replaceUnorderedLists = replaceRegex(
	unorderedListRegex,
	unorderedListReplacer,
);
const replaceOrderedLists = replaceRegex(orderedListRegex, orderedListReplacer);
const replaceParagraphs = replaceRegex(paragraphRegex, paragraphReplacer);
// Fix for tab-indexed code blocks
const codeBlockFixRegex = /\n(<pre>)((\n|.)*)(<\/pre>)/g;
const codeBlockFixer = (
	fullMatch,
	tagStart,
	tagContents,
	lastMatch,
	tagEnd,
) => {
	let lines = "";
	tagContents.split("\n").forEach((line) => {
		lines += `${line.substring(1)}\n`;
	});
	return tagStart + lines + tagEnd;
};
const fixCodeBlocks = replaceRegex(codeBlockFixRegex, codeBlockFixer);
// Replacement rule order function for Markdown
// Do not use as-is, prefer parseMarkdown as seen below
const replaceMarkdown = (str) =>
	replaceParagraphs(
		replaceOrderedLists(
			replaceUnorderedLists(
				replaceHorizontalRules(
					replaceBlockquotes(
						replaceceStrikethrough(
							replaceBoldItalics(
								replaceHeadings(
									replaceLinks(
										replaceImages(replaceInlineCodes(replaceCodeBlocks(str))),
									),
								),
							),
						),
					),
				),
			),
		),
	);
// Parser for Markdown (fixes code, adds empty lines around for parsing)
// Usage: parseMarkdown(strVar)
export const parseMarkdown = (str) =>
	fixCodeBlocks(replaceMarkdown(`\n${str}\n`)).trim();