| 1 | // Utility function to wrap text in HTML tags |
| 1 | 2 | function wrap(text, tag) { |
| 2 | 3 | return `<${tag}>${text}</${tag}>`; |
| 3 | 4 | } |
| 4 | 5 | |
| 5 | | function isTag(text, tag) { |
| 6 | | return text.startsWith(`<${tag}>`); |
| 7 | | } |
| 8 | | |
| 9 | | function parser(markdown, delimiter, tag) { |
| 6 | // Generic parser for inline markdown formatting (bold, italic, etc.) |
| 7 | function parseInlineFormatting(markdown, delimiter, tag) { |
| 10 | 8 | const pattern = new RegExp(`${delimiter}(.+)${delimiter}`); |
| 11 | 9 | const replacement = `<${tag}>$1</${tag}>`; |
| 12 | 10 | return markdown.replace(pattern, replacement); |
| 13 | 11 | } |
| 14 | 12 | |
| 15 | | function parse__(markdown) { |
| 16 | | return parser(markdown, '__', 'strong'); |
| 13 | // Parse bold text using __ delimiter |
| 14 | function parseBold(markdown) { |
| 15 | return parseInlineFormatting(markdown, '__', 'strong'); |
| 17 | 16 | } |
| 18 | 17 | |
| 19 | | function parse_(markdown) { |
| 20 | | return parser(markdown, '_', 'em'); |
| 18 | // Parse italic text using _ delimiter |
| 19 | function parseItalic(markdown) { |
| 20 | return parseInlineFormatting(markdown, '_', 'em'); |
| 21 | 21 | } |
| 22 | 22 | |
| 23 | | function parseText(markdown, list) { |
| 24 | | const parsedText = parse_(parse__(markdown)); |
| 25 | | if (list) { |
| 26 | | return parsedText; |
| 27 | | } else { |
| 28 | | return wrap(parsedText, 'p'); |
| 29 | | } |
| 23 | // Parse inline formatting (bold and italic) within text |
| 24 | function parseInlineText(markdown, isListItem = false) { |
| 25 | const parsedText = parseItalic(parseBold(markdown)); |
| 26 | return isListItem ? parsedText : wrap(parsedText, 'p'); |
| 30 | 27 | } |
| 31 | 28 | |
| 32 | | function parseHeader(markdown, list) { |
| 33 | | let count = 0; |
| 34 | | for (let i = 0; i < markdown.length; i++) { |
| 35 | | if (markdown[i] === '#') { |
| 36 | | count += 1; |
| 29 | // Parse markdown headers (# to ######) |
| 30 | function parseHeader(markdown, isInList) { |
| 31 | // Count leading # characters |
| 32 | let headerLevel = 0; |
| 33 | for (const char of markdown) { |
| 34 | if (char === '#') { |
| 35 | headerLevel++; |
| 37 | 36 | } else { |
| 38 | 37 | break; |
| 39 | 38 | } |
| 40 | 39 | } |
| 41 | | if (count === 0 || count > 6) { |
| 42 | | return [null, list]; |
| 40 | |
| 41 | // Valid headers are 1-6 # characters |
| 42 | if (headerLevel === 0 || headerLevel > 6) { |
| 43 | return [null, isInList]; |
| 43 | 44 | } |
| 44 | | const headerTag = `h${count}`; |
| 45 | | const headerHtml = wrap(markdown.substring(count + 1), headerTag); |
| 46 | | if (list) { |
| 45 | |
| 46 | const headerTag = `h${headerLevel}`; |
| 47 | const headerContent = markdown.substring(headerLevel + 1).trim(); |
| 48 | const headerHtml = wrap(headerContent, headerTag); |
| 49 | |
| 50 | // Close any open list before starting header |
| 51 | if (isInList) { |
| 47 | 52 | return [`</ul>${headerHtml}`, false]; |
| 48 | | } else { |
| 49 | | return [headerHtml, false]; |
| 50 | 53 | } |
| 54 | |
| 55 | return [headerHtml, false]; |
| 51 | 56 | } |
| 52 | 57 | |
| 53 | | function parseLineItem(markdown, list) { |
| 54 | | if (markdown.startsWith('*')) { |
| 55 | | const innerHtml = wrap(parseText(markdown.substring(2), true), 'li'); |
| 56 | | if (list) { |
| 57 | | return [innerHtml, true]; |
| 58 | | } else { |
| 59 | | return [`<ul>${innerHtml}`, true]; |
| 60 | | } |
| 58 | // Parse list items starting with * |
| 59 | function parseListItem(markdown, isInList) { |
| 60 | if (!markdown.startsWith('*')) { |
| 61 | return [null, isInList]; |
| 62 | } |
| 63 | |
| 64 | const itemContent = markdown.substring(2); // Remove '* ' |
| 65 | const parsedItemContent = parseInlineText(itemContent, true); |
| 66 | const listItemHtml = wrap(parsedItemContent, 'li'); |
| 67 | |
| 68 | if (isInList) { |
| 69 | return [listItemHtml, true]; |
| 61 | 70 | } |
| 62 | | return [null, list]; |
| 71 | |
| 72 | // Start a new list |
| 73 | return [`<ul>${listItemHtml}`, true]; |
| 63 | 74 | } |
| 64 | 75 | |
| 65 | | function parseParagraph(markdown, list) { |
| 66 | | if (!list) { |
| 67 | | return [parseText(markdown, false), false]; |
| 68 | | } else { |
| 69 | | return [`</ul>${parseText(markdown, false)}`, false]; |
| 76 | // Parse regular paragraphs |
| 77 | function parseParagraph(markdown, isInList) { |
| 78 | const paragraphHtml = parseInlineText(markdown, false); |
| 79 | |
| 80 | if (isInList) { |
| 81 | // Close the list before starting paragraph |
| 82 | return [`</ul>${paragraphHtml}`, false]; |
| 70 | 83 | } |
| 84 | |
| 85 | return [paragraphHtml, false]; |
| 71 | 86 | } |
| 72 | 87 | |
| 73 | | function parseLine(markdown, list) { |
| 74 | | let [result, inListAfter] = parseHeader(markdown, list); |
| 75 | | if (result === null) { |
| 76 | | [result, inListAfter] = parseLineItem(markdown, list); |
| 88 | // Parse a single line of markdown, determining its type and converting to HTML |
| 89 | function parseLine(markdown, isInList) { |
| 90 | // Try to parse as header first |
| 91 | let [html, stillInList] = parseHeader(markdown, isInList); |
| 92 | |
| 93 | // If not a header, try to parse as list item |
| 94 | if (html === null) { |
| 95 | [html, stillInList] = parseListItem(markdown, isInList); |
| 77 | 96 | } |
| 78 | | if (result === null) { |
| 79 | | [result, inListAfter] = parseParagraph(markdown, list); |
| 97 | |
| 98 | // If not a list item, parse as paragraph |
| 99 | if (html === null) { |
| 100 | [html, stillInList] = parseParagraph(markdown, isInList); |
| 80 | 101 | } |
| 81 | | if (result === null) { |
| 82 | | throw new Error('Remove this line and implement the function'); |
| 102 | |
| 103 | // This should never happen as one of the parsers should always match |
| 104 | if (html === null) { |
| 105 | throw new Error('Unable to parse markdown line'); |
| 83 | 106 | } |
| 84 | | return [result, inListAfter]; |
| 107 | |
| 108 | return [html, stillInList]; |
| 85 | 109 | } |
| 86 | 110 | |
| 87 | 111 | /** |
| 88 | | * @param {string} markdown |
| 89 | | * @returns {string} |
| 112 | * Parse markdown text and convert it to HTML |
| 113 | * @param {string} markdown - The markdown text to parse |
| 114 | * @returns {string} The equivalent HTML |
| 90 | 115 | */ |
| 91 | 116 | export function parse(markdown) { |
| 92 | 117 | const lines = markdown.split('\n'); |
| 93 | | let result = ''; |
| 94 | | let list = false; |
| 95 | | for (let i = 0; i < lines.length; i++) { |
| 96 | | let [lineResult, newList] = parseLine(lines[i], list); |
| 97 | | result += lineResult; |
| 98 | | list = newList; |
| 118 | let htmlResult = ''; |
| 119 | let isInList = false; |
| 120 | |
| 121 | // Process each line of markdown |
| 122 | for (const line of lines) { |
| 123 | const [lineHtml, stillInList] = parseLine(line, isInList); |
| 124 | htmlResult += lineHtml; |
| 125 | isInList = stillInList; |
| 99 | 126 | } |
| 100 | | if (list) { |
| 101 | | return result + '</ul>'; |
| 102 | | } else { |
| 103 | | return result; |
| 127 | |
| 128 | // Close any unclosed list at the end |
| 129 | if (isInList) { |
| 130 | htmlResult += '</ul>'; |
| 104 | 131 | } |
| 132 | |
| 133 | return htmlResult; |
| 105 | 134 | } |