| 1 | | function wrap(text, tag) { |
| 2 | | return `<${tag}>${text}</${tag}>`; |
| 1 | // Helper function to wrap content in HTML tags |
| 2 | function wrap(content, tag) { |
| 3 | return `<${tag}>${content}</${tag}>`; |
| 3 | 4 | } |
| 4 | 5 | |
| 5 | | function isTag(text, tag) { |
| 6 | | return text.startsWith(`<${tag}>`); |
| 7 | | } |
| 8 | | |
| 9 | | function parser(markdown, delimiter, tag) { |
| 10 | | const pattern = new RegExp(`${delimiter}(.+)${delimiter}`); |
| 11 | | const replacement = `<${tag}>$1</${tag}>`; |
| 6 | // Parse bold text (__text__ -> <strong>text</strong>) |
| 7 | function parseBold(markdown) { |
| 8 | const pattern = new RegExp('__' + '(.+)' + '__'); |
| 9 | const replacement = '<strong>$1</strong>'; |
| 12 | 10 | return markdown.replace(pattern, replacement); |
| 13 | 11 | } |
| 14 | 12 | |
| 15 | | function parse__(markdown) { |
| 16 | | return parser(markdown, '__', 'strong'); |
| 17 | | } |
| 18 | | |
| 19 | | function parse_(markdown) { |
| 20 | | return parser(markdown, '_', 'em'); |
| 13 | // Parse italic text (_text_ -> <em>text</em>) |
| 14 | function parseItalic(markdown) { |
| 15 | const pattern = new RegExp('_' + '(.+)' + '_'); |
| 16 | const replacement = '<em>$1</em>'; |
| 17 | return markdown.replace(pattern, replacement); |
| 21 | 18 | } |
| 22 | 19 | |
| 23 | | function parseText(markdown, list) { |
| 24 | | const parsedText = parse_(parse__(markdown)); |
| 25 | | if (list) { |
| 26 | | return parsedText; |
| 20 | // Parse text with inline formatting (bold and italic) |
| 21 | function parseFormattedText(text, inList) { |
| 22 | // Apply formatting: bold first, then italic |
| 23 | const formattedText = parseItalic(parseBold(text)); |
| 24 | |
| 25 | // If we're inside a list, don't wrap in <p> tags |
| 26 | if (inList) { |
| 27 | return formattedText; |
| 27 | 28 | } else { |
| 28 | | return wrap(parsedText, 'p'); |
| 29 | return wrap(formattedText, 'p'); |
| 29 | 30 | } |
| 30 | 31 | } |
| 31 | 32 | |
| 32 | | function parseHeader(markdown, list) { |
| 33 | | let count = 0; |
| 34 | | for (let i = 0; i < markdown.length; i++) { |
| 35 | | if (markdown[i] === '#') { |
| 36 | | count += 1; |
| 33 | // Parse header (# Header -> <h1>Header</h1>) |
| 34 | function parseHeader(line, inList) { |
| 35 | // Count consecutive # characters at the start of the line |
| 36 | let hashCount = 0; |
| 37 | for (let i = 0; i < line.length; i++) { |
| 38 | if (line[i] === '#') { |
| 39 | hashCount += 1; |
| 37 | 40 | } else { |
| 38 | 41 | break; |
| 39 | 42 | } |
| 40 | 43 | } |
| 41 | | if (count === 0 || count > 6) { |
| 42 | | return [null, list]; |
| 44 | |
| 45 | // Valid headers have 1-6 # characters |
| 46 | // Note: The original implementation has a bug where it doesn't check for a space after # |
| 47 | // but we preserve this behavior to maintain compatibility |
| 48 | if (hashCount === 0 || hashCount > 6) { |
| 49 | return [null, inList]; |
| 43 | 50 | } |
| 44 | | const headerTag = `h${count}`; |
| 45 | | const headerHtml = wrap(markdown.substring(count + 1), headerTag); |
| 46 | | if (list) { |
| 51 | |
| 52 | // Create the appropriate header tag (h1-h6) |
| 53 | const headerTag = `h${hashCount}`; |
| 54 | const headerContent = line.substring(hashCount + 1); |
| 55 | const headerHtml = wrap(headerContent, headerTag); |
| 56 | |
| 57 | // If we were in a list, close it before the header |
| 58 | if (inList) { |
| 47 | 59 | return [`</ul>${headerHtml}`, false]; |
| 48 | 60 | } else { |
| 49 | 61 | return [headerHtml, false]; |
| 50 | 62 | } |
| 51 | 63 | } |
| 52 | 64 | |
| 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]; |
| 65 | // Parse list items (* item -> <li>item</li>) |
| 66 | function parseListItem(line, inList) { |
| 67 | if (line.startsWith('*')) { |
| 68 | // Extract the list item content (skip '* ') |
| 69 | const itemContent = line.substring(2); |
| 70 | const formattedItem = parseFormattedText(itemContent, true); |
| 71 | const listItemHtml = wrap(formattedItem, 'li'); |
| 72 | |
| 73 | // If we're already in a list, just add the item |
| 74 | // Otherwise, start a new list |
| 75 | if (inList) { |
| 76 | return [listItemHtml, true]; |
| 58 | 77 | } else { |
| 59 | | return [`<ul>${innerHtml}`, true]; |
| 78 | return [`<ul>${listItemHtml}`, true]; |
| 60 | 79 | } |
| 61 | 80 | } |
| 62 | | return [null, list]; |
| 81 | return [null, inList]; |
| 63 | 82 | } |
| 64 | 83 | |
| 65 | | function parseParagraph(markdown, list) { |
| 66 | | if (!list) { |
| 67 | | return [parseText(markdown, false), false]; |
| 84 | // Parse regular paragraph text |
| 85 | function parseParagraph(line, inList) { |
| 86 | // Parse the text with formatting |
| 87 | const paragraphHtml = parseFormattedText(line, false); |
| 88 | |
| 89 | // If we were in a list, close it before the paragraph |
| 90 | if (inList) { |
| 91 | return [`</ul>${paragraphHtml}`, false]; |
| 68 | 92 | } else { |
| 69 | | return [`</ul>${parseText(markdown, false)}`, false]; |
| 93 | return [paragraphHtml, false]; |
| 70 | 94 | } |
| 71 | 95 | } |
| 72 | 96 | |
| 73 | | function parseLine(markdown, list) { |
| 74 | | let [result, inListAfter] = parseHeader(markdown, list); |
| 97 | // Parse a single line of markdown |
| 98 | function parseLine(line, inList) { |
| 99 | // Try to parse as header first |
| 100 | let [result, newListState] = parseHeader(line, inList); |
| 101 | |
| 102 | // If not a header, try to parse as list item |
| 75 | 103 | if (result === null) { |
| 76 | | [result, inListAfter] = parseLineItem(markdown, list); |
| 104 | [result, newListState] = parseListItem(line, inList); |
| 77 | 105 | } |
| 106 | |
| 107 | // If not a list item, parse as regular paragraph |
| 78 | 108 | if (result === null) { |
| 79 | | [result, inListAfter] = parseParagraph(markdown, list); |
| 109 | [result, newListState] = parseParagraph(line, inList); |
| 80 | 110 | } |
| 111 | |
| 112 | // This should never happen with valid input |
| 81 | 113 | if (result === null) { |
| 82 | | throw new Error('Remove this line and implement the function'); |
| 114 | throw new Error('Unable to parse line: ' + line); |
| 83 | 115 | } |
| 84 | | return [result, inListAfter]; |
| 116 | |
| 117 | return [result, newListState]; |
| 85 | 118 | } |
| 86 | 119 | |
| 87 | 120 | /** |
| 88 | | * @param {string} markdown |
| 89 | | * @returns {string} |
| 121 | * Parse markdown text and convert it to HTML |
| 122 | * @param {string} markdown - The markdown text to parse |
| 123 | * @returns {string} - The resulting HTML |
| 90 | 124 | */ |
| 91 | 125 | export function parse(markdown) { |
| 126 | // Split the markdown into individual lines |
| 92 | 127 | const lines = markdown.split('\n'); |
| 128 | |
| 129 | // Track whether we're currently inside a list |
| 130 | let inList = false; |
| 131 | |
| 132 | // Build the result HTML |
| 93 | 133 | 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; |
| 134 | |
| 135 | // Process each line |
| 136 | for (const line of lines) { |
| 137 | const [lineHtml, newListState] = parseLine(line, inList); |
| 138 | result += lineHtml; |
| 139 | inList = newListState; |
| 99 | 140 | } |
| 100 | | if (list) { |
| 101 | | return result + '</ul>'; |
| 102 | | } else { |
| 103 | | return result; |
| 141 | |
| 142 | // If we're still in a list at the end, close it |
| 143 | if (inList) { |
| 144 | result += '</ul>'; |
| 104 | 145 | } |
| 146 | |
| 147 | return result; |
| 105 | 148 | } |