markdown-magic 3.0.0 → 3.0.1

package/lib/utils/toc.js

@@ -0,0 +1,315 @@
+const { transform } = require('@technote-space/doctoc')
+const { removeLeadingAndTrailingLineBreaks, escapeRegexString } = require('./regex')
+const { findMinIndent } = require('./text')
+const { readFile } = require('./fs')
+
+// Alt https://github.com/thlorenz/doctoc
+
+/**
+ * @typedef {object} TocOptions
+ * @property {boolean} [collapse = false] - Collapse toc in <details> pane
+ * @property {string} [collapseText] - Text in expand pane
+ * @property {string} [excludeText] - Text to exclude from toc
+ * @property {boolean} [firsth1 = true] - Exclude first heading from toc
+ * @property {boolean} [sub = false] - Mark as sub section table of contents
+ * @property {number} [maxDepth = 4] - Max depth of headings to add to toc.
+ */
+
+/**
+ * @typedef {object} TocAPI
+ * @property {TocOptions} options - Toc generation options
+ * @property {string} fileContent - Markdown contents
+ * @property {string} srcPath - Path to markdown file
+ * @property {function} getBlockDetails - Get block details for sub tables
+ */
+
+/**
+ * Generate a table of contents
+ * @param {TocAPI} api 
+ * @returns 
+ */
+async function generateToc({
+  options = {},
+  fileContent,
+  srcPath,
+  getBlockDetails
+}) {
+  // const currentBlock = api.getCurrentBlock()
+  // console.log('originalBlock', originalBlock)
+
+  // process.exit(1)
+  const opts = options || {}
+  const isSub = opts.sub
+  opts.firsth1 = (opts.firsth1) ? true : false
+  let contents = fileContent
+  // console.log('contents', contents)
+
+  let collapseText = opts.collapseText
+
+  let debugFileMatch
+  // console.log('config', config.originalPath)
+  // debugFileMatch = config.originalPath.match(/packages\/analytics\/README\.md/)
+
+  // https://www.npmjs.com/package/@technote-space/toc-generator
+  const tocOptions = {
+    // mode: 'github.com', // github.com | bitbucket.org | gitlab.com | nodejs.org | ghost.org (default: github.com)
+    isNotitle: true,
+    //isCustomMode: true,
+    openingComment: '',
+    closingComment: '',
+    maxHeaderLevel: (opts.maxDepth) ? Number(opts.maxDepth) : 4,
+    // maxHeaderLevel: 2, // default: 4
+    // title: '**Table of Contents**',
+    // isNotitle: true,
+    // isFolding: true,
+    // entryPrefix: '*',
+    // processAll: true,
+    // updateOnly: true,
+    // openingComment: '<!-- toc -->',
+    // closingComment: '<!-- tocstop --> ',
+    // checkOpeningComments: ['<!-- toc '],
+    // checkClosingComments: ['<!-- tocstop '],
+    // isCustomMode: false,
+    // customTemplate: '<p align="center">${ITEMS}</p>',
+    // itemTemplate: '<a href="${LINK}">${TEXT}</a>',
+    // separator: '<span>|</span>',
+    // footer: 'end',
+  }
+  const t = await transform(contents, tocOptions)
+  let outputText = t.wrappedToc || ''
+
+  if (debugFileMatch) {
+    console.log('before firsth1 removal', outputText)
+  }
+
+  /* remove first h1 */
+  if (!opts.firsth1) {
+    // console.log('outputText', outputText)
+    const lines = outputText.split('\n')
+    let firstH1
+    let firstLine
+    const countOfH1s = lines.reduce((acc, line, index) => {
+      if (line !== '' && !firstLine) {
+        firstLine = index
+      }
+      if (line.match(/^-\s+/)) {
+        if (!firstH1) {
+          const headingText = line.match(/\[(.*)\]/)
+          if (headingText) {
+            const rawText = headingText[1]
+            firstH1 = [line, index, firstLine === index, rawText]
+          }
+        }
+        acc = acc + 1
+      }
+      return acc
+    }, 0)
+
+    const firstHeadingData =  (firstH1 && Array.isArray(firstH1)) ? firstH1 : []
+    const [ lineText, lineIndex, isFirst, matchText ] = firstHeadingData
+    
+    let docHasHeading = false
+    if (matchText) {
+      // verify its h1
+      const matchTextEscaped = escapeRegexString(matchText)
+      // console.log('matchTextEscaped', matchTextEscaped)
+      // /^#{1}\s+(.*)/
+      const FIRST_H1_REGEX = new RegExp(`^#\\s*\\[?${matchTextEscaped}\\]?(?:.*)?`, 'gim')
+      // /^<h1\b[^>]*>([\s\S]*?)<\/h1>/
+      const FIRST_HTML_H1_REGEX = new RegExp(`^<h1\\b[^>]*>[\\s]*?(${matchTextEscaped})[\\s]*?<\\/h1>`, 'gim')
+      // /^(.*)\n={3,}/
+      const FIRST_UNDERSCORE_H1 = new RegExp(`^(${matchTextEscaped})\n={3,}`, 'gim')
+  
+      if (contents.match(FIRST_H1_REGEX)) {
+        docHasHeading = matchText
+      } else if (contents.match(FIRST_HTML_H1_REGEX)) {
+        docHasHeading = matchText
+      } else if (contents.match(FIRST_UNDERSCORE_H1)) {
+        docHasHeading = matchText
+      }
+    }
+  
+    if (debugFileMatch) {
+      console.log('matchText', matchText)
+      if (docHasHeading) {
+        console.log('Found heading 1', docHasHeading)
+      }
+    }
+
+    if (debugFileMatch) {
+      console.log('top level toc item count', countOfH1s)
+      if (docHasHeading) {
+        console.log('Found heading 1', docHasHeading)
+        console.log('firstH1', firstH1)
+      }
+    }
+
+    let firstItemWithContent
+    let foundFirstH1
+
+    if (docHasHeading) {
+      outputText = lines.reduce((acc, line, i) => {
+        const isLineEmpty = line === ''
+        if (!isLineEmpty && !firstItemWithContent) {
+          firstItemWithContent = i
+        }
+        if (!foundFirstH1 && (firstItemWithContent === i) && line.match(/^-\s+/)) {
+          foundFirstH1 = true
+          return acc
+        }
+        // Add back line and remove level
+        let newLineValue = line
+        if (lineText) {
+          // const untilFirstH1 = i < lineIndex
+          /* @TODO make option? flatten all non first h1 tags */
+          if (countOfH1s > 0 && !isLineEmpty && newLineValue.match(/^\s+-/)) {
+            newLineValue = line.substring(2)
+          }
+        }
+        
+        acc = acc.concat(newLineValue)
+        return acc
+      }, []).join('\n')
+    }
+ 
+    // console.log('outputText', outputText)
+    if (debugFileMatch) {
+      console.log('after firsth1 removal', outputText)
+    }
+  }
+  // console.log(t)
+  // Fix <h1> with multiple lines
+  const spaces = outputText.match(/\[\n([\s\S]*?)\]/gm)
+  if (spaces) {
+    const fixed = spaces[0]
+      // all new lines
+      .replace(/\n/g, '')
+      // leading spaces
+      .replace(/\[\s+/, '[')
+      .trim()
+    outputText = outputText.replace(spaces[0], fixed)
+  }
+  // console.log(outputText)
+  outputText = outputText
+    .replace(/(.*)\[Table of Contents\]\(.*\)\n?/i, '')
+    .replace(/(.*)\[toc\]\(.*\)\n?/i, '')
+
+
+  if (opts.excludeText) {
+
+    outputText = excludeTocItem(outputText, opts.excludeText)
+    // // (\s+)?-(.*)\[Usage\]\(.*\)
+    // const regex = new RegExp(`\(\\s+\)?-(.*)\\[${opts.excludeText}\\]\\(.*\\)`, 'i')
+    // // /(\s+)?-(.*)\[Usage\]\(.*\)(\n\s+(.*))+/im
+    // const nestedRegex = new RegExp(`\(\\s+\)?-(.*)\\[${opts.excludeText}\\]\\(.*\\)\(\\n\\s+\(.*\)\)+`, 'i')
+    
+    // const hasNested = nestedRegex.exec(outputText)
+    // if (hasNested) {
+    //   // Count indentation of spaces
+    //   const numberOfSpaces = hasNested[1].replace(/\n/g, '').length
+    //   const subItems = numberOfSpaces + 1
+    //   // Update regex to only remove sub items
+    //   const nestedRegexSpaces = new RegExp(`\(\\s+\)?-(.*)\\[${opts.excludeText}\\]\\(.*\\)\(\\n\\s{${subItems},}\(.*\)\)+`, 'i')
+    //   // console.log('nestedRegexSpaces', nestedRegexSpaces)
+    //   // If exclude value has nested sections remove them as well.
+    //   outputText = outputText.replace(nestedRegexSpaces, '')
+    //   outputText = outputText.replace(regex, '')
+    // } else {
+    //   outputText = outputText.replace(regex, '')
+    // }
+  }
+
+  /* Sub table of contents */
+  if (isSub) {
+    // const start = fileContent.indexOf(open.value)
+    // const linesBefore = fileContent.substr(0, start).split('\n')
+    // const closestHeading = linesBefore.reverse().find((line) => {
+    //   return line.match((/^#+/))
+    // })
+    const originalContents = await readFile(srcPath, { encoding: 'utf8' })
+    const originalBlock = getBlockDetails(originalContents)
+    const linesBefore = originalContents.substr(0, originalBlock.block.start).split('\n')
+    const closestHeading = linesBefore.reverse().find((line) => {
+      return line.match((/^#+/))
+    })
+
+    if (closestHeading) {
+      const headingText = closestHeading.replace(/^#*\s*/, '')
+      // console.log('BEFORE', linesBefore)
+      // console.log('closest parent heading', closestHeading)
+      // console.log('headingText', headingText)
+      collapseText = collapseText || `${headingText} contents`
+      // https://regex101.com/r/MB85zm/1
+
+      const findSubToc = new RegExp(`^\(\\s+\)?-(.*)\\[${headingText}\\]\\(.*\\)(\\n\\s.*)+`, 'gm')
+      // console.log('findSubToc', findSubToc)
+      const single = singleLinePattern(headingText)
+      // console.log('single', single)
+      const subItems = outputText.match(findSubToc)
+      if (subItems) {
+        const items = subItems[0].replace(single, '').split('\n')
+          .filter(Boolean)
+        // console.log('items', items)
+        const finalItems = items // .slice(1, items.length)
+        if (finalItems) {
+          const indent = findMinIndent(finalItems.join('\n'))
+          // console.log('min indent', indent)
+          // console.log('finalItems', finalItems)
+          outputText = finalItems.map((thing) => thing.substring(indent)).join('\n')
+        } else {
+          // console.log('No sub items')
+        }
+      }
+    }
+  }
+
+  // If collapse wrap in <details>
+  if (opts.collapse) {
+    return `<details>
+<summary>${collapseText || 'Table of Contents'}</summary>
+
+${outputText   
+  // Replace leading double spaces
+  .replace(/^\n*/g, '')
+  // Replace trailing double spaces
+  .replace(/\n*$/g, '')
+}
+
+</details>`
+  }
+
+  return outputText.replace(removeLeadingAndTrailingLineBreaks, '')
+}
+
+function singleLinePattern(text) {
+  /* (\s+)?-(.*)\[Usage\]\(.*\) */
+  return new RegExp(`\(\\s+\)?-(.*)\\[${text}\\]\\(.*\\)`, 'i')
+}
+
+function excludeTocItem(str, excludeText) {
+  const matchTextEscaped = escapeRegexString(excludeText)
+  /* (\s+)?-(.*)\[Usage\]\(.*\) */
+  const regex = singleLinePattern(matchTextEscaped) // new RegExp(`\(\\s+\)?-(.*)\\[${matchTextEscaped}\\]\\(.*\\)`, 'i')
+  /* /(\s+)?-(.*)\[Usage\]\(.*\)(\n\s+(.*))+/im */
+  const nestedRegex = new RegExp(`\(\\s+\)?-(.*)\\[${matchTextEscaped}\\]\\(.*\\)\(\\n\\s+\(.*\)\)+`, 'i')
+  
+  const hasNested = nestedRegex.exec(str)
+  if (hasNested) {
+    // Count indentation of spaces
+    const numberOfSpaces = (hasNested[1] || '').replace(/\n/g, '').length
+    const subItems = numberOfSpaces + 1
+    // Update regex to only remove sub items
+    const nestedRegexSpaces = new RegExp(`\(\\s+\)?-(.*)\\[${matchTextEscaped}\\]\\(.*\\)\(\\n\\s{${subItems},}\(.*\)\)+`, 'i')
+    // console.log('nestedRegexSpaces', nestedRegexSpaces)
+    // If exclude value has nested sections remove them as well.
+    str = str.replace(nestedRegexSpaces, '')
+    str = str.replace(regex, '')
+  } else {
+    str = str.replace(regex, '')
+  }
+  return str
+}
+
+module.exports = {
+  generateToc
+}

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "markdown-magic",
-  "version": "3.0.0",
+  "version": "3.0.1",
   "description": "Automatically update markdown files with content from external sources",
   "main": "lib/index.js",
   "bin": {
@@ -15,6 +15,8 @@
     "/lib"
   ],
   "scripts": {
+    "types": "tsc",
+    "emit-types": "tsc --noEmit false --emitDeclarationOnly true",
     "docs": "node examples/generate-readme.js",
     "test": "npm run test:lib && npm run test:test && echo 'tests done'",
     "test:lib": "uvu lib '.test.([mc]js|[jt]sx?)$'",
@@ -48,17 +50,19 @@
     "is-glob": "^4.0.3",
     "is-local-path": "^0.1.6",
     "is-valid-path": "^0.1.1",
-    "json-alexander": "^0.1.8",
+    "micro-mdx-parser": "^1.1.0",
     "mri": "^1.2.0",
+    "oparser": "^2.1.1",
     "smart-glob": "^1.0.2",
     "sync-request": "^6.1.0"
   },
   "devDependencies": {
     "ansi-styles": "^4.2.1",
     "concordance": "^5.0.1",
-    "doxxx": "^1.0.0",
+    "doxxx": "^2.0.1",
     "rimraf": "^3.0.2",
     "safe-chalk": "^1.0.0",
+    "typescript": "^5.0.2",
     "uvu": "^0.5.1"
   }
 }