feat: add tests for build post list script

scripts/build-post-list.js

@@ -1,8 +1,7 @@
-const { readdirSync, statSync, existsSync, readFileSync, writeFileSync } = require('fs')
-const { resolve, basename } = require('path')
+const { readdir, stat, pathExists, readFile, writeFile } = require('fs-extra')
+const { basename, join, normalize, sep, posix, relative, parse } = require('path')
 const frontMatter = require('gray-matter')
 const toc = require('markdown-toc')
-const { slugify } = require('markdown-toc/lib/utils')
 const readingTime = require('reading-time')
 const { markdownToTxt } = require('markdown-to-txt')
 const { buildNavTree, addDocButtons } = require('./build-docs')
@@ -15,51 +14,88 @@ const result = {
   docsTree: {}
 }
 const releaseNotes = []
-const basePath = 'pages'
-const postDirectories = [
-  // order of these directories is important, as the blog should come before docs, to create a list of available release notes, which will later be used to release-note-link for spec docs
-  [`${basePath}/blog`, '/blog'],
-  [`${basePath}/docs`, '/docs'],
-  [`${basePath}/about`, '/about']
-];
 
 const addItem = (details) => {
-  if(details.slug.startsWith('/docs'))
-    result["docs"].push(details)
-  else if(details.slug.startsWith('/blog'))
-    result["blog"].push(details)
-  else if(details.slug.startsWith('/about'))
-    result["about"].push(details)
-  else {}
+  const sectionMap = {
+    '/docs': 'docs',
+    '/blog': 'blog',
+    '/about': 'about'
+  };
+  const section = Object.keys(sectionMap).find(key => details.slug.startsWith(key));
+  if (section) {
+    result[sectionMap[section]].push(details);
+  }
+};
+
+function getVersionDetails(slug, weight) {
+  const fileBaseName = basename(slug);
+  const versionName = fileBaseName.split('-')[0];
+  return {
+    title: versionName.startsWith('v')
+      ? capitalize(versionName.slice(1))
+      : capitalize(versionName),
+    weight
+  };
 }
 
-module.exports = async function buildPostList() {
-  walkDirectories(postDirectories, result)
-  const treePosts = buildNavTree(result["docs"].filter((p) => p.slug.startsWith('/docs/')))
-  result["docsTree"] = treePosts
-  result["docs"] = addDocButtons(result["docs"], treePosts)
-  if (process.env.NODE_ENV === 'production') {
-    // console.log(inspect(result, { depth: null, colors: true }))
+/**
+ * Builds a list of posts from the specified directories and writes it to a file
+ * @param {Array<Array<string>>} postDirectories - Array of [directory, slug] tuples
+ * @param {string} basePath - Base path for resolving relative paths
+ * @param {string} writeFilePath - Path where the output JSON will be written
+ * @throws {Error} If required parameters are missing or if any operation fails
+ * @returns {Promise<void>}
+ */
+async function buildPostList(postDirectories, basePath, writeFilePath) {
+  try {
+
+    if (!basePath) {
+      throw new Error('Error while building post list: basePath is required');
+    }
+
+    if (!writeFilePath) {
+      throw new Error('Error while building post list: writeFilePath is required');
+    }
+
+    if (postDirectories.length === 0) {
+      throw new Error('Error while building post list: postDirectories array is empty');
+    }
+    const normalizedBasePath = normalize(basePath)
+    await walkDirectories(postDirectories, result, normalizedBasePath)
+    const treePosts = buildNavTree(result.docs.filter((p) => p.slug.startsWith('/docs/')))
+    result.docsTree = treePosts
+    result.docs = addDocButtons(result.docs, treePosts)
+    await writeFile(writeFilePath, JSON.stringify(result, null, '  '))
+  } catch (error) {
+    throw new Error(`Error while building post list: ${error.message}`, { cause: error });
   }
-  writeFileSync(resolve(__dirname, '..', 'config', 'posts.json'), JSON.stringify(result, null, '  '))
 }
 
-function walkDirectories(directories, result, sectionWeight = 0, sectionTitle, sectionId, rootSectionId) {
+async function walkDirectories(
+  directories,
+  resultObj,
+  basePath,
+  sectionTitle,
+  sectionId,
+  rootSectionId,
+  sectionWeight = 0
+) {
   for (let dir of directories) {
-    let directory = dir[0]
-    let sectionSlug = dir[1] || ''
-    let files = readdirSync(directory);
+    const directory = posix.normalize(dir[0]);
+    const sectionSlug = dir[1] || '';
+    const files = await readdir(directory)
 
     for (let file of files) {
-      let details
-      const fileName = [directory, file].join('/')
-      const fileNameWithSection = [fileName, '_section.mdx'].join('/')
-      const slug = fileName.replace(new RegExp(`^${basePath}`), '')
-      const slugElements = slug.split('/');
-      if (isDirectory(fileName)) {
-        if (existsSync(fileNameWithSection)) {
+      let details;
+      const fileName = normalize(join(directory, file));
+      const fileNameWithSection = normalize(join(fileName, '_section.mdx'))
+      const slug = `/${normalize(relative(basePath, fileName)).replace(/\\/g, '/')}`
+      const slugElements = slug.split('/')
+
+      if (await isDirectory(fileName)) {
+        if (await pathExists(fileNameWithSection)) {
           // Passing a second argument to frontMatter disables cache. See https://github.com/asyncapi/website/issues/1057
-          details = frontMatter(readFileSync(fileNameWithSection, 'utf-8'), {}).data
+          details = frontMatter(await readFile(fileNameWithSection, 'utf-8'), {}).data
           details.title = details.title || capitalize(basename(fileName))
         } else {
           details = {
@@ -68,8 +104,8 @@ function walkDirectories(directories, result, sectionWeight = 0, sectionTitle, s
         }
         details.isSection = true
         if (slugElements.length > 3) {
-           details.parent = slugElements[slugElements.length - 2]
-           details.sectionId = slugElements[slugElements.length - 1]
+          details.parent = slugElements[slugElements.length - 2]
+          details.sectionId = slugElements[slugElements.length - 1]
         }
         if (!details.parent) {
           details.isRootSection = true
@@ -79,9 +115,9 @@ function walkDirectories(directories, result, sectionWeight = 0, sectionTitle, s
         details.slug = slug
         addItem(details)
         const rootId = details.parent || details.rootSectionId
-        walkDirectories([[fileName, slug]], result, details.weight, details.title, details.sectionId, rootId)
-      } else if (file.endsWith('.mdx') && !fileName.endsWith('/_section.mdx')) {
-        const fileContent = readFileSync(fileName, 'utf-8')
+        await walkDirectories([[fileName, slug]], resultObj, basePath, details.title, details.sectionId, rootId, details.sectionWeight)
+      } else if (file.endsWith('.mdx') && !fileName.endsWith(sep + '_section.mdx')) {
+        const fileContent = await readFile(fileName, 'utf-8')
         // Passing a second argument to frontMatter disables cache. See https://github.com/asyncapi/website/issues/1057
         const { data, content } = frontMatter(fileContent, {})
         details = data
@@ -93,21 +129,16 @@ function walkDirectories(directories, result, sectionWeight = 0, sectionTitle, s
         details.sectionTitle = sectionTitle
         details.sectionId = sectionId
         details.rootSectionId = rootSectionId
-        details.id = fileName
-        details.isIndex = fileName.endsWith('/index.mdx')
+        details.id = fileName.replace(/\\/g, '/')
+        details.isIndex = fileName.endsWith(join('index.mdx'))
         details.slug = details.isIndex ? sectionSlug : slug.replace(/\.mdx$/, '')
-        if(details.slug.includes('/reference/specification/') && !details.title) {
-          const fileBaseName = basename(data.slug)  // ex. v2.0.0 | v2.1.0-next-spec.1
-          const fileName = fileBaseName.split('-')[0] // v2.0.0 | v2.1.0
-          details.weight = specWeight--
-
-          if (fileName.startsWith('v')) {
-             details.title = capitalize(fileName.slice(1)) 
-          } else {
-            details.title = capitalize(fileName)
-          }
+        if (details.slug.includes('/reference/specification/') && !details.title) {
+          const fileBaseName = basename(details.slug)
+          const versionDetails = getVersionDetails(details.slug, specWeight--);
+          details.title = versionDetails.title;
+          details.weight = versionDetails.weight;
 
-          if(releaseNotes.includes(details.title)){
+          if (releaseNotes.includes(details.title)) {
             details.releaseNoteLink = `/blog/release-notes-${details.title}`
           }
 
@@ -122,14 +153,10 @@ function walkDirectories(directories, result, sectionWeight = 0, sectionTitle, s
         }
 
         // To create a list of available ReleaseNotes list, which will be used to add details.releaseNoteLink attribute.
-        if(file.startsWith("release-notes") && dir[1] === "/blog"){
-          const fileName_without_extension = file.slice(0,-4)
-          // removes the file extension. For example, release-notes-2.1.0.md -> release-notes-2.1.0
-          const version = fileName_without_extension.slice(fileName_without_extension.lastIndexOf("-")+1)
-
-          // gets the version from the name of the releaseNote .md file (from /blog). For example, version = 2.1.0 if fileName_without_extension = release-notes-2.1.0
-          releaseNotes.push(version)
-          // releaseNotes is the list of all available releaseNotes
+        if (file.startsWith('release-notes') && dir[1] === '/blog') {
+          const { name } = parse(file);
+          const version = name.split('-').pop();
+          releaseNotes.push(version);
         }
 
         addItem(details)
@@ -138,24 +165,38 @@ function walkDirectories(directories, result, sectionWeight = 0, sectionTitle, s
   }
 }
 
+// Matches heading IDs in two formats:
+// 1. {#my-heading-id}
+// 2. <a name="my-heading-id">
+const HEADING_ID_REGEX = /[\s]*(?:\{#([a-zA-Z0-9\-_]+)\}|<a[\s]+name="([a-zA-Z0-9\-_]+)")/;
+
+/**
+ * Extracts heading IDs from markdown headings
+ * @param {string} str - The heading text containing potential ID
+ * @returns {string} The extracted ID or empty string if no valid ID found
+ */
 function slugifyToC(str) {
-  let slug
-  // Try to match heading ids like {# myHeadingId}
-  const headingIdMatch = str.match(/[\s]?\{\#([\w\d\-_]+)\}/)
-  if (headingIdMatch && headingIdMatch.length >= 2) {
-    slug = headingIdMatch[1]
-  } else {
-    // Try to match heading ids like {<a name="myHeadingId"/>}
-    const anchorTagMatch = str.match(/[\s]*<a[\s]+name="([\w\d\s\-_]+)"/)
-    if (anchorTagMatch && anchorTagMatch.length >= 2) slug = anchorTagMatch[1]
-  }
-  return slug || slugify(str, { firsth1: true, maxdepth: 6 })
+  if (typeof str !== 'string') return '';
+  if (!str.trim()) return '';
+  let slug = '';
+
+  // Match heading IDs like {# myHeadingId}
+  const idMatch = str.match(HEADING_ID_REGEX);
+  const [, headingId, anchorId] = idMatch || [];
+  slug = (headingId || anchorId || '').trim();
+
+  // If no valid ID is found, return an empty string
+  return slug;
 }
 
-function isDirectory(dir) {
-  return statSync(dir).isDirectory()
+async function isDirectory(dir) {
+  return (await stat(dir)).isDirectory()
 }
 
 function capitalize(text) {
-  return text.split(/[\s\-]/g).map(word => `${word[0].toUpperCase()}${word.substr(1)}`).join(' ')
+  return text.split(/[\s-]/)
+    .map(word => word.charAt(0).toUpperCase() + word.slice(1))
+    .join(' ');
 }
+
+module.exports = { slugifyToC, buildPostList }

scripts/index.js

@@ -1,13 +1,23 @@
 const { resolve } = require('path');
 const fs = require('fs');
 const rssFeed = require('./build-rss');
-const buildPostList = require('./build-post-list');
+const { buildPostList } = require('./build-post-list');
 const buildCaseStudiesList = require('./casestudies');
 const buildAdoptersList = require('./adopters');
 const buildFinanceInfoList = require('./finance');
 
 async function start() {
-  await buildPostList();
+
+  const postDirectories = [
+    ['pages/blog', '/blog'],
+    ['pages/docs', '/docs'],
+    ['pages/about', '/about']
+  ];
+  const basePath = 'pages';
+  const writeFilePath = resolve(__dirname, '../config', 'posts.json');
+
+  await buildPostList(postDirectories, basePath, writeFilePath);
+
   rssFeed(
     'blog',
     'AsyncAPI Initiative Blog RSS Feed',