refactor: enforce named capture groups; clean up regexes (#6524)

* refactor: enforce named capture groups; clean up regexes

* fixes

* fix

packages/docusaurus-utils/src/__tests__/markdownParser.test.ts

@@ -116,7 +116,7 @@ describe('createExcerpt', () => {
 
           export function ItemCol(props) { return <Item {...props} className={'col col--6 margin-bottom--lg'}/> };
 
-          Lorem **ipsum** dolor sit \`amet\`[^1], consectetur _adipiscing_ elit. [**Vestibulum**](https://wiktionary.org/wiki/vestibulum) ex urna[^bignote], ~molestie~ et sagittis ut, varius ac justo :wink:.
+          Lorem **ipsum** dolor sit \`amet\`[^1], consectetur _adipiscing_ elit. [**Vestibulum**](https://wiktionary.org/wiki/vestibulum) ex urna[^bignote], ~~molestie~~ et sagittis ut, varius ac justo :wink:.
 
           Nunc porttitor libero nec vulputate venenatis. Nam nec rhoncus mauris. Morbi tempus est et nibh maximus, tempus venenatis arcu lobortis.
         `),

packages/docusaurus-utils/src/index.ts

@@ -69,8 +69,8 @@ export async function generate(
   }
 }
 
-const indexRE = /(^|.*\/)index\.(md|mdx|js|jsx|ts|tsx)$/i;
-const extRE = /\.(md|mdx|js|jsx|ts|tsx)$/;
+const indexRE = /(?<dirname>^|.*\/)index\.(?:mdx?|jsx?|tsx?)$/i;
+const extRE = /\.(?:mdx?|jsx?|tsx?)$/;
 
 /**
  * Convert filepath to url path.

packages/docusaurus-utils/src/markdownLinks.ts

@@ -67,11 +67,11 @@ export function replaceMarkdownLinks<T extends ContentPaths>({
     // ink
     // [doc1]: doc1.md -> we replace this doc1.md with correct link
     const mdRegex =
-      /(?:(?:\]\()|(?:\]:\s?))(?!https?:\/\/|@site\/)([^'")\]\s>]+\.mdx?)/g;
+      /(?:(?:\]\()|(?:\]:\s?))(?!https?:\/\/|@site\/)(?<filename>[^'")\]\s>]+\.mdx?)/g;
     let mdMatch = mdRegex.exec(modifiedLine);
     while (mdMatch !== null) {
       // Replace it to correct html link.
-      const mdLink = mdMatch[1];
+      const mdLink = mdMatch.groups!.filename;
 
       const sourcesToTry = [
         path.resolve(path.dirname(filePath), decodeURIComponent(mdLink)),

packages/docusaurus-utils/src/markdownParser.ts

@@ -14,12 +14,12 @@ export function parseMarkdownHeadingId(heading: string): {
   text: string;
   id?: string;
 } {
-  const customHeadingIdRegex = /^(.*?)\s*\{#([\w-]+)\}$/;
+  const customHeadingIdRegex = /^(?<text>.*?)\s*\{#(?<id>[\w-]+)\}$/;
   const matches = customHeadingIdRegex.exec(heading);
   if (matches) {
     return {
-      text: matches[1],
-      id: matches[2],
+      text: matches.groups!.text,
+      id: matches.groups!.id,
     };
   }
   return {text: heading, id: undefined};
@@ -46,7 +46,7 @@ export function createExcerpt(fileString: string): string | undefined {
     }
 
     // Skip import/export declaration.
-    if (/^\s*?import\s.*(from.*)?;?|export\s.*{.*};?/.test(fileLine)) {
+    if (/^(?:import|export)\s.*/.test(fileLine)) {
       continue;
     }
 
@@ -71,25 +71,27 @@ export function createExcerpt(fileString: string): string | undefined {
       // Remove HTML tags.
       .replace(/<[^>]*>/g, '')
       // Remove Title headers
-      .replace(/^#\s*([^#]*)\s*#?/gm, '')
+      .replace(/^#\s*[^#]*\s*#?/gm, '')
       // Remove Markdown + ATX-style headers
-      .replace(/^#{1,6}\s*([^#]*)\s*(#{1,6})?/gm, '$1')
-      // Remove emphasis and strikethroughs.
-      .replace(/([*_~]{1,3})(\S.*?\S{0,1})\1/g, '$2')
+      .replace(/^#{1,6}\s*(?<text>[^#]*)\s*(?:#{1,6})?/gm, '$1')
+      // Remove emphasis.
+      .replace(/(?<opening>[*_]{1,3})(?<text>.*?)\1/g, '$2')
+      // Remove strikethroughs.
+      .replace(/~~(?<text>\S.*\S)~~/g, '$1')
       // Remove images.
-      .replace(/!\[(.*?)\][[(].*?[\])]/g, '$1')
+      .replace(/!\[(?<alt>.*?)\][[(].*?[\])]/g, '$1')
       // Remove footnotes.
-      .replace(/\[\^.+?\](: .*?$)?/g, '')
+      .replace(/\[\^.+?\](?:: .*?$)?/g, '')
       // Remove inline links.
-      .replace(/\[(.*?)\][[(].*?[\])]/g, '$1')
+      .replace(/\[(?<alt>.*?)\][[(].*?[\])]/g, '$1')
       // Remove inline code.
-      .replace(/`(.+?)`/g, '$1')
+      .replace(/`(?<text>.+?)`/g, '$1')
       // Remove blockquotes.
       .replace(/^\s{0,3}>\s?/g, '')
       // Remove admonition definition.
-      .replace(/(:{3}.*)/, '')
+      .replace(/:::.*/, '')
       // Remove Emoji names within colons include preceding whitespace.
-      .replace(/\s?(:(::|[^:\n])+:)/g, '')
+      .replace(/\s?:(?:::|[^:\n])+:/g, '')
       // Remove custom Markdown heading id.
       .replace(/{#*[\w-]+}/, '')
       .trim();
@@ -134,9 +136,9 @@ export function parseMarkdownContentTitle(
   const content = contentUntrimmed.trim();
 
   const IMPORT_STATEMENT =
-    /import\s+(([\w*{}\s\n,]+)from\s+)?["'\s]([@\w/_.-]+)["'\s];?|\n/.source;
+    /import\s+(?:[\w*{}\s\n,]+from\s+)?["'\s][@\w/_.-]+["'\s];?|\n/.source;
   const REGULAR_TITLE =
-    /(?<pattern>#\s*(?<title>[^#\n{]*)+[ \t]*(?<suffix>({#*[\w-]+})|#)?\n*?)/
+    /(?<pattern>#\s*(?<title>[^#\n{]*)+[ \t]*(?<suffix>(?:{#*[\w-]+})|#)?\n*?)/
       .source;
   const ALTERNATE_TITLE = /(?<pattern>\s*(?<title>[^\n]*)\s*\n[=]+)/.source;
 

packages/docusaurus-utils/src/urlUtils.ts

@@ -27,7 +27,7 @@ export function normalizeUrl(rawUrls: string[]): string {
   // There must be two or three slashes in the file protocol,
   // two slashes in anything else.
   const replacement = urls[0].match(/^file:\/\/\//) ? '$1:///' : '$1://';
-  urls[0] = urls[0].replace(/^([^/:]+):\/*/, replacement);
+  urls[0] = urls[0].replace(/^(?<protocol>[^/:]+):\/*/, replacement);
 
   for (let i = 0; i < urls.length; i += 1) {
     let component = urls[i];
@@ -70,14 +70,14 @@ export function normalizeUrl(rawUrls: string[]): string {
   // except the possible first plain protocol part.
 
   // Remove trailing slash before parameters or hash.
-  str = str.replace(/\/(\?|&|#[^!])/g, '$1');
+  str = str.replace(/\/(?<search>\?|&|#[^!])/g, '$1');
 
   // Replace ? in parameters with &.
   const parts = str.split('?');
   str = parts.shift() + (parts.length > 0 ? '?' : '') + parts.join('&');
 
   // Dedupe forward slashes in the entire path, avoiding protocol slashes.
-  str = str.replace(/([^:/]\/)\/+/g, '$1');
+  str = str.replace(/(?<textBefore>[^:/]\/)\/+/g, '$1');
 
   // Dedupe forward slashes at the beginning of the path.
   str = str.replace(/^\/+/g, '/');

packages/docusaurus-utils/src/webpackUtils.ts

@@ -79,12 +79,12 @@ export function getFileLoaderUtils(): FileLoaderUtils {
      */
     images: () => ({
       use: [loaders.url({folder: 'images'})],
-      test: /\.(ico|jpg|jpeg|png|gif|webp)(\?.*)?$/,
+      test: /\.(?:ico|jpe?g|png|gif|webp)(?:\?.*)?$/i,
     }),
 
     fonts: () => ({
       use: [loaders.url({folder: 'fonts'})],
-      test: /\.(woff|woff2|eot|ttf|otf)$/,
+      test: /\.(?:woff2?|eot|ttf|otf)$/i,
     }),
 
     /**
@@ -93,11 +93,11 @@ export function getFileLoaderUtils(): FileLoaderUtils {
      */
     media: () => ({
       use: [loaders.url({folder: 'medias'})],
-      test: /\.(mp4|webm|ogv|wav|mp3|m4a|aac|oga|flac)$/,
+      test: /\.(?:mp4|webm|ogv|wav|mp3|m4a|aac|oga|flac)$/i,
     }),
 
     svg: () => ({
-      test: /\.svg?$/,
+      test: /\.svg$/i,
       oneOf: [
         {
           use: [
@@ -126,7 +126,7 @@ export function getFileLoaderUtils(): FileLoaderUtils {
           // We don't want to use SVGR loader for non-React source code
           // ie we don't want to use SVGR for CSS files...
           issuer: {
-            and: [/\.(ts|tsx|js|jsx|md|mdx)$/],
+            and: [/\.(?:tsx?|jsx?|mdx?)$/i],
           },
         },
         {
@@ -137,7 +137,7 @@ export function getFileLoaderUtils(): FileLoaderUtils {
 
     otherAssets: () => ({
       use: [loaders.file({folder: 'files'})],
-      test: /\.(pdf|doc|docx|xls|xlsx|zip|rar)$/,
+      test: /\.(?:pdf|docx?|xlsx?|zip|rar)$/i,
     }),
   };