refactor: install eslint-plugin-regexp (#6906)

* refactor: install eslint-plugin-regexp

* simplify

* simplify

* fix

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

@@ -169,14 +169,16 @@ describe('reportMessage', () => {
       `"Unexpected \\"reportingSeverity\\" value: foo."`,
     );
     expect(consoleLog).toBeCalledTimes(1);
-    expect(consoleLog).toBeCalledWith(expect.stringMatching(/.*\[INFO].* hey/));
+    expect(consoleLog).toBeCalledWith(
+      expect.stringMatching(/.*\[INFO\].* hey/),
+    );
     expect(consoleWarn).toBeCalledTimes(1);
     expect(consoleWarn).toBeCalledWith(
-      expect.stringMatching(/.*\[WARNING].* hey/),
+      expect.stringMatching(/.*\[WARNING\].* hey/),
     );
     expect(consoleError).toBeCalledTimes(1);
     expect(consoleError).toBeCalledWith(
-      expect.stringMatching(/.*\[ERROR].* hey/),
+      expect.stringMatching(/.*\[ERROR\].* hey/),
     );
   });
 });

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

@@ -404,7 +404,7 @@ Lorem Ipsum
     expect(
       parseMarkdownContentTitle(markdown, {removeContentTitle: true}),
     ).toEqual({
-      content: markdown.trim().replace('# Markdown Title', ''),
+      content: markdown.trim().replace('# Markdown Title\n', ''),
       contentTitle: 'Markdown Title',
     });
   });
@@ -444,7 +444,7 @@ Lorem Ipsum
     expect(
       parseMarkdownContentTitle(markdown, {removeContentTitle: true}),
     ).toEqual({
-      content: markdown.replace('Markdown Title\n==============\n\n', ''),
+      content: markdown.replace('Markdown Title\n==============\n', ''),
       contentTitle: 'Markdown Title',
     });
   });
@@ -547,7 +547,7 @@ Lorem Ipsum
     expect(
       parseMarkdownContentTitle(markdown, {removeContentTitle: true}),
     ).toEqual({
-      content: markdown.replace('# Markdown Title', ''),
+      content: markdown.replace('# Markdown Title\n', ''),
       contentTitle: 'Markdown Title',
     });
   });

packages/docusaurus-utils/src/markdownLinks.ts

@@ -68,7 +68,7 @@ export function replaceMarkdownLinks<T extends ContentPaths>({
     // ink
     // [doc1]: doc1.md -> we replace this doc1.md with correct link
     const mdRegex =
-      /(?:(?:\]\()|(?:\]:\s*))(?!https?:\/\/|@site\/)(?<filename>[^'")\]\s>]+\.mdx?)/g;
+      /(?:\]\(|\]:\s*)(?!https?:\/\/|@site\/)(?<filename>[^'")\]\s>]+\.mdx?)/g;
     let mdMatch = mdRegex.exec(modifiedLine);
     while (mdMatch !== null) {
       // Replace it to correct html link.

packages/docusaurus-utils/src/markdownParser.ts

@@ -14,11 +14,11 @@ export function parseMarkdownHeadingId(heading: string): {
   text: string;
   id?: string;
 } {
-  const customHeadingIdRegex = /^(?<text>.*?)\s*\{#(?<id>[\w-]+)\}$/;
+  const customHeadingIdRegex = /\s*\{#(?<id>[\w-]+)\}$/;
   const matches = customHeadingIdRegex.exec(heading);
   if (matches) {
     return {
-      text: matches.groups!.text!,
+      text: heading.replace(matches[0]!, ''),
       id: matches.groups!.id!,
     };
   }
@@ -69,9 +69,9 @@ export function createExcerpt(fileString: string): string | undefined {
       // Remove HTML tags.
       .replace(/<[^>]*>/g, '')
       // Remove Title headers
-      .replace(/^#\s*[^#]*\s*#?/gm, '')
+      .replace(/^#[^#]+#?/gm, '')
       // Remove Markdown + ATX-style headers
-      .replace(/^#{1,6}\s*(?<text>[^#]*)\s*(?:#{1,6})?/gm, '$1')
+      .replace(/^#{1,6}\s*(?<text>[^#]*)\s*#{0,6}/gm, '$1')
       // Remove emphasis.
       .replace(/(?<opening>[*_]{1,3})(?<text>.*?)\1/g, '$2')
       // Remove strikethroughs.
@@ -79,7 +79,7 @@ export function createExcerpt(fileString: string): string | undefined {
       // Remove images.
       .replace(/!\[(?<alt>.*?)\][[(].*?[\])]/g, '$1')
       // Remove footnotes.
-      .replace(/\[\^.+?\](?:: .*?$)?/g, '')
+      .replace(/\[\^.+?\](?:: .*$)?/g, '')
       // Remove inline links.
       .replace(/\[(?<alt>.*?)\][[(].*?[\])]/g, '$1')
       // Remove inline code.
@@ -91,7 +91,7 @@ export function createExcerpt(fileString: string): string | undefined {
       // Remove Emoji names within colons include preceding whitespace.
       .replace(/\s?:(?:::|[^:\n])+:/g, '')
       // Remove custom Markdown heading id.
-      .replace(/{#*[\w-]+}/, '')
+      .replace(/\{#*[\w-]+\}/, '')
       .trim();
 
     if (cleanedLine) {
@@ -132,35 +132,42 @@ export function parseMarkdownContentTitle(
   const removeContentTitleOption = options?.removeContentTitle ?? false;
 
   const content = contentUntrimmed.trim();
+  // We only need to detect import statements that will be parsed by MDX as
+  // `import` nodes, as broken syntax can't render anyways. That means any block
+  // that has `import` at the very beginning and surrounded by empty lines.
+  const contentWithoutImport = content
+    .replace(/^(?:import\s(?:.|\n(?!\n))*\n{2,})*/, '')
+    .trim();
 
-  const IMPORT_STATEMENT =
-    /import\s+(?:[\w*{}\s\n,]+from\s+)?["'\s][@\w/_.-]+["'\s];?|\n/.source;
-  const REGULAR_TITLE =
-    /(?<pattern>#\s*(?<title>[^#\n{]*)+[ \t]*(?<suffix>(?:{#*[\w-]+})|#)?\n*?)/
-      .source;
-  const ALTERNATE_TITLE = /(?<pattern>\s*(?<title>[^\n]*)\s*\n[=]+)/.source;
-
-  const regularTitleMatch = new RegExp(
-    `^(?:${IMPORT_STATEMENT})*?${REGULAR_TITLE}`,
-    'g',
-  ).exec(content);
-  const alternateTitleMatch = new RegExp(
-    `^(?:${IMPORT_STATEMENT})*?${ALTERNATE_TITLE}`,
-    'g',
-  ).exec(content);
+  const regularTitleMatch = /^#[ \t]+(?<title>[^ \t].*)(?:\n|$)/.exec(
+    contentWithoutImport,
+  );
+  const alternateTitleMatch = /^(?<title>.*)\n=+(?:\n|$)/.exec(
+    contentWithoutImport,
+  );
 
   const titleMatch = regularTitleMatch ?? alternateTitleMatch;
-  const {pattern, title} = titleMatch?.groups ?? {};
-
-  if (!pattern || !title) {
+  if (!titleMatch) {
     return {content, contentTitle: undefined};
   }
   const newContent = removeContentTitleOption
-    ? content.replace(pattern, '')
+    ? content.replace(titleMatch[0]!, '')
     : content;
+  if (regularTitleMatch) {
+    return {
+      content: newContent.trim(),
+      contentTitle: toTextContentTitle(
+        regularTitleMatch
+          .groups!.title!.trim()
+          .replace(/\s*(?:\{#*[\w-]+\}|#+)$/, ''),
+      ).trim(),
+    };
+  }
   return {
     content: newContent.trim(),
-    contentTitle: toTextContentTitle(title.trim()).trim(),
+    contentTitle: toTextContentTitle(
+      alternateTitleMatch!.groups!.title!.trim().replace(/\s*=+$/, ''),
+    ).trim(),
   };
 }
 

packages/docusaurus-utils/src/urlUtils.ts

@@ -58,7 +58,7 @@ export function normalizeUrl(rawUrls: string[]): string {
       if (i > 0) {
         // Removing the starting slashes for each component but the first.
         component = component.replace(
-          /^[/]+/,
+          /^\/+/,
           // Special case where the first element of rawUrls is empty
           // ["", "/hello"] => /hello
           component[0] === '/' && !hasStartingSlash ? '/' : '',
@@ -68,7 +68,7 @@ export function normalizeUrl(rawUrls: string[]): string {
       hasEndingSlash = component[component.length - 1] === '/';
       // Removing the ending slashes for each component but the last. For the
       // last component we will combine multiple slashes to a single one.
-      component = component.replace(/[/]+$/, i < urls.length - 1 ? '' : '/');
+      component = component.replace(/\/+$/, i < urls.length - 1 ? '' : '/');
     }
 
     hasStartingSlash = true;