@docusaurus/utils - npm Package Compare versions

Comparing version 2.0.0-alpha.54 to 2.0.0-alpha.55

lib/index.d.ts

@@ -46,2 +46,3 @@ /**
 export declare function getSubFolder(file: string, refDir: string): string | null;
+export declare function createExcerpt(fileString: string): string | undefined;
 export declare function parse(fileString: string): {
@@ -48,0 +49,0 @@ frontMatter: {

lib/index.js

@@ -169,16 +169,48 @@ "use strict";
 // Regex for an import statement.
-const importRegexString = '^(.*import){1}(.+){0,1}\\s[\'"](.+)[\'"];';
+const importRegexString = '^(.*import){1}(.+){0,1}\\s[\'"](.+)[\'"];?';
+function createExcerpt(fileString) {
+    let fileContent = fileString.trimLeft();
+    if (RegExp(importRegexString).test(fileContent)) {
+        fileContent = fileContent
+            .replace(RegExp(importRegexString, 'gm'), '')
+            .trimLeft();
+    }
+    const fileLines = fileContent.split('\n');
+    for (let fileLine of fileLines) {
+        const cleanedLine = fileLine
+            // Remove HTML tags.
+            .replace(/<[^>]*>/g, '')
+            // Remove 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')
+            // Remove inline links.
+            .replace(/\[(.*?)\][\[\(].*?[\]\)]/g, '$1')
+            // Remove inline code.
+            .replace(/`(.+?)`/g, '$1')
+            // Remove images.
+            .replace(/\!\[(.*?)\][\[\(].*?[\]\)]/g, '')
+            // Remove blockquotes.
+            .replace(/^\s{0,3}>\s?/g, '')
+            // Remove footnotes.
+            .replace(/\[\^.+?\](\: .*?$)?/g, '')
+            // Remove admonition definition.
+            .replace(/(:{3}.*)/, '')
+            // Remove Emoji names within colons include preceding whitespace.
+            .replace(/\s?(:(::|[^:\n])+:)/g, '')
+            .trim();
+        if (cleanedLine) {
+            return cleanedLine;
+        }
+    }
+    return undefined;
+}
+exports.createExcerpt = createExcerpt;
 function parse(fileString) {
     const options = {
         excerpt: (file) => {
-            let fileContent = file.content.trimLeft();
             // Hacky way of stripping out import statements from the excerpt
             // TODO: Find a better way to do so, possibly by compiling the Markdown content,
             // stripping out HTML tags and obtaining the first line.
-            if (RegExp(importRegexString).test(fileContent)) {
-                fileContent = fileContent
-                    .replace(RegExp(importRegexString, 'gm'), '')
-                    .trimLeft();
-            }
-            file.excerpt = fileContent.split('\n', 1).shift();
+            file.excerpt = createExcerpt(file.content);
         },
@@ -185,0 +217,0 @@ };

package.json

 {
   "name": "@docusaurus/utils",
-  "version": "2.0.0-alpha.54",
+  "version": "2.0.0-alpha.55",
   "description": "Node utility functions for Docusaurus packages",
@@ -24,3 +24,3 @@ "main": "./lib/index.js",
   },
-  "gitHead": "753d117025935834af856f4751d0051911254d7b"
+  "gitHead": "d94a549cfbf054acf80e92fda0fbe8072f8e1d9d"
 }

src/__tests__/index.test.ts

@@ -20,2 +20,3 @@ /**
   aliasedSitePath,
+  createExcerpt,
 } from '../index';
@@ -296,2 +297,58 @@
   });
+
+  test('createExcerpt', () => {
+    const asserts = [
+      // Regular content
+      {
+        input: `
+          Lorem ipsum dolor sit amet, consectetur adipiscing elit. Vestibulum ex urna, molestie et sagittis ut, varius ac justo.
+
+          Nunc porttitor libero nec vulputate venenatis. Nam nec rhoncus mauris. Morbi tempus est et nibh maximus, tempus venenatis arcu lobortis.
+        `,
+        output:
+          'Lorem ipsum dolor sit amet, consectetur adipiscing elit. Vestibulum ex urna, molestie et sagittis ut, varius ac justo.',
+      },
+      // Content with imports declarations and Markdown markup, as well as Emoji
+      {
+        input: `
+          import Component from '@site/src/components/Component';
+          import Component from '@site/src/components/Component'
+
+          Lorem **ipsum** dolor sit \`amet\`, consectetur _adipiscing_ elit. [**Vestibulum**](https://wiktionary.org/wiki/vestibulum) ex urna, ~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.
+        `,
+        output:
+          'Lorem ipsum dolor sit amet, consectetur adipiscing elit. Vestibulum ex urna, molestie et sagittis ut, varius ac justo.',
+      },
+      // Content beginning with admonitions
+      {
+        input: `
+          import Component from '@site/src/components/Component'
+
+          :::caution
+
+          Lorem ipsum dolor sit amet, consectetur adipiscing elit.
+
+          :::
+
+          Nunc porttitor libero nec vulputate venenatis. Nam nec rhoncus mauris. Morbi tempus est et nibh maximus, tempus venenatis arcu lobortis.
+        `,
+        output: 'Lorem ipsum dolor sit amet, consectetur adipiscing elit.',
+      },
+      // Content beginning with heading
+      {
+        input: `
+          ## Lorem ipsum dolor sit amet
+
+          Nunc porttitor libero nec vulputate venenatis. Nam nec rhoncus mauris. Morbi tempus est et nibh maximus, tempus venenatis arcu lobortis.
+        `,
+        output: 'Lorem ipsum dolor sit amet',
+      },
+    ];
+
+    asserts.forEach((testCase) => {
+      expect(createExcerpt(testCase.input)).toEqual(testCase.output);
+    });
+  });
 });

src/index.ts

@@ -186,4 +186,47 @@ /**
 // Regex for an import statement.
-const importRegexString = '^(.*import){1}(.+){0,1}\\s[\'"](.+)[\'"];';
+const importRegexString = '^(.*import){1}(.+){0,1}\\s[\'"](.+)[\'"];?';
 
+export function createExcerpt(fileString: string): string | undefined {
+  let fileContent = fileString.trimLeft();
+
+  if (RegExp(importRegexString).test(fileContent)) {
+    fileContent = fileContent
+      .replace(RegExp(importRegexString, 'gm'), '')
+      .trimLeft();
+  }
+
+  const fileLines = fileContent.split('\n');
+
+  for (let fileLine of fileLines) {
+    const cleanedLine = fileLine
+      // Remove HTML tags.
+      .replace(/<[^>]*>/g, '')
+      // Remove 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')
+      // Remove inline links.
+      .replace(/\[(.*?)\][\[\(].*?[\]\)]/g, '$1')
+      // Remove inline code.
+      .replace(/`(.+?)`/g, '$1')
+      // Remove images.
+      .replace(/\!\[(.*?)\][\[\(].*?[\]\)]/g, '')
+      // Remove blockquotes.
+      .replace(/^\s{0,3}>\s?/g, '')
+      // Remove footnotes.
+      .replace(/\[\^.+?\](\: .*?$)?/g, '')
+      // Remove admonition definition.
+      .replace(/(:{3}.*)/, '')
+      // Remove Emoji names within colons include preceding whitespace.
+      .replace(/\s?(:(::|[^:\n])+:)/g, '')
+      .trim();
+
+    if (cleanedLine) {
+      return cleanedLine;
+    }
+  }
+
+  return undefined;
+}
+
 export function parse(
@@ -200,14 +243,6 @@ fileString: string,
     excerpt: (file: matter.GrayMatterFile<string>): void => {
-      let fileContent = file.content.trimLeft();
-
       // Hacky way of stripping out import statements from the excerpt
       // TODO: Find a better way to do so, possibly by compiling the Markdown content,
       // stripping out HTML tags and obtaining the first line.
-      if (RegExp(importRegexString).test(fileContent)) {
-        fileContent = fileContent
-          .replace(RegExp(importRegexString, 'gm'), '')
-          .trimLeft();
-      }
-
-      file.excerpt = fileContent.split('\n', 1).shift();
+      file.excerpt = createExcerpt(file.content);
     },
@@ -214,0 +249,0 @@ };

No alert changes

Improved metrics

Total package byte prevSize

1225735

increased by186.79%

Number of package files

11

increased by10%

Lines of code

965

increased by13%

No dependency changes