feat(blog): warn duplicate and inline authors (#10224)

Co-authored-by: OzakIOne <OzakIOne@users.noreply.github.com>
Co-authored-by: sebastien <lorber.sebastien@gmail.com>

packages/docusaurus-plugin-content-blog/src/__tests__/authors.test.ts

@@ -5,7 +5,7 @@
  * LICENSE file in the root directory of this source tree.
  */
 
-import path from 'path';
+import * as path from 'path';
 import {
   type AuthorsMap,
   getAuthorsMap,

packages/docusaurus-plugin-content-blog/src/__tests__/authorsProblems.test.ts

@@ -0,0 +1,178 @@
+/**
+ * Copyright (c) Facebook, Inc. and its affiliates.
+ *
+ * This source code is licensed under the MIT license found in the
+ * LICENSE file in the root directory of this source tree.
+ */
+
+import {jest} from '@jest/globals';
+import {reportDuplicateAuthors, reportInlineAuthors} from '../authorsProblems';
+import type {Author} from '@docusaurus/plugin-content-blog';
+
+const blogSourceRelative = 'doc.md';
+
+describe('duplicate authors', () => {
+  function testReport({authors}: {authors: Author[]}) {
+    reportDuplicateAuthors({
+      authors,
+      blogSourceRelative,
+    });
+  }
+
+  it('allows duplicated inline authors', () => {
+    const authors: Author[] = [
+      {
+        name: 'Sébastien Lorber',
+      },
+      {
+        name: 'Sébastien Lorber',
+      },
+    ];
+
+    expect(() =>
+      testReport({
+        authors,
+      }),
+    ).not.toThrow();
+  });
+
+  it('rejects duplicated key authors', () => {
+    const authors: Author[] = [
+      {
+        key: 'slorber',
+        name: 'Sébastien Lorber 1',
+        title: 'some title',
+      },
+      {
+        key: 'slorber',
+        name: 'Sébastien Lorber 2',
+        imageURL: '/slorber.png',
+      },
+    ];
+
+    expect(() =>
+      testReport({
+        authors,
+      }),
+    ).toThrowErrorMatchingInlineSnapshot(`
+      "Duplicate blog post authors were found in blog post "doc.md" front matter:
+      - {"key":"slorber","name":"Sébastien Lorber 2","imageURL":"/slorber.png"}"
+    `);
+  });
+});
+
+describe('inline authors', () => {
+  type Options = Parameters<typeof reportInlineAuthors>[0]['options'];
+
+  function testReport({
+    authors,
+    options = {},
+  }: {
+    authors: Author[];
+    options?: Partial<Options>;
+  }) {
+    const defaultOptions: Options = {
+      onInlineAuthors: 'throw',
+      authorsMapPath: 'authors.yml',
+    };
+
+    reportInlineAuthors({
+      authors,
+      options: {
+        ...defaultOptions,
+        ...options,
+      },
+      blogSourceRelative,
+    });
+  }
+
+  it('allows predefined authors', () => {
+    const authors: Author[] = [
+      {
+        key: 'slorber',
+        name: 'Sébastien Lorber',
+      },
+      {
+        key: 'ozaki',
+        name: 'Clément Couriol',
+      },
+    ];
+
+    expect(() =>
+      testReport({
+        authors,
+      }),
+    ).not.toThrow();
+  });
+
+  it('rejects inline authors', () => {
+    const authors: Author[] = [
+      {
+        key: 'slorber',
+        name: 'Sébastien Lorber',
+      },
+      {name: 'Inline author 1'},
+      {
+        key: 'ozaki',
+        name: 'Clément Couriol',
+      },
+      {imageURL: '/inline-author2.png'},
+    ];
+
+    expect(() =>
+      testReport({
+        authors,
+      }),
+    ).toThrowErrorMatchingInlineSnapshot(`
+      "Some blog authors used in "doc.md" are not defined in "authors.yml":
+      - {"name":"Inline author 1"}
+      - {"imageURL":"/inline-author2.png"}
+
+      Note that we recommend to declare authors once in a "authors.yml" file and reference them by key in blog posts front matter to avoid author info duplication.
+      But if you want to allow inline blog authors, you can disable this message by setting onInlineAuthors: 'ignore' in your blog plugin options.
+      More info at https://docusaurus.io/docs/blog
+      "
+    `);
+  });
+
+  it('warn inline authors', () => {
+    const authors: Author[] = [
+      {
+        key: 'slorber',
+        name: 'Sébastien Lorber',
+      },
+      {name: 'Inline author 1'},
+      {
+        key: 'ozaki',
+        name: 'Clément Couriol',
+      },
+      {imageURL: '/inline-author2.png'},
+    ];
+
+    const consoleMock = jest
+      .spyOn(console, 'warn')
+      .mockImplementation(() => {});
+
+    expect(() =>
+      testReport({
+        authors,
+        options: {
+          onInlineAuthors: 'warn',
+        },
+      }),
+    ).not.toThrow();
+    expect(consoleMock).toHaveBeenCalledTimes(1);
+    expect(consoleMock.mock.calls[0]).toMatchInlineSnapshot(`
+      [
+        "[WARNING] Some blog authors used in "doc.md" are not defined in "authors.yml":
+      - {"name":"Inline author 1"}
+      - {"imageURL":"/inline-author2.png"}
+
+      Note that we recommend to declare authors once in a "authors.yml" file and reference them by key in blog posts front matter to avoid author info duplication.
+      But if you want to allow inline blog authors, you can disable this message by setting onInlineAuthors: 'ignore' in your blog plugin options.
+      More info at https://docusaurus.io/docs/blog
+      ",
+      ]
+    `);
+  });
+});

packages/docusaurus-plugin-content-blog/src/__tests__/feed.test.ts

@@ -101,6 +101,7 @@ describe.each(['atom', 'rss', 'json'])('%s', (feedType) => {
           defaultReadingTime({content}),
         truncateMarker: /<!--\s*truncate\s*-->/,
         onInlineTags: 'ignore',
+        onInlineAuthors: 'ignore',
       } as PluginOptions,
     );
 
@@ -143,6 +144,7 @@ describe.each(['atom', 'rss', 'json'])('%s', (feedType) => {
           defaultReadingTime({content}),
         truncateMarker: /<!--\s*truncate\s*-->/,
         onInlineTags: 'ignore',
+        onInlineAuthors: 'ignore',
       } as PluginOptions,
     );
 
@@ -197,6 +199,7 @@ describe.each(['atom', 'rss', 'json'])('%s', (feedType) => {
           defaultReadingTime({content}),
         truncateMarker: /<!--\s*truncate\s*-->/,
         onInlineTags: 'ignore',
+        onInlineAuthors: 'ignore',
       } as PluginOptions,
     );
 
@@ -242,6 +245,7 @@ describe.each(['atom', 'rss', 'json'])('%s', (feedType) => {
           defaultReadingTime({content}),
         truncateMarker: /<!--\s*truncate\s*-->/,
         onInlineTags: 'ignore',
+        onInlineAuthors: 'ignore',
       } as PluginOptions,
     );
 
@@ -287,6 +291,7 @@ describe.each(['atom', 'rss', 'json'])('%s', (feedType) => {
           defaultReadingTime({content}),
         truncateMarker: /<!--\s*truncate\s*-->/,
         onInlineTags: 'ignore',
+        onInlineAuthors: 'ignore',
       } as PluginOptions,
     );
 

packages/docusaurus-plugin-content-blog/src/authorsProblems.ts

@@ -0,0 +1,72 @@
+/**
+ * Copyright (c) Facebook, Inc. and its affiliates.
+ *
+ * This source code is licensed under the MIT license found in the
+ * LICENSE file in the root directory of this source tree.
+ */
+
+import _ from 'lodash';
+import logger from '@docusaurus/logger';
+import type {Author, PluginOptions} from '@docusaurus/plugin-content-blog';
+
+export function reportAuthorsProblems(params: {
+  authors: Author[];
+  blogSourceRelative: string;
+  options: Pick<PluginOptions, 'onInlineAuthors' | 'authorsMapPath'>;
+}): void {
+  reportInlineAuthors(params);
+  reportDuplicateAuthors(params);
+}
+
+export function reportInlineAuthors({
+  authors,
+  blogSourceRelative,
+  options: {onInlineAuthors, authorsMapPath},
+}: {
+  authors: Author[];
+  blogSourceRelative: string;
+  options: Pick<PluginOptions, 'onInlineAuthors' | 'authorsMapPath'>;
+}): void {
+  if (onInlineAuthors === 'ignore') {
+    return;
+  }
+  const inlineAuthors = authors.filter((author) => !author.key);
+  if (inlineAuthors.length > 0) {
+    logger.report(onInlineAuthors)(
+      logger.interpolate`Some blog authors used in path=${blogSourceRelative} are not defined in path=${authorsMapPath}:
+- ${inlineAuthors.map(authorToString).join('\n- ')}
+
+Note that we recommend to declare authors once in a path=${authorsMapPath} file and reference them by key in blog posts front matter to avoid author info duplication.
+But if you want to allow inline blog authors, you can disable this message by setting onInlineAuthors: 'ignore' in your blog plugin options.
+More info at url=${'https://docusaurus.io/docs/blog'}
+`,
+    );
+  }
+}
+
+export function reportDuplicateAuthors({
+  authors,
+  blogSourceRelative,
+}: {
+  authors: Author[];
+  blogSourceRelative: string;
+}): void {
+  const duplicateAuthors = _(authors)
+    // for now we only check for predefined authors duplicates
+    .filter((author) => !!author.key)
+    .groupBy((author) => author.key)
+    .pickBy((authorsByKey) => authorsByKey.length > 1)
+    // We only keep the "test" of all the duplicate groups
+    // The first author of a group is not really a duplicate...
+    .flatMap(([, ...rest]) => rest)
+    .value();
+
+  if (duplicateAuthors.length > 0) {
+    throw new Error(logger.interpolate`Duplicate blog post authors were found in blog post path=${blogSourceRelative} front matter:
+- ${duplicateAuthors.map(authorToString).join('\n- ')}`);
+  }
+}
+
+function authorToString(author: Author) {
+  return JSON.stringify(author);
+}

packages/docusaurus-plugin-content-blog/src/blogUtils.ts

@@ -30,6 +30,7 @@ import {
 import {getTagsFile} from '@docusaurus/utils-validation';
 import {validateBlogPostFrontMatter} from './frontMatter';
 import {type AuthorsMap, getAuthorsMap, getBlogPostAuthors} from './authors';
+import {reportAuthorsProblems} from './authorsProblems';
 import type {TagsFile} from '@docusaurus/utils';
 import type {LoadContext, ParseFrontMatter} from '@docusaurus/types';
 import type {
@@ -317,7 +318,13 @@ async function processBlogSourceFile(
     routeBasePath,
     tagsRouteBasePath,
   ]);
+
   const authors = getBlogPostAuthors({authorsMap, frontMatter, baseUrl});
+  reportAuthorsProblems({
+    authors,
+    blogSourceRelative,
+    options,
+  });
 
   const tags = normalizeTags({
     options,

packages/docusaurus-plugin-content-blog/src/options.ts

@@ -58,6 +58,7 @@ export const DEFAULT_OPTIONS: PluginOptions = {
   processBlogPosts: async () => undefined,
   onInlineTags: 'warn',
   tags: undefined,
+  onInlineAuthors: 'warn',
 };
 
 const PluginOptionSchema = Joi.object<PluginOptions>({
@@ -156,6 +157,9 @@ const PluginOptionSchema = Joi.object<PluginOptions>({
     .disallow('')
     .allow(null, false)
     .default(() => DEFAULT_OPTIONS.tags),
+  onInlineAuthors: Joi.string()
+    .equal('ignore', 'log', 'warn', 'throw')
+    .default(DEFAULT_OPTIONS.onInlineAuthors),
 }).default(DEFAULT_OPTIONS);
 
 export function validateOptions({

packages/docusaurus-plugin-content-blog/src/plugin-content-blog.d.ts

@@ -44,6 +44,8 @@ yarn workspace v1.22.19image` is a collocated image path, this entry will be the
   };
 
   export type Author = {
+    key?: string; // TODO temporary, need refactor
+
     /**
      * If `name` doesn't exist, an `imageURL` is expected.
      */
@@ -442,6 +444,8 @@ yarn workspace v1.22.19image` is a collocated image path, this entry will be the
        *  (filter, modify, delete, etc...).
        */
       processBlogPosts: ProcessBlogPostsFn;
+      /** The behavior of Docusaurus when it finds inline authors. */
+      onInlineAuthors: 'ignore' | 'log' | 'warn' | 'throw';
     };
 
   /**

packages/docusaurus-utils/src/tags.ts

@@ -28,7 +28,7 @@ export type TagsPluginOptions = {
   // TODO allow option tags later? | TagsFile;
   /** Path to the tags file. */
   tags: string | false | null | undefined;
-  /** The behavior of Docusaurus when it found inline tags. */
+  /** The behavior of Docusaurus when it finds inline tags. */
   onInlineTags: 'ignore' | 'log' | 'warn' | 'throw';
 };
 

website/_dogfooding/_docs tests/standalone.mdx

@@ -1,8 +1,7 @@
 ---
 tags:
   - b
-  - label: d
+  - d
-    permalink: d-custom-permalink
 ---
 
 # Standalone doc

website/_dogfooding/_docs tests/tags.yml

@@ -4,6 +4,7 @@ b:
   label: 'Label for tag b'
 c:
   permalink: '/permalink-for-tag-c'
+d:
 e:
   label: 'Label for tag e'
   description: 'Description for tag e'

website/_dogfooding/dogfooding.config.ts

@@ -93,6 +93,7 @@ export const dogfoodingPluginInstances: PluginConfig[] = [
           ? undefined
           : defaultReadingTime({content, options: {wordsPerMinute: 5}}),
       onInlineTags: 'warn',
+      onInlineAuthors: 'ignore',
       tags: 'tags.yml',
     } satisfies BlogOptions,
   ],

website/blog/2018/04-30-How-I-Converted-Profilo-To-Docusaurus.mdx

@@ -1,9 +1,6 @@
 ---
 title: How I Converted Profilo to Docusaurus in Under 2 Hours
-author: Christine Abernathy
+authors: [abernathyca]
-authorURL: http://twitter.com/abernathyca
-authorImageURL: https://github.com/caabernathy.png
-authorTwitter: abernathyca
 tags: [profilo, adoption]
 ---
 

website/blog/2018/09-11-Towards-Docusaurus-2.mdx

@@ -1,10 +1,6 @@
 ---
 title: Towards Docusaurus 2
-author: Endilie Yacop Sucipto
+authors: endiliey
-authorTitle: Maintainer of Docusaurus
-authorURL: https://github.com/endiliey
-authorImageURL: https://github.com/endiliey.png
-authorTwitter: endiliey
 tags: [new, adoption]
 ---
 

website/blog/2021/11-21-algolia-docsearch-migration/index.mdx

@@ -1,13 +1,6 @@
 ---
 title: DocSearch migration
-authors:
+authors: [shortcuts, slorber]
-  - name: Clément Vannicatte
-    title: Software Engineer @ Algolia
-    url: https://github.com/shortcuts
-    image_url: https://github.com/shortcuts.png
-    twitter: sh0rtcts
-  - key: slorber
-
 tags: [search]
 image: ./img/social-card.png
 ---

website/blog/authors.yml

@@ -43,3 +43,23 @@ Josh-Cena:
   url: https://joshcena.com/
   image_url: https://github.com/josh-cena.png
   email: sidachen2003@gmail.com
+
+endiliey:
+  name: Endilie Yacop Sucipto
+  title: Maintainer of Docusaurus
+  url: https://github.com/endiliey
+  image_url: https://github.com/endiliey.png
+  twitter: endiliey
+
+abernathyca:
+  name: Christine Abernathy
+  url: http://twitter.com/abernathyca
+  image_url: https://github.com/caabernathy.png
+  twitter: abernathyca
+
+shortcuts:
+  name: Clément Vannicatte
+  title: Software Engineer @ Algolia
+  url: https://github.com/shortcuts
+  image_url: https://github.com/shortcuts.png
+  twitter: sh0rtcts

website/docusaurus.config.ts

@@ -291,6 +291,7 @@ export default async function createConfigAsync() {
             copyright: `Copyright © ${new Date().getFullYear()} Facebook, Inc.`,
             language: defaultLocale,
           },
+          onInlineAuthors: 'warn',
         },
       ],
       [