@fullstackdatasolutions/articles 1.2.2 → 1.3.0

package/src/__tests__/seoUtils.test.ts

@@ -1,4 +1,5 @@
 import { ArticlesConfig } from '../articlesConfig'
+import type { Article } from '../articleTypes'
 import {
   generateArticleStaticParams,
   generateArticlesIndexMetadata,
@@ -11,6 +12,8 @@ import {
   generateSeriesMetadata,
   getArticleSitemapEntries,
   generateRssFeed,
+  generateLlmsTxt,
+  generateLlmsFullTxt,
   resolveSearchMetadata,
   resolveSocialMetadata,
 } from '../seoUtils'
@@ -102,6 +105,15 @@ jest.mock('../server-articles', () => ({
       return `${cfg.siteUrl.replace(/\/$/, '')}/articles/${article.slug}.md`
     }
   ),
+  categoryToSlug: (category: string) =>
+    category
+      .toLowerCase()
+      .replaceAll(/\s+/g, '-')
+      .replaceAll(/[^a-z0-9-]/g, ''),
+  getArticleMarkdown: jest.fn(async (slug: string) =>
+    slug === 'missing-body' ? null : `Body of ${slug}.`
+  ),
+  buildMarkdownTwinHeader: jest.fn((article: { title: string }) => `HEADER(${article.title})\n\n`),
   getArticlesBySeries: jest.fn(async (seriesSlug: string) => {
     if (seriesSlug === 'new-gm-path') {
       return [
@@ -563,6 +575,71 @@ describe('getArticleSitemapEntries', () => {
     expect(categoryEntry.changeFrequency).toBe('weekly')
   })
 
+  it('derives category lastModified from the newest article in it, not now', async () => {
+    const entries = await getArticleSitemapEntries('https://example.com')
+    const campaigns = entries.find((e) => e.url.endsWith('/articles/category/campaigns'))
+    // getAllArticles mock returns article-one (2025-01-15) and article-two
+    // (2025-01-16), both uncategorized in the mock, so no date is derivable.
+    expect(campaigns).toBeDefined()
+    expect(campaigns!.lastModified).toBeUndefined()
+  })
+
+  it('emits series entries for every distinct seriesSlug', async () => {
+    const { getAllArticles } = jest.requireMock('../server-articles')
+    getAllArticles.mockResolvedValueOnce([
+      {
+        slug: 'one',
+        title: 'One',
+        excerpt: '',
+        author: '',
+        category: 'Campaigns',
+        categories: ['Campaigns'],
+        readTime: '',
+        featuredImage: '',
+        date: '2025-01-15',
+        seriesSlug: 'new-gm',
+      },
+      {
+        slug: 'two',
+        title: 'Two',
+        excerpt: '',
+        author: '',
+        category: 'Campaigns',
+        categories: ['Campaigns'],
+        readTime: '',
+        featuredImage: '',
+        date: '2025-03-01',
+        seriesSlug: 'new-gm',
+      },
+    ])
+
+    const entries = await getArticleSitemapEntries('https://example.com')
+    const series = entries.filter((e) => e.url.includes('/articles/series/'))
+    expect(series).toHaveLength(1)
+    expect(series[0].url).toBe('https://example.com/articles/series/new-gm')
+    expect(series[0].lastModified).toEqual(new Date('2025-03-01'))
+  })
+
+  it('omits paginated entries unless listingPagination is pages', async () => {
+    const entries = await getArticleSitemapEntries({
+      siteUrl: 'https://example.com',
+      siteName: 'Example',
+    })
+    expect(entries.some((e) => e.url.includes('/page/'))).toBe(false)
+  })
+
+  it('emits paginated listing entries starting at page 2 in pages mode', async () => {
+    const entries = await getArticleSitemapEntries({
+      siteUrl: 'https://example.com',
+      siteName: 'Example',
+      listingPagination: 'pages',
+      pageSize: 1,
+    })
+    const indexPages = entries.filter((e) => e.url.startsWith('https://example.com/articles/page/'))
+    // Two articles at pageSize 1 -> pages 1 and 2; page 1 is the listing URL.
+    expect(indexPages.map((e) => e.url)).toEqual(['https://example.com/articles/page/2'])
+  })
+
   it('returns empty array when fetching throws', async () => {
     const { getAllArticles } = jest.requireMock('../server-articles')
     getAllArticles.mockRejectedValueOnce(new Error('filesystem error'))
@@ -931,3 +1008,205 @@ describe('generateSeriesMetadata', () => {
     expect(meta.alternates?.canonical).toBe('https://example.com/articles/series/new-gm-path')
   })
 })
+
+describe('generateRssFeed full content', () => {
+  const baseConfig: ArticlesConfig = {
+    siteUrl: 'https://example.com',
+    siteName: 'Example Site',
+    description: 'Test description',
+  }
+
+  const article = {
+    slug: 'my-article',
+    title: 'My Article',
+    excerpt: 'A short excerpt',
+    date: '2024-01-15',
+    author: 'Jane Doe',
+    category: 'Campaigns',
+    categories: ['Campaigns'],
+    readTime: '3 min read',
+    featuredImage: '',
+    htmlContent: '<p>Full body.</p>',
+  } as Article
+
+  it('declares the content namespace', () => {
+    expect(generateRssFeed([], baseConfig)).toContain(
+      'xmlns:content="http://purl.org/rss/1.0/modules/content/"'
+    )
+  })
+
+  it('omits content:encoded by default', () => {
+    expect(generateRssFeed([article], baseConfig)).not.toContain('<content:encoded>')
+  })
+
+  it('emits content:encoded when fullContent is set', () => {
+    expect(generateRssFeed([article], baseConfig, { fullContent: true })).toContain(
+      '<content:encoded><![CDATA[<p>Full body.</p>]]></content:encoded>'
+    )
+  })
+
+  it('skips content:encoded for articles with no htmlContent', () => {
+    const xml = generateRssFeed([{ ...article, htmlContent: undefined }], baseConfig, {
+      fullContent: true,
+    })
+    expect(xml).not.toContain('<content:encoded>')
+  })
+
+  it('uses the configured language for the channel', () => {
+    expect(generateRssFeed([], baseConfig)).toContain('<language>en</language>')
+    expect(generateRssFeed([], { ...baseConfig, language: 'es' })).toContain(
+      '<language>es</language>'
+    )
+  })
+})
+
+describe('generateLlmsTxt', () => {
+  const baseConfig: ArticlesConfig = {
+    siteUrl: 'https://example.com/',
+    siteName: 'Example Site',
+    description: 'Test description',
+  }
+
+  const article = (overrides: Partial<Article> & Pick<Article, 'slug' | 'title'>): Article =>
+    ({
+      excerpt: `Excerpt for ${overrides.slug}.`,
+      date: '2025-01-15',
+      author: 'Jane Doe',
+      category: 'Campaigns',
+      categories: ['Campaigns'],
+      readTime: '3 min read',
+      featuredImage: '',
+      aiCrawl: true,
+      ...overrides,
+    }) as Article
+
+  it('lists only aiCrawl articles, grouped by category, linking to the .md twin', () => {
+    const output = generateLlmsTxt(
+      [
+        article({ slug: 'one', title: 'One' }),
+        article({ slug: 'two', title: 'Two', category: 'Volunteers', categories: ['Volunteers'] }),
+        article({ slug: 'blocked', title: 'Blocked', aiCrawl: false }),
+      ],
+      baseConfig
+    )
+
+    expect(output).toBe(
+      [
+        '# Example Site',
+        '',
+        '> Test description',
+        '',
+        '## Campaigns',
+        '',
+        '- [One](https://example.com/articles/one.md): Excerpt for one.',
+        '## Volunteers',
+        '',
+        '- [Two](https://example.com/articles/two.md): Excerpt for two.',
+        '## Collections',
+        '',
+        '- [Campaigns](https://example.com/articles/category/campaigns.md)',
+        '- [Volunteers](https://example.com/articles/category/volunteers.md)',
+        '',
+      ].join('\n')
+    )
+    expect(output).not.toContain('Blocked')
+  })
+
+  it('falls back to siteName when description is unset and handles an empty corpus', () => {
+    const output = generateLlmsTxt([], { siteUrl: 'https://example.com', siteName: 'Example Site' })
+    expect(output).toContain('> Example Site articles')
+    expect(output).toContain('_No articles available._')
+  })
+
+  it('groups uncategorized articles under Articles and omits an empty excerpt', () => {
+    const output = generateLlmsTxt(
+      [article({ slug: 'bare', title: 'Bare', category: '', excerpt: '' })],
+      baseConfig
+    )
+    expect(output).toContain('## Articles')
+    expect(output).toContain('- [Bare](https://example.com/articles/bare.md)\n')
+    expect(output).toContain('## Collections')
+    expect(output).toContain('- [Campaigns](https://example.com/articles/category/campaigns.md)')
+  })
+})
+
+describe('generateLlmsFullTxt', () => {
+  const baseConfig: ArticlesConfig = {
+    siteUrl: 'https://example.com',
+    siteName: 'Example Site',
+    description: 'Test description',
+  }
+
+  const article = (slug: string, aiCrawl = true): Article =>
+    ({
+      slug,
+      title: slug.toUpperCase(),
+      excerpt: '',
+      author: '',
+      category: 'Campaigns',
+      categories: ['Campaigns'],
+      readTime: '',
+      featuredImage: '',
+      aiCrawl,
+    }) as Article
+
+  it('concatenates header-prefixed bodies for opted-in articles only', async () => {
+    const output = await generateLlmsFullTxt(
+      [article('one'), article('blocked', false)],
+      baseConfig
+    )
+
+    expect(output).toBe(
+      ['# Example Site', '', '> Test description', '', 'HEADER(ONE)\n\nBody of one.'].join('\n')
+    )
+  })
+
+  it('skips articles whose markdown cannot be read', async () => {
+    const output = await generateLlmsFullTxt([article('missing-body')], baseConfig)
+    expect(output).toBe(['# Example Site', '', '> Test description', ''].join('\n'))
+  })
+
+  it('falls back to siteName when description is unset', async () => {
+    const output = await generateLlmsFullTxt([], {
+      siteUrl: 'https://example.com',
+      siteName: 'Example Site',
+    })
+    expect(output).toContain('> Example Site articles')
+  })
+})
+
+describe('titleTemplate', () => {
+  const base: ArticlesConfig = { siteUrl: 'https://example.com', siteName: 'Example Site' }
+
+  it('appends the site name by default', async () => {
+    const meta = await generateArticleMetadata('article-one', base)
+    expect(meta.title).toBe('Article One | Example Site')
+  })
+
+  it('drops the suffix with {title}', async () => {
+    const meta = await generateArticleMetadata('article-one', { ...base, titleTemplate: '{title}' })
+    expect(meta.title).toBe('Article One')
+  })
+
+  it('supports a custom arrangement of both placeholders', async () => {
+    const meta = await generateArticleMetadata('article-one', {
+      ...base,
+      titleTemplate: '{siteName}: {title}',
+    })
+    expect(meta.title).toBe('Example Site: Article One')
+  })
+
+  it('applies to index, category, series, and author titles too', async () => {
+    const config = { ...base, titleTemplate: '{title}' }
+    expect(generateArticlesIndexMetadata(config).title).toBe('Articles')
+    expect((await generateCategoryMetadata('campaigns', config)).title).toBe('Campaigns Articles')
+    expect((await generateSeriesMetadata('new-gm-path', config)).title).toBe('New GM Path Series')
+  })
+
+  it('leaves those titles suffixed under the default template', async () => {
+    expect(generateArticlesIndexMetadata(base).title).toBe('Articles | Example Site')
+    expect((await generateCategoryMetadata('campaigns', base)).title).toBe(
+      'Campaigns Articles | Example Site'
+    )
+  })
+})