// Title Extractor Unit Tests
// Test Type: Unit Test
import { describe, expect, it } from 'vitest'
import {
extractDocxTitle,
extractHtmlTitle,
extractMarkdownTitle,
extractPdfTitle,
extractTxtTitle,
fileNameToTitle,
} from '../title-extractor.js'
// ============================================
// Tests
// ============================================
describe('Title Extractor', () => {
// --------------------------------------------
// fileNameToTitle helper
// --------------------------------------------
describe('fileNameToTitle', () => {
it('should strip extension and replace hyphens/underscores with spaces', () => {
expect(fileNameToTitle('2024-annual-report.pdf')).toBe('2024 annual report')
})
it('should handle file names with multiple dots', () => {
expect(fileNameToTitle('report.v2.final.pdf')).toBe('report.v2.final')
})
it('should handle file names with underscores', () => {
expect(fileNameToTitle('my_document_title.md')).toBe('my document title')
})
it('should handle file names with mixed hyphens and underscores', () => {
expect(fileNameToTitle('project-plan_v2.txt')).toBe('project plan v2')
})
it('should handle file names with no extension', () => {
expect(fileNameToTitle('README')).toBe('README')
})
})
// --------------------------------------------
// extractMarkdownTitle
// --------------------------------------------
describe('extractMarkdownTitle', () => {
it('should extract title from YAML frontmatter', () => {
const text = '---\ntitle: My Document\ndate: 2024-01-01\n---\n\nContent here.'
const result = extractMarkdownTitle(text, 'test.md')
expect(result.title).toBe('My Document')
expect(result.source).toBe('metadata')
})
it('should extract title from YAML frontmatter with double quotes', () => {
const text = '---\ntitle: "My Quoted Document"\n---\n\nContent here.'
const result = extractMarkdownTitle(text, 'test.md')
expect(result.title).toBe('My Quoted Document')
expect(result.source).toBe('metadata')
})
it('should extract title from YAML frontmatter with single quotes', () => {
const text = "---\ntitle: 'My Single Quoted Document'\n---\n\nContent here."
const result = extractMarkdownTitle(text, 'test.md')
expect(result.title).toBe('My Single Quoted Document')
expect(result.source).toBe('metadata')
})
it('should extract first H1 heading when no frontmatter', () => {
const text = '# My Title\n\nContent here.'
const result = extractMarkdownTitle(text, 'test.md')
expect(result.title).toBe('My Title')
expect(result.source).toBe('content')
})
it('should prefer frontmatter over H1', () => {
const text = '---\ntitle: Frontmatter Title\n---\n\n# Heading Title\n\nContent here.'
const result = extractMarkdownTitle(text, 'test.md')
expect(result.title).toBe('Frontmatter Title')
expect(result.source).toBe('metadata')
})
it('should fall back to file name when no title found', () => {
const text = 'Just some plain text without any title markers.'
const result = extractMarkdownTitle(text, 'my-notes.md')
expect(result.title).toBe('my notes')
expect(result.source).toBe('filename')
})
it('should return source metadata for frontmatter, content for H1, filename for fallback', () => {
const frontmatter = extractMarkdownTitle('---\ntitle: Test\n---\n', 'test.md')
expect(frontmatter.source).toBe('metadata')
const h1 = extractMarkdownTitle('# Test\n', 'test.md')
expect(h1.source).toBe('content')
const fallback = extractMarkdownTitle('no title here', 'test.md')
expect(fallback.source).toBe('filename')
})
})
// --------------------------------------------
// extractTxtTitle
// --------------------------------------------
describe('extractTxtTitle', () => {
it('should extract first line as title when followed by empty line', () => {
const text = 'Document Title\n\nThis is the body text.'
const result = extractTxtTitle(text, 'document.txt')
expect(result.title).toBe('Document Title')
expect(result.source).toBe('content')
})
it('should fall back to file name when first line has no empty line after', () => {
const text = 'Line one\nLine two\nLine three'
const result = extractTxtTitle(text, 'my-notes.txt')
expect(result.title).toBe('my notes')
expect(result.source).toBe('filename')
})
it('should fall back to file name for empty text', () => {
const result = extractTxtTitle('', 'empty-file.txt')
expect(result.title).toBe('empty file')
expect(result.source).toBe('filename')
})
})
// --------------------------------------------
// extractHtmlTitle
// --------------------------------------------
describe('extractHtmlTitle', () => {
it('should use readability title when available', () => {
const result = extractHtmlTitle('Article Title', 'page.html')
expect(result.title).toBe('Article Title')
expect(result.source).toBe('content')
})
it('should fall back to file name when readability title is empty', () => {
const result = extractHtmlTitle('', 'my-page.html')
expect(result.title).toBe('my page')
expect(result.source).toBe('filename')
})
it('should fall back to file name when readability title is whitespace only', () => {
const result = extractHtmlTitle(' ', 'my-page.html')
expect(result.title).toBe('my page')
expect(result.source).toBe('filename')
})
})
// --------------------------------------------
// extractPdfTitle
// --------------------------------------------
describe('extractPdfTitle', () => {
it('should use PDF metadata title when available', () => {
const result = extractPdfTitle('Annual Report 2024', 'Some chunk text', 'report.pdf')
expect(result.title).toBe('Annual Report 2024')
expect(result.source).toBe('metadata')
})
it('should use first page chunk text when no metadata title', () => {
const result = extractPdfTitle(undefined, 'The Unity Game Designer Playbook', 'report.pdf')
expect(result.title).toBe('The Unity Game Designer Playbook')
expect(result.source).toBe('content')
})
it('should fall back to file name when no metadata and no chunk text', () => {
const result = extractPdfTitle(undefined, undefined, 'annual-report.pdf')
expect(result.title).toBe('annual report')
expect(result.source).toBe('filename')
})
it('should ignore metadata title if it looks like a file path', () => {
const result = extractPdfTitle('/home/user/document.pdf', undefined, 'my-doc.pdf')
expect(result.title).toBe('my doc')
expect(result.source).toBe('filename')
})
it('should ignore metadata title if it contains backslash path', () => {
const result = extractPdfTitle('C:\\Users\\doc.pdf', undefined, 'my-doc.pdf')
expect(result.title).toBe('my doc')
expect(result.source).toBe('filename')
})
it('should ignore metadata title if it is empty or whitespace', () => {
const result = extractPdfTitle(' ', undefined, 'my-doc.pdf')
expect(result.title).toBe('my doc')
expect(result.source).toBe('filename')
})
it('should prefer metadata over chunk text when both available', () => {
const result = extractPdfTitle('Metadata Title', 'Chunk Title', 'fallback.pdf')
expect(result.title).toBe('Metadata Title')
expect(result.source).toBe('metadata')
})
it('should fall back from file-path metadata to chunk text', () => {
const result = extractPdfTitle('/usr/local/doc.pdf', 'Real Title From Content', 'my-doc.pdf')
expect(result.title).toBe('Real Title From Content')
expect(result.source).toBe('content')
})
})
// --------------------------------------------
// extractDocxTitle
// --------------------------------------------
describe('extractDocxTitle', () => {
it('should extract first h1 from mammoth HTML output', () => {
const html = '<h1>Document Title</h1><p>Some content here.</p>'
const result = extractDocxTitle(html, 'document.docx')
expect(result.title).toBe('Document Title')
expect(result.source).toBe('content')
})
it('should fall back to file name when no h1 found', () => {
const html = '<p>Some content without heading.</p>'
const result = extractDocxTitle(html, 'my-document.docx')
expect(result.title).toBe('my document')
expect(result.source).toBe('filename')
})
it('should handle HTML with no heading tags', () => {
const html = '<p>Just a paragraph.</p><p>Another paragraph.</p>'
const result = extractDocxTitle(html, 'notes.docx')
expect(result.title).toBe('notes')
expect(result.source).toBe('filename')
})
it('should extract only the first h1 when multiple exist', () => {
const html = '<h1>First Title</h1><h1>Second Title</h1><p>Content.</p>'
const result = extractDocxTitle(html, 'document.docx')
expect(result.title).toBe('First Title')
expect(result.source).toBe('content')
})
})
})
