kivv

// ============================================================================= // kivv - arXiv Client Tests // ============================================================================= // Comprehensive unit tests for ArxivClient and ArxivQueryBuilder // Tests rate limiting, XML parsing, error handling, and query building // ============================================================================= import { describe, it, expect, beforeEach, vi } from 'vitest'; import { ArxivClient, ArxivQueryBuilder, type ArxivPaper } from '../../shared/arxiv-client'; // ============================================================================= // Mock Data - Sample arXiv API Responses // ============================================================================= const SAMPLE_ATOM_XML = `<?xml version="1.0" encoding="UTF-8"?> <feed xmlns="http://www.w3.org/2005/Atom"> <title>ArXiv Query: cat:cs.AI</title> <entry> <id>http://arxiv.org/abs/2101.12345v1</id> <title>Attention Is All You Need</title> <summary>We propose a new simple network architecture, the Transformer, based solely on attention mechanisms.</summary> <published>2017-06-12T17:57:34Z</published> <author><name>Ashish Vaswani</name></author> <author><name>Noam Shazeer</name></author> <category term="cs.AI"/> <category term="cs.LG"/> <link href="http://arxiv.org/abs/2101.12345" rel="alternate"/> <link href="http://arxiv.org/pdf/2101.12345" rel="related"/> </entry> <entry> <id>http://arxiv.org/abs/2202.54321v2</id> <title>BERT: Pre-training of Deep Bidirectional Transformers</title> <summary>We introduce a new language representation model called BERT.</summary> <published>2018-10-11T18:21:03Z</published> <author><name>Jacob Devlin</name></author> <author><name>Ming-Wei Chang</name></author> <author><name>Kenton Lee</name></author> <category term="cs.CL"/> <category term="cs.AI"/> <link href="http://arxiv.org/abs/2202.54321" rel="alternate"/> <link href="http://arxiv.org/pdf/2202.54321" rel="related"/> </entry> </feed>`; const MALFORMED_XML = `<?xml version="1.0" encoding="UTF-8"?> <feed xmlns="http://www.w3.org/2005/Atom"> <entry> <id>http://arxiv.org/abs/9999.99999v1</id> <title>Incomplete Entry</title> <!-- Missing summary, authors, categories --> </entry> </feed>`; const EMPTY_FEED_XML = `<?xml version="1.0" encoding="UTF-8"?> <feed xmlns="http://www.w3.org/2005/Atom"> <title>ArXiv Query: cat:nonexistent</title> </feed>`; // ============================================================================= // ArxivQueryBuilder Tests // ============================================================================= describe('ArxivQueryBuilder', () => { it('should build category query', () => { const query = new ArxivQueryBuilder() .addCategory('cs.AI') .build(); expect(query).toBe('cat:cs.AI'); }); it('should build multiple category query with OR', () => { const query = new ArxivQueryBuilder() .addCategory('cs.AI') .addCategory('cs.LG') .build('OR'); expect(query).toBe('cat:cs.AI OR cat:cs.LG'); }); it('should build multiple category query with AND', () => { const query = new ArxivQueryBuilder() .addCategory('cs.AI') .addCategory('cs.LG') .build('AND'); expect(query).toBe('cat:cs.AI AND cat:cs.LG'); }); it('should build keyword query', () => { const query = new ArxivQueryBuilder() .addKeyword('transformers') .build(); expect(query).toBe('all:transformers'); }); it('should build multiple keyword query', () => { const query = new ArxivQueryBuilder() .addKeyword('transformers') .addKeyword('attention') .build('OR'); expect(query).toBe('all:transformers OR all:attention'); }); it('should build title keyword query', () => { const query = new ArxivQueryBuilder() .addTitleKeyword('attention') .build(); expect(query).toBe('ti:attention'); }); it('should build author query', () => { const query = new ArxivQueryBuilder() .addAuthor('Vaswani') .build(); expect(query).toBe('au:Vaswani'); }); it('should build abstract keyword query', () => { const query = new ArxivQueryBuilder() .addAbstractKeyword('neural networks') .build(); expect(query).toBe('abs:neural networks'); }); it('should build complex mixed query', () => { const query = new ArxivQueryBuilder() .addCategory('cs.AI') .addKeyword('transformers') .addAuthor('Vaswani') .build('AND'); expect(query).toBe('cat:cs.AI AND all:transformers AND au:Vaswani'); }); it('should reset builder to empty state', () => { const builder = new ArxivQueryBuilder() .addCategory('cs.AI') .addKeyword('transformers'); const firstQuery = builder.build(); expect(firstQuery).toBe('cat:cs.AI OR all:transformers'); builder.reset(); const secondQuery = builder.addCategory('cs.LG').build(); expect(secondQuery).toBe('cat:cs.LG'); }); it('should default to OR operator', () => { const query = new ArxivQueryBuilder() .addCategory('cs.AI') .addCategory('cs.LG') .build(); // No operator specified expect(query).toBe('cat:cs.AI OR cat:cs.LG'); }); }); // ============================================================================= // ArxivClient - XML Parsing Tests // ============================================================================= describe('ArxivClient - XML Parsing', () => { let client: ArxivClient; beforeEach(() => { client = new ArxivClient(); }); it('should parse valid Atom XML correctly', async () => { // Mock fetch to return sample XML global.fetch = vi.fn(() => Promise.resolve({ ok: true, text: () => Promise.resolve(SAMPLE_ATOM_XML), } as Response) ); const papers = await client.search({ query: 'cat:cs.AI' }); expect(papers).toHaveLength(2); // Verify first paper expect(papers[0].arxiv_id).toBe('2101.12345'); expect(papers[0].title).toBe('Attention Is All You Need'); expect(papers[0].authors).toBe('Ashish Vaswani, Noam Shazeer'); expect(papers[0].abstract).toContain('Transformer'); expect(papers[0].published_date).toBe('2017-06-12T17:57:34Z'); expect(papers[0].categories).toBe('cs.AI, cs.LG'); expect(papers[0].arxiv_url).toBe('http://arxiv.org/abs/2101.12345'); expect(papers[0].pdf_url).toBe('http://arxiv.org/pdf/2101.12345'); // Verify second paper expect(papers[1].arxiv_id).toBe('2202.54321'); expect(papers[1].title).toBe('BERT: Pre-training of Deep Bidirectional Transformers'); expect(papers[1].authors).toBe('Jacob Devlin, Ming-Wei Chang, Kenton Lee'); expect(papers[1].abstract).toContain('BERT'); expect(papers[1].published_date).toBe('2018-10-11T18:21:03Z'); expect(papers[1].categories).toBe('cs.CL, cs.AI'); }); it('should strip version suffix from arXiv ID', async () => { global.fetch = vi.fn(() => Promise.resolve({ ok: true, text: () => Promise.resolve(SAMPLE_ATOM_XML), } as Response) ); const papers = await client.search({ query: 'cat:cs.AI' }); // XML contains "2101.12345v1" but should return "2101.12345" expect(papers[0].arxiv_id).toBe('2101.12345'); expect(papers[0].arxiv_id).not.toContain('v1'); // XML contains "2202.54321v2" but should return "2202.54321" expect(papers[1].arxiv_id).toBe('2202.54321'); expect(papers[1].arxiv_id).not.toContain('v2'); }); it('should handle multiple authors correctly', async () => { global.fetch = vi.fn(() => Promise.resolve({ ok: true, text: () => Promise.resolve(SAMPLE_ATOM_XML), } as Response) ); const papers = await client.search({ query: 'cat:cs.AI' }); expect(papers[0].authors).toBe('Ashish Vaswani, Noam Shazeer'); expect(papers[1].authors).toBe('Jacob Devlin, Ming-Wei Chang, Kenton Lee'); }); it('should handle multiple categories correctly', async () => { global.fetch = vi.fn(() => Promise.resolve({ ok: true, text: () => Promise.resolve(SAMPLE_ATOM_XML), } as Response) ); const papers = await client.search({ query: 'cat:cs.AI' }); expect(papers[0].categories).toBe('cs.AI, cs.LG'); expect(papers[1].categories).toBe('cs.CL, cs.AI'); }); it('should handle malformed XML gracefully', async () => { global.fetch = vi.fn(() => Promise.resolve({ ok: true, text: () => Promise.resolve(MALFORMED_XML), } as Response) ); const papers = await client.search({ query: 'cat:cs.AI' }); // Should skip entries with missing required fields expect(papers).toHaveLength(0); }); it('should handle empty feed XML', async () => { global.fetch = vi.fn(() => Promise.resolve({ ok: true, text: () => Promise.resolve(EMPTY_FEED_XML), } as Response) ); const papers = await client.search({ query: 'cat:nonexistent' }); expect(papers).toHaveLength(0); }); it('should use fallback URLs when links missing', async () => { const xmlWithoutLinks = `<?xml version="1.0" encoding="UTF-8"?> <feed xmlns="http://www.w3.org/2005/Atom"> <entry> <id>http://arxiv.org/abs/1234.56789v1</id> <title>Test Paper</title> <summary>Test abstract</summary> <published>2024-01-01T00:00:00Z</published> <author><name>Test Author</name></author> <category term="cs.AI"/> </entry> </feed>`; global.fetch = vi.fn(() => Promise.resolve({ ok: true, text: () => Promise.resolve(xmlWithoutLinks), } as Response) ); const papers = await client.search({ query: 'cat:cs.AI' }); expect(papers).toHaveLength(1); expect(papers[0].arxiv_url).toBe('http://arxiv.org/abs/1234.56789'); expect(papers[0].pdf_url).toBe('http://arxiv.org/pdf/1234.56789'); }); }); // ============================================================================= // ArxivClient - Rate Limiting Tests // ============================================================================= describe('ArxivClient - Rate Limiting', () => { let client: ArxivClient; beforeEach(() => { client = new ArxivClient(); vi.clearAllMocks(); // Mock fetch for all rate limit tests global.fetch = vi.fn(() => Promise.resolve({ ok: true, text: () => Promise.resolve(EMPTY_FEED_XML), } as Response) ); }); it('should enforce 3-second rate limit between requests', async () => { const startTime = Date.now(); // First request (no delay) await client.search({ query: 'cat:cs.AI' }); const firstRequestTime = Date.now() - startTime; // Second request (should wait ~3 seconds + jitter) await client.search({ query: 'cat:cs.LG' }); const secondRequestTime = Date.now() - startTime; // Second request should take at least 3000ms longer const timeBetweenRequests = secondRequestTime - firstRequestTime; expect(timeBetweenRequests).toBeGreaterThanOrEqual(3000); // Should also have jitter (100-500ms), so total should be <= 3500ms expect(timeBetweenRequests).toBeLessThanOrEqual(3600); }); it('should add random jitter to rate limit', async () => { const delays: number[] = []; // Make 3 requests and measure delays (reduced from 5 to avoid timeout) for (let i = 0; i < 3; i++) { const startTime = Date.now(); await client.search({ query: `cat:cs.AI${i}` }); if (i > 0) { delays.push(Date.now() - startTime); } } // All delays should be at least 3000ms for (const delay of delays) { expect(delay).toBeGreaterThanOrEqual(3000); } // Delays should have jitter variation (within 100ms buckets) // Just verify delays are in expected range (3000-3600ms) for (const delay of delays) { expect(delay).toBeLessThanOrEqual(3600); } }, 15000); // Increase timeout to 15 seconds it('should NOT delay first request', async () => { const startTime = Date.now(); await client.search({ query: 'cat:cs.AI' }); const elapsed = Date.now() - startTime; // First request should be immediate (< 100ms for network + processing) expect(elapsed).toBeLessThan(1000); }); it('should enforce rate limit across multiple sequential requests', async () => { const startTime = Date.now(); // Make 3 requests sequentially await client.search({ query: 'cat:cs.AI' }); await client.search({ query: 'cat:cs.LG' }); await client.search({ query: 'cat:cs.CL' }); const totalTime = Date.now() - startTime; // Total time should be at least 6 seconds (2 delays of 3s each) // First request is immediate, second waits 3s, third waits another 3s expect(totalTime).toBeGreaterThanOrEqual(6000); // Should also be less than 7.5s (accounting for jitter) expect(totalTime).toBeLessThan(7500); }, 15000); // Increase timeout to 15 seconds }); // ============================================================================= // ArxivClient - Error Handling Tests // ============================================================================= describe('ArxivClient - Error Handling', () => { let client: ArxivClient; beforeEach(() => { client = new ArxivClient(); }); it('should return empty array on HTTP error', async () => { global.fetch = vi.fn(() => Promise.resolve({ ok: false, status: 500, statusText: 'Internal Server Error', } as Response) ); const papers = await client.search({ query: 'cat:cs.AI' }); expect(papers).toHaveLength(0); }); it('should return empty array on network error', async () => { global.fetch = vi.fn(() => Promise.reject(new Error('Network error')) ); const papers = await client.search({ query: 'cat:cs.AI' }); expect(papers).toHaveLength(0); }); it('should return empty array on invalid XML', async () => { global.fetch = vi.fn(() => Promise.resolve({ ok: true, text: () => Promise.resolve('not valid xml at all'), } as Response) ); const papers = await client.search({ query: 'cat:cs.AI' }); expect(papers).toHaveLength(0); }); }); // ============================================================================= // ArxivClient - Search Parameters Tests // ============================================================================= describe('ArxivClient - Search Parameters', () => { let client: ArxivClient; beforeEach(() => { client = new ArxivClient(); }); it('should build URL with correct query parameters', async () => { let capturedUrl = ''; global.fetch = vi.fn((url) => { capturedUrl = url.toString(); return Promise.resolve({ ok: true, text: () => Promise.resolve(EMPTY_FEED_XML), } as Response); }); await client.search({ query: 'cat:cs.AI', maxResults: 20, start: 10, sortBy: 'relevance', sortOrder: 'ascending', }); expect(capturedUrl).toContain('search_query=cat%3Acs.AI'); expect(capturedUrl).toContain('max_results=20'); expect(capturedUrl).toContain('start=10'); expect(capturedUrl).toContain('sortBy=relevance'); expect(capturedUrl).toContain('sortOrder=ascending'); }); it('should use default parameters when not specified', async () => { let capturedUrl = ''; global.fetch = vi.fn((url) => { capturedUrl = url.toString(); return Promise.resolve({ ok: true, text: () => Promise.resolve(EMPTY_FEED_XML), } as Response); }); await client.search({ query: 'cat:cs.AI' }); expect(capturedUrl).toContain('max_results=10'); expect(capturedUrl).toContain('start=0'); expect(capturedUrl).toContain('sortBy=submittedDate'); expect(capturedUrl).toContain('sortOrder=descending'); }); });