auction-scrapper/tests/unit/services/scraper_service.spec.ts

import { test } from '@japa/runner'
import { ScraperService, ScraperError } from '#services/scraper_service'
import { readFile } from 'node:fs/promises'
import { fileURLToPath } from 'node:url'
import { dirname, join } from 'node:path'

// Get the directory path for fixtures
const __filename = fileURLToPath(import.meta.url)
const __dirname = dirname(__filename)
const fixturesPath = join(__dirname, '..', '..', 'fixtures')

test.group('ScraperService - parsePage', () => {
  test('should parse valid auction HTML correctly', async ({ assert }) => {
    const scraper = new ScraperService()
    const html = await readFile(join(fixturesPath, 'sample_auctions.html'), 'utf-8')

    const auctions = scraper.parsePage(html)

    assert.lengthOf(auctions, 3, 'Should parse all 3 auctions')

    // Verify first auction
    const firstAuction = auctions[0]
    assert.equal(firstAuction.auctionNum, 'AUC-2024-001')
    assert.equal(firstAuction.title, 'Поставка офисной мебели')
    assert.equal(firstAuction.organization, 'ООО "Тестовая Компания"')
    assert.equal(firstAuction.status, 'Активный')
    assert.equal(firstAuction.deadline, '15.03.2025')
    assert.equal(firstAuction.link, 'https://icetrade.by/trades/view/12345')
    assert.equal(firstAuction.description, 'Комплект офисной мебели для нового офиса')
  })

  test('should handle relative URLs and convert to absolute', async ({ assert }) => {
    const scraper = new ScraperService()
    const html = await readFile(join(fixturesPath, 'sample_auctions.html'), 'utf-8')

    const auctions = scraper.parsePage(html)

    // First two auctions have relative URLs
    assert.isTrue(auctions[0].link.startsWith('https://icetrade.by/'))
    assert.isTrue(auctions[1].link.startsWith('https://icetrade.by/'))

    // Third auction already has absolute URL
    assert.equal(auctions[2].link, 'https://icetrade.by/trades/view/12347')
  })

  test('should handle null deadline gracefully', async ({ assert }) => {
    const scraper = new ScraperService()
    const html = await readFile(join(fixturesPath, 'sample_auctions.html'), 'utf-8')

    const auctions = scraper.parsePage(html)

    // Third auction has empty deadline
    assert.isNull(auctions[2].deadline)
  })

  test('should handle null description when not present', async ({ assert }) => {
    const scraper = new ScraperService()
    const html = await readFile(join(fixturesPath, 'sample_auctions.html'), 'utf-8')

    const auctions = scraper.parsePage(html)

    // Second auction has no description
    assert.isNull(auctions[1].description)
  })

  test('should return empty array when no auctions table found', async ({ assert }) => {
    const scraper = new ScraperService()
    const html = await readFile(join(fixturesPath, 'empty_page.html'), 'utf-8')

    const auctions = scraper.parsePage(html)

    assert.lengthOf(auctions, 0, 'Should return empty array')
  })

  test('should skip invalid auction rows and continue parsing', async ({ assert }) => {
    const scraper = new ScraperService()
    const html = await readFile(join(fixturesPath, 'malformed_auction.html'), 'utf-8')

    const auctions = scraper.parsePage(html)

    // Both rows in malformed_auction.html should fail validation
    // First row: empty auction number
    // Second row: no href (empty string fails URL validation)
    assert.lengthOf(auctions, 0, 'Should skip all malformed rows')
  })

  test('should throw ScraperError for completely invalid HTML', ({ assert }) => {
    const scraper = new ScraperService()
    const invalidHtml = 'This is not valid HTML at all {{{'

    // Should not throw - cheerio is very lenient
    // But should return empty array since no table found
    const auctions = scraper.parsePage(invalidHtml)
    assert.lengthOf(auctions, 0)
  })
})

test.group('ScraperService - fetchPage', () => {
  test('should build correct URL with all parameters', ({ assert }) => {
    const scraper = new ScraperService()

    // Access private method via type assertion for testing
    const buildUrl = (scraper as any).buildUrl.bind(scraper)
    const url = buildUrl(1)

    assert.include(url, 'https://icetrade.by/trades/index')
    assert.include(url, 'p=1')
    assert.include(url, 'onPage=100')
    assert.include(url, 'sort=num%3Adesc')
    assert.include(url, 'zakup_type%5B1%5D=1')
    assert.include(url, 'zakup_type%5B2%5D=1')
    assert.include(url, 'r%5B1%5D=1')
    assert.include(url, 't%5BTrade%5D=1')
  })

  test('should include correct page number in URL', ({ assert }) => {
    const scraper = new ScraperService()
    const buildUrl = (scraper as any).buildUrl.bind(scraper)

    const url1 = buildUrl(1)
    const url5 = buildUrl(5)

    assert.include(url1, 'p=1')
    assert.include(url5, 'p=5')
  })
})

test.group('ScraperService - scrapeAuctions', () => {
  test('should throw error if maxPages is less than 1', async ({ assert }) => {
    const scraper = new ScraperService()

    await assert.rejects(
      async () => await scraper.scrapeAuctions(0),
      'maxPages must be at least 1'
    )

    await assert.rejects(
      async () => await scraper.scrapeAuctions(-5),
      'maxPages must be at least 1'
    )
  })

  // Note: Full integration tests for scrapeAuctions would require:
  // 1. Mocking fetch API
  // 2. Providing mock responses
  // These should be in integration tests, not unit tests
})

test.group('ScraperService - validation', () => {
  test('should validate auction number is not empty', ({ assert }) => {
    const scraper = new ScraperService()
    const html = `
      <table class="auctions w100">
        <tbody>
          <tr>
            <td>   </td>
            <td><a href="https://test.com">Title</a></td>
            <td>Org</td>
            <td>Status</td>
            <td></td>
          </tr>
        </tbody>
      </table>
    `

    const auctions = scraper.parsePage(html)
    assert.lengthOf(auctions, 0, 'Should skip auction with empty number')
  })

  test('should validate title is not empty', ({ assert }) => {
    const scraper = new ScraperService()
    const html = `
      <table class="auctions w100">
        <tbody>
          <tr>
            <td>AUC-001</td>
            <td><a href="https://test.com">   </a></td>
            <td>Org</td>
            <td>Status</td>
            <td></td>
          </tr>
        </tbody>
      </table>
    `

    const auctions = scraper.parsePage(html)
    assert.lengthOf(auctions, 0, 'Should skip auction with empty title')
  })

  test('should validate organization is not empty', ({ assert }) => {
    const scraper = new ScraperService()
    const html = `
      <table class="auctions w100">
        <tbody>
          <tr>
            <td>AUC-001</td>
            <td><a href="https://test.com">Title</a></td>
            <td>   </td>
            <td>Status</td>
            <td></td>
          </tr>
        </tbody>
      </table>
    `

    const auctions = scraper.parsePage(html)
    assert.lengthOf(auctions, 0, 'Should skip auction with empty organization')
  })

  test('should validate link is a valid URL', ({ assert }) => {
    const scraper = new ScraperService()
    const html = `
      <table class="auctions w100">
        <tbody>
          <tr>
            <td>AUC-001</td>
            <td><a href="">Title</a></td>
            <td>Org</td>
            <td>Status</td>
            <td></td>
          </tr>
        </tbody>
      </table>
    `

    const auctions = scraper.parsePage(html)
    assert.lengthOf(auctions, 0, 'Should skip auction with invalid URL')
  })

  test('should trim whitespace from all fields', ({ assert }) => {
    const scraper = new ScraperService()
    const html = `
      <table class="auctions w100">
        <tbody>
          <tr>
            <td>  AUC-001  </td>
            <td><a href="https://test.com">  Test Title  </a></td>
            <td>  Test Org  </td>
            <td>  Active  </td>
            <td>  2025-03-15  </td>
          </tr>
        </tbody>
      </table>
    `

    const auctions = scraper.parsePage(html)
    assert.lengthOf(auctions, 1)

    const auction = auctions[0]
    assert.equal(auction.auctionNum, 'AUC-001')
    assert.equal(auction.title, 'Test Title')
    assert.equal(auction.organization, 'Test Org')
    assert.equal(auction.status, 'Active')
    assert.equal(auction.deadline, '2025-03-15')
  })
})

test.group('ScraperService - ScraperError', () => {
  test('should create ScraperError with message', ({ assert }) => {
    const error = new ScraperError('Test error message')

    assert.equal(error.message, 'Test error message')
    assert.equal(error.name, 'ScraperError')
    assert.instanceOf(error, Error)
  })

  test('should create ScraperError with cause', ({ assert }) => {
    const originalError = new Error('Original error')
    const error = new ScraperError('Wrapped error', originalError)

    assert.equal(error.message, 'Wrapped error')
    assert.equal(error.cause, originalError)
  })
})