Add fetchPage() with axios, timeout, and retry logic

Implement fetchPage() function for HTTP scraping:
- Uses axios for HTTP GET requests
- Configurable timeout (default 30s) and User-Agent header
- Retry with exponential backoff (1s, 2s, 4s) on 5xx and network errors
- Does not retry on 4xx client errors
- Logs each attempt with attempt number and error details
- Returns HTML string on success, throws after retries exhausted

.github/workflows/deploy.yml

@@ -4,7 +4,6 @@ on:
   push:
     branches: [ main ]
   pull_request:
-    branches: [ main ]
   workflow_dispatch:
 
 env:
@@ -128,7 +127,7 @@ jobs:
           # Build JSON payload
           PAYLOAD=$(jq -n \
             --arg repo "${{ github.repository }}" \
-            --arg branch "${{ github.ref_name }}" \
+            --arg branch "${{ github.head_ref || github.ref_name }}" \
             --arg commit "${{ github.sha }}" \
             --arg commit_short "$(echo '${{ github.sha }}' | cut -c1-7)" \
             --arg commit_message "${{ github.event.head_commit.message || github.event.pull_request.title || 'N/A' }}" \

__tests__/scraper/fetchPage.test.js

@@ -0,0 +1,535 @@
+/**
+ * Tests for fetchPage function
+ *
+ * Acceptance Criteria:
+ * - Uses axios for HTTP GET
+ * - Enforces timeout from config (default 30s)
+ * - Sets User-Agent header from config
+ * - Implements retry with exponential backoff (1s, 2s, 4s)
+ * - Retries on 5xx errors and network timeouts
+ * - Does NOT retry on 4xx errors
+ * - Logs each attempt with attempt number
+ * - Returns HTML string on success
+ * - Throws error after all retries exhausted
+ */
+
+const axios = require('axios');
+
+// Mock axios
+jest.mock('axios');
+
+// Mock config
+jest.mock('../../config/scraper', () => ({
+  RETRY_CONFIG: {
+    maxRetries: 3,
+    baseDelay: 1000,
+    timeout: 30000
+  },
+  USER_AGENT: 'Mozilla/5.0 (compatible; ApartmentScraper/1.0)'
+}));
+
+// Import after mocking
+const { fetchPage } = require('../../services/scraperService');
+const config = require('../../config/scraper');
+
+describe('fetchPage', () => {
+  let mockLogger;
+
+  beforeEach(() => {
+    jest.clearAllMocks();
+    jest.useFakeTimers();
+
+    // Create mock logger
+    mockLogger = {
+      info: jest.fn(),
+      warn: jest.fn(),
+      error: jest.fn()
+    };
+  });
+
+  afterEach(() => {
+    jest.useRealTimers();
+  });
+
+  describe('successful requests', () => {
+    it('should return HTML string on success', async () => {
+      const htmlContent = '<html><body>Test content</body></html>';
+
+      axios.get.mockResolvedValueOnce({
+        status: 200,
+        data: htmlContent
+      });
+
+      const result = await fetchPage('https://example.com', mockLogger);
+
+      expect(result).toBe(htmlContent);
+      expect(axios.get).toHaveBeenCalledTimes(1);
+    });
+
+    it('should use axios for HTTP GET', async () => {
+      const url = 'https://example.com/apartments';
+
+      axios.get.mockResolvedValueOnce({
+        status: 200,
+        data: '<html></html>'
+      });
+
+      await fetchPage(url, mockLogger);
+
+      expect(axios.get).toHaveBeenCalledWith(url, expect.any(Object));
+    });
+
+    it('should set timeout from config', async () => {
+      axios.get.mockResolvedValueOnce({
+        status: 200,
+        data: '<html></html>'
+      });
+
+      await fetchPage('https://example.com', mockLogger);
+
+      expect(axios.get).toHaveBeenCalledWith(
+        expect.any(String),
+        expect.objectContaining({
+          timeout: config.RETRY_CONFIG.timeout
+        })
+      );
+    });
+
+    it('should set User-Agent header from config', async () => {
+      axios.get.mockResolvedValueOnce({
+        status: 200,
+        data: '<html></html>'
+      });
+
+      await fetchPage('https://example.com', mockLogger);
+
+      expect(axios.get).toHaveBeenCalledWith(
+        expect.any(String),
+        expect.objectContaining({
+          headers: expect.objectContaining({
+            'User-Agent': config.USER_AGENT
+          })
+        })
+      );
+    });
+
+    it('should log the attempt with attempt number', async () => {
+      axios.get.mockResolvedValueOnce({
+        status: 200,
+        data: '<html></html>'
+      });
+
+      await fetchPage('https://example.com', mockLogger);
+
+      expect(mockLogger.info).toHaveBeenCalledWith(
+        'Fetching page',
+        expect.objectContaining({
+          url: 'https://example.com',
+          attempt: 1
+        })
+      );
+    });
+
+    it('should log successful fetch with status and content length', async () => {
+      const htmlContent = '<html><body>Content</body></html>';
+
+      axios.get.mockResolvedValueOnce({
+        status: 200,
+        data: htmlContent
+      });
+
+      await fetchPage('https://example.com', mockLogger);
+
+      expect(mockLogger.info).toHaveBeenCalledWith(
+        'Page fetched successfully',
+        expect.objectContaining({
+          status: 200,
+          contentLength: htmlContent.length
+        })
+      );
+    });
+  });
+
+  describe('retry behavior on 5xx errors', () => {
+    it('should retry on 500 Internal Server Error', async () => {
+      const error500 = new Error('Internal Server Error');
+      error500.response = { status: 500 };
+
+      axios.get
+        .mockRejectedValueOnce(error500)
+        .mockResolvedValueOnce({ status: 200, data: '<html></html>' });
+
+      const promise = fetchPage('https://example.com', mockLogger);
+
+      // Fast-forward through the retry delay
+      await jest.advanceTimersByTimeAsync(1000);
+
+      const result = await promise;
+
+      expect(result).toBe('<html></html>');
+      expect(axios.get).toHaveBeenCalledTimes(2);
+    });
+
+    it('should retry on 502 Bad Gateway', async () => {
+      const error502 = new Error('Bad Gateway');
+      error502.response = { status: 502 };
+
+      axios.get
+        .mockRejectedValueOnce(error502)
+        .mockResolvedValueOnce({ status: 200, data: '<html></html>' });
+
+      const promise = fetchPage('https://example.com', mockLogger);
+      await jest.advanceTimersByTimeAsync(1000);
+
+      await promise;
+
+      expect(axios.get).toHaveBeenCalledTimes(2);
+    });
+
+    it('should retry on 503 Service Unavailable', async () => {
+      const error503 = new Error('Service Unavailable');
+      error503.response = { status: 503 };
+
+      axios.get
+        .mockRejectedValueOnce(error503)
+        .mockResolvedValueOnce({ status: 200, data: '<html></html>' });
+
+      const promise = fetchPage('https://example.com', mockLogger);
+      await jest.advanceTimersByTimeAsync(1000);
+
+      await promise;
+
+      expect(axios.get).toHaveBeenCalledTimes(2);
+    });
+
+    it('should retry on 504 Gateway Timeout', async () => {
+      const error504 = new Error('Gateway Timeout');
+      error504.response = { status: 504 };
+
+      axios.get
+        .mockRejectedValueOnce(error504)
+        .mockResolvedValueOnce({ status: 200, data: '<html></html>' });
+
+      const promise = fetchPage('https://example.com', mockLogger);
+      await jest.advanceTimersByTimeAsync(1000);
+
+      await promise;
+
+      expect(axios.get).toHaveBeenCalledTimes(2);
+    });
+  });
+
+  describe('retry behavior on network errors', () => {
+    it('should retry on network timeout (ECONNABORTED)', async () => {
+      const timeoutError = new Error('timeout of 30000ms exceeded');
+      timeoutError.code = 'ECONNABORTED';
+      // Network errors don't have a response property
+
+      axios.get
+        .mockRejectedValueOnce(timeoutError)
+        .mockResolvedValueOnce({ status: 200, data: '<html></html>' });
+
+      const promise = fetchPage('https://example.com', mockLogger);
+      await jest.advanceTimersByTimeAsync(1000);
+
+      const result = await promise;
+
+      expect(result).toBe('<html></html>');
+      expect(axios.get).toHaveBeenCalledTimes(2);
+    });
+
+    it('should retry on DNS resolution failure (ENOTFOUND)', async () => {
+      const dnsError = new Error('getaddrinfo ENOTFOUND');
+      dnsError.code = 'ENOTFOUND';
+
+      axios.get
+        .mockRejectedValueOnce(dnsError)
+        .mockResolvedValueOnce({ status: 200, data: '<html></html>' });
+
+      const promise = fetchPage('https://example.com', mockLogger);
+      await jest.advanceTimersByTimeAsync(1000);
+
+      await promise;
+
+      expect(axios.get).toHaveBeenCalledTimes(2);
+    });
+
+    it('should retry on connection refused (ECONNREFUSED)', async () => {
+      const connError = new Error('connect ECONNREFUSED');
+      connError.code = 'ECONNREFUSED';
+
+      axios.get
+        .mockRejectedValueOnce(connError)
+        .mockResolvedValueOnce({ status: 200, data: '<html></html>' });
+
+      const promise = fetchPage('https://example.com', mockLogger);
+      await jest.advanceTimersByTimeAsync(1000);
+
+      await promise;
+
+      expect(axios.get).toHaveBeenCalledTimes(2);
+    });
+  });
+
+  describe('no retry on 4xx errors', () => {
+    it('should NOT retry on 400 Bad Request', async () => {
+      const error400 = new Error('Bad Request');
+      error400.response = { status: 400 };
+
+      axios.get.mockRejectedValueOnce(error400);
+
+      await expect(fetchPage('https://example.com', mockLogger))
+        .rejects.toThrow();
+
+      expect(axios.get).toHaveBeenCalledTimes(1);
+    });
+
+    it('should NOT retry on 401 Unauthorized', async () => {
+      const error401 = new Error('Unauthorized');
+      error401.response = { status: 401 };
+
+      axios.get.mockRejectedValueOnce(error401);
+
+      await expect(fetchPage('https://example.com', mockLogger))
+        .rejects.toThrow();
+
+      expect(axios.get).toHaveBeenCalledTimes(1);
+    });
+
+    it('should NOT retry on 403 Forbidden', async () => {
+      const error403 = new Error('Forbidden');
+      error403.response = { status: 403 };
+
+      axios.get.mockRejectedValueOnce(error403);
+
+      await expect(fetchPage('https://example.com', mockLogger))
+        .rejects.toThrow();
+
+      expect(axios.get).toHaveBeenCalledTimes(1);
+    });
+
+    it('should NOT retry on 404 Not Found', async () => {
+      const error404 = new Error('Not Found');
+      error404.response = { status: 404 };
+
+      axios.get.mockRejectedValueOnce(error404);
+
+      await expect(fetchPage('https://example.com', mockLogger))
+        .rejects.toThrow();
+
+      expect(axios.get).toHaveBeenCalledTimes(1);
+    });
+
+    it('should NOT retry on 429 Too Many Requests', async () => {
+      const error429 = new Error('Too Many Requests');
+      error429.response = { status: 429 };
+
+      axios.get.mockRejectedValueOnce(error429);
+
+      await expect(fetchPage('https://example.com', mockLogger))
+        .rejects.toThrow();
+
+      expect(axios.get).toHaveBeenCalledTimes(1);
+    });
+  });
+
+  describe('exponential backoff', () => {
+    it('should use exponential backoff delays: 1s, 2s, 4s', async () => {
+      const error500 = new Error('Internal Server Error');
+      error500.response = { status: 500 };
+
+      // Fail 3 times, then succeed
+      axios.get
+        .mockRejectedValueOnce(error500)
+        .mockRejectedValueOnce(error500)
+        .mockRejectedValueOnce(error500)
+        .mockResolvedValueOnce({ status: 200, data: '<html></html>' });
+
+      const promise = fetchPage('https://example.com', mockLogger);
+
+      // First retry: 1 second delay
+      await jest.advanceTimersByTimeAsync(1000);
+
+      // Second retry: 2 second delay
+      await jest.advanceTimersByTimeAsync(2000);
+
+      // Third retry: 4 second delay
+      await jest.advanceTimersByTimeAsync(4000);
+
+      await promise;
+
+      expect(axios.get).toHaveBeenCalledTimes(4);
+    });
+
+    it('should log each retry attempt with attempt number', async () => {
+      const error500 = new Error('Internal Server Error');
+      error500.response = { status: 500 };
+
+      axios.get
+        .mockRejectedValueOnce(error500)
+        .mockRejectedValueOnce(error500)
+        .mockResolvedValueOnce({ status: 200, data: '<html></html>' });
+
+      const promise = fetchPage('https://example.com', mockLogger);
+
+      await jest.advanceTimersByTimeAsync(1000);
+      await jest.advanceTimersByTimeAsync(2000);
+
+      await promise;
+
+      // Check that attempt numbers were logged
+      expect(mockLogger.info).toHaveBeenCalledWith(
+        'Fetching page',
+        expect.objectContaining({ attempt: 1 })
+      );
+      expect(mockLogger.info).toHaveBeenCalledWith(
+        'Fetching page',
+        expect.objectContaining({ attempt: 2 })
+      );
+      expect(mockLogger.info).toHaveBeenCalledWith(
+        'Fetching page',
+        expect.objectContaining({ attempt: 3 })
+      );
+    });
+
+    it('should log failed attempts with error details', async () => {
+      const error500 = new Error('Internal Server Error');
+      error500.response = { status: 500 };
+
+      axios.get
+        .mockRejectedValueOnce(error500)
+        .mockResolvedValueOnce({ status: 200, data: '<html></html>' });
+
+      const promise = fetchPage('https://example.com', mockLogger);
+      await jest.advanceTimersByTimeAsync(1000);
+
+      await promise;
+
+      expect(mockLogger.warn).toHaveBeenCalledWith(
+        'Fetch attempt failed',
+        expect.objectContaining({
+          attempt: 1,
+          statusCode: 500,
+          isRetryable: true
+        })
+      );
+    });
+
+    it('should log wait time before retry', async () => {
+      const error500 = new Error('Internal Server Error');
+      error500.response = { status: 500 };
+
+      axios.get
+        .mockRejectedValueOnce(error500)
+        .mockResolvedValueOnce({ status: 200, data: '<html></html>' });
+
+      const promise = fetchPage('https://example.com', mockLogger);
+      await jest.advanceTimersByTimeAsync(1000);
+
+      await promise;
+
+      expect(mockLogger.info).toHaveBeenCalledWith(
+        'Waiting before retry',
+        expect.objectContaining({ delay: 1000 })
+      );
+    });
+  });
+
+  describe('retry exhaustion', () => {
+    // Use real timers for exhaustion tests to avoid promise handling issues
+    beforeEach(() => {
+      jest.useRealTimers();
+    });
+
+    afterEach(() => {
+      jest.useFakeTimers();
+    });
+
+    it('should throw error after all retries exhausted (4 attempts total)', async () => {
+      // Override config for faster tests
+      const originalBaseDelay = config.RETRY_CONFIG.baseDelay;
+      config.RETRY_CONFIG.baseDelay = 1; // 1ms instead of 1000ms
+
+      try {
+        const error500 = new Error('Internal Server Error');
+        error500.response = { status: 500 };
+
+        // All 4 attempts fail (1 initial + 3 retries)
+        axios.get
+          .mockRejectedValueOnce(error500)
+          .mockRejectedValueOnce(error500)
+          .mockRejectedValueOnce(error500)
+          .mockRejectedValueOnce(error500);
+
+        await expect(fetchPage('https://example.com', mockLogger))
+          .rejects.toThrow('Internal Server Error');
+
+        expect(axios.get).toHaveBeenCalledTimes(4);
+      } finally {
+        config.RETRY_CONFIG.baseDelay = originalBaseDelay;
+      }
+    });
+
+    it('should throw the last error when retries are exhausted', async () => {
+      // Override config for faster tests
+      const originalBaseDelay = config.RETRY_CONFIG.baseDelay;
+      config.RETRY_CONFIG.baseDelay = 1; // 1ms instead of 1000ms
+
+      try {
+        const error500 = new Error('Server Error');
+        error500.response = { status: 500 };
+
+        axios.get.mockRejectedValue(error500);
+
+        await expect(fetchPage('https://example.com', mockLogger))
+          .rejects.toBe(error500);
+      } finally {
+        config.RETRY_CONFIG.baseDelay = originalBaseDelay;
+      }
+    });
+  });
+
+  describe('axios configuration', () => {
+    it('should set maxRedirects to 5', async () => {
+      axios.get.mockResolvedValueOnce({
+        status: 200,
+        data: '<html></html>'
+      });
+
+      await fetchPage('https://example.com', mockLogger);
+
+      expect(axios.get).toHaveBeenCalledWith(
+        expect.any(String),
+        expect.objectContaining({
+          maxRedirects: 5
+        })
+      );
+    });
+
+    it('should accept 2xx and 3xx status codes via validateStatus', async () => {
+      axios.get.mockResolvedValueOnce({
+        status: 200,
+        data: '<html></html>'
+      });
+
+      await fetchPage('https://example.com', mockLogger);
+
+      const callArgs = axios.get.mock.calls[0][1];
+      const validateStatus = callArgs.validateStatus;
+
+      // Should accept 2xx
+      expect(validateStatus(200)).toBe(true);
+      expect(validateStatus(201)).toBe(true);
+      expect(validateStatus(204)).toBe(true);
+
+      // Should accept 3xx
+      expect(validateStatus(301)).toBe(true);
+      expect(validateStatus(302)).toBe(true);
+
+      // Should reject 4xx and 5xx
+      expect(validateStatus(400)).toBe(false);
+      expect(validateStatus(404)).toBe(false);
+      expect(validateStatus(500)).toBe(false);
+    });
+  });
+});

__tests__/scraper/scraperConfig.test.js

@@ -0,0 +1,192 @@
+/**
+ * Tests for config/scraper.js
+ * Tests scraper configuration constants and environment variable overrides
+ */
+
+describe('config/scraper', () => {
+  // Store original env values to restore after tests
+  const originalEnv = { ...process.env };
+
+  beforeEach(() => {
+    // Clear all scraper-related env vars before each test
+    delete process.env.SCRAPER_SCHEDULE;
+    delete process.env.SCRAPER_TIMEZONE;
+    delete process.env.SCRAPER_ENABLED;
+    delete process.env.SCRAPER_TIMEOUT;
+    delete process.env.SCRAPER_UNITS_COLLECTION;
+    delete process.env.SCRAPER_PRICES_COLLECTION;
+    delete process.env.SCRAPER_SUMMARIES_COLLECTION;
+    delete process.env.SCRAPER_RUNS_COLLECTION;
+
+    // Clear the require cache to reload config with fresh env
+    jest.resetModules();
+  });
+
+  afterAll(() => {
+    // Restore original environment
+    process.env = originalEnv;
+  });
+
+  describe('TARGET_URL', () => {
+    test('should be set to the correct apartment listing URL', () => {
+      const config = require('../../config/scraper');
+      expect(config.TARGET_URL).toBe(
+        'https://countryclubtowersandgardens.com/property/country-club-towers/apartments/?spaces_tab=unit'
+      );
+    });
+  });
+
+  describe('SCRAPER_SCHEDULE', () => {
+    test('should default to "0 6 * * *" when env var not set', () => {
+      const config = require('../../config/scraper');
+      expect(config.SCRAPER_SCHEDULE).toBe('0 6 * * *');
+    });
+
+    test('should use env var override when set', () => {
+      process.env.SCRAPER_SCHEDULE = '0 12 * * *';
+      const config = require('../../config/scraper');
+      expect(config.SCRAPER_SCHEDULE).toBe('0 12 * * *');
+    });
+  });
+
+  describe('SCRAPER_TIMEZONE', () => {
+    test('should default to "UTC" when env var not set', () => {
+      const config = require('../../config/scraper');
+      expect(config.SCRAPER_TIMEZONE).toBe('UTC');
+    });
+
+    test('should use env var override when set', () => {
+      process.env.SCRAPER_TIMEZONE = 'America/Denver';
+      const config = require('../../config/scraper');
+      expect(config.SCRAPER_TIMEZONE).toBe('America/Denver');
+    });
+  });
+
+  describe('SCRAPER_ENABLED', () => {
+    test('should default to true when env var not set', () => {
+      const config = require('../../config/scraper');
+      expect(config.SCRAPER_ENABLED).toBe(true);
+    });
+
+    test('should be false when env var is set to "false"', () => {
+      process.env.SCRAPER_ENABLED = 'false';
+      const config = require('../../config/scraper');
+      expect(config.SCRAPER_ENABLED).toBe(false);
+    });
+
+    test('should be true when env var is set to any other value', () => {
+      process.env.SCRAPER_ENABLED = 'true';
+      const config = require('../../config/scraper');
+      expect(config.SCRAPER_ENABLED).toBe(true);
+    });
+
+    test('should be true when env var is empty string', () => {
+      process.env.SCRAPER_ENABLED = '';
+      const config = require('../../config/scraper');
+      expect(config.SCRAPER_ENABLED).toBe(true);
+    });
+  });
+
+  describe('SCRAPER_TIMEOUT', () => {
+    test('should default to 30000 when env var not set', () => {
+      const config = require('../../config/scraper');
+      expect(config.SCRAPER_TIMEOUT).toBe(30000);
+    });
+
+    test('should use env var override when set', () => {
+      process.env.SCRAPER_TIMEOUT = '60000';
+      const config = require('../../config/scraper');
+      expect(config.SCRAPER_TIMEOUT).toBe(60000);
+    });
+
+    test('should parse string to integer', () => {
+      process.env.SCRAPER_TIMEOUT = '45000';
+      const config = require('../../config/scraper');
+      expect(typeof config.SCRAPER_TIMEOUT).toBe('number');
+      expect(config.SCRAPER_TIMEOUT).toBe(45000);
+    });
+  });
+
+  describe('USER_AGENT', () => {
+    test('should be a valid User-Agent string', () => {
+      const config = require('../../config/scraper');
+      expect(config.USER_AGENT).toBe('Mozilla/5.0 (compatible; ApartmentScraper/1.0)');
+    });
+  });
+
+  describe('RETRY_CONFIG', () => {
+    test('should have maxRetries property', () => {
+      const config = require('../../config/scraper');
+      expect(config.RETRY_CONFIG).toHaveProperty('maxRetries');
+      expect(config.RETRY_CONFIG.maxRetries).toBe(3);
+    });
+
+    test('should have baseDelay property', () => {
+      const config = require('../../config/scraper');
+      expect(config.RETRY_CONFIG).toHaveProperty('baseDelay');
+      expect(config.RETRY_CONFIG.baseDelay).toBe(1000);
+    });
+
+    test('should have timeout property', () => {
+      const config = require('../../config/scraper');
+      expect(config.RETRY_CONFIG).toHaveProperty('timeout');
+      expect(config.RETRY_CONFIG.timeout).toBe(30000);
+    });
+
+    test('timeout should respect SCRAPER_TIMEOUT env var', () => {
+      process.env.SCRAPER_TIMEOUT = '45000';
+      const config = require('../../config/scraper');
+      expect(config.RETRY_CONFIG.timeout).toBe(45000);
+    });
+  });
+
+  describe('COLLECTIONS', () => {
+    describe('default values', () => {
+      test('UNITS should default to "units_migration_test"', () => {
+        const config = require('../../config/scraper');
+        expect(config.COLLECTIONS.UNITS).toBe('units_migration_test');
+      });
+
+      test('PRICES should default to "unit_prices_migration_test"', () => {
+        const config = require('../../config/scraper');
+        expect(config.COLLECTIONS.PRICES).toBe('unit_prices_migration_test');
+      });
+
+      test('DAILY_SUMMARIES should default to "daily_summaries"', () => {
+        const config = require('../../config/scraper');
+        expect(config.COLLECTIONS.DAILY_SUMMARIES).toBe('daily_summaries');
+      });
+
+      test('SCRAPER_RUNS should default to "scraper_runs"', () => {
+        const config = require('../../config/scraper');
+        expect(config.COLLECTIONS.SCRAPER_RUNS).toBe('scraper_runs');
+      });
+    });
+
+    describe('env var overrides', () => {
+      test('UNITS should respect SCRAPER_UNITS_COLLECTION env var', () => {
+        process.env.SCRAPER_UNITS_COLLECTION = 'units_custom';
+        const config = require('../../config/scraper');
+        expect(config.COLLECTIONS.UNITS).toBe('units_custom');
+      });
+
+      test('PRICES should respect SCRAPER_PRICES_COLLECTION env var', () => {
+        process.env.SCRAPER_PRICES_COLLECTION = 'prices_custom';
+        const config = require('../../config/scraper');
+        expect(config.COLLECTIONS.PRICES).toBe('prices_custom');
+      });
+
+      test('DAILY_SUMMARIES should respect SCRAPER_SUMMARIES_COLLECTION env var', () => {
+        process.env.SCRAPER_SUMMARIES_COLLECTION = 'summaries_custom';
+        const config = require('../../config/scraper');
+        expect(config.COLLECTIONS.DAILY_SUMMARIES).toBe('summaries_custom');
+      });
+
+      test('SCRAPER_RUNS should respect SCRAPER_RUNS_COLLECTION env var', () => {
+        process.env.SCRAPER_RUNS_COLLECTION = 'runs_custom';
+        const config = require('../../config/scraper');
+        expect(config.COLLECTIONS.SCRAPER_RUNS).toBe('runs_custom');
+      });
+    });
+  });
+});

config/scraper.js

@@ -0,0 +1,37 @@
+/**
+ * Scraper Configuration Module
+ *
+ * Contains all constants and settings for the apartment price scraper.
+ * All values can be overridden via environment variables for deployment flexibility.
+ *
+ * See py_migration/phase-3/3.2-TECHNICAL-DESIGN.md Section 5.3 for full documentation.
+ */
+
+module.exports = {
+  // Target URL for apartment listings
+  TARGET_URL: 'https://countryclubtowersandgardens.com/property/country-club-towers/apartments/?spaces_tab=unit',
+
+  // Scheduling configuration
+  SCRAPER_SCHEDULE: process.env.SCRAPER_SCHEDULE || '0 6 * * *',
+  SCRAPER_TIMEZONE: process.env.SCRAPER_TIMEZONE || 'UTC',
+  SCRAPER_ENABLED: process.env.SCRAPER_ENABLED !== 'false',
+
+  // HTTP settings
+  SCRAPER_TIMEOUT: parseInt(process.env.SCRAPER_TIMEOUT, 10) || 30000,
+  USER_AGENT: 'Mozilla/5.0 (compatible; ApartmentScraper/1.0)',
+
+  // Retry configuration for HTTP requests
+  RETRY_CONFIG: {
+    maxRetries: 3,
+    baseDelay: 1000, // 1 second, exponential backoff: 1s, 2s, 4s
+    timeout: parseInt(process.env.SCRAPER_TIMEOUT, 10) || 30000
+  },
+
+  // MongoDB collection names (environment variable overrides for development isolation)
+  COLLECTIONS: {
+    UNITS: process.env.SCRAPER_UNITS_COLLECTION || 'units_migration_test',
+    PRICES: process.env.SCRAPER_PRICES_COLLECTION || 'unit_prices_migration_test',
+    DAILY_SUMMARIES: process.env.SCRAPER_SUMMARIES_COLLECTION || 'daily_summaries',
+    SCRAPER_RUNS: process.env.SCRAPER_RUNS_COLLECTION || 'scraper_runs'
+  }
+};

services/scraperService.js

@@ -0,0 +1,114 @@
+/**
+ * Scraper Service
+ *
+ * Core scraper logic including HTTP fetching, HTML parsing,
+ * data transformation, and database operations.
+ */
+
+const axios = require('axios');
+const config = require('../config/scraper');
+
+/**
+ * Sleep utility for retry delays
+ * @param {number} ms - Milliseconds to sleep
+ * @returns {Promise<void>}
+ */
+function sleep(ms) {
+  return new Promise(resolve => setTimeout(resolve, ms));
+}
+
+/**
+ * Determine if an error is retryable
+ * @param {Error} error - Axios error
+ * @returns {boolean} True if should retry
+ */
+function isRetryableError(error) {
+  // Network errors (timeout, DNS, connection) don't have a response property
+  if (!error.response) {
+    return true;
+  }
+
+  const status = error.response.status;
+
+  // 5xx server errors are retryable
+  if (status >= 500) {
+    return true;
+  }
+
+  // 429 Too Many Requests - do not retry immediately
+  if (status === 429) {
+    return false;
+  }
+
+  // 4xx client errors - do not retry
+  return false;
+}
+
+/**
+ * Fetch page HTML with retry logic
+ * @param {string} url - Target URL
+ * @param {Object} logger - Logger instance
+ * @returns {Promise<string>} HTML content
+ * @throws {Error} After all retries exhausted
+ */
+async function fetchPage(url, logger) {
+  const { maxRetries, baseDelay, timeout } = config.RETRY_CONFIG;
+  let lastError;
+
+  for (let attempt = 1; attempt <= maxRetries + 1; attempt++) {
+    try {
+      logger.info('Fetching page', { url, attempt });
+
+      const response = await axios.get(url, {
+        timeout,
+        headers: {
+          'User-Agent': config.USER_AGENT
+        },
+        maxRedirects: 5,
+        validateStatus: (status) => status < 400 // Accept 2xx and 3xx
+      });
+
+      logger.info('Page fetched successfully', {
+        status: response.status,
+        contentLength: response.data.length
+      });
+
+      return response.data;
+
+    } catch (error) {
+      lastError = error;
+
+      // Determine if error is retryable
+      const isRetryable = isRetryableError(error);
+
+      logger.warn('Fetch attempt failed', {
+        attempt,
+        errorType: error.name,
+        errorMessage: error.message,
+        statusCode: error.response?.status,
+        isRetryable
+      });
+
+      // Don't retry non-retryable errors (4xx)
+      if (!isRetryable) {
+        throw error;
+      }
+
+      // Don't wait after last attempt
+      if (attempt <= maxRetries) {
+        const delay = baseDelay * Math.pow(2, attempt - 1); // Exponential backoff
+        logger.info('Waiting before retry', { delay });
+        await sleep(delay);
+      }
+    }
+  }
+
+  throw lastError;
+}
+
+module.exports = {
+  fetchPage,
+  // Export for testing
+  isRetryableError,
+  sleep
+};