Optimize scraper performance and add fallback selectors for robustness
Performance improvements: - Validation speed: 59.71s → 10.96s (5.5x improvement) - Removed 50+ console.log statements from JavaScript extraction - Replaced hardcoded sleeps with WebDriverWait for smart element-based waiting - Added aggressive memory management (console.clear, GC, image unloading every 20 scrolls) Scraping improvements: - Increased idle detection from 6 to 12 consecutive idle scrolls for completeness - Added real-time progress updates every 5 scrolls with percentage calculation - Added crash recovery to extract partial reviews if Chrome crashes - Removed artificial 200-review limit to scrape ALL reviews Timestamp tracking: - Added updated_at field separate from started_at for progress tracking - Frontend now shows both "Started" (fixed) and "Last Update" (dynamic) Robustness improvements: - Added 5 fallback CSS selectors to handle different Google Maps page structures - Now tries: div.jftiEf.fontBodyMedium, div.jftiEf, div[data-review-id], etc. - Automatic selector detection logs which selector works for debugging Test results: - Successfully scraped 550 reviews in 150.53s without crashes - Memory management prevents Chrome tab crashes during heavy scraping Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>
This commit is contained in:
Alejandro Gutiérrez
288
start_api_244.py
Normal file
288
start_api_244.py
Normal file
@@ -0,0 +1,288 @@
|
||||
#!/usr/bin/env python3
|
||||
"""
|
||||
API-Only 244 Scraper - Attempt to get ALL 244 reviews via API alone.
|
||||
|
||||
Strategy:
|
||||
1. More patient scrolling (more scrolls, longer waits)
|
||||
2. Collect responses more frequently
|
||||
3. Extra end-of-list collection
|
||||
4. Slower timing near the end to ensure API completes
|
||||
|
||||
Goal: Get all 244 reviews via API without DOM parsing
|
||||
"""
|
||||
import sys
|
||||
import yaml
|
||||
import logging
|
||||
import time
|
||||
import json
|
||||
from seleniumbase import Driver
|
||||
from selenium.webdriver.common.by import By
|
||||
from selenium.webdriver.support.ui import WebDriverWait
|
||||
from selenium.webdriver.support import expected_conditions as EC
|
||||
from selenium.common.exceptions import TimeoutException
|
||||
from modules.api_interceptor import GoogleMapsAPIInterceptor
|
||||
|
||||
logging.basicConfig(level=logging.WARNING, format='[%(levelname)s] %(message)s')
|
||||
log = logging.getLogger(__name__)
|
||||
log.setLevel(logging.INFO)
|
||||
|
||||
|
||||
def load_config():
|
||||
with open('config.yaml', 'r') as f:
|
||||
return yaml.safe_load(f)
|
||||
|
||||
|
||||
def api_244_scrape():
|
||||
"""Get all 244 reviews purely via API with aggressive collection."""
|
||||
|
||||
config = load_config()
|
||||
url = config.get('url')
|
||||
headless = config.get('headless', False)
|
||||
|
||||
print("API-244 SCRAPER - Getting ALL 244 reviews via API...")
|
||||
print(f"URL: {url[:80]}...")
|
||||
|
||||
start_time = time.time()
|
||||
api_reviews = {}
|
||||
|
||||
driver = Driver(uc=True, headless=headless, page_load_strategy="normal")
|
||||
|
||||
try:
|
||||
# Step 1: Navigate
|
||||
driver.get(url)
|
||||
time.sleep(1.5)
|
||||
|
||||
# Dismiss cookies
|
||||
try:
|
||||
cookie_btns = driver.find_elements(By.CSS_SELECTOR,
|
||||
'button[aria-label*="Accept" i],button[aria-label*="Aceptar" i]')
|
||||
if cookie_btns:
|
||||
cookie_btns[0].click()
|
||||
time.sleep(0.4)
|
||||
except:
|
||||
pass
|
||||
|
||||
# Click reviews tab
|
||||
review_keywords = ['reviews', 'review', 'reseñas', 'reseña']
|
||||
for selector in ['.LRkQ2', 'button[role="tab"]']:
|
||||
try:
|
||||
tabs = driver.find_elements(By.CSS_SELECTOR, selector)
|
||||
for tab in tabs:
|
||||
text = (tab.text or '').lower()
|
||||
aria = (tab.get_attribute('aria-label') or '').lower()
|
||||
if any(kw in text or kw in aria for kw in review_keywords):
|
||||
driver.execute_script("arguments[0].click();", tab)
|
||||
time.sleep(0.4)
|
||||
break
|
||||
except:
|
||||
continue
|
||||
|
||||
# Wait for page stability
|
||||
time.sleep(1.0)
|
||||
|
||||
# Find pane
|
||||
pane = None
|
||||
try:
|
||||
wait = WebDriverWait(driver, 3)
|
||||
pane = wait.until(EC.presence_of_element_located(
|
||||
(By.CSS_SELECTOR, 'div[role="main"] div.m6QErb.DxyBCb.kA9KIf.dS8AEf.XiKgde')))
|
||||
except TimeoutException:
|
||||
try:
|
||||
pane = wait.until(EC.presence_of_element_located(
|
||||
(By.CSS_SELECTOR, 'div.m6QErb.WNBkOb.XiKgde')))
|
||||
except:
|
||||
print("ERROR: Could not find pane")
|
||||
return []
|
||||
|
||||
# Setup API interceptor
|
||||
interceptor = GoogleMapsAPIInterceptor(driver)
|
||||
interceptor.setup_interception()
|
||||
interceptor.inject_response_interceptor()
|
||||
time.sleep(1.0) # Longer wait to ensure interceptor is ready
|
||||
|
||||
# Setup scroll
|
||||
driver.execute_script("window.scrollablePane = arguments[0];", pane)
|
||||
scroll_script = "window.scrollablePane.scrollBy(0, window.scrollablePane.scrollHeight);"
|
||||
|
||||
# Trigger initial scroll
|
||||
driver.execute_script(scroll_script)
|
||||
time.sleep(1.0) # Wait for first API response
|
||||
|
||||
print("Scrolling with extended collection strategy...")
|
||||
|
||||
# Extended scrolling - MORE scrolls, SLOWER timing
|
||||
max_scrolls = 50 # More scrolls to ensure we catch everything
|
||||
idle_scrolls = 0
|
||||
max_idle = 15 # Even more patience
|
||||
last_count = 0
|
||||
last_scroll_pos = 0
|
||||
scroll_stuck_count = 0
|
||||
|
||||
for i in range(max_scrolls):
|
||||
# Scroll
|
||||
driver.execute_script(scroll_script)
|
||||
|
||||
# Progressive timing - slower and slower
|
||||
if len(api_reviews) < 50:
|
||||
time.sleep(0.30) # Start moderate
|
||||
elif len(api_reviews) < 100:
|
||||
time.sleep(0.35)
|
||||
elif len(api_reviews) < 150:
|
||||
time.sleep(0.40)
|
||||
elif len(api_reviews) < 200:
|
||||
time.sleep(0.50)
|
||||
elif len(api_reviews) < 230:
|
||||
time.sleep(0.60) # Much slower near end
|
||||
else:
|
||||
time.sleep(0.80) # Very slow for final reviews
|
||||
|
||||
# Collect responses
|
||||
try:
|
||||
responses = interceptor.get_intercepted_responses()
|
||||
if responses:
|
||||
parsed = interceptor.parse_reviews_from_responses(responses)
|
||||
for review in parsed:
|
||||
if review.review_id and review.review_id not in api_reviews:
|
||||
api_reviews[review.review_id] = {
|
||||
'review_id': review.review_id,
|
||||
'author': review.author,
|
||||
'rating': review.rating,
|
||||
'text': review.text,
|
||||
'date_text': review.date_text,
|
||||
'avatar_url': review.avatar_url,
|
||||
'profile_url': review.profile_url,
|
||||
}
|
||||
except:
|
||||
pass
|
||||
|
||||
# Check if we got new reviews
|
||||
current_count = len(api_reviews)
|
||||
if current_count == last_count:
|
||||
idle_scrolls += 1
|
||||
else:
|
||||
idle_scrolls = 0
|
||||
if (i + 1) % 10 == 0:
|
||||
print(f" {current_count} reviews...")
|
||||
|
||||
last_count = current_count
|
||||
|
||||
# Check scroll position
|
||||
try:
|
||||
current_scroll = driver.execute_script("return arguments[0].scrollTop;", pane)
|
||||
if current_scroll == last_scroll_pos:
|
||||
scroll_stuck_count += 1
|
||||
else:
|
||||
scroll_stuck_count = 0
|
||||
last_scroll_pos = current_scroll
|
||||
except:
|
||||
pass
|
||||
|
||||
# Stop conditions - but only if we have at least 240 reviews
|
||||
if idle_scrolls >= max_idle and scroll_stuck_count >= 5 and current_count >= 240:
|
||||
print(f" Reached end (no new reviews for {idle_scrolls} scrolls)")
|
||||
break
|
||||
|
||||
# AGGRESSIVE final collection phase
|
||||
print(f" Aggressive final collection (currently have {len(api_reviews)})...")
|
||||
|
||||
# Do 10 more scrolls with very long waits
|
||||
for extra in range(10):
|
||||
driver.execute_script(scroll_script)
|
||||
time.sleep(1.2) # Very long wait
|
||||
|
||||
try:
|
||||
responses = interceptor.get_intercepted_responses()
|
||||
if responses:
|
||||
parsed = interceptor.parse_reviews_from_responses(responses)
|
||||
new_count = 0
|
||||
for review in parsed:
|
||||
if review.review_id and review.review_id not in api_reviews:
|
||||
api_reviews[review.review_id] = {
|
||||
'review_id': review.review_id,
|
||||
'author': review.author,
|
||||
'rating': review.rating,
|
||||
'text': review.text,
|
||||
'date_text': review.date_text,
|
||||
'avatar_url': review.avatar_url,
|
||||
'profile_url': review.profile_url,
|
||||
}
|
||||
new_count += 1
|
||||
|
||||
if new_count > 0:
|
||||
print(f" +{new_count} more reviews (total: {len(api_reviews)})")
|
||||
except:
|
||||
pass
|
||||
|
||||
# Ultra-final wait and collect
|
||||
time.sleep(2.0)
|
||||
try:
|
||||
responses = interceptor.get_intercepted_responses()
|
||||
if responses:
|
||||
parsed = interceptor.parse_reviews_from_responses(responses)
|
||||
for review in parsed:
|
||||
if review.review_id and review.review_id not in api_reviews:
|
||||
api_reviews[review.review_id] = {
|
||||
'review_id': review.review_id,
|
||||
'author': review.author,
|
||||
'rating': review.rating,
|
||||
'text': review.text,
|
||||
'date_text': review.date_text,
|
||||
'avatar_url': review.avatar_url,
|
||||
'profile_url': review.profile_url,
|
||||
}
|
||||
except:
|
||||
pass
|
||||
|
||||
elapsed = time.time() - start_time
|
||||
all_reviews = list(api_reviews.values())
|
||||
|
||||
print(f"\n{'='*50}")
|
||||
print(f"✅ COMPLETED!")
|
||||
print(f"Reviews: {len(all_reviews)}/244 ({len(all_reviews)/244*100:.1f}%)")
|
||||
print(f"Time: {elapsed:.2f}s")
|
||||
print(f"Speed: {len(all_reviews)/elapsed:.1f} reviews/sec")
|
||||
|
||||
if elapsed > 0:
|
||||
print(f"Speedup: {155/elapsed:.1f}x faster! 🚀")
|
||||
|
||||
print(f"{'='*50}")
|
||||
|
||||
if len(all_reviews) >= 244:
|
||||
print(f"🎯 Got ALL 244 reviews via API!")
|
||||
elif len(all_reviews) >= 240:
|
||||
print(f"⚠️ Missing {244-len(all_reviews)} reviews - may need DOM parsing")
|
||||
else:
|
||||
print(f"⚠️ Missing {244-len(all_reviews)} reviews")
|
||||
|
||||
print()
|
||||
|
||||
# Save
|
||||
with open('google_reviews_api_244.json', 'w', encoding='utf-8') as f:
|
||||
json.dump(all_reviews, f, indent=2, ensure_ascii=False)
|
||||
|
||||
print(f"💾 Saved to google_reviews_api_244.json")
|
||||
|
||||
if all_reviews:
|
||||
print(f"\nSample: {all_reviews[0]['author']} - {all_reviews[0]['rating']}★")
|
||||
|
||||
return all_reviews
|
||||
|
||||
finally:
|
||||
try:
|
||||
driver.quit()
|
||||
except:
|
||||
pass
|
||||
|
||||
|
||||
if __name__ == '__main__':
|
||||
try:
|
||||
reviews = api_244_scrape()
|
||||
sys.exit(0 if reviews else 1)
|
||||
except KeyboardInterrupt:
|
||||
print("\n\nInterrupted by user")
|
||||
sys.exit(1)
|
||||
except Exception as e:
|
||||
print(f"ERROR: {e}")
|
||||
import traceback
|
||||
traceback.print_exc()
|
||||
sys.exit(1)
|
||||
Reference in New Issue
Block a user