a03e68fa2f
Implement dynamic configuration selection based on URL patterns to optimize crawling for different content types. This feature enables users to apply different crawling strategies (PDF extraction, content filtering, JavaScript execution) based on URL matching patterns. Key additions: - Add url_matcher and match_mode parameters to CrawlerRunConfig - Implement is_match() method supporting string patterns, functions, and mixed lists - Add MatchMode enum for OR/AND logic when combining multiple matchers - Update AsyncWebCrawler.arun_many() to accept List[CrawlerRunConfig] - Add select_config() method to dispatchers for runtime config selection - First matching config wins, with fallback to default Pattern matching supports: - Glob-style strings: *.pdf, */blog/*, *api* - Lambda functions: lambda url: 'github.com' in url - Mixed patterns with AND/OR logic for complex matching This enables optimal per-URL configuration: - PDFs: Use PDFContentScrapingStrategy without JavaScript - Blogs: Apply content filtering to reduce noise - APIs: Skip JavaScript, use JSON extraction - Dynamic sites: Execute only necessary JavaScript Breaking changes: None - fully backward compatible
43 lines
1.3 KiB
Python
43 lines
1.3 KiB
Python
"""
|
|
Test example for multiple crawler configs feature
|
|
"""
|
|
import asyncio
|
|
import sys
|
|
from pathlib import Path
|
|
|
|
# Add parent directory to path for imports
|
|
sys.path.insert(0, str(Path(__file__).parent.parent))
|
|
|
|
from crawl4ai import AsyncWebCrawler, CrawlerRunConfig, CacheMode
|
|
from crawl4ai.processors.pdf import PDFContentScrapingStrategy
|
|
|
|
|
|
async def test_run_many():
|
|
default_config = CrawlerRunConfig(
|
|
cache_mode=CacheMode.BYPASS,
|
|
# scraping_strategy=PDFContentScrapingStrategy()
|
|
)
|
|
|
|
test_urls = [
|
|
# "https://blog.python.org/", # Blog URL
|
|
"https://www.python.org/", # Generic HTTPS page
|
|
"https://www.kidocode.com/", # Generic HTTPS page
|
|
"https://www.example.com/", # Generic HTTPS page
|
|
# "https://www.w3.org/WAI/ER/tests/xhtml/testfiles/resources/pdf/dummy.pdf",
|
|
]
|
|
|
|
async with AsyncWebCrawler() as crawler:
|
|
# Single config - traditional usage still works
|
|
print("Test 1: Single config (backwards compatible)")
|
|
result = await crawler.arun_many(
|
|
urls=test_urls[:2],
|
|
config=default_config
|
|
)
|
|
print(f"Crawled {len(result)} URLs with single config\n")
|
|
for item in result:
|
|
print(f" {item.url} -> {item.status_code}")
|
|
|
|
|
|
if __name__ == "__main__":
|
|
asyncio.run(test_run_many())
|