Spaces:
Sleeping
Sleeping
Update main.py
Browse files
main.py
CHANGED
@@ -8,11 +8,13 @@ from fastapi.templating import Jinja2Templates
|
|
8 |
from fastapi.staticfiles import StaticFiles
|
9 |
from typing import List, Dict
|
10 |
import asyncio
|
|
|
11 |
|
12 |
# Set up logging
|
13 |
logging.basicConfig(level=logging.INFO, format='%(asctime)s - %(levelname)s - %(message)s')
|
|
|
14 |
|
15 |
-
app = FastAPI(title="Website Scraper API
|
16 |
|
17 |
# Mount static files
|
18 |
app.mount("/static", StaticFiles(directory="static"), name="static")
|
@@ -26,21 +28,34 @@ MAX_PAGES = 20
|
|
26 |
async def scrape_page(url: str, visited: set, base_domain: str) -> tuple[Dict, set]:
|
27 |
"""Scrape a single page for text, images, and links using Playwright."""
|
28 |
try:
|
|
|
29 |
async with async_playwright() as p:
|
30 |
browser = await p.chromium.launch(headless=True)
|
31 |
context = await browser.new_context(
|
32 |
user_agent="Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/118.0.0.0 Safari/537.36",
|
33 |
-
viewport={"width":
|
|
|
34 |
)
|
35 |
page = await context.new_page()
|
36 |
-
|
37 |
-
#
|
38 |
-
|
39 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
40 |
# Scroll to trigger lazy-loaded images
|
41 |
await page.evaluate("window.scrollTo(0, document.body.scrollHeight)")
|
42 |
await page.wait_for_timeout(2000) # Wait for lazy-loaded content
|
43 |
-
|
44 |
# Extract text content
|
45 |
text_content = await page.evaluate(
|
46 |
"""() => document.body.innerText"""
|
@@ -74,6 +89,7 @@ async def scrape_page(url: str, visited: set, base_domain: str) -> tuple[Dict, s
|
|
74 |
links = set(urljoin(url, link) for link in links if urlparse(urljoin(url, link)).netloc == base_domain and urljoin(url, link) not in visited)
|
75 |
|
76 |
await browser.close()
|
|
|
77 |
|
78 |
page_data = {
|
79 |
"url": url,
|
@@ -83,13 +99,14 @@ async def scrape_page(url: str, visited: set, base_domain: str) -> tuple[Dict, s
|
|
83 |
return page_data, links
|
84 |
|
85 |
except Exception as e:
|
86 |
-
|
87 |
return {}, set()
|
88 |
|
89 |
@app.get("/scrape")
|
90 |
async def crawl_website(url: HttpUrl):
|
91 |
"""Crawl the website starting from the given URL and return scraped data for up to 10 pages as JSON."""
|
92 |
try:
|
|
|
93 |
visited = set()
|
94 |
to_visit = {str(url)}
|
95 |
base_domain = urlparse(str(url)).netloc
|
@@ -100,27 +117,31 @@ async def crawl_website(url: HttpUrl):
|
|
100 |
if current_url in visited:
|
101 |
continue
|
102 |
|
103 |
-
|
104 |
visited.add(current_url)
|
105 |
|
106 |
page_data, new_links = await scrape_page(current_url, visited, base_domain)
|
107 |
if page_data:
|
108 |
results.append(page_data)
|
109 |
to_visit.update(new_links)
|
110 |
-
|
111 |
# Small delay to avoid overwhelming the server
|
112 |
await asyncio.sleep(0.5)
|
113 |
|
|
|
114 |
return JSONResponse(content={"pages": results})
|
115 |
|
116 |
except Exception as e:
|
|
|
117 |
raise HTTPException(status_code=500, detail=f"Scraping failed: {str(e)}")
|
118 |
|
119 |
@app.get("/")
|
120 |
async def serve_home(request: Request):
|
121 |
"""Serve the frontend HTML page."""
|
|
|
122 |
return templates.TemplateResponse("index.html", {"request": request})
|
123 |
|
124 |
if __name__ == "__main__":
|
|
|
125 |
import uvicorn
|
126 |
uvicorn.run(app, host="0.0.0.0", port=7860)
|
|
|
8 |
from fastapi.staticfiles import StaticFiles
|
9 |
from typing import List, Dict
|
10 |
import asyncio
|
11 |
+
import os
|
12 |
|
13 |
# Set up logging
|
14 |
logging.basicConfig(level=logging.INFO, format='%(asctime)s - %(levelname)s - %(message)s')
|
15 |
+
logger = logging.getLogger(__name__)
|
16 |
|
17 |
+
app = FastAPI(title="Website Scraper API with Frontend")
|
18 |
|
19 |
# Mount static files
|
20 |
app.mount("/static", StaticFiles(directory="static"), name="static")
|
|
|
28 |
async def scrape_page(url: str, visited: set, base_domain: str) -> tuple[Dict, set]:
|
29 |
"""Scrape a single page for text, images, and links using Playwright."""
|
30 |
try:
|
31 |
+
logger.info(f"Starting Playwright for URL: {url}")
|
32 |
async with async_playwright() as p:
|
33 |
browser = await p.chromium.launch(headless=True)
|
34 |
context = await browser.new_context(
|
35 |
user_agent="Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/118.0.0.0 Safari/537.36",
|
36 |
+
viewport={"width": 800, "height": 600}, # Reduced viewport for performance
|
37 |
+
bypass_csp=True # Bypass Content Security Policy
|
38 |
)
|
39 |
page = await context.new_page()
|
40 |
+
|
41 |
+
# Retry navigation with fallback
|
42 |
+
for attempt in range(2): # Try up to 2 times
|
43 |
+
try:
|
44 |
+
logger.info(f"Navigating to {url} (Attempt {attempt + 1})")
|
45 |
+
await page.goto(url, wait_until="domcontentloaded", timeout=30000) # 30s timeout
|
46 |
+
break # Success, exit retry loop
|
47 |
+
except Exception as e:
|
48 |
+
logger.warning(f"Navigation attempt {attempt + 1} failed for {url}: {str(e)}")
|
49 |
+
if attempt == 1: # Last attempt
|
50 |
+
logger.error(f"All navigation attempts failed for {url}")
|
51 |
+
await browser.close()
|
52 |
+
return {}, set()
|
53 |
+
await asyncio.sleep(1) # Wait before retry
|
54 |
+
|
55 |
# Scroll to trigger lazy-loaded images
|
56 |
await page.evaluate("window.scrollTo(0, document.body.scrollHeight)")
|
57 |
await page.wait_for_timeout(2000) # Wait for lazy-loaded content
|
58 |
+
|
59 |
# Extract text content
|
60 |
text_content = await page.evaluate(
|
61 |
"""() => document.body.innerText"""
|
|
|
89 |
links = set(urljoin(url, link) for link in links if urlparse(urljoin(url, link)).netloc == base_domain and urljoin(url, link) not in visited)
|
90 |
|
91 |
await browser.close()
|
92 |
+
logger.info(f"Successfully scraped {url}")
|
93 |
|
94 |
page_data = {
|
95 |
"url": url,
|
|
|
99 |
return page_data, links
|
100 |
|
101 |
except Exception as e:
|
102 |
+
logger.error(f"Error scraping {url}: {str(e)}")
|
103 |
return {}, set()
|
104 |
|
105 |
@app.get("/scrape")
|
106 |
async def crawl_website(url: HttpUrl):
|
107 |
"""Crawl the website starting from the given URL and return scraped data for up to 10 pages as JSON."""
|
108 |
try:
|
109 |
+
logger.info(f"Starting crawl for {url}")
|
110 |
visited = set()
|
111 |
to_visit = {str(url)}
|
112 |
base_domain = urlparse(str(url)).netloc
|
|
|
117 |
if current_url in visited:
|
118 |
continue
|
119 |
|
120 |
+
logger.info(f"Scraping: {current_url}")
|
121 |
visited.add(current_url)
|
122 |
|
123 |
page_data, new_links = await scrape_page(current_url, visited, base_domain)
|
124 |
if page_data:
|
125 |
results.append(page_data)
|
126 |
to_visit.update(new_links)
|
127 |
+
|
128 |
# Small delay to avoid overwhelming the server
|
129 |
await asyncio.sleep(0.5)
|
130 |
|
131 |
+
logger.info(f"Crawl completed for {url}")
|
132 |
return JSONResponse(content={"pages": results})
|
133 |
|
134 |
except Exception as e:
|
135 |
+
logger.error(f"Scraping failed for {url}: {str(e)}")
|
136 |
raise HTTPException(status_code=500, detail=f"Scraping failed: {str(e)}")
|
137 |
|
138 |
@app.get("/")
|
139 |
async def serve_home(request: Request):
|
140 |
"""Serve the frontend HTML page."""
|
141 |
+
logger.info("Serving home page")
|
142 |
return templates.TemplateResponse("index.html", {"request": request})
|
143 |
|
144 |
if __name__ == "__main__":
|
145 |
+
logger.info("Starting FastAPI server on port 7860")
|
146 |
import uvicorn
|
147 |
uvicorn.run(app, host="0.0.0.0", port=7860)
|