Spaces:
Sleeping
Sleeping
Upload website_capturer.py
Browse files- utils/website_capturer.py +138 -0
utils/website_capturer.py
ADDED
|
@@ -0,0 +1,138 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
"""
|
| 2 |
+
Enhanced Website Capturer - Full Page Screenshot
|
| 3 |
+
Captures complete page height while maintaining fixed width
|
| 4 |
+
"""
|
| 5 |
+
|
| 6 |
+
import asyncio
|
| 7 |
+
from pathlib import Path
|
| 8 |
+
from typing import Dict, Tuple
|
| 9 |
+
import logging
|
| 10 |
+
|
| 11 |
+
logger = logging.getLogger(__name__)
|
| 12 |
+
|
| 13 |
+
|
| 14 |
+
async def capture_website_fullpage(
|
| 15 |
+
website_url: str,
|
| 16 |
+
output_dir: str = "./reports",
|
| 17 |
+
desktop_width: int = 1440,
|
| 18 |
+
mobile_width: int = 375
|
| 19 |
+
) -> Dict[str, str]:
|
| 20 |
+
import os
|
| 21 |
+
# Ensure local browser path is set
|
| 22 |
+
if "PLAYWRIGHT_BROWSERS_PATH" not in os.environ:
|
| 23 |
+
os.environ["PLAYWRIGHT_BROWSERS_PATH"] = os.path.join(os.getcwd(), ".playwright-browsers")
|
| 24 |
+
"""
|
| 25 |
+
Capture full-page screenshots of website at multiple viewports.
|
| 26 |
+
|
| 27 |
+
Args:
|
| 28 |
+
website_url: URL of the website to capture
|
| 29 |
+
output_dir: Directory to save screenshots
|
| 30 |
+
desktop_width: Desktop viewport width (height auto-calculated)
|
| 31 |
+
mobile_width: Mobile viewport width (height auto-calculated)
|
| 32 |
+
|
| 33 |
+
Returns:
|
| 34 |
+
Dictionary with paths to captured screenshots
|
| 35 |
+
"""
|
| 36 |
+
try:
|
| 37 |
+
from playwright.async_api import async_playwright
|
| 38 |
+
except ImportError:
|
| 39 |
+
import sys
|
| 40 |
+
import os
|
| 41 |
+
# Common paths for HF Spaces
|
| 42 |
+
sys.path.append("/usr/local/lib/python3.10/site-packages")
|
| 43 |
+
sys.path.append("/home/user/.local/lib/python3.10/site-packages")
|
| 44 |
+
from playwright.async_api import async_playwright
|
| 45 |
+
|
| 46 |
+
Path(output_dir).mkdir(parents=True, exist_ok=True)
|
| 47 |
+
screenshots = {}
|
| 48 |
+
|
| 49 |
+
async with async_playwright() as p:
|
| 50 |
+
browser = await p.chromium.launch(headless=True)
|
| 51 |
+
|
| 52 |
+
try:
|
| 53 |
+
# Desktop capture - Full page height
|
| 54 |
+
print(" 📱 Capturing desktop (1440px width, full height)...")
|
| 55 |
+
page = await browser.new_page(viewport={"width": desktop_width, "height": 1080})
|
| 56 |
+
await page.goto(website_url, wait_until="networkidle")
|
| 57 |
+
|
| 58 |
+
# Get full page height
|
| 59 |
+
desktop_height = await page.evaluate("() => document.documentElement.scrollHeight")
|
| 60 |
+
print(f" ℹ️ Desktop full height: {desktop_height}px")
|
| 61 |
+
|
| 62 |
+
# Set viewport to full height and capture
|
| 63 |
+
await page.set_viewport_size({"width": desktop_width, "height": desktop_height})
|
| 64 |
+
desktop_path = f"{output_dir}/desktop_{desktop_width}x{desktop_height}.png"
|
| 65 |
+
await page.screenshot(path=desktop_path, full_page=True)
|
| 66 |
+
screenshots["desktop"] = desktop_path
|
| 67 |
+
print(f" ✓ Saved: {desktop_path}")
|
| 68 |
+
|
| 69 |
+
await page.close()
|
| 70 |
+
|
| 71 |
+
# Mobile capture - Full page height
|
| 72 |
+
print(" 📱 Capturing mobile (375px width, full height)...")
|
| 73 |
+
page = await browser.new_page(viewport={"width": mobile_width, "height": 812})
|
| 74 |
+
await page.goto(website_url, wait_until="networkidle")
|
| 75 |
+
|
| 76 |
+
# Get full page height
|
| 77 |
+
mobile_height = await page.evaluate("() => document.documentElement.scrollHeight")
|
| 78 |
+
print(f" ℹ️ Mobile full height: {mobile_height}px")
|
| 79 |
+
|
| 80 |
+
# Set viewport to full height and capture
|
| 81 |
+
await page.set_viewport_size({"width": mobile_width, "height": mobile_height})
|
| 82 |
+
mobile_path = f"{output_dir}/mobile_{mobile_width}x{mobile_height}.png"
|
| 83 |
+
await page.screenshot(path=mobile_path, full_page=True)
|
| 84 |
+
screenshots["mobile"] = mobile_path
|
| 85 |
+
print(f" ✓ Saved: {mobile_path}")
|
| 86 |
+
|
| 87 |
+
await page.close()
|
| 88 |
+
|
| 89 |
+
finally:
|
| 90 |
+
await browser.close()
|
| 91 |
+
|
| 92 |
+
return screenshots
|
| 93 |
+
|
| 94 |
+
|
| 95 |
+
def capture_website_sync(
|
| 96 |
+
website_url: str,
|
| 97 |
+
output_dir: str = "./reports",
|
| 98 |
+
desktop_width: int = 1440,
|
| 99 |
+
mobile_width: int = 375
|
| 100 |
+
) -> Dict[str, str]:
|
| 101 |
+
"""
|
| 102 |
+
Synchronous wrapper for capturing full-page website screenshots.
|
| 103 |
+
|
| 104 |
+
Args:
|
| 105 |
+
website_url: URL of the website to capture
|
| 106 |
+
output_dir: Directory to save screenshots
|
| 107 |
+
desktop_width: Desktop viewport width
|
| 108 |
+
mobile_width: Mobile viewport width
|
| 109 |
+
|
| 110 |
+
Returns:
|
| 111 |
+
Dictionary with paths to captured screenshots
|
| 112 |
+
"""
|
| 113 |
+
return asyncio.run(
|
| 114 |
+
capture_website_fullpage(
|
| 115 |
+
website_url,
|
| 116 |
+
output_dir,
|
| 117 |
+
desktop_width,
|
| 118 |
+
mobile_width
|
| 119 |
+
)
|
| 120 |
+
)
|
| 121 |
+
|
| 122 |
+
|
| 123 |
+
if __name__ == "__main__":
|
| 124 |
+
# Test the function
|
| 125 |
+
import sys
|
| 126 |
+
|
| 127 |
+
if len(sys.argv) < 2:
|
| 128 |
+
print("Usage: python website_capturer_fullpage.py <url> [output_dir]")
|
| 129 |
+
sys.exit(1)
|
| 130 |
+
|
| 131 |
+
url = sys.argv[1]
|
| 132 |
+
output = sys.argv[2] if len(sys.argv) > 2 else "./reports"
|
| 133 |
+
|
| 134 |
+
print(f"Capturing website: {url}")
|
| 135 |
+
result = capture_website_sync(url, output)
|
| 136 |
+
print(f"\nScreenshots saved:")
|
| 137 |
+
for key, path in result.items():
|
| 138 |
+
print(f" {key}: {path}")
|