WACZ image finder

README.md

WACZ Webcam Image Extractor

A Python script to extract webcam images from WACZ (Web Archive Collection Zipped) files.

Purpose

Extracts Belarus customs webcam images (OSH1.jpg, OSH2.jpg, OSH3.jpg, OSH4.jpg) from archived web data collected between August 5-27, 2025.

Requirements

pip install warcio

Usage

python extract_webcam.py

What it does

• Searches all .wacz files in the data/ directory
• Extracts images matching pattern /webcam/OSH*.jpg
• Saves images to extracted_webcam/ directory
• Names files as {timestamp}_{image_name}.jpg

Output

The script processes WACZ files and reports progress:

Found 377 WACZ files
✓ filename1: extracted 4 images
✓ filename2: extracted 0 images
...
Extracted 1112 total images to extracted_webcam

Results

From 377 WACZ files (40GB total):

• 1,112 webcam images extracted
• 278 timestamps with webcam data
• Date range: August 5-27, 2025
• Source: customs.gov.by/webcam/ monitoring

Notes

• Only processes files containing Belarus customs webcam URLs
• Validates JPEG format with magic byte checking
• Handles compressed WARC files within WACZ archives
• Skips files without relevant webcam content

extract_webcam.py

#!/usr/bin/env python3

import zipfile
import os
from pathlib import Path
from warcio.archiveiterator import ArchiveIterator
from urllib.parse import urlparse

def extract_images_from_warc(warc_stream, output_dir, wacz_name):
    """Extract webcam/OSH1.jpg images from a WARC stream."""
    images = []
    
    for record in ArchiveIterator(warc_stream):
        if record.rec_type == 'response':
            url = record.rec_headers.get_header('WARC-Target-URI', '')
            if '/webcam/OSH' in url and url.endswith('.jpg'):
                content = record.content_stream().read()
                if content and content.startswith(b'\xff\xd8'):  # JPEG magic bytes
                    image_name = Path(urlparse(url).path).name
                    filename = f"{wacz_name}_{image_name}"
                    output_path = Path(output_dir) / filename
                    output_path.write_bytes(content)
                    images.append(str(output_path))
    
    return images

def process_wacz_file(wacz_path, output_dir):
    """Process a single WACZ file and extract webcam images."""
    wacz_name = Path(wacz_path).stem
    images = []
    
    try:
        with zipfile.ZipFile(wacz_path, 'r') as zf:
            warc_files = [name for name in zf.namelist() if name.startswith('archive/') and '.warc' in name]
            
            for warc_name in warc_files:
                with zf.open(warc_name) as warc_stream:
                    images.extend(extract_images_from_warc(warc_stream, output_dir, wacz_name))
        
        print(f"✓ {wacz_name}: extracted {len(images)} images")
        return images
        
    except Exception as e:
        print(f"✗ {wacz_name}: {e}")
        return []

def main():
    data_dir = Path("data")
    output_dir = Path("extracted_webcam")
    output_dir.mkdir(exist_ok=True)
    
    wacz_files = list(data_dir.glob("*.wacz"))
    print(f"Found {len(wacz_files)} WACZ files")
    
    all_images = []
    for wacz_file in wacz_files:
        images = process_wacz_file(wacz_file, output_dir)
        all_images.extend(images)
    
    print(f"\nExtracted {len(all_images)} total images to {output_dir}")

if __name__ == "__main__":
    main()