AndroidFileHost Scraper

Simple and powerful Python script to look for keywords or download direct files from AndroidFileHost by imitating their API calls. Supports downloading from multiple pages, direct FID / URL downloads, automatic mirror failover, dry-run inspection, JSON output, and sorting.

Why?

AFH is slowly dying, which means a lot of ROMs, kernels, and recoveries will be forever gone if we don't archive them elsewhere. This script makes bulk archiving and single-file retrieval fast and easy.

Features

• Keyword Search & Archiving: Search and download files across multiple pages with popularity or newest sorting.
• Direct Downloads: Download specific files directly by AFH URL or numeric FID (-u / --url, --fid).
• Mirror Failover: Automatic fallback to alternate mirrors if the preferred mirror (USA/Germany) fails or is unreachable.
• Dry Run & Inspection: Preview search matches or target file metadata without downloading (--list / --dry-run).
• Structured JSON Output: Machine-readable JSON output mode (--json) for scripting and pipelines.
• Async HTTP/2 Engine: High-throughput concurrent downloads with real-time transfer speed and progress tracking.
• Integrity Validation: Automatic SHA-256 and MD5 hash verification with .part atomic file staging.
• Safe Path Resolution: Default archive directory (afh_archive) is placed in the current working directory, preventing site-packages pollution when installed via pip.

Installation

Install dependencies:

pip install httpx h2 beautifulsoup4

You can also check the releases page to download precompiled standalone executables for Windows, Linux, and macOS without installing Python.

Alternatively, install globally via PyPI:

pip install AndroidFileHostScraper

You can then run it from anywhere using either androidfilehostscraper or afhscraper.

Usage

Interactive Mode

Simply run the tool without arguments:

python AFHscraper.py

You will be prompted to choose between searching for keywords or entering direct file URLs/FIDs, followed by mirror and concurrency preferences.

CLI Mode

AFHScraper provides a command-line interface for automation and scripting:

options:
  -h, --help            show this help message and exit
  -s, --search SEARCH   Search terms (comma-separated)
  -u, --url URL         Direct download by AFH URL or FID (comma-separated for multiple)
  --fid FID             Direct download by FID (comma-separated for multiple)
  --sort {newest,popular}
                        Sort order (newest or popular)
  -n, --files FILES     Maximum files to download per search term
  -m, --mirror {usa,germany}
                        Preferred mirror location
  -t, --threads THREADS
                        Number of concurrent downloads (default: 1)
  -o, --output OUTPUT   Download directory
  --list, --dry-run     Search and display matching files without downloading
  --json                Output results in JSON format
  -l, --log-level {DEBUG,INFO,WARNING,ERROR}
                        Logging level (default: INFO)

CLI Examples

• Search and download files:

  python AFHscraper.py -s "twrp,lineage" --sort popular -n 20 -t 5 -m usa

• Direct download by URL or FID:

  python AFHscraper.py -u "https://androidfilehost.com/?fid=1234567890"
  python AFHscraper.py --fid 1234567890,9876543210 -t 2 -o ./downloads

• Dry-run / list files without downloading:

  python AFHscraper.py -s "lineage" --list
  python AFHscraper.py --fid 1234567890 --list

• Output in JSON format:

  python AFHscraper.py -s "lineage" --list --json
  python AFHscraper.py --fid 1234567890 --json

Download Path Resolution

By default, files are downloaded into an afh_archive directory inside your current working directory (or adjacent to the executable if running a frozen binary). This ensures that running afhscraper when installed in global Python environments never writes downloaded files into site-packages. Use -o / --output to set a custom destination directory.

Mirror Failover

AFHScraper first attempts to download from your selected mirror region (usa or germany). If that mirror fails, responds with an HTTP error, or returns an invalid stream, AFHScraper automatically falls back to remaining available mirrors until the download succeeds or all options are exhausted.

Running Tests

Unit tests are written using Python's built-in unittest framework and require no external test runner:

python -m unittest discover -s tests
# or simply
python -m unittest

Extras

Enjoy my work? Please consider a small donation!