中文文档

HomDGCat Wiki Mirror

A complete offline mirror of homdgcat.wiki, covering characters, weapons, artifacts/relics and more from Genshin Impact and Honkai: Star Rail.

Contents

• 15,357 files, ~2.5 GB
• 575+ pages (Genshin + Star Rail characters/weapons/artifacts, etc.)
• 4,580+ dynamic subpages (quests, items, monsters, GCG cards, etc.)
• Data files in 5 languages (CH / EN / JP / KR / RU)
• All image assets (avatars, skill icons, weapon art, gacha art, etc.)
• TextMap localization data

Quick Start

Get Site Data

The site data (site/ directory, ~2.5 GB) is not included in this repository. Three ways to obtain it:

Option 1: Download from GitHub Releases (Recommended)

Download site.7z from the latest release and extract it to the project root:

# Download (~2 GB)
curl -LO https://github.com/NewbieXvwu/HomDGCat/releases/download/initial/site.7z

# Extract (requires 7-Zip)
7z x site.7z

Option 2: Download from Internet Archive

If GitHub is unavailable, download from Internet Archive:

ia download homdgcat-wiki-mirror site.7z
7z x site.7z

Option 3: Pull from the source site

Download all files directly from homdgcat.wiki (may fail if the source site is unstable):

# Default (10 threads)
python main.py download

# 16 threads
python main.py download --workers 16

# Check progress
python main.py status

The script reads file paths from filelist.txt and downloads all missing files into site/. Existing files are skipped automatically; supports resuming interrupted downloads.

Browse Locally

# Default port 9000
python main.py serve

# Custom port
python main.py serve --port 3000

Open http://localhost:9000 in your browser to browse the full site.

Server Engines

The script supports two server engines and auto-detects availability:

stdlib mode (zero dependencies)

Works out of the box with no third-party packages. Based on Python's http.server:

• HTTP/1.1 + keep-alive
• Multi-threaded request handling
• gzip compression (with LRU cache)
• ETag / 304 conditional requests
• Tiered Cache-Control headers

Suitable for personal local browsing.

ASGI mode (recommended)

Auto-enabled when dependencies are installed:

pip install starlette hypercorn h2

Based on hypercorn + starlette:

• Async I/O, far better concurrency than threading
• HTTP/2 support (requires TLS certificate)
• Production-grade connection management and timeouts
• gzip compression + ETag / 304
• Path traversal protection

Enable HTTPS + HTTP/2:

python main.py serve --cert cert.pem --key key.pem

File Structure

main.py               # Unified tool script (download / serve / status)
filelist.txt         # Complete file list (15,357 entries)
site/                # Site files directory
  index/             # Homepage
  sr/char/           # Star Rail character pages
  gi/char/           # Genshin character pages
  data/{CH,EN,...}/  # Star Rail multi-language data
  gi/{CH,EN,...}/    # Genshin multi-language data
  images/            # Star Rail image assets
  homdgcat-res/      # Genshin image assets
  TextMap/           # TextMap localization data

Command Reference

Command	Description
download	Download all missing files
download -w 16	Use 16 concurrent threads
download -r 5	Retry failed downloads 5 times
serve	Start local server
serve -p 3000	Custom port
serve --cert X --key Y	Enable HTTPS/HTTP2
status	Show download progress
--lang en	Force English output
--lang zh	Force Chinese output

Requirements

• Python 3.8+
• Download: no third-party dependencies
• Server (stdlib mode): no third-party dependencies
• Server (ASGI mode): starlette, hypercorn, h2

License

This tool is intended for personal offline backup of homdgcat.wiki. All site content is copyrighted by its original authors.