Phone Brand & Model Summary

Phone manufacturers often expose only a raw, cryptic model code (e.g. MI-ONE PLUS) at the OS/analytics level instead of the human-readable name (e.g. "小米 1 联通版"). KHwang9883's MobileModels crowdsources the mapping between these codes and their real names, one markdown file per brand.

This repo pulls that project in as a git submodule, parses every brand's markdown file, and flattens the result into a single CSV — so you can just load one file instead of parsing 20+ markdown files yourself.

Output

brand_model.csv — one row per device, with columns:

Column	Description
brand	Brand slug (e.g. xiaomi, huawei, apple)
model	Raw model code as reported by the device (e.g. MI-ONE PLUS)
area	cn or en, depending on which regional markdown file the entry came from
brand_name	Brand name in Chinese (e.g. 小米)
model_name	Human-readable model name (e.g. 小米 1 联通版)

Auto-update

A GitHub Actions workflow (update.yml) runs daily:

1. Pulls the MobileModels submodule to its latest commit
2. Regenerates brand_model.csv by running phoneModel.py
3. Commits and pushes if the data changed

No manual token setup is needed — it authenticates with the built-in GITHUB_TOKEN. If you fork this repo and the push step fails with a permissions error, check Settings → Actions → General → Workflow permissions and set it to "Read and write permissions".

Running locally

git clone --recurse-submodules https://github.com/jantacy/mobilemodels.git
cd mobilemodels
pip install -r requirements.txt
python phoneModel.py