ghlang 2.4.1

Generate language statistics and visualizations from GitHub repositories or local files

See what languages you've been coding in

Generate pretty charts from your GitHub repos or local files

Table of Contents

1. About The Project
   • Built With
2. Getting Started
   • Prerequisites
   • Installation
3. Usage
4. Shell Completion
5. Configuration
6. Output
7. Themes
8. License

About The Project

Ever wondered what languages you actually use? ghlang makes pretty charts to show you:

my actual language stats across all repos

• GitHub mode: Pulls stats from all your repos via the API (counts bytes)
• Local mode: Analyzes files on your machine using tokount (counts lines)

Why ghlang?

Unlike tools like github-readme-stats (which generate SVG cards for your README), ghlang is a CLI tool that:

• Runs locally on your machine (Python-based)
• Analyzes local files, not just GitHub repos
• Generates downloadable charts (PNG/SVG) you can use anywhere
• Exports raw JSON data for further analysis
• Works offline for local analysis
• Gives you full control over the data

If you want embedded GitHub stats for your README, use github-readme-stats. If you want to analyze your actual codebase and generate charts you can save, share, or customize, use ghlang.

(that said, ghlang does support SVG output with --format svg, so you can totally embed your charts in READMEs too - would be cool to see people do that!)

(back to top)

Built With

• Python
• Typer
• Matplotlib
• Pillow
• Requests
• Rich
• tokount (for local analysis)

(back to top)

Getting Started

Getting this running is pretty straightforward.

What You'll Need

• Python 3.10+
• For GitHub mode: a GitHub token
• For local mode: tokount

Installation

# with pipx (recommended)
pipx install ghlang

# or with pip
pip install ghlang

# or with yay (AUR)
yay -S python-ghlang

# or with paru (AUR)
paru -S python-ghlang

# or install from source
pip install git+https://github.com/MihaiStreames/ghlang.git

For local mode, you'll also need tokount:

# with cargo
cargo install tokount

# or with yay (AUR)
yay -S tokount

Setting Up GitHub Mode

1. Get a token from GitHub Settings

   • Pick repo for private repos, or just public_repo for public only

2. Run it once to create the config file:

   ghlang github
   

   Config lives at ~/.config/ghlang/config.toml (or %LOCALAPPDATA%\ghlang\config.toml on Windows)

3. Add your token to the config:

   [github]
   token = "ghp_your_token_here"
   

4. Run it again and you're good:

   ghlang github
   

(back to top)

Usage

All the Flags

Both github and local commands share the same options:

Flag	Short	What it does
--config		use a different config file
--output-dir		where to save the charts (directory)
--output	-o	custom output filename (creates _pie and _bar variants)
--title	-t	custom chart title
--top-n		how many languages in the bar chart (default: 5)
--save-json		save raw stats as JSON files
--theme		chart color theme (default: light)
--format	-f	output format, overrides --output extension (png or svg)
--json-only		output JSON only, skip chart generation
--stdout		output stats to stdout (implies --json-only --quiet)
--quiet	-q	suppress log output (only show errors)
--verbose	-v	show more details

The local command also takes an optional [PATH] argument (defaults to .) and has one extra flag:

Flag	What it does
--follow-links	follow symlinks when analyzing (unix only)

The config command has its own options:

Flag	What it does
--show	print config as formatted table
--path	print config file path
--raw	print raw TOML contents

(back to top)

Shell Completion

ghlang has built-in shell completion. To enable it:

# install completion for your shell
ghlang --install-completion

# or just view the completion script
ghlang --show-completion

After installing, restart your shell or source your config file.

(back to top)

What You Get

Charts end up in your output directory (.png by default, or .svg with --format svg):

File	What it is
language_pie.png	pie chart with all languages
language_bar.png	bar chart with top N languages
language_stats.json	raw stats (with --save-json)
tokount_stats.json	detailed tokount output (local mode, with --save-json)
repositories.json	list of repos analyzed (GitHub mode, with --save-json)
github_colors.json	language colors from GitHub (with --save-json)

(back to top)

Config Options

Everything lives in config.toml:

[github]

Option	Default	What it does
token	-	your GitHub token
affiliation	"owner,collaborator,organization_member"	which repos to include
visibility	"all"	all, public, or private
ignored_repos	[]	repos to skip (e.g. "org/*", "https://github.com/user/repo")

[tokount]

Option	Default	What it does
ignored_dirs	["node_modules", "vendor", ...]	directories to skip

[output]

Option	Default	What it does
directory	"~/Documents/ghlang-stats"	where to save charts

[preferences]

Option	Default	What it does
verbose	false	more logging
theme	"light"	chart color theme

(back to top)

Themes

ghlang comes with built-in themes and supports community themes:

Theme	Preview	Source
light		built-in
dark		built-in
monokai		community

Using themes:

# use a theme
ghlang github --theme dark

Set default in config.toml:

[preferences]
theme = "dark"

(back to top)

License

MIT. Do whatever you want with it. See LICENSE for more information.

(back to top)

---

Made with ❤️