nyuuzyou/jihulab-code

---
annotations_creators:
- machine-generated
language_creators:
- found
language:
- code
- zh
- en
license: other
multilinguality:
- multilingual
pretty_name: JihuLab Code Dataset
size_categories:
- 1M<n<10M
source_datasets:
- original
task_categories:
- text-generation
tags:
- code
- chinese
configs:
- config_name: default
data_files:
- split: train
path: "data.parquet"
default: true
dataset_info:
features:
- name: code
dtype: string
- name: repo_name
dtype: string
- name: path
dtype: string
- name: language
dtype: string
- name: license
dtype: string
- name: size
dtype: int64
---

# JihuLab Code Dataset

## Dataset Description

This dataset was compiled from code repositories hosted on [JihuLab](https://jihulab.com), a GitLab-based code hosting platform operated by JiHu (GitLab's Chinese joint venture). JihuLab serves as the primary GitLab instance for Chinese developers and enterprises, offering localized services and compliance with Chinese regulations. This dataset is particularly valuable for training code models with Chinese language understanding and enterprise-level coding practices.

### Dataset Summary

| Statistic | Value |
|-----------|-------|
| **Total Files** | 1,853,253 |
| **Total Repositories** | 11,589 |
| **Total Size** | 1.5 GB (compressed Parquet) / 12.76 GB (uncompressed) |
| **Programming Languages** | 304 |
| **File Format** | Parquet with Zstd compression |

### Key Features

- **Chinese developer ecosystem**: Contains code from JihuLab, GitLab's official Chinese distribution, featuring Chinese comments, documentation, and variable names
- **Diverse language coverage**: Spans 304 programming languages identified by [go-enry](https://github.com/go-enry/go-enry) (based on GitHub Linguist rules)
- **Rich metadata**: Includes repository name, file path, detected language, license information, and file size
- **Enterprise and open-source projects**: Includes code from both individual developers and Chinese enterprises using GitLab
- **Quality filtered**: Extensive filtering to remove vendor code, build artifacts, generated files, and low-quality content

### Languages

The dataset includes 304 programming languages. The top 30 languages by file count:

| Rank | Language | File Count |
|------|----------|------------|
| 1 | Java | 348,517 |
| 2 | C | 209,924 |
| 3 | JavaScript | 191,164 |
| 4 | Python | 172,798 |
| 5 | C++ | 136,046 |
| 6 | Go | 80,000 |
| 7 | TypeScript | 79,067 |
| 8 | HTML | 69,173 |
| 9 | C# | 64,511 |
| 10 | Rust | 50,515 |
| 11 | Shell | 43,352 |
| 12 | Vue | 40,687 |
| 13 | TSX | 36,844 |
| 14 | CSS | 34,779 |
| 15 | Makefile | 26,227 |
| 16 | Ruby | 25,812 |
| 17 | PHP | 21,401 |
| 18 | CMake | 15,292 |
| 19 | Kotlin | 14,220 |
| 20 | BitBake | 13,060 |
| 21 | SCSS | 10,957 |
| 22 | Scala | 9,333 |
| 23 | Dart | 9,125 |
| 24 | Lua | 7,413 |
| 25 | ASP.NET | 7,005 |
| 26 | Vim Script | 5,710 |
| 27 | Unix Assembly | 5,239 |
| 28 | Starlark | 5,134 |
| 29 | Objective-C | 4,931 |
| 30 | Factor | 4,920 |

### Licenses

The dataset includes files from repositories with various licenses. Repositories with restrictive licenses (CC-BY-ND variants, Commons Clause, SSPL) were excluded:

| License | File Count |
|---------|------------|
| apache-2.0 | 551,008 |
| unknown | 535,320 |
| mit | 320,834 |
| agpl-3.0 | 169,922 |
| gpl-2.0 | 112,829 |
| bsd | 65,104 |
| cc0-1.0 | 13,557 |
| lgpl-3.0 | 12,871 |
| lgpl-2.1 | 9,960 |
| bsd-3-clause | 9,109 |
| bsl-1.1 | 8,972 |
| epl-1.0 | 7,494 |
| gpl-3.0 | 7,476 |
| unlicense | 6,265 |
| cc-by-3.0 | 4,717 |
| cc-by-nc-sa | 4,339 |
| mpl-2.0 | 3,847 |
| cc-by-4.0 | 2,459 |
| cc-by-nc-sa-4.0 | 1,715 |
| cc-by-sa-4.0 | 1,701 |
| bsd-2-clause | 1,599 |
| cc-by-nc-nd-4.0 | 1,222 |
| isc | 520 |
| wtfpl | 274 |
| cc-by-nc-4.0 | 122 |
| cc-by-sa | 13 |
| cc-by-sa-3.0 | 4 |

## Dataset Structure

### Data Fields

| Field | Type | Description |
|-------|------|-------------|
| `code` | string | Content of the source file (UTF-8 encoded) |
| `repo_name` | string | Name of the JihuLab repository (format: `username/repo` or `group/subgroup/repo`) |
| `path` | string | Path of the file within the repository (relative to repo root) |
| `language` | string | Programming language as identified by [go-enry](https://github.com/go-enry/go-enry) |
| `license` | string | License of the repository (SPDX identifier or "unknown") |
| `size` | int64 | Size of the source file in bytes |

### Data Format

- **Format**: Apache Parquet with Zstd compression
- **File Structure**: Single consolidated file (`data.parquet`)

### Data Splits

All examples are in the train split. There is no validation or test split.

### Example Data Point

```
{
'code': 'package com.example.demo;\n\nimport org.springframework.boot.*;\nimport org.springframework.boot.autoconfigure.*;\n...',
'repo_name': 'SmallQ/demo',
'path': 'src/main/java/com/example/demo/DemoApplication.java',
'language': 'Java',
'license': 'unknown',
'size': 400
}
```

## Dataset Creation

### Pipeline Overview

The dataset was created through a multi-stage pipeline:

1. **Repository Discovery**: Paginated API requests to JihuLab's GitLab API (`/api/v4/projects`) to enumerate public repositories
2. **Branch Selection**: Using the repository's default branch (typically `main` or `master`)
3. **Repository Downloading**: Downloading repository archives via JihuLab's archive endpoint
4. **Content Extraction**: Extracting and filtering source code files
5. **Parquet Generation**: Writing filtered records to Parquet with Zstd compression

### Language Detection

Programming languages are detected using [go-enry](https://github.com/go-enry/go-enry), a Go port of GitHub's Linguist library. Only files classified as **Programming** or **Markup** language types are included (Data and Prose types are excluded).

### License Detection

Licenses are detected by:
1. Scanning for license files (`LICENSE`, `LICENSE.txt`, `LICENSE.md`, `COPYING`, etc.)
2. Matching license text against known patterns (MIT, Apache 2.0, GPL variants, BSD, Creative Commons, etc.)
3. Defaulting to "unknown" if no license can be detected

**Blocked Licenses**: The following restrictive licenses are excluded from the dataset:
- `cc-by-nd`, `cc-by-nd-2.0`, `cc-by-nd-3.0`, `cc-by-nd-4.0` (Creative Commons No-Derivatives)
- `commons-clause`
- `sspl`, `sspl-1.0` (Server Side Public License)

### File Filtering

Extensive filtering is applied to ensure data quality:

#### Size Limits
| Limit | Value |
|-------|-------|
| Max repository ZIP size | 48 MB |
| Max single file size | 1 MB |
| Max line length | 1,000 characters |

#### Excluded Directories
- **Configuration**: `.git/`, `.github/`, `.gitlab/`, `.vscode/`, `.idea/`, `.vs/`, `.settings/`, `.eclipse/`, `.project/`, `.metadata/`
- **Vendor/Dependencies**: `node_modules/`, `bower_components/`, `jspm_packages/`, `vendor/`, `third_party/`, `3rdparty/`, `external/`, `packages/`, `deps/`, `lib/vendor/`, `target/dependency/`, `Pods/`
- **Build Output**: `build/`, `dist/`, `out/`, `bin/`, `target/`, `release/`, `debug/`, `.next/`, `.nuxt/`, `_site/`, `_build/`, `__pycache__/`, `.pytest_cache/`, `cmake-build-*`, `.gradle/`, `.maven/`

#### Excluded Files
- **Lock Files**: `package-lock.json`, `yarn.lock`, `pnpm-lock.yaml`, `Gemfile.lock`, `Cargo.lock`, `poetry.lock`, `Pipfile.lock`, `composer.lock`, `go.sum`, `mix.lock`
- **Minified Files**: Any file containing `.min.` in the name
- **Binary Files**: `.exe`, `.dll`, `.so`, `.dylib`, `.a`, `.lib`, `.o`, `.obj`, `.jar`, `.war`, `.ear`, `.class`, `.pyc`, `.pyo`, `.wasm`, `.bin`, `.dat`, `.pdf`, `.doc`, `.docx`, `.xls`, `.xlsx`, `.ppt`, `.pptx`, `.zip`, `.tar`, `.gz`, `.bz2`, `.7z`, `.rar`, `.jpg`, `.jpeg`, `.png`, `.gif`, `.bmp`, `.ico`, `.svg`, `.mp3`, `.mp4`, `.avi`, `.mov`, `.wav`, `.flac`, `.ttf`, `.otf`, `.woff`, `.woff2`, `.eot`
- **System Files**: `.DS_Store`, `thumbs.db`

#### Content Filtering
- **UTF-8 Validation**: Files must be valid UTF-8 encoded text
- **Binary Detection**: Files detected as binary by go-enry are excluded
- **Generated Files**: Files with generation markers in the first 500 bytes are excluded:
- `generated by`, `do not edit`, `auto-generated`, `autogenerated`, `automatically generated`, `code generator`, `generated code`, `this file is generated`, `@generated`, `<auto-generated`
- **Empty Files**: Files that are empty or contain only whitespace are excluded
- **Long Lines**: Files with any line exceeding 1,000 characters are excluded
- **go-enry Filters**: Additional filtering using go-enry's `IsVendor()`, `IsImage()`, `IsDotFile()`, `IsTest()`, and `IsGenerated()` functions
- **Documentation-only Repos**: Repositories containing only documentation files (no actual code) are skipped

### Source Data

All data originates from public repositories hosted on [JihuLab](https://jihulab.com).

## Considerations for Using the Data

### Personal and Sensitive Information

The dataset may contain:
- Email addresses in code comments or configuration files
- API keys or credentials that were accidentally committed
- Personal information in comments or documentation

Users should exercise caution and implement appropriate filtering when using this data.

### Licensing Information

This dataset is a collection of source code from repositories with various licenses. Any use of all or part of the code gathered in this dataset must abide by the terms of the original licenses, including attribution clauses when relevant. The license field in each data point indicates the license of the source repository.