Skip to main content

jcgraph

PyPI version Python Version License: MIT

Java 代码知识图谱构建工具 - 通过静态分析构建 Java 代码知识图谱,为 AI 助手提供精准的代码理解能力

✨ 核心特性

  • 🤖 MCP Server - 为 Claude、Cursor 等 AI 工具提供 Java 代码理解能力
  • 🔍 精准解析 - 基于 Tree-sitter 的语法分析,支持类、方法、字段、注解
  • 🕸️ 调用链分析 - 完整追踪方法调用关系,支持接口多实现、Lombok
  • 📊 可视化界面 - Web UI 展示调用时序图和代码结构
  • 🚀 快速查询 - SQLite 本地存储,秒级响应

📦 安装

# 基础安装
pip install jcgraph

# 完整功能(包含 Web 服务)
pip install jcgraph[full]

# MCP Server(需要 Python >= 3.10)
pip install jcgraph[mcp]

提示: 如果 pip 命令无法使用,请使用 python3 -m pip install jcgraph 或参考安装文档

🚀 快速开始

1. 扫描 Java 项目

cd /path/to/java/project
jcgraph scan . --name my-project

2. 配置 AI 助手(MCP Server)

# 配置到 Claude Desktop
jcgraph mcp setup --client claude-desktop

# 配置到 Cursor
jcgraph mcp setup --client cursor

配置后重启 AI 工具,即可使用以下能力:

  • 搜索 Java 类和方法
  • 获取方法调用时序图(支持智能过滤)
  • 查看方法源代码
  • 分析代码结构

3. 启动 Web 界面(可选)

jcgraph serve --port 8080

访问 http://localhost:8080 查看可视化界面

🤖 MCP Server 工具

配置后,AI 助手可使用以下工具:

工具 功能 参数
get_call_sequence 获取方法调用时序图 method_id, depth, min_importance
search_code 搜索类/方法/字段 query, node_type, limit
get_method_code 查看方法源代码 method_id

核心功能 - 智能过滤调用时序:

通过 min_importance 参数控制展示粒度:

  • 30 - 展开大部分节点(包括辅助方法)
  • 50(默认)- 只展开重要节点
  • 60 - 只展开高重要度节点
  • 80 - 只展开关键路径

详见:MCP Server 使用指南

📚 使用示例

在 Claude/Cursor 中提问

请分析 UserService.login 方法的调用链,深度3层,只展示重要节点

AI 会自动调用 jcgraph 工具,返回结构化的调用时序:

login() → validate() → findByUsername() → checkPassword()

命令行查询

# 查看统计信息
jcgraph stats

# 检查类信息
jcgraph check class UserService

# 诊断导出
jcgraph diagnose

🏗️ 支持特性

语法解析:

  • ✅ 类(class/interface/enum/abstract)
  • ✅ 方法(静态/抽象/构造函数)
  • ✅ Lombok 注解(@Data, @Getter, @Setter)
  • ✅ Lambda 表达式
  • ✅ 泛型参数

关系分析:

  • ✅ 继承(extends)
  • ✅ 实现(implements)
  • ✅ 方法调用(包括链式调用)
  • ✅ 依赖注入(@Autowired, @Resource)
  • ✅ 接口多实现识别
  • ✅ 循环检测(for/while)
  • ✅ 分支检测(if/switch)

📖 文档

🔧 系统要求

  • Python >= 3.9(基础功能)
  • Python >= 3.10(MCP Server)
  • Java 8+ 项目

📝 更新日志

v0.2.4

  • ✨ 解耦 MCP Server 和 FastAPI 依赖
  • ✨ 新增智能节点过滤(min_importance 参数)
  • 📚 完善安装和配置文档
  • 🐛 修复静态文件打包问题

v0.2.0

  • ✨ 新增 MCP Server 支持
  • ✨ 新增日志和诊断工具
  • ✨ 新增自动配置命令
  • 📚 完整的使用文档

🤝 贡献

欢迎提交 Issue 和 Pull Request!

📄 License

MIT License

🔗 链接


为 Java 开发者和 AI 辅助编程而生

Metadata

Release files for jcgraph 0.2.7

For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.

Source distribution (sdist)

Source distribution for jcgraph 0.2.7
File Size Uploaded Provenance Actions
jcgraph-0.2.7.tar.gz 441.3 kB Without provenance Details for jcgraph-0.2.7.tar.gz

Built distribution (wheel)

Table of built distributions (wheels) for jcgraph 0.2.7
File Interpreter ABI Platform Actions
jcgraph-0.2.7-py3-none-any.whl Python 3 none any Details for jcgraph-0.2.7-py3-none-any.whl

Total release size: 878.4 kB

Release files / jcgraph-0.2.7.tar.gz

Download URL jcgraph-0.2.7.tar.gz
Size 441.3 kB
Tags Source
SHA-256 checksum
How to use checksums
561e6fc960cd6ef8663cf07053d36eda545a47b59f0d5974ce957ac7de86cd13
BLAKE2b-256 checksum
How to use checksums
c43f373d52c1009cded00aa7c1eb32ed5ba9c740374a8796d6ed0bc543f6be2e
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
No
Uploaded via twine/6.2.0 CPython/3.9.6

Release files / jcgraph-0.2.7-py3-none-any.whl

Download URL jcgraph-0.2.7-py3-none-any.whl
Size 437.0 kB
Tags Python 3
SHA-256 checksum
How to use checksums
088024a8cde9b6ae84780251d813ef4038ecf6f0ad9b0258f6070900414bf7cc
BLAKE2b-256 checksum
How to use checksums
4050bfe9f309d638d12b7242924909d873ceb5e6f5a0533bda67253740892317
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
No
Uploaded via twine/6.2.0 CPython/3.9.6

Security for jcgraph 0.2.7

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

No provenance found

None of the files in this release were uploaded with provenance metadata.

A note on package security

PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.

To learn more about protecting yourself from malicious actors, visit our package security documentation.

Anthropic, PBC Visionary sponsor Bloomberg Visionary sponsor Hudson River Trading Visionary sponsor Meta Visionary sponsor NVIDIA Visionary sponsor Microsoft Sustainability sponsor Depot Continuous Integration AWS Cloud computing and Security Sponsor Datadog Monitoring Fastly CDN Google Download Analytics Sentry Error logging StatusPage Status page