refactor: ready for open (#12)

* refactor: remove compress implementations

* refactor: move out main

* go mod tidy

* readme

* feat: comand write

* doc: add docs

* remove

* opt doc

* fix:(go_ast) get receiver name bug

* remove rust ci

* tmp remove line CI

* add NewIdentityFromString()

* move example json to huggingface

* fix

* fix

Author: AsterDY

.github/workflows/rust.yml

@@ -74,3 +74,4 @@ src/lang/testdata
 *.json
 
 tools
+abcoder

.github/workflows/simple_checks.yml

@@ -1,126 +1,64 @@
-<!--
- Copyright 2025 CloudWeGo Authors
- 
- Licensed under the Apache License, Version 2.0 (the "License");
- you may not use this file except in compliance with the License.
- You may obtain a copy of the License at
- 
-     https://www.apache.org/licenses/LICENSE-2.0
- 
- Unless required by applicable law or agreed to in writing, software
- distributed under the License is distributed on an "AS IS" BASIS,
- WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
- See the License for the specific language governing permissions and
- limitations under the License.
--->
-
 # ABCoder: AI-Based Coder(AKA: A Brand-new Coder)
 
 ![ABCoder](images/ABCoder.png)
 
-ABCoder, an AI-powered tool, streamlines coding by keeping real-time status updates, providing lossless code compression, and giving development guidance. It enhances testing by identifying quality, generating reports, and auto-creating test cases. It also offers guidance for refactoring, including language stack switches.
-
-# Table of Contents
-
-- [ABCoder: AI-Based Coder(AKA: A Brand-new Coder)](#abcoder-ai-based-coderaka-a-brand-new-coder)
-- [Table of Contents](#table-of-contents)
-- [Overview](#overview)
-- [Quick Start](#quick-start)
-  - [Prerequisites](#prerequisites)
-  - [Running through Coze OpenAPI](#running-through-coze-openapi)
-- [Status Update](#status-update)
-- [Lossless Compression](#lossless-compression)
-- [Development Guide](#development-guide)
-- [Testing Enhancements](#testing-enhancements)
-- [Refactor/Rewrite Guide](#refactorrewrite-guide)
-- [Getting Involved](#getting-involved)
-
 # Overview
+ABCoder, an AI-oriented code-processing SDK, is designed to enhance coding context for Large-Language-Model (LLM), and boost developing AI-assisted-coding applications.
 
-ABCoder is a comprehensive open-source software development tool that aims to utilize artificial intelligence to enhance
-the process of coding. This project focuses on various aspects of software development ranging from repository analysis,
-issue and pull request tracking, to automated code compression, development guidance, testing enhancement, and
-refactoring guidance.
 
-# Quick Start
+## Features
 
-## Prerequisites
-- install git and set your access token for github on cmd-line
-- install [rust-toolchain](https://www.rust-lang.org/tools/install) (stable)
-- (optional) install [ollama](https://github.com/ollama/ollama) and run your LLM
-- (optional) create a [Coze](https://www.coze.com/docs/developer_guides/coze_api_overview?_lang=en) agent and set its OpenAPI key
+-  Universal Abstract Syntax Tree (UniAST), an language-independent, AI-friendly specification of code information, providing a flexible and structrual coding context for both AI and hunman.
+  
+-  General Parser, parses abitary-language codes to UniAST.
 
-## Running through Coze OpenAPI
-1. Set .env file for configuration on ABCoder's working directory. Taking Coze as an example:
-```
-# cache for repo，AST and so on
-WORK_DIR=tmp_abcoder
+-  General Writer, transforms UniAST back to codes.
+  
+- (Comming Soon) General Iterator, a framework for visiting the UniAST easily and implementing batch-code-processing workflows.
 
-# exclude dirs for repo parsing, separated by comma
-EXCLUDE_DIRS=target,gen-codes 
+- (Comming Soon) Code RAG, provides a set of tools and functions to help the LLM understand your codes much deeper than ever.
 
-# LLM's api type
-API_TYPE=coze # coze|ollama 
+Based on these features, developers can easily implement or enhance their AI-assisted-coding applications, such as reviewing, optimizing, translating, etc.
 
-# LLM's output language
-LANGUAGE=zh 
 
-# Coze options
-COZE_API_TOKEN="{YOUR_COZE_API_TOKEN}"
-COZE_BOT_ID={YOUR_COZE_BOT_ID}
-```
+## Universal-Abstract-Syntax-Tree Specification
 
-2. compile the parsers
-```
-./script/make_parser.sh
-```
+see [UniAST Specification](docs/uniast-zh.md)
 
-3. compile and run ABCoder
-```
-cargo run --bin cmd compress https://xxx.git
-```
 
-4. Once triggered, ABCoder will take three steps:
-   1. Download the repository in {REPO_DIR}
-   2. Parse the repository and store the AST in {CACHE_DIR}
-   3. Call the LLM to compress the repository codes, and refresh the AST for each call.
-You can stop the process at anytime after step 2. You can restart the compressing by running the same command.
+# Getting Started
 
-5. Export the compressed results
+1. Install ABCoder:
+```bash
+go install github.com/cloudwego/abcoder@latest
 ```
-cargo run --bin cmd export https://xxx.git --out-dir {OUTPUT_DIR}
+2. Use ABCoder to parse a repository to UniAST (JSON)
+```bash
+abcoder parse {language} {repo-path} > ast.json
+```
+3. Do your magic with UniAST...
+4. Use ABCoder to write a UniAST back to codes
+```bash
+abcoder write {language} ast.json
 ```
 
-# Status Update
-
-The system is designed to automatically fetch the latest data from Github upon triggering relevant tasks, ensuring the
-repository status is always up-to-date. It can answer queries related to function, defects based on issue and PR
-information. For more details, check out our Issues and Pull Requests sections on Github.
-
-# Lossless Compression
-
-The system also offers a lossless compression feature for repository code. The specific implementation methods are being
-optimized, and more details will be available soon.
-
-# Development Guide
 
-We welcome all developers wishing to contribute to ABCoder. Our system provides detailed guidance for manual development
-and also supports auto-generation of instructions. Check out our Contribution Guide for more information.
+# Supported Languages
 
-# Testing Enhancements
+ABCoder currently supports the following languages:
 
-The system is designed to analyze existing functions and corresponding tests, identify the overall quality of testing,
-produce reports, and automatically generate test cases for weakly covered items. Our goal is to help repositories
-enhance and perfect their test cases.
+| Language | Parser      | Writer      |
+| -------- | ----------- | ----------- |
+| Go       | ✅           | ✅           |
+| Rust     | ✅           | Coming Soon |
+| C        | Coming Soon | ❌           |
 
-# Refactor/Rewrite Guide
 
-We offer guidance for both small-scale feature iterations and large-scale rewrites, including language stack switches.
-Our system provides a detailed guide for manual development and also supports automated guidance generation.
 
 # Getting Involved
 
 We encourage developers to contribute and make this tool more powerful. If you are interested in contributing to ABCoder
-project, kindly check out our Getting Involved Guide.
+project, kindly check out our Getting Involved Guide:
+- [Parser Extension](docs/parser-zh.md)
 
 > Note: This is a dynamic README and is subject to changes as the project evolves.

.gitignore

@@ -0,0 +1,102 @@
+# ABCoder - Language Parser 介绍
+
+当前 ABCoder 基于 [LSP](https://microsoft.github.io/language-server-protocol/) 协议实现 Parser ，以达到精确依赖收集，并方便后续多语言扩展
+
+## 代码结构
+
+位于 [lang](/lang) 包下，包括：
+
+- uniast：统一 AST 结构的 golang 定义
+- lsp：LSP 协议处理 client，提供了 文件解析、引用查找、语法树解析、定义查找等接口，以及**通用的语言规范 LanguageSpec 接口**
+- collect：负责基于 LSP 符号收集和导出 UniAST，是核心运算逻辑
+- {language}：主体是对 lsp#Spec 接口的对应 {language} 规范的实现。此外还有具体 LSP server 的一些具体调用逻辑
+
+## 运算过程
+
+![lang-parser](../images/lang-parser.png)
+
+1. 通过命令行参数识别语言启动对应 LSP server，并传入初始化参数
+2. 遍历仓库文件，调用 `textDocument/documentSymbol` 方法获取每个文件的所有符号。对于每个符号
+   1. 调用 `textDocument/semanticTokens/range` 方法获取符号代码中的 tokens
+   2. 识别出有效实体的 token，调用 `textDocument/definition` 跳转到对应符号位置，从而建立节点依赖关系
+3. 循环 2 直文件结束。最终将收集到的 lsp symbols 转换为 UniAST 格式并输出
+
+## 扩展其它语言实现
+
+由于 UniAST 并不完全等价 LSP， 因此需要实现一些特定语言专属的行为接口才能进行转换。参考 lang/rust 包，大体需要实现以下能力：
+
+- GetDefaultLSP()：映射用户输入 language 到具体的 lsp.Language，以及对应的 LSP 名称
+- CheckRepo()：检查用户仓库情况，根据各语言规范额处理工具链等问题，并返回默认打开的第一个文件（用于触发 LSP server），以及等候 sever 初始化完成的时间（根据仓库大小来决定）
+- **LanguageSpec interface**: 核心模块，用于处理非 LSP 通用的语法信息、比如判断一个 token 是否是标准库的符号、函数签名解析等：
+- ModulePatcher: 后处理模块，用于处理语言特殊的信息收集。比如 rust 的 use 符号收集（LSP 不收集）。可以不实现
+
+### LaunguageSpec
+
+```
+用于在 LSP 符号收集过程中转换为 UniAST 所需信息，并且这些信息非 LSP 通用定义
+
+```go
+
+// Detailed implementation used for collect LSP symbols and transform them to UniAST
+type LanguageSpec interface {
+    // initialize a root workspace, and return all modules [modulename=>abs-path] inside
+    WorkSpace(root string) (map[string]string, error)
+
+    // give an absolute file path and returns its module name and package path
+    // external path should alse be supported
+    // FIXEM: some language (like rust) may have sub-mods inside a file, but we still consider it as a unity mod here
+    NameSpace(path string) (string, string, error)
+
+    // tells if a file belang to language AST
+    ShouldSkip(path string) bool
+
+    // return the first declaration token of a symbol, as Type-Name
+    DeclareTokenOfSymbol(sym DocumentSymbol) int
+
+    // tells if a token is an AST entity
+    IsEntityToken(tok Token) bool
+
+    // tells if a token is a std token
+    IsStdToken(tok Token) bool
+
+    // return the SymbolKind of a token
+    TokenKind(tok Token) SymbolKind
+
+    // tells if a symbol is a main function
+    IsMainFunction(sym DocumentSymbol) bool
+
+    // tells if a symbol is a language symbol (func, type, variable, etc) in workspace
+    IsEntitySymbol(sym DocumentSymbol) bool
+
+    // tells if a symbol is public in workspace
+    IsPublicSymbol(sym DocumentSymbol) bool
+
+    // declare if the language has impl symbol
+    // if it return true, the ImplSymbol() will be called
+    HasImplSymbol() bool
+    // if a symbol is an impl symbol, return the token index of interface type, receiver type and first-method start (-1 means not found)
+    // ortherwise the collector will use FunctionSymbol() as receiver type token index (-1 means not found)
+    ImplSymbol(sym DocumentSymbol) (int, int, int)
+
+    // if a symbol is a Function or Method symbol,  return the token index of Receiver (-1 means not found),TypeParameters, InputParameters and Outputs
+    FunctionSymbol(sym DocumentSymbol) (int, []int, []int, []int)
+}
+```
+
+- Rust-parser 实现位置：[RustSpec](/lang/rust/spec.go)
+
+```
+
+### ModulePatcher
+
+用于后处理收集完成的模块信息
+
+```go
+// ModulePatcher supplements some information for module
+type ModulePatcher interface {
+    // Patch is called after collect all symbols
+    Patch(ast *parse.Module)
+}
+```
+
+- Rust-parser 实现: [RustModulePatcher](/lang/rust/patch.go)