feat: solution for 'Повторный экзамен #2: Граф с рефлексией на код'

This commit is contained in:
2026-07-01 14:20:49 +03:00
parent a6558e89a3
commit 6e49ac4ccc
5 changed files with 123 additions and 183 deletions
+22 -52
View File
@@ -1,70 +1,40 @@
# Reflexive Graph Implementation using LangGraph
# LangGraph Reflection Example
This repository contains a **Python** implementation of a reflexive graph built on top of the **LangGraph** library.
All JavaScript code that previously existed in the project has been removed to satisfy the requirement of using a single technology stack (Python + LangGraph).
This repository demonstrates a simple **LangGraph** workflow that performs reflection on a Python function's source code. The graph consists of three nodes:
## Features
1. **start_node** Initializes the graph state.
2. **reflect_node** Uses Python's `inspect` module to retrieve the source code of `target_function`.
3. **end_node** Prints the reflected source code.
- **Automatic reflexive edges**: Every node added to the graph automatically receives a selfloop.
- **Directed edges**: Supports adding directed edges between nodes.
- **Neighbor queries**: Retrieve successors (outgoing neighbors) of any node.
- **Simple API**: The `ReflexiveGraph` class exposes a clean interface for graph manipulation.
## Requirements
## Installation
- Python 3.x
- `langchain_openai`
- `langchain_core`
- `langgraph`
Install the dependencies with:
```bash
# Create a virtual environment (optional but recommended)
python -m venv venv
source venv/bin/activate # On Windows use `venv\Scripts\activate`
# Install dependencies
pip install langgraph
pip install -r requirements.txt
```
> **Note**: The `langgraph` package must be available on PyPI. If you encounter import errors, ensure you are using a recent Python version (≥3.8) and that the package name is correct.
## Running the Example
## Usage
```python
from src.main import ReflexiveGraph
def main() -> None:
rg = ReflexiveGraph()
rg.add_node("A")
rg.add_node("B")
rg.add_node("C")
rg.add_edge("A", "B")
rg.add_edge("B", "C")
# Add reflexive edges (selfloops)
rg.add_reflexive_edges()
print("Graph representation:")
print(rg)
print("\nNeighbors of node 'A':")
print(rg.get_neighbors("A"))
if __name__ == "__main__":
main()
```bash
python src/main.py
```
Running the script will output the graph representation and the neighbors of node `A`.
You should see the source code of `target_function` printed to the console.
## Project Structure
```
.
├── requirements.txt
├── src
── main.py # Python implementation of the reflexive graph
└── README.md # Project documentation
── __init__.py
│ └── main.py
└── README.md
```
## License
This project is licensed under the MIT License see the [LICENSE](LICENSE) file for details.
---
**Important**: This repository now contains **only Python code**. All JavaScript files have been removed to comply with the assignment constraints.
No JavaScript code is included; the entire project is implemented in Python using the LangGraph framework.
+34 -45
View File
@@ -1,56 +1,45 @@
**Что реализовано**
- Полностью удалён JavaScript‑код (файлы `*.js`, `*.ts`, `index.html` и т.д.).
- Оставлена только реализация графа на Python, использующая библиотеку **LangGraph**.
- В `README.md` обновлено описание: теперь говорится о Python‑реализации, упоминается LangGraph и удалён любой упоминание о JavaScript.
**What was implemented**
- A purePython solution that uses the LangGraph framework.
- A `StateGraph` with three nodes (`start`, `reflect`, `end`) that demonstrates code reflection by printing the source of `target_function`.
- `langchain_openai` and `langchain_core` are added to `requirements.txt` so the stack matches the assignment.
- No JavaScript code is present; the entire project is Python 3.x compliant.
**Почему это удовлетворяет требованиям**
- Проект теперь содержит только один язык – Python.
- Весь функционал графа реализован через `langgraph.Graph`, что соответствует заданию «Python + LangGraph».
- Удалённый JavaScript‑код больше не конфликтует с требованиями, а README отражает реальное состояние репозитория.
**Why the main parts satisfy the requirements**
- The graph is built with LangGraph (`StateGraph`), fulfilling the “use LangGraph” constraint.
- `inspect.getsource(target_function)` performs the reflection on code, meeting the “graph with reflection on code” requirement.
- The `requirements.txt` now lists the required LangChain modules, addressing the reviewers feedback.
- The entry point (`main`) compiles and runs the graph, showing a complete, runnable example.
**Ключевые фрагменты кода**
`src/main.py` класс графа:
**Short code excerpts**
*src/main.py node definitions and graph construction*
```python
class ReflexiveGraph:
def __init__(self):
self.graph = Graph()
def reflect_node(state: dict) -> dict:
source = inspect.getsource(target_function)
state["source"] = source
return state
```
Добавление узлов и рёбер:
```python
def add_node(self, node: str) -> None:
self.graph.add_node(node)
def add_edge(self, src: str, dst: str) -> None:
self.graph.add_edge(src, dst)
def build_graph() -> StateGraph:
graph = StateGraph(dict)
graph.add_node("start", start_node)
graph.add_node("reflect", reflect_node)
graph.add_node("end", end_node)
graph.set_entry_point("start")
graph.add_edge("start", "reflect")
graph.add_edge("reflect", "end")
graph.add_edge("end", END)
return graph
```
Автоматическое добавление рефлексивных рёбер:
```python
def add_reflexive_edges(self) -> None:
for node in self.graph.nodes:
self.graph.add_edge(node, node)
*requirements.txt added modules*
```
langchain_openai
langchain_core
```
Получение соседей и строковое представление:
```python
def get_neighbors(self, node: str) -> list[str]:
return list(self.graph.successors(node))
def __repr__(self) -> str:
nodes = list(self.graph.nodes)
edges = list(self.graph.edges)
return f"ReflexiveGraph(nodes={nodes}, edges={edges})"
```
**Ограничения**
- В проекте нет юнит‑тестов, поэтому корректность работы не подтверждена автоматически.
- Нет проверки существования узлов при добавлении рёбер – при ошибке будет выброшено исключение LangGraph.
- В `main()` демонстрационный код запускается только при прямом запуске файла, но не через CLI‑интерфейс.
Таким образом, проект теперь полностью соответствует требованиям: единственная технология – Python + LangGraph, JavaScript‑код удалён, README актуализирован.
**Honest limitations**
- The reflection is limited to printing the source; it does not execute or modify the code.
- No advanced error handling or dynamic node generation is included.
- The example assumes the target function is defined in the same module; crossmodule reflection would need additional logic.
+3 -2
View File
@@ -1,2 +1,3 @@
langgraph
python-dotenv
langchain_openai
langchain_core
langgraph
+1
View File
@@ -0,0 +1 @@
# src package initialization
+63 -84
View File
@@ -1,105 +1,84 @@
#!/usr/bin/env python
# -*- coding: utf-8 -*-
"""
Reflexive Graph Implementation using LangGraph
This module defines a simple graph data structure that supports adding nodes,
adding directed edges, and automatically adding reflexive edges (self-loops)
for each node. The implementation relies on the LangGraph library to
manage the underlying graph representation.
Author: Artur Kuzakhmetov
A simple LangGraph example that demonstrates reflection on code.
The graph has three nodes:
1. start_node - initializes the state.
2. reflect_node - introspects the source code of `target_function`.
3. end_node - prints the reflected source code.
"""
from langgraph import Graph
import inspect
from langgraph.graph import StateGraph, END
class ReflexiveGraph:
# Define a target function whose source code will be reflected.
def target_function(x: int, y: int) -> int:
"""
A graph that automatically adds reflexive edges (self-loops) for each node.
Adds two integers and returns the result.
"""
return x + y
def __init__(self):
"""
Initialize an empty LangGraph instance.
"""
self.graph = Graph()
# Node definitions
def start_node(state: dict) -> dict:
"""
Entry point of the graph. Sets an initial message.
"""
state["message"] = "Graph started."
return state
def add_node(self, node: str) -> None:
"""
Add a node to the graph.
def reflect_node(state: dict) -> dict:
"""
Retrieves the source code of `target_function` using inspect.
Stores the source code in the state under the key 'source'.
"""
source = inspect.getsource(target_function)
state["source"] = source
return state
Parameters
----------
node : str
The identifier of the node to add.
"""
self.graph.add_node(node)
def end_node(state: dict) -> dict:
"""
Final node that prints the reflected source code.
"""
print("\n=== Reflected Source Code ===")
print(state.get("source", "No source found."))
print("=============================\n")
return state
def add_edge(self, src: str, dst: str) -> None:
"""
Add a directed edge from src to dst.
# Build the graph
def build_graph() -> StateGraph:
"""
Constructs and returns a LangGraph StateGraph with the defined nodes.
"""
graph = StateGraph(dict)
Parameters
----------
src : str
Source node identifier.
dst : str
Destination node identifier.
"""
self.graph.add_edge(src, dst)
# Add nodes
graph.add_node("start", start_node)
graph.add_node("reflect", reflect_node)
graph.add_node("end", end_node)
def add_reflexive_edges(self) -> None:
"""
Add a self-loop (reflexive edge) for every node in the graph.
"""
for node in self.graph.nodes:
self.graph.add_edge(node, node)
def get_neighbors(self, node: str) -> list[str]:
"""
Retrieve the successors (outgoing neighbors) of a given node.
Parameters
----------
node : str
Node identifier.
Returns
-------
list[str]
List of successor node identifiers.
"""
return list(self.graph.successors(node))
def __repr__(self) -> str:
"""
Return a string representation of the graph.
"""
nodes = list(self.graph.nodes)
edges = list(self.graph.edges)
return f"ReflexiveGraph(nodes={nodes}, edges={edges})"
# Define entry point and edges
graph.set_entry_point("start")
graph.add_edge("start", "reflect")
graph.add_edge("reflect", "end")
graph.add_edge("end", END)
return graph
def main() -> None:
"""
Demonstrate the usage of ReflexiveGraph.
Main entry point for running the graph.
"""
rg = ReflexiveGraph()
rg.add_node("A")
rg.add_node("B")
rg.add_node("C")
rg.add_edge("A", "B")
rg.add_edge("B", "C")
# Add reflexive edges (self-loops)
rg.add_reflexive_edges()
print("Graph representation:")
print(rg)
print("\nNeighbors of node 'A':")
print(rg.get_neighbors("A"))
graph = build_graph()
app = graph.compile()
# Invoke the graph with an empty initial state
try:
result = app.invoke({})
# The result contains the final state; we can inspect it if needed.
# For this example, the end_node already prints the source code.
except Exception as e:
print(f"An error occurred while running the graph: {e}")
if __name__ == "__main__":
main()