Metadata-Version: 2.4
Name: textstructparser
Version: 2025.12.20202423
Summary: A new package designed to parse and structure information from unstructured text. It takes textual input and utilizes advanced language models with pattern matching to extract specific data points, en
Author-email: textstructparser <hi@eugene.plus>
License: MIT
Project-URL: Homepage, https://github.com/chigwell/textstructparser
Requires-Python: >=3.9
Description-Content-Type: text/markdown
Requires-Dist: langchain-llm7>=0.0.0
Requires-Dist: llmatch-messages>=0.0.0
Requires-Dist: langchain-core>=0.3.0

# **textstructparser** 🔍

**Parse and structure information from unstructured text using advanced language models.**

[![PyPI version](https://badge.fury.io/py/textstructparser.svg)](https://badge.fury.io/py/textstructparser)
[![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](https://opensource.org/licenses/MIT)
[![Downloads](https://pepy.tech/badge/textstructparser)](https://pepy.tech/project/textstructparser)
[![LinkedIn](https://img.shields.io/badge/LinkedIn-Connect-blue)](https://www.linkedin.com/in/chigwell/)

---

## **Overview**
`textstructparser` is a Python package designed to extract structured data from unstructured text using **pattern matching** and **large language models (LLMs)**. It ensures consistent, machine-readable output for downstream applications like data processing, NLP pipelines, or automation workflows.

By default, it uses **ChatLLM7** (via `langchain_llm7`), but developers can easily integrate their preferred LLM (e.g., OpenAI, Anthropic, Google) for flexibility.

---

## **Installation**
```bash
pip install textstructparser
```

---

## **Features**
✅ **Flexible LLM Integration** – Works with any `BaseChatModel` from LangChain.
✅ **Pattern-Based Extraction** – Uses regex constraints for structured output.
✅ **Default LLM7 Support** – Free-tier rate limits sufficient for most use cases.
✅ **Customizable** – Override defaults with your own API keys or LLMs.

---

## **Usage Examples**

### **Basic Usage (Default LLM7)**
```python
from textstructparser import textstructparser

user_input = "Extract dates and names from this text: John Doe visited New York on 2023-10-15."
response = textstructparser(user_input)
print(response)  # Returns structured data matching the predefined pattern
```

### **Using a Custom LLM (OpenAI)**
```python
from langchain_openai import ChatOpenAI
from textstructparser import textstructparser

llm = ChatOpenAI(model="gpt-4")
response = textstructparser(user_input, llm=llm)
```

### **Using a Custom LLM (Anthropic)**
```python
from langchain_anthropic import ChatAnthropic
from textstructparser import textstructparser

llm = ChatAnthropic(model="claude-2")
response = textstructparser(user_input, llm=llm)
```

### **Using a Custom LLM (Google Vertex AI)**
```python
from langchain_google_genai import ChatGoogleGenerativeAI
from textstructparser import textstructparser

llm = ChatGoogleGenerativeAI(model="gemini-pro")
response = textstructparser(user_input, llm=llm)
```

### **Passing a Custom API Key (LLM7)**
```python
from textstructparser import textstructparser

# Option 1: Via environment variable
import os
os.environ["LLM7_API_KEY"] = "your_api_key_here"

# Option 2: Directly in function call
response = textstructparser(user_input, api_key="your_api_key_here")
```

---

## **Parameters**
| Parameter | Type | Description |
|-----------|------|-------------|
| `user_input` | `str` | The unstructured text to parse. |
| `api_key` | `Optional[str]` | LLM7 API key (defaults to `LLM7_API_KEY` env var). |
| `llm` | `Optional[BaseChatModel]` | Custom LangChain LLM (defaults to `ChatLLM7`). |

---

## **How It Works**
1. **Input Processing**: Takes raw text and applies a **predefined regex pattern** for structured extraction.
2. **LLM Integration**: Uses `llmatch` (from `llmatch_messages`) to query the LLM with system/human prompts.
3. **Output Validation**: Ensures extracted data matches the regex pattern before returning.

---

## **Default LLM (LLM7)**
- **Provider**: [LLM7](https://token.llm7.io/)
- **Free Tier**: Sufficient for most use cases (rate limits apply).
- **Upgrade**: Pass a custom API key via `api_key` or `LLM7_API_KEY` env var.

**Get a free API key**: [https://token.llm7.io/](https://token.llm7.io/)

---

## **Custom LLM Support**
For **OpenAI**, **Anthropic**, **Google**, or other LangChain-compatible LLMs, simply pass your model instance:
```python
from langchain_openai import ChatOpenAI
from textstructparser import textstructparser

llm = ChatOpenAI(model="gpt-3.5-turbo")
response = textstructparser(user_input, llm=llm)
```

---

## **Contributing**
Contributions are welcome! Please open an issue or submit a PR:
🔗 [GitHub Issues](https://github.com/chigwell/textstructparser/issues)

---

## **License**
MIT – See [LICENSE](https://github.com/chigwell/textstructparser/blob/main/LICENSE) for details.

---

## **Author**
👤 **Eugene Evstafev**
📧 [hi@euegne.plus](mailto:hi@euegne.plus)
🔗 [LinkedIn](https://www.linkedin.com/in/chigwell/)
🐙 [GitHub](https://github.com/chigwell)

---
