Metadata-Version: 2.5
Name: langchain-anytomarkdown
Version: 0.1.0
Summary: LangChain document loader for the AnyToMarkdown API: PDF, Word, Excel, PowerPoint and web pages to Markdown.
Project-URL: Homepage, https://anymarkdownapi.gtgroup.dev/
Project-URL: Documentation, https://anymarkdownapi.gtgroup.dev/api/
License: MIT
Keywords: document loader,langchain,markdown,pdf,rag
Requires-Python: >=3.9
Requires-Dist: langchain-core>=0.2
Requires-Dist: requests>=2.31
Description-Content-Type: text/markdown

# langchain-anytomarkdown

LangChain document loader for the [AnyToMarkdown API](https://anymarkdownapi.gtgroup.dev/api/). It turns PDF, Word, Excel, PowerPoint, e-mail and web pages into Markdown, with tables, headings and lists kept. Scanned pages are read with OCR.

```bash
pip install langchain-anytomarkdown
```

Get a free API key on [RapidAPI](https://rapidapi.com/controller2042000/api/anytomarkdown-pdf-to-markdown-and-document-parser-api) (the Basic plan includes 1,000 conversions a month).

```python
import os
from langchain_anytomarkdown import AnyToMarkdownLoader

docs = AnyToMarkdownLoader(file_path="handbook.pdf", api_key=os.environ["RAPIDAPI_KEY"]).load()
print(docs[0].metadata)  # title, source_format, tokens, pages_processed
```

Load a web page instead:

```python
docs = AnyToMarkdownLoader(url="https://en.wikipedia.org/wiki/Markdown", api_key=os.environ["RAPIDAPI_KEY"]).load()
```

Files are processed in memory and are not stored. Every call uses one conversion from your plan.
