Metadata-Version: 2.5
Name: langchain-cellprose
Version: 0.1.0
Summary: LangChain document loader for Excel workbooks as Cellprose text (formulas, every sheet, cell addresses)
Project-URL: Homepage, https://cellprose.com
Project-URL: Documentation, https://cellprose.com/guide/
Project-URL: Repository, https://github.com/cellprose/cellprose
Project-URL: Issues, https://github.com/cellprose/cellprose/issues
Author: Utkarsh Gaikwad
License-Expression: Apache-2.0
Requires-Python: >=3.10
Requires-Dist: cellprose>=0.3
Requires-Dist: langchain-core>=0.3
Description-Content-Type: text/markdown

# langchain-cellprose

A LangChain document loader that reads `.xlsx` and `.xlsm` workbooks as [Cellprose](https://cellprose.com) text,
keeping formulas, cell addresses and every sheet in valid Markdown.

```bash
pip install langchain-cellprose   # not yet on PyPI: pip install -e integrations/langchain-cellprose from a clone
```

```python
from langchain_cellprose import CellproseLoader

docs = CellproseLoader("book.xlsx").load()                 # one Document for the whole workbook
docs = CellproseLoader("book.xlsx", mode="sheet").load()   # one per sheet, metadata["sheet"] set
docs = CellproseLoader("book.xlsx", max_rows=200).load()   # options pass to cellprose.dumps
```

For question answering over a workbook, passing the whole workbook in one Document usually beats chunking: the
benchmark answers came from reading every sheet at once. Use `mode="sheet"` when the workbook is too large for
the context window.
