parser/xlsx_parser.py

from typing import Iterator
from langchain_core.documents import Document
from langchain_community.document_loaders.base import BaseBlobParser
from langchain_community.document_loaders.blob_loaders import Blob


class XlsxParser(BaseBlobParser):
    """Parse Microsoft Excel spreadsheets from a blob."""

    def lazy_parse(self, blob: Blob) -> Iterator[Document]:
        """Parse a Microsoft Excel document into the Document iterator.

        Args:
            blob: The blob to parse.

        Returns: An iterator of Documents.
        """
        try:
            from openpyxl import load_workbook
        except ImportError as e:
            raise ImportError(
                "Could not import openpyxl, please install with `pip install openpyxl`."
            ) from e

        supported_mime_types = [
            "application/vnd.openxmlformats-officedocument.spreadsheetml.sheet"  # .xlsx
        ]

        # Debugging: Print MIME type
        print(f"Blob MIME type: {blob.mimetype}")

        if blob.mimetype not in supported_mime_types:
            raise ValueError(
                f"This blob type is not supported for this parser. Supported types are: {supported_mime_types}"
            )

        with blob.as_bytes_io() as xlsx_file:
            workbook = load_workbook(xlsx_file, data_only=True)
            for sheet in workbook.sheetnames:
                worksheet = workbook[sheet]
                text = ""
                for row in worksheet.iter_rows(values_only=True):
                    row_data = "\t".join([str(cell) if cell is not None else "" for cell in row])
                    text += row_data + "\n"
                metadata = {"source": blob.source, "sheet": sheet}
                yield Document(page_content=text, metadata=metadata)