docling-project/docling · error · DocumentLoadError

MsPowerpointDocumentBackend could not load document with has

Error message

MsPowerpointDocumentBackend could not load document with hash {self.document_hash}

What it means

DocumentLoadError raised by MsPowerpointDocumentBackend when python-pptx's Presentation() constructor throws on the input. Any exception while opening the file (invalid zip, unreadable presentation.xml, wrong extension content, encrypted file) is wrapped, with valid left False and the openpyxl-style chained cause preserved for diagnosis.

Source

Thrown at docling/backend/mspowerpoint_backend.py:167

        super().__init__(in_doc, path_or_stream, options)
        self.path_or_stream: Union[BytesIO, Path] = path_or_stream
        self.page_range = in_doc.limits.page_range

        self.pptx_to_pdf_converter: Optional[Callable] = None
        self.pptx_to_pdf_converter_init: bool = False
        self._render_charts: bool = False

        self.pptx_obj: Optional[presentation.Presentation] = None
        self.valid: bool = False
        try:
            if isinstance(self.path_or_stream, BytesIO):
                self.pptx_obj = Presentation(self.path_or_stream)
            elif isinstance(self.path_or_stream, Path):
                self.pptx_obj = Presentation(str(self.path_or_stream))

            self.valid = True
        except Exception as e:
            raise DocumentLoadError(
                f"MsPowerpointDocumentBackend could not load document with hash {self.document_hash}"
            ) from e

        return

    def page_count(self) -> int:
        if self.is_valid():
            assert self.pptx_obj is not None
            return len(self.pptx_obj.slides)
        else:
            return 0

    @override
    def is_valid(self) -> bool:
        return self.valid

    @classmethod
    @override

View on GitHub (pinned to 61d76f1ff3)

Solutions

1. Read e.__cause__ from the caught DocumentLoadError to see python-pptx's actual complaint.
2. Re-save the deck from PowerPoint/Keynote as standard .pptx and retry.
3. Remove password protection before conversion.
4. Sniff the real format (filetype/python-magic) and route ODP/PDF/legacy inputs to their proper backends; for .ppt ensure soffice is on PATH.

Example fix

# before
result = converter.convert('deck.pptx')  # actually ODP renamed

# after
import filetype
mime = filetype.guess('deck.pptx').mime
assert mime in ('application/vnd.openxmlformats-officedocument.presentationml.presentation',)
result = converter.convert('deck.pptx')

import filetype

def is_real_pptx(path: str) -> bool:
    kind = filetype.guess(path)
    return kind is not None and kind.mime == (
        'application/vnd.openxmlformats-officedocument.presentationml.presentation'
    )

from docling.core.exceptions import DocumentLoadError
try:
    result = converter.convert(pptx_path)
except DocumentLoadError as e:
    log.error('deck load failed (%s): %s', pptx_path, e.__cause__)
    quarantine(pptx_path)

When it happens

Trigger: Converting a file that pptx cannot open: .ppt renamed to .pptx without conversion, a .pptx that is actually a PDF/ODP, an Office-encrypted (password) deck, a macro-enabled file python-pptx rejects, or a truncated upload. Also .ppt input when the soffice-based pre-conversion step produced an invalid pptx.

Common situations: Upload pipelines receiving mislabeled or password-protected decks, Keynote/Google-Slides exports with quirks, corrupted email attachments, or missing/broken LibreOffice for legacy .ppt.

Related errors

• docling-parse could not load document {self.document_hash}:
• MsExcelDocumentBackend could not load document with hash {se
• CsvDocumentBackend could not load document with hash {self.d
• docling-parse could not load document {self.document_hash}.
• Could not initialize the EBCDIC backend for file with hash {

AI-assisted analysis of docling-project/docling@61d76f1ff3 (2026-08-14). Data as JSON: /api/errors/3eea7e62675f0187. Report an issue: GitHub.