zylon-ai/private-gpt · error · ValueError

No valid document content found after conversion

Error message

No valid document content found after conversion

What it means

Raised by DoclingApiReader.lazy_load_data after a successful conversion when every page-content string produced by _get_content is empty (valid_contents filters out falsy entries and the list ends up empty). The server returned content-bearing fields, but after page splitting on the PAGE_PLACEHOLDER marker each fragment is blank/whitespace — so there is genuinely nothing to embed.

Source

Thrown at private_gpt/components/readers/docling/docling_api_reader.py:181

        pages = file_info.config.get("pages", None)

        try:
            conversion_result = await self.client.convert_from_bytes(
                file_name, file_bytes, to_formats=["md"], pages=pages, **load_kwargs
            )
        except Exception as e:
            raise ValueError(f"Document conversion failed: {e}") from e

        if conversion_result.status not in ["success", "partial_success"]:
            raise ValueError(
                f"Document conversion failed with status: {conversion_result.status}. "
                f"Errors: {conversion_result.errors}"
            )

        contents = self._get_content(conversion_result)
        valid_contents = [content for content in contents if content]
        if not valid_contents:
            raise ValueError("No valid document content found after conversion")
        if self._is_extraction_unsuccessful(valid_contents):
            raise ExtractionUnsuccessfulError(
                f"Document extraction unsuccessful for '{file_name}': unmapped-glyph "
                f"ratio exceeded threshold ({self.config.failure_threshold})."
            )

        docs = [
            self._page_to_doc(
                content=content,
                index=idx,
                include_page_metadata=len(valid_contents) > 1,
                extra_info=extra_info,
            )
            for idx, content in enumerate(valid_contents)
        ]

        if debug_mode:
            logger.info(f"Loaded document from {file_name}")

View on GitHub (pinned to 4a030776a3)

Solutions

1. Remove or widen the pages filter in the file config so real content pages are included.
2. Verify with an external tool (pdftotext) that the selected pages actually contain text; if not, enable OCR (docling.use_ocr + langs).
3. Skip files that legitimately have no text at the pipeline level (a separate emptiness check) instead of sending them to Docling.

# validate a requested page range actually has text before ingesting
import pypdf

def page_range_has_text(path: str, pages) -> bool:
    reader = pypdf.PdfReader(path)
    idx = pages if pages else range(len(reader.pages))
    return any(reader.pages[i].extract_text().strip() for i in idx)

try:
    nodes = [n async for n in reader.lazy_load_data(file_info)]
except ValueError as e:
    if "No valid document content" in str(e):
        logger.warning("skipping empty document %s", file_info.file_name)
        return []
    raise

When it happens

Trigger: Conversion returns a single empty page or all pages split to empty strings: page-range selections pointing at blank pages (pages config), documents whose text is entirely images with OCR off, placeholder-only content that the post-processing strips.

Common situations: Ingesting with a 'pages' filter that selects trailing blank pages; scanned documents with OCR disabled; test/placeholder PDFs generated with no text; trimmed exports where text lives in annotations not body content.

Related errors

• No valid content found in the conversion result
• Document conversion failed: {e}
• Document conversion failed with status: {conversion_result.s
• Document extraction unsuccessful for '{file_name}': unmapped
• Invalid system item in list (dict): {item}

AI-assisted analysis of zylon-ai/private-gpt@4a030776a3 (2026-08-15). Data as JSON: /api/errors/b0786241a08fe4a7. Report an issue: GitHub.