Tencent/WeKnora · error

MinerU file_parse: %w

Error message

MinerU file_parse: %w

What it means

Public entry point MinerUConverter.Read wraps every failure from callFileParse (the synchronous /file_parse API call) with this message. Any transport, request-construction, or response-decoding failure from the MinerU file parsing endpoint surfaces as 'MinerU file_parse: <cause>'. Inspect the wrapped cause to identify the real problem.

Source

Thrown at internal/infrastructure/docparser/mineru_converter.go:85

	if err := validateMinerUOutboundURL(c.endpoint); err != nil {
		return &types.ReadResult{Error: err.Error()}, nil
	}
	if c.vlmServerURL != "" {
		if err := validateMinerUOutboundURL(c.vlmServerURL); err != nil {
			return &types.ReadResult{Error: err.Error()}, nil
		}
	}

	content := req.FileContent
	if len(content) == 0 {
		return &types.ReadResult{Error: "no file content provided"}, nil
	}

	logger.Infof(context.Background(), "[MinerU] Parsing file=%s size=%d via %s", req.FileName, len(content), c.endpoint)

	mdContent, imagesB64, err := c.callFileParse(ctx, content, req.FileName, req.FileType)
	if err != nil {
		return nil, fmt.Errorf("MinerU file_parse: %w", err)
	}

	// MinerU already returns markdown with embedded HTML blocks (e.g. <table>, <details>).
	// Re-running the whole document through html-to-markdown corrupts valid markdown
	// by escaping headings and image syntax. Only apply narrow compatibility fixes.
	mdContent = normalizeMinerUMarkdown(mdContent)

	// Process images: decode base64, build ImageRef list, replace refs in markdown
	imageRefs, mdContent := c.processImages(mdContent, imagesB64)

	mdContent, imageRefs = ensureOriginalImageRef(req, mdContent, imageRefs)

	logger.Infof(context.Background(), "[MinerU] Parsed successfully, markdown=%d chars, images=%d", len(mdContent), len(imageRefs))

	return &types.ReadResult{
		MarkdownContent: mdContent,
		ImageRefs:       imageRefs,
	}, nil

View on GitHub (pinned to 988cbb0330)

Solutions

1. Unwrap the error (errors.Unwrap / %v of the cause) to see which inner step failed.
2. Verify c.endpoint is correct and reachable (curl $ENDPOINT/file_parse/health if available).
3. Check MinerU server logs for the corresponding request to see the server-side failure.
4. Confirm the file content and FileType are valid for MinerU (supported formats: pdf, images, etc.).

doc, err := conv.Read(ctx, req)
if err != nil {
    var transient bool
    if strings.Contains(err.Error(), "HTTP request") || strings.Contains(err.Error(), "read zip body") {
        transient = true // safe to retry
    }
    log.Printf("mineru parse failed (retryable=%v): %v", transient, err)
    if transient {
        doc, err = conv.Read(ctx, req)
    }
}

When it happens

Trigger: Any error returned by callFileParse(ctx, content, req.FileName, req.FileType): bad multipart construction, HTTP failure, non-200 status, or unparseable response body.

Common situations: MinerU endpoint unreachable or misconfigured (wrong c.endpoint base URL); server returned 4xx/5xx (see companion 'MinerU API status %d' error); response JSON shape changed between MinerU versions.

Related errors

• create request: %w
• HTTP request: %w
• MinerU API status %d: %s
• read response body: %w
• credential verification failed: %w

AI-assisted analysis of Tencent/WeKnora@988cbb0330 (2026-09-02). Data as JSON: /api/errors/9568e13ca05ca979. Report an issue: GitHub.