datawhalechina/hello-agents · error · ValueError

Selected table_id '{requested_table_id}' does not contain an

Error message

Selected table_id '{requested_table_id}' does not contain any numeric columns and cannot be analyzed.

What it means

Raised by the PDF document-ingestion pipeline when the caller explicitly selects a table via selected_table_id, but that table's extracted schema contains zero numeric columns. The pipeline can only run statistical analysis on numeric data, so it refuses up front with a ValueError instead of producing an empty analysis. It is distinct from the sibling error for a table_id that does not exist at all.

Source

Thrown at Co-creation-projects/healer-666-Academic-Data-Agent/src/data_analysis_agent/document_ingestion.py:355

    parsed_document_path = (data_dir / "parsed_document.json").resolve()

    full_text, records = _extract_pdf_payload(
        source_path,
        max_pdf_pages=max_pdf_pages,
        max_candidate_tables=max_candidate_tables,
        extracted_tables_dir=extracted_tables_dir,
    )
    background_literature_context = _extract_background_context(full_text)
    requested_table_id = str(selected_table_id or "").strip()
    requested_record = None
    if requested_table_id:
        requested_record = next((record for record in records if record.table_id == requested_table_id), None)
        if requested_record is None:
            raise ValueError(
                f"Selected table_id '{requested_table_id}' was not found in the extracted candidate tables."
            )
        if not requested_record.numeric_columns:
            raise ValueError(
                f"Selected table_id '{requested_table_id}' does not contain any numeric columns and cannot be analyzed."
            )
    primary_record = requested_record or _select_primary_table(records)
    warnings: list[str] = []

    if primary_record is None:
        summary = (
            "PDF 解析失败：未提取到满足主表路由规则的结构化表格。"
            "V1 暂不支持复杂多表路由或扫描件恢复，请手动裁剪 PDF 或改上传目标表格。"
        )
        parsed_payload = _serialize_parsed_document(
            source_pdf=source_path,
            background_literature_context=background_literature_context,
            full_text_excerpt=full_text[:2000],
            selected_table_id="",
            records=records,
        )
        parsed_document_path.parent.mkdir(parents=True, exist_ok=True)

View on GitHub (pinned to 606a07d341)

Solutions

Pick a different selected_table_id from the candidate records that do have numeric columns (inspect record.numeric_columns for each extracted record).
Omit selected_table_id and let _select_primary_table auto-route to a table with numeric data.
If the target table genuinely holds numbers, fix the source PDF or extraction so numeric cells are recognized (cleaner PDF export, remove currency/unit symbols from cells, or improve the numeric-coercion step).
Manually crop the PDF to just the target table, as the pipeline's own failure summary suggests for complex multi-table documents.

Example fix

# before
result = ingest_and_analyze(pdf_path, selected_table_id="table_03")  # table_03 is text-only

# after
numeric_ids = [r.table_id for r in records if r.numeric_columns]
result = ingest_and_analyze(pdf_path, selected_table_id=numeric_ids[0]) if numeric_ids else ingest_and_analyze(pdf_path)

Defensive patterns

Strategy: validation

Validate before calling

# Before passing selected_table_id
record = next((r for r in records if r.table_id == wanted_id), None)
if record is not None and not record.numeric_columns:
    raise SystemExit(f"table {wanted_id} has no numeric columns; pick one of "
                     f"{[r.table_id for r in records if r.numeric_columns]}")

Type guard

def is_analyzable(record) -> bool:
    return record is not None and bool(record.numeric_columns)

Try / catch

try:
    ingest(pdf, selected_table_id=table_id)
except ValueError as e:
    if "does not contain any numeric columns" in str(e):
        table_id = next(r.table_id for r in records if r.numeric_columns)  # fallback pick
        ingest(pdf, selected_table_id=table_id)
    else:
        raise

Prevention

Surface each candidate table's numeric_columns to the user before they choose a table_id.
Auto-filter candidate tables to those with at least one numeric column.
Add a UI hint that non-numeric tables cannot be analyzed.

When it happens

Trigger: Calling the ingestion/analysis function with selected_table_id pointing at an extracted table whose numeric_columns list is empty — e.g. a table of pure text labels, a header-only fragment, or a table where every numeric cell failed type coercion during extraction.

Common situations: PDFs with many tables where the user picks the wrong table_id; scanned or garbled PDFs where numbers are OCR'd as text; tables whose 'numeric' columns are formatted with currency symbols/units that defeat the numeric detector; multi-page PDFs where caption tables get extracted as candidates.

Related errors

AI-assisted analysis of datawhalechina/hello-agents@606a07d341 (2026-08-14). Data as JSON: /api/errors/ddd537f5914c69c1. Report an issue: GitHub.