pathwaycom/pathway · error · ValueError

vector column {vector._name!r} does not belong to the provid

Error message

vector column {vector._name!r} does not belong to the provided table. Pass a column reference from the same table, e.g. vector=table.{vector._name}.

What it means

Raised by pw.io.pinecone.write() when the column passed as vector= belongs to a different table than the table being written. Pathway column references carry their owning table, and the connector refuses to build a dataflow that mixes columns from unrelated tables. The message tells you the exact reference to re-pass, e.g. vector=table.embedding.

Source

Thrown at python/pathway/io/pinecone/__init__.py:346

    ...     table,
    ...     index_name="docs-sparse",
    ...     primary_key=table.doc_id,
    ...     vector=table.bm25,
    ...     metadata_columns=[table.title],
    ...     api_key="YOUR_API_KEY",
    ... )
    >>> pw.run(monitoring_level=pw.MonitoringLevel.NONE)  # doctest: +SKIP
    """
    _check_entitlements("pinecone")

    if primary_key is not None and primary_key._table is not table:
        raise ValueError(
            f"primary_key column {primary_key._name!r} does not belong to the "
            f"provided table. Pass a column reference from the same table, "
            f"e.g. primary_key=table.{primary_key._name}."
        )
    if vector._table is not table:
        raise ValueError(
            f"vector column {vector._name!r} does not belong to the provided "
            f"table. Pass a column reference from the same table, "
            f"e.g. vector=table.{vector._name}."
        )

    pk_name = primary_key._name if primary_key is not None else None
    vector_name = vector._name

    if pk_name == vector_name:
        raise ValueError(
            f"primary_key and vector both reference column {pk_name!r}; they must "
            "be different columns."
        )

    if metadata_columns is None:
        metadata_names = [
            col_name
            for col_name in table.column_names()

View on GitHub (pinned to fa2f74a464)

Solutions

Pass the column reference from the same table object you pass as the first argument: pw.io.pinecone.write(table, ..., vector=table.embedding).
If you transformed the table (select/with_columns/rename), pass the transformed table as the first argument and use its column reference.
Verify identity before the call: assert vector._table is table (useful in tests).

Example fix

# before
pw.io.pinecone.write(original, index_name="docs", vector=embeddings.embedding, api_key=k)
# after
pw.io.pinecone.write(embeddings, index_name="docs", vector=embeddings.embedding, api_key=k)

Defensive patterns

Strategy: validation

Validate before calling

def assert_same_table(table, *cols):
    for c in cols:
        if c._table is not table:
            raise ValueError(f"column {c._name!r} is not from the given table")

assert_same_table(table, vector)  # before pw.io.pinecone.write(...)

Try / catch

try:
    pw.io.pinecone.write(table, index_name="docs", vector=table.emb, api_key=k)
except ValueError as e:
    if "does not belong to the provided table" in str(e):
        raise ValueError(f"wrong table reference: {e}") from e
    raise

Prevention

Keep one variable per table and always reference columns through it at the write call.
After adding select/with_columns steps, update the table passed to write() and all column references together.
Add a tiny unit test that asserts vector._table is table before wiring the connector.

When it happens

Trigger: Calling pw.io.pinecone.write(table, ..., vector=other_table.embedding) where other_table is not the same object as table; or passing a column from a table produced by a transform (e.g. table.select(...), table.with_columns(...)) while still passing the pre-transform table as the first argument.

Common situations: Common after refactoring a pipeline: a select()/with_columns() step is inserted between table creation and the write, and the old variable still names the original table. Also happens when writing multiple derived tables and copy-pasting the write() call.

Related errors

AI-assisted analysis of pathwaycom/pathway@fa2f74a464 (2026-08-15). Data as JSON: /api/errors/fb4acd9bcc5bf1e6. Report an issue: GitHub.