deepset-ai/haystack · error · ValueError

The value of parameter <meta_value_type> must be 'float', 'i

Error message

The value of parameter <meta_value_type> must be 'float', 'int', 'date' or None but is currently set to '{meta_value_type}'.
Change the <meta_value_type> value to 'float', 'int', 'date' or None when initializing the MetaFieldRanker.

What it means

MetaFieldRanker sorts documents by a metadata field, and meta_value_type tells it how to parse the values (as float, int, or date). The library only accepts 'float', 'int', 'date' or None; anything else makes sorting semantics ambiguous, so _validate_params raises a ValueError at init or run time.

Source

Thrown at haystack/components/rankers/meta_field.py:155

        if sort_order not in ["ascending", "descending"]:
            raise ValueError(
                "The value of parameter <sort_order> must be 'ascending' or 'descending', "
                f"but is currently set to '{sort_order}'.\n"
                "Change the <sort_order> value to 'ascending' or 'descending' when initializing the "
                "MetaFieldRanker."
            )

        if missing_meta not in ["drop", "top", "bottom"]:
            raise ValueError(
                "The value of parameter <missing_meta> must be 'drop', 'top', or 'bottom', "
                f"but is currently set to '{missing_meta}'.\n"
                "Change the <missing_meta> value to 'drop', 'top', or 'bottom' when initializing the "
                "MetaFieldRanker."
            )

        if meta_value_type not in ["float", "int", "date", None]:
            raise ValueError(
                "The value of parameter <meta_value_type> must be 'float', 'int', 'date' or None but is "
                f"currently set to '{meta_value_type}'.\n"
                "Change the <meta_value_type> value to 'float', 'int', 'date' or None when initializing the "
                "MetaFieldRanker."
            )

    @component.output_types(documents=list[Document])
    def run(
        self,
        documents: list[Document],
        top_k: int | None = None,
        weight: float | None = None,
        ranking_mode: Literal["reciprocal_rank_fusion", "linear_score"] | None = None,
        sort_order: Literal["ascending", "descending"] | None = None,
        missing_meta: Literal["drop", "top", "bottom"] | None = None,
        meta_value_type: Literal["float", "int", "date"] | None = None,
    ) -> dict[str, Any]:
        """

View on GitHub (pinned to e318778c9b)

Solutions

1. Set meta_value_type to exactly one of 'float', 'int', 'date', or omit it (None)
2. Fix the string casing/spelling (Python string compare is case-sensitive)
3. If values are strings, remove meta_value_type or use 'date' with ISO-formatted date strings

Example fix

// before
ranker = MetaFieldRanker(meta_field="rating", meta_value_type="number")
// after
ranker = MetaFieldRanker(meta_field="rating", meta_value_type="float")

VALID = {"float", "int", "date", None}
if meta_value_type not in VALID:
    raise ValueError(f"meta_value_type must be one of 'float','int','date',None, got {meta_value_type!r}")

def is_valid_meta_value_type(t: object) -> bool:
    return t in ("float", "int", "date", None)

try:
    ranker = MetaFieldRanker(meta_field="rating", meta_value_type=t)
except ValueError as e:
    logger.error("Invalid meta_value_type: %s", e)
    ranker = MetaFieldRanker(meta_field="rating")  # default None

When it happens

Trigger: MetaFieldRanker(meta_value_type="str") or any misspelled/other value such as "number", "Float", "datetime" — validated in __init__ and again in run().

Common situations: Typo in the type string, copying config from another ranker, using a type name from another library (e.g. 'datetime' or 'number'), or programmatic pipeline YAML with a wrong value.

Understand the failure class

Background: "must be positive", "Invalid value": how libraries reject invalid parameter values (ValueError, ArgumentError, INVALID_PARAMETER_VALUE) — this error's family across 28 libraries.

Related errors

• Invalid value for word_count_threshold: {word_count_threshol
• top_k must be > 0, but got {top_k}
• top_k must be > 0, but got {top_k}
• Parameter <weight> must be in range [0,1] but is currently s
• The value of parameter <ranking_mode> must be 'reciprocal_ra

AI-assisted analysis of deepset-ai/haystack@e318778c9b (2026-08-30). Data as JSON: /api/errors/5a10b476d4539c0c. Report an issue: GitHub.