classify_columns

Public callable

Classify multiple columns and return normalized governance suggestions.

Parameters:

Name	Type	Description	Default
profile	Any	Value used by this callable.	required
metadata	Any	Value used by this callable.	None
business_context	Any	Value used by this callable.	None
rules	Any	Value used by this callable.	None
dataset_name	Any	Value used by this callable.	None
table_name	Any	Value used by this callable.	None
run_id	Any	Value used by this callable.	None

Returns:

Type	Description
list[dict]	Structured output produced by this callable.

Source code in src/fabricops_kit/governance.py

def classify_columns(profile: dict | list[dict], metadata: dict | list[dict] | None = None, business_context: str | dict | None = None, rules: list[dict] | None = None, dataset_name: str | None = None, table_name: str | None = None, run_id: str | None = None) -> list[dict]:
    """Classify multiple columns and return normalized governance suggestions.

        Parameters
        ----------
        profile : Any
            Value used by this callable.
        metadata : Any
            Value used by this callable.
        business_context : Any
            Value used by this callable.
        rules : Any
            Value used by this callable.
        dataset_name : Any
            Value used by this callable.
        table_name : Any
            Value used by this callable.
        run_id : Any
            Value used by this callable.

        Returns
        -------
        list[dict]
            Structured output produced by this callable.
    """
    del dataset_name, table_name, run_id
    columns = _normalize_columns(profile)
    meta_lookup: dict[str, dict] = {}
    if isinstance(metadata, dict):
        if all(isinstance(v, dict) for v in metadata.values()):
            meta_lookup = {str(k): v for k, v in metadata.items()}
        else:
            meta_lookup = {str(metadata.get("column_name") or ""): metadata}
    elif isinstance(metadata, list):
        meta_lookup = {str(_column_name(m)): m for m in metadata}

    out = []
    for col in columns:
        name = _column_name(col)
        if not name:
            continue
        out.append(classify_column(name, data_type=str(col.get("data_type") or col.get("dtype") or "") or None, profile=col, metadata=meta_lookup.get(name, {}), business_context=business_context, rules=rules))
    return out