Function _calc_concat_over

xarray/structure/concat.py:371–559

Determine which dataset variables need to be concatenated in the result,

(
    datasets: list[T_Dataset],
    dim: Hashable,
    all_dims: set[Hashable],
    data_vars: T_DataVars | Iterable[Hashable] | CombineKwargDefault,
    coords: ConcatOptions | Iterable[Hashable] | CombineKwargDefault,
    compat: CompatOptions | CombineKwargDefault,
)

def _calc_concat_over(
    datasets: list[T_Dataset],
    dim: Hashable,
    all_dims: set[Hashable],
    data_vars: T_DataVars | Iterable[Hashable] | CombineKwargDefault,
    coords: ConcatOptions | Iterable[Hashable] | CombineKwargDefault,
    compat: CompatOptions | CombineKwargDefault,
) -> tuple[set[Hashable], dict[Hashable, bool], list[int], set[Hashable]]:
    """
    Determine which dataset variables need to be concatenated in the result,
    """
    # variables to be concatenated
    concat_over = set()
    # variables checked for equality
    equals: dict[Hashable, bool] = {}
    # skip merging these variables.
    #   if concatenating over a dimension 'x' that is associated with an index over 2 variables,
    #   'x' and 'y', then we assert join="equals" on `y` and don't need to merge it.
    #   that assertion happens in the align step prior to this function being called
    skip_merge: set[Hashable] = set()

    if dim in all_dims:
        concat_over_existing_dim = True
        concat_over.add(dim)
    else:
        concat_over_existing_dim = False

    if data_vars == "minimal" and coords == "minimal" and not concat_over_existing_dim:
        raise ValueError(
            "Cannot specify both data_vars='minimal' and coords='minimal' when "
            "concatenating over a new dimension."
        )

    if data_vars is None or (
        isinstance(data_vars, CombineKwargDefault) and data_vars._value is None
    ):
        data_vars = "minimal" if concat_over_existing_dim else "all"

    concat_dim_lengths = []
    for ds in datasets:
        if concat_over_existing_dim and dim not in ds.dims and dim in ds:
            ds = ds.set_coords(dim)
        concat_over.update(k for k, v in ds.variables.items() if dim in v.dims)
        for _, idx_vars in ds.xindexes.group_by_index():
            if any(dim in v.dims for v in idx_vars.values()):
                skip_merge.update(idx_vars.keys())
        concat_dim_lengths.append(ds.sizes.get(dim, 1))

    def process_subset_opt(
        opt: ConcatOptions | Iterable[Hashable] | CombineKwargDefault,
        subset: Literal["coords", "data_vars"],
    ) -> None:
        original = set(concat_over)
        compat_str = (
            compat._value if isinstance(compat, CombineKwargDefault) else compat
        )
        assert compat_str is not None
        if isinstance(opt, str | CombineKwargDefault):