Function _dataset_concat

xarray/structure/concat.py:605–863

Concatenate a sequence of datasets along a new or existing dimension

(
    datasets: Iterable[T_Dataset],
    dim: Hashable | T_Variable | T_DataArray | pd.Index,
    data_vars: T_DataVars | CombineKwargDefault,
    coords: ConcatOptions | Iterable[Hashable] | CombineKwargDefault,
    compat: CompatOptions | CombineKwargDefault,
    positions: Iterable[Iterable[int]] | None,
    fill_value: Any,
    join: JoinOptions | CombineKwargDefault,
    combine_attrs: CombineAttrsOptions,
    create_index_for_new_dim: bool,
    *,
    preexisting_dim: bool = False,
)

def _dataset_concat(
    datasets: Iterable[T_Dataset],
    dim: Hashable | T_Variable | T_DataArray | pd.Index,
    data_vars: T_DataVars | CombineKwargDefault,
    coords: ConcatOptions | Iterable[Hashable] | CombineKwargDefault,
    compat: CompatOptions | CombineKwargDefault,
    positions: Iterable[Iterable[int]] | None,
    fill_value: Any,
    join: JoinOptions | CombineKwargDefault,
    combine_attrs: CombineAttrsOptions,
    create_index_for_new_dim: bool,
    *,
    preexisting_dim: bool = False,
) -> T_Dataset:
    """
    Concatenate a sequence of datasets along a new or existing dimension
    """
    from xarray.core.dataarray import DataArray
    from xarray.core.dataset import Dataset

    datasets = list(datasets)

    if not all(isinstance(dataset, Dataset) for dataset in datasets):
        raise TypeError(
            "The elements in the input list need to be either all 'Dataset's or all 'DataArray's"
        )

    dim_var: Variable | None
    if isinstance(dim, DataArray):
        dim_var = dim.variable
    elif isinstance(dim, Variable):
        dim_var = dim
    else:
        dim_var = None

    dim_name, index = _calc_concat_dim_index(dim)

    # Make sure we're working on a copy (we'll be loading variables)
    datasets = [ds.copy() for ds in datasets]
    datasets = list(
        align(
            *datasets, join=join, copy=False, exclude=[dim_name], fill_value=fill_value
        )
    )

    all_dims, dim_coords, dims_sizes, coord_names, data_names, vars_order = (
        _parse_datasets(datasets)
    )
    if preexisting_dim:
        # When concatenating DataTree objects, a dimension may be pre-existing
        # because it exists elsewhere on the trees, even if it does not exist
        # on the dataset objects at this node.
        all_dims.add(dim_name)
    indexed_dim_names = set(dim_coords)

    both_data_and_coords = coord_names & data_names
    if both_data_and_coords:
        raise ValueError(