Merge pull request #47 from TomNicholas/rename_open_dataset_via_kerchunk

Rename open_dataset_via_kerchunk to open_virtual_dataset
TomNicholas · Mar 25, 2024 · 13f40fc · 13f40fc
2 parents 4375ccb + 4b137fe
commit 13f40fc
Show file tree

Hide file tree

Showing 4 changed files with 9 additions and 8 deletions.
diff --git a/docs/api.md b/docs/api.md
@@ -16,6 +16,6 @@
 ## Xarray
 
 ```{eval-rst}
-.. autoclass:: virtualizarr.xarray.open_dataset_via_kerchunk
+.. autoclass:: virtualizarr.xarray.open_virtual_dataset
     :members:
 ```
diff --git a/virtualizarr/__init__.py b/virtualizarr/__init__.py
@@ -1,3 +1,3 @@
 from .manifests import ManifestArray  # type: ignore # noqa
 from .xarray import VirtualiZarrDatasetAccessor  # type: ignore # noqa
-from .xarray import open_dataset_via_kerchunk  # noqa: F401
+from .xarray import open_virtual_dataset  # noqa: F401
diff --git a/virtualizarr/kerchunk.py b/virtualizarr/kerchunk.py
@@ -37,23 +37,23 @@ def read_kerchunk_references_from_file(
     if filetype is None:
         filetype = _automatically_determine_filetype(filepath)
 
-    if filetype == "netCDF3":
+    if filetype.lower() == "netcdf3":
         from kerchunk.netCDF3 import SingleHdf5ToZarr
 
         refs = SingleHdf5ToZarr.NetCDF3ToZarr(filepath).translate()
-    elif filetype == "netCDF4":
+    elif filetype.lower() == "netcdf4":
         from kerchunk.hdf import SingleHdf5ToZarr
 
         refs = SingleHdf5ToZarr(filepath).translate()
     elif filetype == "grib":
         # TODO Grib files should be handled as a DataTree object
         # see https://github.com/TomNicholas/VirtualiZarr/issues/11
         raise NotImplementedError(f"Unsupported file type: {filetype}")
-    elif filetype == "tiff":
+    elif filetype.lower() == "tiff":
         from kerchunk.tiff import tiff_to_zarr
 
         refs = tiff_to_zarr(filepath)
-    elif filetype == "fits":
+    elif filetype.lower() == "fits":
         from kerchunk.fits import process_file
 
         refs = process_file(filepath)

diff --git a/virtualizarr/xarray.py b/virtualizarr/xarray.py
@@ -16,14 +16,14 @@ class ManifestBackendArray(ManifestArray, BackendArray):
     ...
 
 
-def open_dataset_via_kerchunk(
+def open_virtual_dataset(
     filepath: str,
     filetype: str,
     drop_variables: Optional[List[str]] = None,
     virtual_array_class=ManifestArray,
 ) -> xr.Dataset:
     """
-    Use kerchunk to open a single legacy file as an xarray Dataset wrapping virtualized zarr arrays.
+    Open a file or store as an xarray Dataset wrapping virtualized zarr arrays.
 
     It's important that we avoid creating any IndexVariables, as our virtualized zarr array objects don't actually contain a collection that can be turned into a pandas.Index.
 
@@ -33,6 +33,7 @@ def open_dataset_via_kerchunk(
         File path to open as a set of virtualized zarr arrays.
     filetype : str, default None
         Type of file to be opened. Used to determine which kerchunk file format backend to use.
+        Can be one of {'netCDF3', 'netCDF4'}.
         If not provided will attempt to automatically infer the correct filetype from the the filepath's extension.
     drop_variables: list[str], default is None
         Variables in the file to drop before returning.