Allow (local) pathlib input to HDF

kerchunk/hdf.py

@@ -1,6 +1,7 @@
 import base64
 import io
 import logging
+import pathlib
 from typing import Union, BinaryIO
 
 import fsspec.core
@@ -91,7 +92,7 @@ def __init__(
 
         # Open HDF5 file in read mode...
         lggr.debug(f"HDF5 file: {h5f}")
-        if isinstance(h5f, str):
+        if isinstance(h5f, (pathlib.Path, str)):
             fs, path = fsspec.core.url_to_fs(h5f, **(storage_options or {}))
             self.input_file = fs.open(path, "rb")
             url = h5f

kerchunk/tests/test_hdf.py

@@ -199,6 +199,21 @@ def test_string_embed():
     assert (z.vlen_str[1:] == "").all()
 
 
+def test_string_pathlib():
+    # essentially copied test above
+    import pathlib
+
+    fn = osp.join(here, "vlen.h5")
+    h = kerchunk.hdf.SingleHdf5ToZarr(pathlib.Path(fn), vlen_encode="embed")
+    out = h.translate()
+    fs = fsspec.filesystem("reference", fo=out)
+    assert txt in fs.references["vlen_str/0"]
+    z = zarr.open(fs.get_mapper())
+    assert z.vlen_str.dtype == "O"
+    assert z.vlen_str[0] == txt
+    assert (z.vlen_str[1:] == "").all()
+
+
 def test_string_null():
     fn = osp.join(here, "vlen.h5")
     h = kerchunk.hdf.SingleHdf5ToZarr(fn, fn, vlen_encode="null", inline_threshold=0)