From b5c93227d1760399ebcd6e5ebb2fa923c22ed165 Mon Sep 17 00:00:00 2001
From: Oliver Ruebel <oruebel@lbl.gov>
Date: Thu, 11 Jan 2024 01:26:44 -0800
Subject: [PATCH] Add unit test for conversion of byte fillvalue to string

---
 src/hdmf_zarr/utils.py        |  8 ++++----
 tests/unit/test_zarrdataio.py | 18 ++++++++++++++++++
 2 files changed, 22 insertions(+), 4 deletions(-)

diff --git a/src/hdmf_zarr/utils.py b/src/hdmf_zarr/utils.py
index bde22296..19bf75b9 100644
--- a/src/hdmf_zarr/utils.py
+++ b/src/hdmf_zarr/utils.py
@@ -485,14 +485,14 @@ def from_h5py_dataset(h5dataset, **kwargs):
         :returns: ZarrDataIO object wrapping the dataset
         """
         filters = ZarrDataIO.hdf5_to_zarr_filters(h5dataset)
-        fillvalue = h5dataset.fillvalue if 'fillvalue' not in kwargs else kwargs.pop('fillvalue')
-        if isinstance(fillvalue, bytes): # bytes are not JSON serializable so use string instead
-            fillvalue = str(fillvalue)
+        fillval = h5dataset.fillvalue if 'fillvalue' not in kwargs else kwargs.pop('fillvalue')
+        if isinstance(fillval, bytes): # bytes are not JSON serializable so use string instead
+            fillval = fillval.decode("utf-8")
         chunks = h5dataset.chunks if 'chunks' not in kwargs else kwargs.pop('chunks')
         re = ZarrDataIO(
             data=h5dataset,
             filters=filters,
-            fillvalue=fillvalue,
+            fillvalue=fillval,
             chunks=chunks,
             **kwargs)
         return re
diff --git a/tests/unit/test_zarrdataio.py b/tests/unit/test_zarrdataio.py
index 0fc2f88c..bb5a1fe2 100644
--- a/tests/unit/test_zarrdataio.py
+++ b/tests/unit/test_zarrdataio.py
@@ -216,4 +216,22 @@ def test_from_h5py_dataset(self):
         self.assertIsInstance(re_zarrdataio.io_settings['filters'][0], numcodecs.Shuffle)
         self.assertIsInstance(re_zarrdataio.io_settings['filters'][1], numcodecs.Zlib)
         # Close the HDF5 file
+        h5file.close()
+
+    def test_from_h5py_datase_bytes_fillvaluet(self):
+        """
+        Test ZarrDataIO.from_h5py_dataset with a fillvalue that is in bytes, which needs to be handled
+        separately since bytes are not JSON serializable by default
+        """
+        h5file = h5py.File(self.hdf_filename, mode='a')
+        # print(np.arange(10, dtype=np.int8).tobytes())
+        h5dset = h5file.create_dataset(
+            name='test_str',
+            data=[b'hello', b'world', b'go'],
+            fillvalue=b'None')
+        re_zarrdataio = ZarrDataIO.from_h5py_dataset(h5dset)
+        # Test that all settings are being presevered when creating the ZarrDataIO object
+        self.assertIsInstance(re_zarrdataio, ZarrDataIO)
+        self.assertEqual(re_zarrdataio.io_settings['fill_value'], str("None"))
+        # Close the HDF5 file
         h5file.close()
\ No newline at end of file