Add NWB file convert unit test (#161)

* Add NWB file convert unit test
* Fix #160. Resolve bytes dtype on export

CHANGELOG.md

@@ -8,6 +8,7 @@
 
 ### Bug Fixes
 * Fixed bug when converting HDF5 datasets with unlimited dimensions @oruebel [#155](https://github.com/hdmf-dev/hdmf-zarr/pull/155)
+* Fixed bug resolving bytes dtype when exporting from Zarr to Zarr @oruebel [#161](https://github.com/hdmf-dev/hdmf-zarr/pull/161)
 * Adjust gallery tests to not fail on deprecation warnings from pandas. @rly [#157](https://github.com/hdmf-dev/hdmf-zarr/pull/157)
 
 ## 0.5.0 (December 8, 2023)

src/hdmf_zarr/backend.py

@@ -1137,6 +1137,7 @@ def write_dataset(self, **kwargs):  # noqa: C901
         "utf8": str,
         "utf-8": str,
         "ascii": bytes,
+        "bytes":  bytes,
         "str": str,
         "isodatetime": str,
         "string_": bytes,
@@ -1193,7 +1194,9 @@ def __resolve_dtype_helper__(cls, dtype):
     @classmethod
     def get_type(cls, data):
         if isinstance(data, str):
-            return str
+            return cls.__dtypes.get("str")
+        elif isinstance(data, bytes):
+            return cls.__dtypes.get("bytes")
         elif not hasattr(data, '__len__'):
             return type(data)
         else:
@@ -1230,8 +1233,8 @@ def __list_fill__(self, parent, name, data, options=None):  # noqa: C901
             # which Zarr does not allow for dataset shape. Check for the shape attribute first before falling
             # back on get_data_shape
             if hasattr(data, 'shape') and data.shape is not None:
-                data_shape = data.shape  
-            # This is a fall-back just in case. However this should not happen for standard numpy and h5py arrays 
+                data_shape = data.shape
+            # This is a fall-back just in case. However this should not happen for standard numpy and h5py arrays
             else: # pragma: no cover
                 data_shape = get_data_shape(data) # pragma: no cover
         # Deal with object dtype

tests/unit/test_io_convert.py

@@ -36,6 +36,8 @@
 import shutil
 import numpy as np
 import numcodecs
+from datetime import datetime
+from dateutil import tz
 from abc import ABCMeta, abstractmethod
 
 from hdmf_zarr.backend import (ZarrIO,
@@ -49,14 +51,19 @@
 from hdmf.common import DynamicTable
 from hdmf.common import CSRMatrix
 
-
 from tests.unit.utils import (Foo, FooBucket, FooFile, get_foo_buildmanager,
                               Baz, BazData, BazBucket, get_baz_buildmanager,
                               BazCpdData, get_temp_filepath)
 
 from zarr.storage import (DirectoryStore,
                           TempStore,
                           NestedDirectoryStore)
+try:
+    import pynwb
+    PYNWB_AVAILABLE = True
+except ImportError:
+    PYNWB_AVAILABLE = False
+
 
 
 class MixinTestCaseConvert(metaclass=ABCMeta):
@@ -403,6 +410,40 @@ def setUpContainer(self):
         else:
             raise NotImplementedError("FOO_TYPE %i not implemented in test" % self.FOO_TYPE)
 
+############################################
+# HDMF Common test container mixins
+###########################################
+class MixinTestNWBFile():
+    """
+    Mixin class used in conjunction with MixinTestCaseConvert to create conversion tests that
+    test export of a basic NWBFile. This class only defines the setUpContainer function for the test.
+    The roundtripExportContainer function required for the test needs to be defined separately
+    (e.g., by another mixin or the test class itself)
+    """
+    TABLE_TYPE = 0
+
+    def get_manager(self):
+        return pynwb.get_manager()
+
+    def setUpContainer(self):
+        if not PYNWB_AVAILABLE:
+            self.skipTest('Skip test. PyNWB is not installed')
+
+        subject = pynwb.file.Subject(
+            subject_id="001",
+            species="Mus musculus",
+            sex="M",
+            date_of_birth=datetime(2018, 4, 25, 2, 30, 3, tzinfo=tz.gettz("US/Pacific")),
+            age="P1D",
+            description=None,
+        )
+        nwbfile = pynwb.file.NWBFile(
+            session_description="Test File",
+            identifier="0000",
+            session_start_time=datetime(2019, 4, 25, 2, 30, 3, tzinfo=tz.gettz("US/Pacific")),
+            subject=subject,
+        )
+        return nwbfile
 
 ########################################
 # HDMF Baz test dataset of references
@@ -639,6 +680,44 @@ class TestHDF5toZarrFooCase2(MixinTestFoo,
     FOO_TYPE = MixinTestFoo.FOO_TYPES['link_data']
 
 
+class TestZarrToZarrNWBFile(MixinTestNWBFile,
+                            MixinTestZarrToZarr,
+                            MixinTestCaseConvert,
+                            TestCase):
+    """
+    Test the conversion of DynamicTable containers from Zarr to HDF5.
+    See MixinTestDynamicTableContainer.setUpContainer for the container spec.
+    """
+    IGNORE_NAME = True
+    IGNORE_HDMF_ATTRS = True
+    IGNORE_STRING_TO_BYTE = False
+
+
+class TestHDF5ToZarrNWBFile(MixinTestNWBFile,
+                            MixinTestHDF5ToZarr,
+                            MixinTestCaseConvert,
+                            TestCase):
+    """
+    Test the conversion of DynamicTable containers from Zarr to HDF5.
+    See MixinTestDynamicTableContainer.setUpContainer for the container spec.
+    """
+    IGNORE_NAME = True
+    IGNORE_HDMF_ATTRS = True
+    IGNORE_STRING_TO_BYTE = False
+
+class TestZarrToHDF5NWBFile(MixinTestNWBFile,
+                            MixinTestZarrToHDF5,
+                            MixinTestCaseConvert,
+                            TestCase):
+    """
+    Test the conversion of DynamicTable containers from Zarr to HDF5.
+    See MixinTestDynamicTableContainer.setUpContainer for the container spec.
+    """
+    IGNORE_NAME = True
+    IGNORE_HDMF_ATTRS = True
+    IGNORE_STRING_TO_BYTE = False
+
+
 ########################################
 # Test cases for dataset of references
 ########################################