Merge branch 'dev' into another_html_fix

.pre-commit-config.yaml

@@ -18,7 +18,7 @@ repos:
 #     hooks:
 #     -   id: black
 -   repo: https://github.com/astral-sh/ruff-pre-commit
-    rev: v0.7.1
+    rev: v0.7.3
     hooks:
     -   id: ruff
 # -   repo: https://github.com/econchick/interrogate

CHANGELOG.md

@@ -4,10 +4,12 @@
 
 ### Enhancements
 - Added support for expandable datasets of references for untyped and compound data types. @stephprince [#1188](https://github.com/hdmf-dev/hdmf/pull/1188)
-- Improved html representation of data in `Containers` @h-mayorquin [#1100](https://github.com/hdmf-dev/hdmf/pull/1100)
+- Improved html representation of data in `Container` objects. @h-mayorquin [#1100](https://github.com/hdmf-dev/hdmf/pull/1100)
+- Added error when using colon for `Container` name. A colon cannot be used as a group name when writing to Zarr on Windows. @stephprince [#1202](https://github.com/hdmf-dev/hdmf/pull/1202)
 
 ### Bug fixes
 - Fixed inaccurate error message when validating reference data types. @stephprince [#1199](https://github.com/hdmf-dev/hdmf/pull/1199)
+- Fixed incorrect dtype conversion of a StrDataset. @stephprince [#1205](https://github.com/hdmf-dev/hdmf/pull/1205)
 
 ## HDMF 3.14.5 (October 6, 2024)
 

src/hdmf/build/objectmapper.py

@@ -21,7 +21,7 @@
 from ..query import ReferenceResolver
 from ..spec import Spec, AttributeSpec, DatasetSpec, GroupSpec, LinkSpec, RefSpec
 from ..spec.spec import BaseStorageSpec
-from ..utils import docval, getargs, ExtenderMeta, get_docval, get_data_shape
+from ..utils import docval, getargs, ExtenderMeta, get_docval, get_data_shape, StrDataset
 
 _const_arg = '__constructor_arg'
 
@@ -212,7 +212,10 @@ def convert_dtype(cls, spec, value, spec_dtype=None):  # noqa: C901
         if (isinstance(value, np.ndarray) or
                 (hasattr(value, 'astype') and hasattr(value, 'dtype'))):
             if spec_dtype_type is _unicode:
-                ret = value.astype('U')
+                if isinstance(value, StrDataset):
+                    ret = value
+                else:
+                    ret = value.astype('U')
                 ret_dtype = "utf8"
             elif spec_dtype_type is _ascii:
                 ret = value.astype('S')

src/hdmf/container.py

@@ -303,8 +303,8 @@ def __new__(cls, *args, **kwargs):
     @docval({'name': 'name', 'type': str, 'doc': 'the name of this container'})
     def __init__(self, **kwargs):
         name = getargs('name', kwargs)
-        if '/' in name:
-            raise ValueError("name '" + name + "' cannot contain '/'")
+        if ('/' in name or ':' in name) and not self._in_construct_mode:
+            raise ValueError(f"name '{name}' cannot contain a '/' or ':'")
         self.__name = name
         self.__field_values = dict()
         self.__read_io = None

tests/unit/build_tests/test_convert_dtype.py

@@ -1,12 +1,17 @@
 from datetime import datetime, date
 
 import numpy as np
+import h5py
+import unittest
+
 from hdmf.backends.hdf5 import H5DataIO
 from hdmf.build import ObjectMapper
 from hdmf.data_utils import DataChunkIterator
 from hdmf.spec import DatasetSpec, RefSpec, DtypeSpec
 from hdmf.testing import TestCase
+from hdmf.utils import StrDataset
 
+H5PY_3 = h5py.__version__.startswith('3')
 
 class TestConvertDtype(TestCase):
 
@@ -321,6 +326,19 @@ def test_text_spec(self):
                 self.assertIs(ret, value)
                 self.assertEqual(ret_dtype, 'utf8')
 
+    @unittest.skipIf(not H5PY_3, "Use StrDataset only for h5py 3+")
+    def test_text_spec_str_dataset(self):
+        text_spec_types = ['text', 'utf', 'utf8', 'utf-8']
+        for spec_type in text_spec_types:
+            with self.subTest(spec_type=spec_type):
+                with h5py.File("test.h5", "w", driver="core", backing_store=False) as f:
+                    spec = DatasetSpec('an example dataset', spec_type, name='data')
+
+                    value = StrDataset(f.create_dataset('data', data=['a', 'b', 'c']), None)
+                    ret, ret_dtype = ObjectMapper.convert_dtype(spec, value)  # no conversion
+                    self.assertIs(ret, value)
+                    self.assertEqual(ret_dtype, 'utf8')
+
     def test_ascii_spec(self):
         ascii_spec_types = ['ascii', 'bytes']
         for spec_type in ascii_spec_types:

tests/unit/test_container.py

@@ -180,6 +180,17 @@ def test_set_parent_overwrite_proxy(self):
     def test_slash_restriction(self):
         self.assertRaises(ValueError, Container, 'bad/name')
 
+        # check no error raised in construct mode
+        child_obj = Container.__new__(Container, in_construct_mode=True)
+        child_obj.__init__('bad/name')
+
+    def test_colon_restriction(self):
+        self.assertRaises(ValueError, Container, 'bad:name')
+
+        # check no error raised in construct mode
+        child_obj = Container.__new__(Container, in_construct_mode=True)
+        child_obj.__init__('bad:name')
+
     def test_set_modified_parent(self):
         """Test that set modified properly sets parent modified
         """