Merge branch 'dev-hotfix-version-issue' into dev

h5rdmtoolbox/convention/generate.py

@@ -374,7 +374,7 @@ def get_dict(self):
     def write_lines(self, file):
         """Write validator lines to file"""
         file.writelines(f'\n\nimport re\n\n')
-        file.writelines(f'\ndef {self.name}(value, parent=None, attrs=None):')
+        file.writelines(f'\ndef {self.name}(value, handler):')
         file.writelines(f"\n    pattern = re.compile(r'{self.re_pattern}')")
         file.writelines("\n    if not pattern.match(value):")
         file.writelines("\n        raise ValueError('Invalid format for pattern')")

h5rdmtoolbox/data/tutorial_convention.yaml

@@ -39,7 +39,7 @@ long_name:
   target_method: create_dataset
   validator: $regex(^[a-zA-Z].*$)
   alternative_standard_attribute: standard_name
-  description: An comprehensive description of the dataset. If not set, the standard_name attribute must be given.
+  description: A comprehensive description of the dataset. If not set, the standard_name attribute must be given.
 
 comment-file:
   target_method: __init__

h5rdmtoolbox/database/__init__.py

@@ -42,6 +42,16 @@ def rdf_find(source, *,
     ----------
     source: Union[str, pathlib.Path, h5tbx.Group]
         Filename or hdf group
+    rdf_subject: Optional[str]
+        The RDF subject to search for if provided
+    rdf_type: Optional[str]
+        The RDF type to search for if provided
+    rdf_predicate: Optional[str]
+        The RDF predicate to search for if provided
+    rdf_object: Optional[str]
+        The RDF object to search for if provided
+    recursive: bool=True
+        If True, search recursively. If False, only search the current group
     """
     if isinstance(source, (str, pathlib.Path)):
         return FileDB(source).rdf_find(rdf_subject=rdf_subject,

h5rdmtoolbox/database/hdfdb/filedb.py

@@ -18,7 +18,7 @@ def __init__(self, filename: Union[str, pathlib.Path]):
         self.rdf_find = self._instance_rdf_find
 
     @staticmethod
-    def find_one(filename: Union[str, pathlib.Path], *args, **kwargs) -> lazy.LHDFObject:
+    def find_one(filename: Union[str, pathlib.Path], *args, **kwargs) -> lazy.LazyObject:
         """Please refer to the docstring of the find_one method of the ObjDB class"""
         with h5py.File(str(filename), 'r') as h5:
             return ObjDB(h5).find_one(*args, **kwargs)
@@ -45,7 +45,7 @@ def _instance_find_one(self, *args, **kwargs):
             return ObjDB(h5).find_one(*args, **kwargs)
 
     @staticmethod
-    def find(file_or_filename, *args, **kwargs) -> Generator[lazy.LHDFObject, None, None]:
+    def find(file_or_filename, *args, **kwargs) -> List[lazy.LazyObject]:
         """Please refer to the docstring of the find method of the ObjDB class"""
         if isinstance(file_or_filename, (h5py.Group, h5py.Dataset)):
             return list(ObjDB(file_or_filename).find(*args, **kwargs))
@@ -104,7 +104,7 @@ def insert_filename(self, filename: Union[str, pathlib.Path]):
         self.filenames.append(pathlib.Path(filename))
         self.filenames = list(set(self.filenames))
 
-    def find_one(self, *args, **kwargs) -> lazy.LHDFObject:
+    def find_one(self, *args, **kwargs) -> lazy.LazyObject:
         """Call find_one on all the files registered. If more than one file
         contains the object, the first one is returned. If you want to find one per file,
         call find_one_per_file instead."""

h5rdmtoolbox/user.py

@@ -122,14 +122,15 @@ def clear_cache(self, delta_days: int, utime: bool = False):
             return
         if self.user_dirs['cache'].exists():
             for f in self.user_dirs['cache'].iterdir():
-                # get the file creation time
-                if utime:
-                    fct = f.stat().st_atime
-                else:
-                    fct = f.stat().st_ctime
-                dt = _now - fct
-                if dt > delta_days * 86400:
-                    f.unlink()
+                if f.is_file():
+                    # get the file creation time
+                    if utime:
+                        fct = f.stat().st_atime
+                    else:
+                        fct = f.stat().st_ctime
+                    dt = _now - fct
+                    if dt > delta_days * 86400:
+                        f.unlink()
 
     def reset(self):
         """Deletes all user data"""

requirements-dev.txt

@@ -1,8 +1,8 @@
 -e .
 -r requirements.txt
-pytest>=7.1.2
-pytest-cov
+pytest>=8.3.3
+pytest-cov>=5.0.0
 mongomock == 4.1.2
-scikit-image
+scikit-image>=0.21.0
 scikit-learn
-scipy
+scipy>=1.10.1

requirements-doc.txt

@@ -1,11 +1,11 @@
 -e .
 -r requirements.txt
-jupyterlab
-sphinx >= 4.5.0
-sphinx_book_theme
-sphinx-copybutton
-sphinx-design
-myst-nb
-sphinxcontrib-bibtex
-kglab
-scikit-image
+Sphinx==4.5.0
+sphinx_book_theme==0.3.3
+sphinx-copybutton==0.5.2
+sphinx-design==0.4.1
+myst-nb==0.17.2
+sphinxcontrib-bibtex==2.6.3
+
+scikit-image==0.24.0
+scikit-learn

requirements.txt

@@ -8,19 +8,19 @@ xarray>=2022.3.0
 pint==0.21.1
 pint_xarray>=0.2.1
 regex>=2020.7.9
-packaging
+packaging>=24.1
 python-forge==18.6.0
-requests
-rdflib<=7.0.0
+requests>=2.32.3
+rdflib>=7.0.0
 # database:
-pymongo>=4.2.0
+pymongo>=4.2.0,<4.9.0
 # csv
 pandas>=1.4.3
 # snt
 xmltodict<=0.13.0
 tabulate>=0.8.10,<=0.9.0
 python-gitlab
 pypandoc>=1.11
-pydantic >= 2.3.0, < 2.8.0
+pydantic>=2.8.2
 # other:
-ontolutils >= 0.2.23
+ontolutils >= 0.5.0

setup.cfg

@@ -31,19 +31,19 @@ install_requires =
     IPython >= 7.34.0  # colab wants 7.34.0
     pyyaml > 6.0  # 5.4 does not work with python 3.11!
     xarray >= 2022.3.0
-    pint == 0.21.1  # pint >= 0.19.2<0.22.0 Note: 0.22.0 shows bad str-output for quantities with format Lx~
+    pint==0.21.1  # pint >= 0.19.2<0.22.0 Note: 0.22.0 shows bad str-output for quantities with format Lx~
     pint_xarray >= 0.2.1
     regex>=2020.7.9
-    packaging
-    ontolutils>=0.2.20
-    python-forge == 18.6.0
-    requests
-    pydantic >= 2.3.0, < 2.8.0
-    rdflib<=7.0.0
+    packaging>=24.1
+    ontolutils>=0.5.0
+    python-forge==18.6.0
+    requests>=2.32.3
+    pydantic>=2.8.2
+    rdflib>=7.0.0
 
 [options.extras_require]
 database=
-    pymongo>=4.2.0
+    pymongo>=4.2.0,<4.9.0
 layout_validation=
     tabulate>=0.8.10,<=0.9.0
 csv =
@@ -56,32 +56,31 @@ snt =
 gui =
     PyQt5==5.15.10
 test =
-    pytest >= 7.1.2
-    pytest-cov
+    pytest>=8.3.3
+    pytest-cov>=5.0.0
     pylint
-    mongomock == 4.1.2
+    mongomock==4.1.2
     xmltodict<=0.13.0
-    scipy  # provides netcdf4
+    scipy>=1.10.1  # provides netcdf4
     %(csv)s
     %(snt)s
     %(database)s
 docs =
-;     kglab
     %(csv)s
     %(snt)s
     %(database)s
     %(test)s
-    jupyterlab
-    Sphinx >= 3,<5
-    sphinx_book_theme == 0.3.3
-    sphinx-copybutton
-    scikit-image
+;     jupyterlab>4.0.0
+    Sphinx==4.5.0
+    sphinx_book_theme==0.3.3
+    sphinx-copybutton==0.5.2
+    sphinx-design==0.4.1
+    myst-nb==0.17.2
+    sphinxcontrib-bibtex==2.6.3
+    scikit-image>=0.21.0
     scikit-learn
 ;    nbsphinx
-    sphinx-design
-    simplejson
-    myst-nb
-    sphinxcontrib-bibtex
+;     simplejson
 
 complete =
     %(layout_validation)s

tests/conventions/standard_attributes/test_standard_names.py

@@ -221,7 +221,7 @@ def test_to_html(self):
 
     def test_from_zenodo(self):
         if self.connected:
-            snt = StandardNameTable.from_zenodo(siurce=TutorialSNTZenodoRecordID)
+            snt = StandardNameTable.from_zenodo(source=TutorialSNTZenodoRecordID)
             self.assertIsInstance(snt, StandardNameTable)
             filename = h5tbx.UserDir['standard_name_tables'] / f'{TutorialSNTZenodoRecordID}.yaml'
             self.assertTrue(filename.exists())

tests/conventions/test_conventions.py

@@ -509,7 +509,7 @@ def test_from_zenodo(self):
             _ddir = h5tbx.UserDir['convention'] / 'h5rdmtoolbox_tutorial_convention'
             if _ddir.exists():
                 shutil.rmtree(_ddir)
-            h5tbx.convention.from_zenodo(doi_or_recid=TutorialConventionZenodoRecordID)
+            h5tbx.convention.from_zenodo(doi_or_recid=TutorialConventionZenodoRecordID, force_download=True)
             # h5tbx.convention.from_yaml('test_convention.yaml')
             h5tbx.use('h5rdmtoolbox-tutorial-convention')
 

tests/wrapper/test_file.py

@@ -81,6 +81,23 @@ def test_reopen_file(self):
             self.assertTrue('data_type' in h5.attrs.raw)
             self.assertTrue('contact' in h5.attrs.raw)
 
+    def test_grrrr(self):
+        from pydantic.functional_validators import WrapValidator
+        from typing_extensions import Annotated
+        from pydantic import BaseModel
+        import re
+        def regex_0(value, handler):
+            pattern = re.compile(r'^[a-zA-Z].*$')
+            if not pattern.match(value):
+                raise ValueError('Invalid format for pattern')
+            return value
+
+        regex_0 = Annotated[str, WrapValidator(regex_0)]
+
+        class long_name(BaseModel):
+            """A comprehensive description of the dataset. If not set, the standard_name attribute must be given."""
+            long_name: regex_0
+
     def test_dumps(self):
         with h5tbx.File() as h5:
             h5.dumps()