h5rdmtoolbox 2.1.0__tar.gz → 2.2.0__tar.gz

{h5rdmtoolbox-2.1.0 → h5rdmtoolbox-2.2.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: h5rdmtoolbox
-Version: 2.1.0
+Version: 2.2.0
 Summary: Supporting a FAIR Research Data lifecycle using Python and HDF5.
 Home-page: https://h5rdmtoolbox.readthedocs.io/en/latest/
 Author: Matthias Probst
@@ -17,18 +17,18 @@ Requires-Python: <3.13,>=3.9
 Description-Content-Type: text/markdown
 License-File: LICENSE
 Requires-Dist: appdirs>=1.4.4
-Requires-Dist: importlib_resources
+Requires-Dist: importlib_resources<7.0.0,>=6.5.2
 Requires-Dist: numpy<2.0.0,>=1.20
-Requires-Dist: h5py>3.7.0
+Requires-Dist: h5py<4.0.0,>=3.8.0
 Requires-Dist: matplotlib>=3.5.2
 Requires-Dist: IPython>=7.34.0
-Requires-Dist: pyyaml>6.0
+Requires-Dist: pyyaml>=6.0
 Requires-Dist: xarray>=2022.3.0
 Requires-Dist: pint==0.21.1
-Requires-Dist: pint_xarray>=0.2.1
+Requires-Dist: pint_xarray<=0.4.0,>=0.2.1
 Requires-Dist: regex>=2020.7.9
 Requires-Dist: packaging>=24.1
-Requires-Dist: ontolutils>=0.13.3
+Requires-Dist: ontolutils<=1.0.0,>=0.13.3
 Requires-Dist: python-forge==18.6.0
 Requires-Dist: requests>=2.32.3
 Requires-Dist: pydantic>=2.8.2
@@ -111,41 +111,59 @@ Requires-Dist: tabulate<=0.9.0,>=0.8.10; extra == "complete"
 Requires-Dist: python-gitlab; extra == "complete"
 Requires-Dist: pypandoc>=1.11; extra == "complete"
 Requires-Dist: pymongo<=4.10.1,>=4.2.0; extra == "complete"
-Requires-Dist: pandas>=1.4.3; extra == "complete"
-Requires-Dist: xmltodict<=0.13.0; extra == "complete"
-Requires-Dist: tabulate<=0.9.0,>=0.8.10; extra == "complete"
-Requires-Dist: python-gitlab; extra == "complete"
-Requires-Dist: pypandoc>=1.11; extra == "complete"
-Requires-Dist: pymongo<=4.10.1,>=4.2.0; extra == "complete"
-Requires-Dist: pytest>=8.3.3; extra == "complete"
-Requires-Dist: pytest-cov>=5.0.0; extra == "complete"
-Requires-Dist: pylint; extra == "complete"
-Requires-Dist: scikit-image; extra == "complete"
-Requires-Dist: ssnolib>=1.5.1.2; extra == "complete"
-Requires-Dist: mongomock==4.1.2; extra == "complete"
-Requires-Dist: xmltodict<=0.13.0; extra == "complete"
-Requires-Dist: scipy>=1.10.1; extra == "complete"
-Requires-Dist: scikit-image>=0.21.0; extra == "complete"
-Requires-Dist: pandas>=1.4.3; extra == "complete"
-Requires-Dist: xmltodict<=0.13.0; extra == "complete"
-Requires-Dist: tabulate<=0.9.0,>=0.8.10; extra == "complete"
-Requires-Dist: python-gitlab; extra == "complete"
-Requires-Dist: pypandoc>=1.11; extra == "complete"
-Requires-Dist: pymongo<=4.10.1,>=4.2.0; extra == "complete"
-Requires-Dist: Sphinx==8.0.0; extra == "complete"
-Requires-Dist: sphinx_book_theme==1.1.3; extra == "complete"
-Requires-Dist: sphinx-copybutton==0.5.2; extra == "complete"
-Requires-Dist: sphinx-design==0.6.1; extra == "complete"
-Requires-Dist: myst-nb==1.2.0; extra == "complete"
-Requires-Dist: sphinxcontrib-bibtex==2.6.3; extra == "complete"
-Requires-Dist: scikit-image>=0.21.0; extra == "complete"
-Requires-Dist: scikit-learn; extra == "complete"
+Provides-Extra: complete-with-docs
+Requires-Dist: tabulate<=0.9.0,>=0.8.10; extra == "complete-with-docs"
+Requires-Dist: PyQt5==5.15.10; extra == "complete-with-docs"
+Requires-Dist: pytest>=8.3.3; extra == "complete-with-docs"
+Requires-Dist: pytest-cov>=5.0.0; extra == "complete-with-docs"
+Requires-Dist: pylint; extra == "complete-with-docs"
+Requires-Dist: scikit-image; extra == "complete-with-docs"
+Requires-Dist: ssnolib>=1.5.1.2; extra == "complete-with-docs"
+Requires-Dist: mongomock==4.1.2; extra == "complete-with-docs"
+Requires-Dist: xmltodict<=0.13.0; extra == "complete-with-docs"
+Requires-Dist: scipy>=1.10.1; extra == "complete-with-docs"
+Requires-Dist: scikit-image>=0.21.0; extra == "complete-with-docs"
+Requires-Dist: pandas>=1.4.3; extra == "complete-with-docs"
+Requires-Dist: xmltodict<=0.13.0; extra == "complete-with-docs"
+Requires-Dist: tabulate<=0.9.0,>=0.8.10; extra == "complete-with-docs"
+Requires-Dist: python-gitlab; extra == "complete-with-docs"
+Requires-Dist: pypandoc>=1.11; extra == "complete-with-docs"
+Requires-Dist: pymongo<=4.10.1,>=4.2.0; extra == "complete-with-docs"
+Requires-Dist: pandas>=1.4.3; extra == "complete-with-docs"
+Requires-Dist: xmltodict<=0.13.0; extra == "complete-with-docs"
+Requires-Dist: tabulate<=0.9.0,>=0.8.10; extra == "complete-with-docs"
+Requires-Dist: python-gitlab; extra == "complete-with-docs"
+Requires-Dist: pypandoc>=1.11; extra == "complete-with-docs"
+Requires-Dist: pymongo<=4.10.1,>=4.2.0; extra == "complete-with-docs"
+Requires-Dist: pytest>=8.3.3; extra == "complete-with-docs"
+Requires-Dist: pytest-cov>=5.0.0; extra == "complete-with-docs"
+Requires-Dist: pylint; extra == "complete-with-docs"
+Requires-Dist: scikit-image; extra == "complete-with-docs"
+Requires-Dist: ssnolib>=1.5.1.2; extra == "complete-with-docs"
+Requires-Dist: mongomock==4.1.2; extra == "complete-with-docs"
+Requires-Dist: xmltodict<=0.13.0; extra == "complete-with-docs"
+Requires-Dist: scipy>=1.10.1; extra == "complete-with-docs"
+Requires-Dist: scikit-image>=0.21.0; extra == "complete-with-docs"
+Requires-Dist: pandas>=1.4.3; extra == "complete-with-docs"
+Requires-Dist: xmltodict<=0.13.0; extra == "complete-with-docs"
+Requires-Dist: tabulate<=0.9.0,>=0.8.10; extra == "complete-with-docs"
+Requires-Dist: python-gitlab; extra == "complete-with-docs"
+Requires-Dist: pypandoc>=1.11; extra == "complete-with-docs"
+Requires-Dist: pymongo<=4.10.1,>=4.2.0; extra == "complete-with-docs"
+Requires-Dist: Sphinx==8.0.0; extra == "complete-with-docs"
+Requires-Dist: sphinx_book_theme==1.1.3; extra == "complete-with-docs"
+Requires-Dist: sphinx-copybutton==0.5.2; extra == "complete-with-docs"
+Requires-Dist: sphinx-design==0.6.1; extra == "complete-with-docs"
+Requires-Dist: myst-nb==1.2.0; extra == "complete-with-docs"
+Requires-Dist: sphinxcontrib-bibtex==2.6.3; extra == "complete-with-docs"
+Requires-Dist: scikit-image>=0.21.0; extra == "complete-with-docs"
+Requires-Dist: scikit-learn; extra == "complete-with-docs"
 Dynamic: license-file
 
 # HDF5 Research Data Management Toolbox
 
 ![Tests](https://github.com/matthiasprobst/h5RDMtoolbox/actions/workflows/tests.yml/badge.svg)
-![DOCS](https://codecov.io/gh/matthiasprobst/h5RDMtoolbox/branch/dev/graph/badge.svg)
+[![codecov](https://codecov.io/gh/matthiasprobst/h5RDMtoolbox/graph/badge.svg?token=IVG4AQEW47)](https://codecov.io/gh/matthiasprobst/h5RDMtoolbox)
 [![Documentation Status](https://readthedocs.org/projects/h5rdmtoolbox/badge/?version=latest)](https://h5rdmtoolbox.readthedocs.io/en/latest/?badge=latest)
 ![pyvers](https://img.shields.io/badge/python-3.9%20%7C%203.10%20%7C%203.11%20%7C%203.12-blue)
 

{h5rdmtoolbox-2.1.0 → h5rdmtoolbox-2.2.0}/README.md

@@ -1,7 +1,7 @@
 # HDF5 Research Data Management Toolbox
 
 ![Tests](https://github.com/matthiasprobst/h5RDMtoolbox/actions/workflows/tests.yml/badge.svg)
-![DOCS](https://codecov.io/gh/matthiasprobst/h5RDMtoolbox/branch/dev/graph/badge.svg)
+[![codecov](https://codecov.io/gh/matthiasprobst/h5RDMtoolbox/graph/badge.svg?token=IVG4AQEW47)](https://codecov.io/gh/matthiasprobst/h5RDMtoolbox)
 [![Documentation Status](https://readthedocs.org/projects/h5rdmtoolbox/badge/?version=latest)](https://h5rdmtoolbox.readthedocs.io/en/latest/?badge=latest)
 ![pyvers](https://img.shields.io/badge/python-3.9%20%7C%203.10%20%7C%203.11%20%7C%203.12-blue)
 

{h5rdmtoolbox-2.1.0 → h5rdmtoolbox-2.2.0}/h5rdmtoolbox/__init__.py

@@ -129,44 +129,53 @@ def get_ld(
         structural: bool = True,
         contextual: bool = True,
         skipND: int = 1,
-        blank_node_iri_base: Optional[str] = None,
-        **kwargs) -> rdflib.Graph:
+        file_uri: Optional[str] = None,
+        context: Optional[Dict] = None
+) -> rdflib.Graph:
     """Return the HDF file content as a rdflib.Graph object."""
     from h5rdmtoolbox.ld import get_ld
-    return get_ld(hdf_filename, structural=structural, contextual=contextual, skipND=skipND,
-                  blank_node_iri_base=blank_node_iri_base, **kwargs)
+    return get_ld(hdf_filename,
+                  structural=structural,
+                  contextual=contextual,
+                  skipND=skipND,
+                  file_uri=file_uri,
+                  context=context)
 
 
 def dump_jsonld(
         hdf_filename: Union[str, pathlib.Path],
         skipND: int = 1,
+        indent: int = 2,
         structural: bool = True,
         contextual: bool = True,
         context: Optional[Dict] = None,
-        blank_node_iri_base: Optional[str] = None,
-        **kwargs):
+        file_uri: Optional[str] = None
+):
     """Return the file content as a JSON-LD string."""
     from .ld import optimize_context
     context = context or {}
     graph = get_ld(hdf_filename,
                    structural=structural,
                    contextual=contextual,
-                   blank_node_iri_base=blank_node_iri_base,
-                   skipND=skipND,
-                   **kwargs)
+                   file_uri=file_uri,
+                   skipND=skipND)
     context = optimize_context(graph, context)
-    return graph.serialize(format="json-ld", indent=2, auto_compact=True,
-                           context=context)
+    return graph.serialize(
+        format="json-ld",
+        indent=indent,
+        auto_compact=True,
+        context=context
+    )
 
 
 def dump_jsonld_depr(hdf_filename: Union[str, pathlib.Path],
                      skipND: int = 1,
                      structural: bool = True,
-                     semantic: bool = True,
+                     contextual: bool = True,
                      resolve_keys: bool = True,
                      context: Optional[Dict] = None,
-                     blank_node_iri_base: Optional[str] = None,
-                     **kwargs) -> str:
+                     blank_node_iri_base: Optional[str] = None
+                     ) -> str:
     """Dump the JSON-LD representation of the file. With semantic=True and structural=False, the JSON-LD
     represents the semantic content only. To get a pure structural representation, set semantic=False, which
     will ignore any RDF content. If both are set to True, the JSON-LD will contain both structural and semantic.
@@ -198,13 +207,15 @@ def dump_jsonld_depr(hdf_filename: Union[str, pathlib.Path],
         if not isinstance(blank_node_iri_base, (str, HttpUrl)):
             raise ValueError('blank_node_iri_base must be a valid URL')
     from .wrapper import jsonld
-    if not structural and not semantic:
-        raise ValueError('At least one of structural or semantic must be True.')
+    if not structural and not contextual:
+        raise ValueError('At least one of structural or contextual must be True.')
 
     from h5rdmtoolbox.ld.hdf.file import get_serialized_ld
-    if structural and not semantic:
-        return get_serialized_ld(hdf_filename, blank_node_iri_base, format="json-ld", context=context, skipND=skipND)
-    return get_serialized_ld(hdf_filename, blank_node_iri_base, format="json-ld", context=context, skipND=skipND)
+    if structural and not contextual:
+        return get_serialized_ld(hdf_filename, blank_node_iri_base, format="json-ld", context=context,
+                                 contextual=contextual, skipND=skipND)
+    return get_serialized_ld(hdf_filename, blank_node_iri_base, format="json-ld", context=context,
+                             contextual=contextual, skipND=skipND)
     # with File(hdf_filename) as h5:
     #     return jsonld.dumps(
     #         h5,
@@ -221,20 +232,17 @@ def serialize(hdf_filename,
               fmt: str = "ttl",
               skipND: int = 1,
               structural: bool = True,
-              semantic: bool = True,
-              resolve_keys: bool = True,
-              blank_node_iri_base: Optional[str] = None,
+              contextual: bool = True,
+              file_uri: Optional[str] = None,
               **kwargs):
     """Alternative to json-ld but allows multiple serialization options"""
     fmt = kwargs.pop("format", fmt)
     with File(hdf_filename) as h5:
-        return h5.serialize(fmt,
-                            skipND,
+        return h5.serialize(fmt=fmt,
+                            skipND=skipND,
                             structural=structural,
-                            semantic=semantic,
-                            resolve_keys=resolve_keys,
-                            blank_node_iri_base=blank_node_iri_base,
-                            **kwargs)
+                            contextual=contextual,
+                            file_uri=file_uri)
 
 
 def build_pyvis_graph(hdf_filename, output_filename="kg-graph.html", notebook=False,

{h5rdmtoolbox-2.1.0 → h5rdmtoolbox-2.2.0}/h5rdmtoolbox/convention/ontology/hdf_datatypes.py

@@ -1,5 +1,6 @@
 import h5py
 import numpy as np
+import rdflib
 from ontolutils.namespacelib.hdf5 import HDF5
 
 # Mapping HDF5 numpy datatypes to ontology types
@@ -53,12 +54,11 @@ def _get_endianness(dtype):
     return 'LE'  # Default to little-endian if unknown
 
 
-def get_datatype(dataset: h5py.Dataset):
+def get_datatype(dataset: h5py.Dataset) -> rdflib.URIRef:
     """Maps an HDF5 dataset to its ontology-defined datatype."""
     dtype = dataset.dtype
     np_type = np.dtype(dtype).type  # Get numpy type (e.g., np.float32)
 
     endianness = _get_endianness(dtype)
 
-    ontology_type = HDF5_DATATYPE_MAP.get((np_type, endianness), None)
-    return ontology_type
+    return HDF5_DATATYPE_MAP.get((np_type, endianness), None)

h5rdmtoolbox-2.2.0/h5rdmtoolbox/ld/__init__.py

@@ -0,0 +1,76 @@
+import pathlib
+import warnings
+from typing import Union, Optional, Dict
+
+import h5py
+import rdflib
+
+from .hdf.file import get_ld as get_hdf_ld
+from .user.file import get_ld as get_contextual_ld
+from .utils import optimize_context
+
+
+def get_ld(
+        hdf_filename: Union[str, pathlib.Path],
+        structural: bool = True,
+        contextual: bool = True,
+        file_uri: Optional[str] = None,
+        skipND: Optional[int] = 1,
+        context: Optional[Dict] = None) -> rdflib.Graph:
+    """Return the HDF file content as a rdflib.Graph object."""
+
+    graph = None
+    with h5py.File(hdf_filename) as h5:
+        if contextual and structural:
+            graph1 = get_hdf_ld(h5, file_uri=file_uri, skipND=skipND)
+            graph2 = get_contextual_ld(h5, file_uri=file_uri)
+            graph = graph1 + graph2
+        else:
+            if structural:
+                graph = get_hdf_ld(hdf_filename, file_uri=file_uri, skipND=skipND)
+
+            if contextual:
+                graph = get_contextual_ld(hdf_filename, file_uri=file_uri)
+    if graph is None:
+        raise ValueError("structural and semantic cannot be both False.")
+    context = context or {}
+    for prefix, uri in context.items():
+        if not isinstance(uri, rdflib.URIRef):
+            graph.bind("ex", rdflib.URIRef(uri))
+    return graph
+
+
+def hdf2jsonld(
+        filename: Union[str, pathlib.Path],
+        metadata_filename: Optional[Union[str, pathlib.Path]] = None,
+        context: Optional[dict] = None,
+        structural: bool = True,
+        contextual: bool = True,
+        indent: int = 2,
+        file_uri: Optional[str] = None,
+        skipND: Optional[int] = 1,
+):
+    if metadata_filename is None:
+        metadata_filename = pathlib.Path(filename).with_suffix('.jsonld')  # recommended suffix for JSON-LD is .jsonld!
+    else:
+        metadata_filename = pathlib.Path(metadata_filename)
+
+    graph = get_ld(
+        hdf_filename=filename,
+        structural=structural,
+        contextual=contextual,
+        file_uri=file_uri,
+        skipND=skipND
+    )
+    context = context or {}
+    context = optimize_context(graph, context)
+
+    with open(metadata_filename, 'w', encoding='utf-8') as f:
+        f.write(
+            graph.serialize(format='json-ld',
+                            indent=indent,
+                            auto_compact=True,
+                            context=context)
+        )
+
+    return metadata_filename

h5rdmtoolbox-2.2.0/h5rdmtoolbox/ld/globals.py

@@ -0,0 +1,3 @@
+_CACHE = {
+    "FILE_MTIME_ID": {}
+}

{h5rdmtoolbox-2.1.0 → h5rdmtoolbox-2.2.0}/h5rdmtoolbox/ld/hdf/attributes.py

@@ -3,14 +3,20 @@ from ontolutils.namespacelib.hdf5 import HDF5
 from rdflib import Namespace, Literal
 from rdflib import XSD, RDF
 
+from ..utils import get_attr_node
+
 HDF = Namespace(str(HDF5))
 
 
-def process_attribute(name, value, graph, parent_uri):
+def process_attribute(*, name, value, graph, parent, parent_uri, blank_node_iri_base):
     """Process an HDF5 attribute, adding it to the RDF graph."""
     if name.isupper() or name.startswith('@'):
         return
-    attr_uri = rdflib.BNode()
+    attr_uri = get_attr_node(
+        parent,
+        name,
+        blank_node_iri_base
+    )
 
     if isinstance(value, str):
         graph.add((attr_uri, RDF.type, HDF5.StringAttribute))

{h5rdmtoolbox-2.1.0 → h5rdmtoolbox-2.2.0}/h5rdmtoolbox/ld/hdf/datasets.py

@@ -10,6 +10,7 @@ from rdflib import XSD, RDF
 
 from h5rdmtoolbox.convention.ontology.hdf_datatypes import get_datatype
 from h5rdmtoolbox.ld.hdf.attributes import process_attribute
+from h5rdmtoolbox.ld.utils import get_property_node
 
 HDF = Namespace(str(HDF5))
 HDF5_FILTER_ONTOLOGY = {
@@ -71,11 +72,11 @@ def process_dataset(
     else:
         graph.add((dataset_uri, HDF5.maximumSize, rdflib.Literal(-1, datatype=XSD.integer)))
 
-    datatype = get_datatype(dataset)
+    datatype_node = get_datatype(dataset)
 
-    if datatype:
-        graph.add((datatype, RDF.type, HDF5.Datatype))
-        graph.add((dataset_uri, HDF5.datatype, datatype))
+    if datatype_node:
+        graph.add((datatype_node, RDF.type, HDF5.Datatype))
+        graph.add((dataset_uri, HDF5.datatype, datatype_node))
 
     is_string_dataset = False
     if dataset.dtype.kind == 'S':
@@ -97,22 +98,36 @@ def process_dataset(
         graph.add((dataset_uri, HDF5.layout, HDF5.H5D_VIRTUAL))
 
     if dataset.chunks:
-        chunk_dimension_uri = rdflib.BNode()
+        chunk_dimension_uri = get_property_node(
+            dataset,
+            name='chunk_dimensions',
+            blank_node_iri_base=blank_node_iri_base
+        )
         graph.add((chunk_dimension_uri, RDF.type, HDF5.ChunkDimension))
         graph.add((dataset_uri, HDF5.chunk, chunk_dimension_uri))
 
         for ichunk, chunk in enumerate(dataset.chunks):
-            dimension_index_uri = rdflib.BNode()
+            dimension_index_uri = get_property_node(
+                dataset,
+                name=f'chunk_dimension_{ichunk}',
+                blank_node_iri_base=blank_node_iri_base
+            )
             graph.add((dimension_index_uri, RDF.type, HDF5.DataspaceDimension))
             graph.add((chunk_dimension_uri, HDF5.dimension, dimension_index_uri))
             graph.add((dimension_index_uri, HDF5.size, rdflib.Literal(chunk, datatype=XSD.integer)))
             graph.add((dimension_index_uri, HDF5.dimensionIndex, rdflib.Literal(ichunk, datatype=XSD.integer)))
 
+    dataspace_uri = get_property_node(
+        obj=dataset, name='dataspace', blank_node_iri_base=blank_node_iri_base
+    )
     if dataset.ndim > 0:
-        dataspace_uri = rdflib.BNode()
         graph.add((dataspace_uri, RDF.type, HDF5.SimpleDataspace))
         for idim, dim in enumerate(dataset.shape):
-            dataspace_dimension_node = rdflib.BNode()
+            dataspace_dimension_node = get_property_node(
+                obj=dataset,
+                name=f'dataspace_dimension_{idim}',
+                blank_node_iri_base=blank_node_iri_base
+            )
             graph.add((dataspace_dimension_node, RDF.type, HDF5.DataspaceDimension))
             graph.add((dataspace_uri, HDF5.dimension, dataspace_dimension_node))
             graph.add((dataspace_dimension_node, HDF5.size, rdflib.Literal(dim, datatype=XSD.integer)))
@@ -125,8 +140,7 @@ def process_dataset(
             else:
                 graph.add((dataset_uri, HDF5.value, rdflib.Literal(data)))
     else:
-        dataspace_uri = HDF5.scalarDataspace
-        graph.add((dataspace_uri, RDF.type, HDF5.scalarDataspace))
+        graph.add((dataspace_uri, RDF.type, HDF5.ScalarDataspace))
 
         if skipND and dataset.ndim < skipND:
             data = dataset[()]
@@ -139,4 +153,5 @@ def process_dataset(
 
     # Process attributes of the dataset
     for attr, value in dataset.attrs.items():
-        process_attribute(attr, value, graph, dataset_uri)
+        process_attribute(name=attr, value=value, graph=graph, parent=dataset, parent_uri=dataset_uri,
+                          blank_node_iri_base=blank_node_iri_base)

{h5rdmtoolbox-2.1.0 → h5rdmtoolbox-2.2.0}/h5rdmtoolbox/ld/hdf/file.py

@@ -14,29 +14,30 @@ from ..utils import optimize_context, get_obj_bnode, get_file_bnode
 HDF = Namespace(str(HDF5))
 
 
-def get_ld(source: Union[str, h5py.File], blank_node_iri_base: Optional[str] = None,
+def get_ld(source: Union[str, h5py.File],
+           file_uri: Optional[str] = None,
            skipND: int = 1) -> rdflib.Graph:
     """Convert an HDF5 file into an RDF graph."""
 
     if not isinstance(source, h5py.File):
         with h5py.File(source) as h5f:
-            return get_ld(h5f, blank_node_iri_base=blank_node_iri_base, skipND=skipND)
+            return get_ld(h5f, file_uri=file_uri, skipND=skipND)
 
     graph = Graph()
     graph.bind("hdf", HDF)
 
     file_frdf_manager = FileRDFManager(source.attrs)
     if file_frdf_manager.subject:
-        file_uri = rdflib.URIRef(file_frdf_manager.subject)
+        _file_uri = rdflib.URIRef(file_frdf_manager.subject)
     else:
-        file_uri = get_file_bnode(source, blank_node_iri_base=blank_node_iri_base)
+        _file_uri = get_file_bnode(source, file_uri=file_uri)
 
-    graph.add((file_uri, RDF.type, HDF.File))
+    graph.add((_file_uri, RDF.type, HDF.File))
 
-    root_group_uri = get_obj_bnode(source["/"], blank_node_iri_base=blank_node_iri_base)
-    graph.add((file_uri, HDF5.rootGroup, root_group_uri))
+    root_group_uri = get_obj_bnode(source["/"], blank_node_iri_base=file_uri)
+    graph.add((_file_uri, HDF5.rootGroup, root_group_uri))
 
-    process_group(source["/"], graph, file_uri, blank_node_iri_base=blank_node_iri_base, skipND=skipND)
+    process_group(source["/"], graph, parent_uri=_file_uri, blank_node_iri_base=file_uri, skipND=skipND)
 
     return graph
 

{h5rdmtoolbox-2.1.0 → h5rdmtoolbox-2.2.0}/h5rdmtoolbox/ld/hdf/groups.py

@@ -13,7 +13,13 @@ from h5rdmtoolbox.ld.utils import get_obj_bnode
 HDF = Namespace(str(HDF5))
 
 
-def process_group(group, graph, parent_uri, blank_node_iri_base: Optional[str] = None, skipND: int = 1):
+def process_group(
+        group,
+        graph,
+        parent_uri,
+        blank_node_iri_base: Optional[str] = None,
+        skipND: int = 1
+):
     """Recursively process HDF5 groups and datasets, adding them to the RDF graph."""
     group_uri = get_obj_bnode(group, blank_node_iri_base=blank_node_iri_base)
     graph.add((group_uri, RDF.type, HDF.Group))
@@ -25,10 +31,18 @@ def process_group(group, graph, parent_uri, blank_node_iri_base: Optional[str] =
         graph.add((group_uri, HDF.member, item_uri))
 
         if isinstance(item, h5py.Group):
-            process_group(item, graph, group_uri, blank_node_iri_base=blank_node_iri_base)
+            process_group(item, graph, parent_uri=group_uri, blank_node_iri_base=blank_node_iri_base)
         elif isinstance(item, h5py.Dataset):
-            process_dataset(item, graph, group_uri, item_uri, blank_node_iri_base=blank_node_iri_base, skipND=skipND)
+            process_dataset(
+                item,
+                graph,
+                parent_uri=group_uri,
+                dataset_uri=item_uri,
+                blank_node_iri_base=blank_node_iri_base,
+                skipND=skipND
+            )
 
     # Process attributes of the group
     for attr, value in group.attrs.items():
-        process_attribute(attr, value, graph, group_uri)
+        process_attribute(name=attr, value=value, graph=graph, parent=group, parent_uri=group_uri,
+                          blank_node_iri_base=blank_node_iri_base)

{h5rdmtoolbox-2.1.0 → h5rdmtoolbox-2.2.0}/h5rdmtoolbox/ld/user/attributes.py

@@ -1,5 +1,6 @@
 import json
 import warnings
+from typing import Union
 
 import numpy as np
 import rdflib
@@ -9,7 +10,7 @@ from h5rdmtoolbox.ld.utils import get_attr_dtype_as_XSD, get_obj_bnode
 from ..rdf import FileRDFManager, RDFManager
 
 
-def process_file_attribute(parent_obj, name, data, graph, blank_node_iri_base, file_uri):
+def process_file_attribute(parent_obj, name, data, graph, file_uri: Union[rdflib.URIRef, rdflib.BNode]):
     rdf_user_type = FileRDFManager(parent_obj.attrs).type
     if rdf_user_type is not None:
         if isinstance(rdf_user_type, (list, np.ndarray)):
@@ -48,18 +49,18 @@ def process_file_attribute(parent_obj, name, data, graph, blank_node_iri_base, f
                                 f"Error parsing JSON-LD object for attribute. name={name}, data={data}. Orig. Error: {e}")
 
             elif isinstance(rdf_user_object, dict):
-                    try:
-                        obj_graph = rdflib.Graph().parse(data=json.loads(json.dumps(rdf_user_object)), format="json-ld")
-                        # relate the obj_graph with the predicate:
-                        _subjects = set(obj_graph.subjects())
-                        if len(_subjects) != 1:
-                            warnings.warn(f"Error parsing JSON-LD object for attribute. name={name}, data={data}. "
-                                          f"Expected exactly one subject, found {len(_subjects)}")
-                        obj_graph.add((file_uri, rdflib.URIRef(rdf_user_predicate), list(_subjects)[0]))
-                        graph += obj_graph
-                    except Exception as e:
-                        warnings.warn(
-                            f"Error parsing JSON-LD object for attribute. name={name}, data={data}. Orig. Error: {e}")
+                try:
+                    obj_graph = rdflib.Graph().parse(data=json.loads(json.dumps(rdf_user_object)), format="json-ld")
+                    # relate the obj_graph with the predicate:
+                    _subjects = set(obj_graph.subjects())
+                    if len(_subjects) != 1:
+                        warnings.warn(f"Error parsing JSON-LD object for attribute. name={name}, data={data}. "
+                                      f"Expected exactly one subject, found {len(_subjects)}")
+                    obj_graph.add((file_uri, rdflib.URIRef(rdf_user_predicate), list(_subjects)[0]))
+                    graph += obj_graph
+                except Exception as e:
+                    warnings.warn(
+                        f"Error parsing JSON-LD object for attribute. name={name}, data={data}. Orig. Error: {e}")
         else:
             graph.add(
                 (file_uri,
@@ -70,7 +71,6 @@ def process_file_attribute(parent_obj, name, data, graph, blank_node_iri_base, f
 
 
 def process_attribute(parent_obj, name, data, graph, blank_node_iri_base):
-
     rdf_manager = RDFManager(parent_obj.attrs)
 
     parent_uri = rdf_manager.subject

{h5rdmtoolbox-2.1.0 → h5rdmtoolbox-2.2.0}/h5rdmtoolbox/ld/user/file.py

@@ -15,12 +15,14 @@ from ..rdf import FileRDFManager
 HDF = Namespace(str(HDF5))
 
 
-def get_ld(source: Union[str, h5py.File], blank_node_iri_base: Optional[str] = None) -> rdflib.Graph:
+def get_ld(
+        source: Union[str, h5py.File],
+        file_uri: Optional[str] = None) -> rdflib.Graph:
     """Convert an HDF5 file into an RDF graph."""
 
     if not isinstance(source, h5py.File):
         with h5py.File(source) as h5f:
-            return get_ld(h5f)
+            return get_ld(h5f, file_uri=file_uri)
 
     graph = Graph()
     graph.bind("m4i", rdflib.URIRef("http://w3id.org/nfdi4ing/metadata4ing#"))
@@ -30,27 +32,25 @@ def get_ld(source: Union[str, h5py.File], blank_node_iri_base: Optional[str] = N
     file_frdf_manager = FileRDFManager(source.attrs)
 
     if file_frdf_manager.subject:
-        file_uri = rdflib.URIRef(file_frdf_manager.subject)
+        file_node = rdflib.URIRef(file_frdf_manager.subject)
     else:
-        file_uri = get_file_bnode(source, blank_node_iri_base=blank_node_iri_base)
-
+        file_node = get_file_bnode(source, file_uri=file_uri)
 
     if source.name == "/":
         file_rdf = file_frdf_manager.type
     else:
         file_rdf = None
 
-
     if file_rdf:
         if isinstance(file_rdf, list):
             for rdf_type in file_rdf:
-                graph.add((file_uri, rdflib.RDF.type, rdflib.URIRef(rdf_type)))
+                graph.add((file_node, rdflib.RDF.type, rdflib.URIRef(rdf_type)))
         else:
-            graph.add((file_uri, rdflib.RDF.type, rdflib.URIRef(file_rdf)))
+            graph.add((file_node, rdflib.RDF.type, rdflib.URIRef(file_rdf)))
 
     for ak, av in source.attrs.items():
-        process_file_attribute(source, ak, av, graph, blank_node_iri_base, file_uri)
+        process_file_attribute(source, ak, av, graph, file_node)
 
-    process_group(source, graph, blank_node_iri_base=blank_node_iri_base)
+    process_group(source, graph, blank_node_iri_base=file_uri)
 
     return graph