Generalize Validation

.github/workflows/generate_test_files.yml

@@ -1,6 +1,7 @@
 name: Generate test files
 on:
   workflow_dispatch:
+  # pull_request:
 
 jobs:
   gen-test-files:

CHANGELOG.md

@@ -1,8 +1,13 @@
 # PyNWB Changelog
 
-## PyNWB 2.2.0 (October 11, 2022)
+## PyNWB 2.2.0 (October 19, 2022)
 
-### Internal enhancements:
+### Enhancements and minor changes
+- Enhanced `pynwb.validate` API function to accept a list of file paths as well as the ability to operate on cached
+  namespaces. Also adjusted the validate CLI to directly use the API function. @CodyCBakerPhD
+  [#1511](https://github.com/NeurodataWithoutBorders/pynwb/pull/1511)
+
+### Internal enhancements
 - Moved CI to GitHub Actions. @rly [#1560](https://github.com/NeurodataWithoutBorders/pynwb/pull/1560),
   [#1566](https://github.com/NeurodataWithoutBorders/pynwb/pull/1566)
 

src/pynwb/__init__.py

@@ -11,15 +11,14 @@
 from hdmf.utils import docval, getargs, popargs, get_docval
 from hdmf.backends.io import HDMFIO
 from hdmf.backends.hdf5 import HDF5IO as _HDF5IO
-from hdmf.validate import ValidatorMap
 from hdmf.build import BuildManager, TypeMap
 import hdmf.common
 
-
 CORE_NAMESPACE = 'core'
 __core_ns_file_name = 'nwb.namespace.yaml'
 
 from .spec import NWBDatasetSpec, NWBGroupSpec, NWBNamespace  # noqa E402
+from .validate import validate  # noqa: F401, E402
 
 
 def __get_resources():
@@ -186,18 +185,6 @@ def get_sum(self, a, b):
     return __TYPE_MAP.get_dt_container_cls(neurodata_type, namespace)
 
 
-@docval({'name': 'io', 'type': HDMFIO, 'doc': 'the HDMFIO object to read from'},
-        {'name': 'namespace', 'type': str, 'doc': 'the namespace to validate against', 'default': CORE_NAMESPACE},
-        returns="errors in the file", rtype=list,
-        is_method=False)
-def validate(**kwargs):
-    """Validate an NWB file against a namespace"""
-    io, namespace = getargs('io', 'namespace', kwargs)
-    builder = io.read_builder()
-    validator = ValidatorMap(io.manager.namespace_catalog.get_namespace(name=namespace))
-    return validator.validate(builder)
-
-
 class NWBHDF5IO(_HDF5IO):
 
     @docval({'name': 'path', 'type': (str, Path), 'doc': 'the path to the HDF5 file', 'default': None},

src/pynwb/testing/make_test_files.py

@@ -1,9 +1,10 @@
+from datetime import datetime
 import numpy as np
 from pathlib import Path
-
-from datetime import datetime
-from pynwb import NWBFile, NWBHDF5IO, __version__, TimeSeries
+from pynwb import NWBFile, NWBHDF5IO, __version__, TimeSeries, get_class, load_namespaces
 from pynwb.image import ImageSeries
+from pynwb.spec import NWBNamespaceBuilder, export_spec, NWBGroupSpec, NWBAttributeSpec
+
 
 # pynwb 1.0.2 should be installed with hdmf 1.0.3
 # pynwb 1.0.3 should be installed with hdmf 1.0.5
@@ -149,12 +150,61 @@ def _make_imageseries_nonmatch_starting_frame():
     _write(test_name, nwbfile)
 
 
+def _make_empty_with_extension():
+    ns_builder = NWBNamespaceBuilder(
+        doc="An NWB test extension",
+        name="ndx-testextension",
+        version="0.1.0",
+        author="PyNWB Test File Generator",
+        contact="[email protected]",
+    )
+
+    ns_builder.include_type('TimeSeries', namespace='core')
+    tetrode_series = NWBGroupSpec(
+        neurodata_type_def='TimeSeriesWithID',
+        neurodata_type_inc='TimeSeries',
+        doc=('An extension of TimeSeries to include an ID.'),
+        attributes=[
+            NWBAttributeSpec(
+                name='id',
+                doc='The time series ID.',
+                dtype='int32'
+            ),
+        ],
+    )
+
+    new_data_types = [tetrode_series]
+
+    # export the spec to yaml files in the current directory
+    export_spec(ns_builder, new_data_types, output_dir=".")
+
+    nwbfile = NWBFile(session_description='ADDME',
+                      identifier='ADDME',
+                      session_start_time=datetime.now().astimezone())
+
+    load_namespaces("ndx-testextension.namespace.yaml")  # load from the current directory
+    TimeSeriesWithID = get_class("TimeSeriesWithID", "ndx-testextension")
+    ts = TimeSeriesWithID(
+        name="test_ts",
+        data=[1., 2., 3.],
+        description="ADDME",
+        unit="ADDME",
+        rate=1.,
+        id=1,
+    )
+    nwbfile.add_acquisition(ts)
+    test_name = 'nwbfile_with_extension'
+    _write(test_name, nwbfile)
+
+
 if __name__ == '__main__':
     # install these versions of PyNWB and run this script to generate new files
     # python src/pynwb/testing/make_test_files.py
     # files will be made in src/pynwb/testing/
     # files should be moved to tests/back_compat/
 
+    # NOTE: this script is run in the GitHub Actions workflow generate_test_files.yml
+
     if __version__ == '1.1.2':
         _make_empty()
         _make_str_experimenter()
@@ -170,3 +220,4 @@ def _make_imageseries_nonmatch_starting_frame():
         _make_imageseries_no_data()
         _make_imageseries_non_external_format()
         _make_imageseries_nonmatch_starting_frame()
+        _make_empty_with_extension()