def test_classification(self):
cases = [
('', False, False, False, False, True, True),
('#', False, False, False, False, True, True),
('#f', False, False, False, False, True, True),
('?', False, False, False, False, True, False),
('?q', False, False, False, False, True, False),
('p', False, False, False, False, True, False),
('/p', False, False, False, True, False, False),
('/p?', False, False, False, True, False, False),
('/p?q', False, False, False, True, False, False),
('/p#', False, False, False, True, False, False),
('/p#f', False, False, False, True, False, False),
('/p?q#f', False, False, False, True, False, False),
('//', False, False, True, False, False, False),
('//n?', False, False, True, False, False, False),
('//n?q', False, False, True, False, False, False),
('//n#', False, False, True, False, False, False),
('//n#f', False, False, True, False, False, False),
('//n?q#f', False, False, True, False, False, False),
('s:', True, True, False, False, False, False),
('s:p', True, True, False, False, False, False),
('s:p?', True, True, False, False, False, False),
('s:p?q', True, True, False, False, False, False),
('s:p#', True, False, False, False, False, False),
('s:p#f', True, False, False, False, False, False),
('s://', True, True, False, False, False, False),
('s://h', True, True, False, False, False, False),
('s://h/', True, True, False, False, False, False),
('s://h/p', True, True, False, False, False, False),
('s://h/p?', True, True, False, False, False, False),
('s://h/p?q', True, True, False, False, False, False),
('s://h/p#', True, False, False, False, False, False),
('s://h/p#f', True, False, False, False, False, False),
]
for s, uri, absuri, netpath, abspath, relpath, samedoc in cases:
for ref in [s, s.encode('ascii')]:
parts = uritools.urisplit(ref)
self.assertEqual(parts.isuri(), uri)
self.assertEqual(parts.isabsuri(), absuri)
self.assertEqual(parts.isnetpath(), netpath)
self.assertEqual(parts.isabspath(), abspath)
self.assertEqual(parts.isrelpath(), relpath)
self.assertEqual(parts.issamedoc(), samedoc)
self.assertEqual(uritools.isuri(ref), uri)
self.assertEqual(uritools.isabsuri(ref), absuri)
self.assertEqual(uritools.isnetpath(ref), netpath)
self.assertEqual(uritools.isabspath(ref), abspath)
self.assertEqual(uritools.isrelpath(ref), relpath)
self.assertEqual(uritools.issamedoc(ref), samedoc)
def _load_configuration(self, uri: str) -> None:
"""
Loads in the configuration from the specified file resource.
#see `adobe himl` package for more information:
https://github.com/adobe/himl
:param uri:
RFC-3986 Uniform Resource Identifier (URI)
"""
if not uritools.isuri(uri):
raise ValueError(URI_ERROR_MSG)
uri_tuple = uritools.urisplit(uri)
yaml_file = uri_tuple[2][1:]
try:
self.settings = dict(self.yaml.process(path=yaml_file, output_format="yaml", print_data=True))
except OSError:
self._log.critical(OS_ERROR_MSG, yaml_file, exc_info=True)
sys.exit()
def validate_type_URI(value):
""" Uses the distutils library to validate date time. Returns True/False """
return uritools.isuri(value)
def validate_biosimulations_metadata_for_uri(metadata,
validate_minimal_metadata=False,
archive=None,
working_dir=None):
""" Validate BioSimulations metadata for a file in a COMBINE/OMEX archive
Args:
metadata (:obj:`dict`): BioSimulations metadata
validate_minimal_metadata (:obj:`bool`, optional): whether to check that all required metadata attributes
are defined
archive (:obj:`CombineArchive`, optional): parent COMBINE archive
working_dir (:obj:`str`, optional): working directory (e.g., directory of the parent COMBINE/OMEX archive)
Returns:
:obj:`tuple`:
* nested :obj:`list` of :obj:`str`: nested list of errors with the metadata
* nested :obj:`list` of :obj:`str`: nested list of warnings with the metadata
"""
errors = []
warnings = []
# required attributes are present
if validate_minimal_metadata:
for predicate_type in BIOSIMULATIONS_PREDICATE_TYPES.values():
if predicate_type['required'] and (
(not predicate_type['multiple_allowed']
and metadata[predicate_type['attribute']] is None) or
(predicate_type['multiple_allowed']
and metadata[predicate_type['attribute']] == [])):
errors.append([
'Attribute `{}` ({}) is required.'.format(
predicate_type['attribute'], predicate_type['uri'])
])
# URIs are URLs
# Identifiers.org URLs point to valid identifiers
for predicate_type in BIOSIMULATIONS_PREDICATE_TYPES.values():
if predicate_type['has_uri'] and predicate_type['has_label']:
if predicate_type['multiple_allowed']:
objects = metadata[predicate_type['attribute']]
else:
objects = [metadata[predicate_type['attribute']]]
for object in objects:
if object and object['uri']:
if not uritools.isuri(object['uri']):
errors.append([
'URI `{}` of attribute `{}` ({}) is not a valid URI.'
.format(object['uri'], predicate_type['attribute'],
predicate_type['uri'])
])
else:
match = re.match(
r'^https?://identifiers\.org/(([^/:]+/[^/:]+|[^/:]+)[/:](.+))$',
object['uri'])
if match:
try:
validate_identifiers_org_uri(object['uri'])
except InvalidIdentifiersOrgUri as exception:
msg = (
'URI `{}` of attribute `{}` ({}) is not a valid Identifiers.org identifier.'
).format(object['uri'],
predicate_type['attribute'],
predicate_type['uri'])
errors.append([msg, [[str(exception)]]])
# thumbnail is a file; file type is checked by COMBINE validation
if working_dir:
for thumbnail in metadata['thumbnails']:
thumbnail = os.path.relpath(thumbnail, '.')
thumbnail_filename = os.path.join(working_dir, thumbnail)
if os.path.isfile(thumbnail_filename):
if archive:
for content in archive.contents:
if (content and content.location
and thumbnail == os.path.relpath(
content.location, '.')):
is_valid = False
for format in BIOSIMULATIONS_THUMBNAIL_FORMATS:
if content.format and re.match(
CombineArchiveContentFormatPattern[
format], content.format):
is_valid = True
break
if not is_valid:
errors.append([
'The format of thumbnail `{}` must be one of the following:'
.format(thumbnail),
sorted([[format] for format in
BIOSIMULATIONS_THUMBNAIL_FORMATS])
])
else:
errors.append(
['Thumbnail `{}` is not a file.'.format(thumbnail)])
else:
if metadata['thumbnails']:
warnings.append([
('The locations of the thumbnails could not be validated '
'because a working directory was not provided.')
])
# created is a date
if metadata['created'] is not None:
try:
dateutil.parser.parse(metadata['created'])
except dateutil.parser.ParserError:
errors.append([
'Created date `{}` is not a valid date.'.format(
metadata['created'])
])
# modified are dates
for date in metadata['modified']:
try:
dateutil.parser.parse(date)
except dateutil.parser.ParserError:
errors.append(
['Modified date `{}` is not a valid date.'.format(date)])
# return errors and warnings
return (errors, warnings)