def test_installs_en_core_web_sm_if_not_found(self):
""" Due to questionable PyPI security policies, check en_core_web_sm installation is fired if not present """
nlp = spacy.load('en_core_web_sm')
with mock.patch('PatternOmatic.api.pkg_resources.working_set'
) as patch_working_set:
with mock.patch('PatternOmatic.api.spacy_download'
) as patch_spacy_download:
with mock.patch(
'PatternOmatic.api.spacy_load') as patch_spacy_load:
patch_working_set.return_value = []
patch_spacy_download.return_value = 'I\'ve been fired'
patch_spacy_load.return_value = nlp
find_patterns(['Hi'])
super().assertTrue(patch_spacy_download.called)
def test_find_patterns_when_valid_configuration_file_provided(self):
""" Checks that providing a valid configuration file path loads configuration from that file """
config_file_path = \
os.path.join(os.path.join(os.path.dirname(os.path.abspath(__file__)), os.pardir), 'config.ini')
_ = find_patterns(self.my_samples, configuration=config_file_path)
super().assertEqual(config_file_path, Config().file_path)
def test_find_patterns_when_bad_language_provided(self):
""" Checks that providing an imaginary language model makes find_patterns use en_core_web_sm """
with super().assertLogs(LOG) as cm:
bad_model = 'Something'
_ = find_patterns(self.my_samples,
spacy_language_model_name=bad_model)
super().assertEqual(
f'WARNING:PatternOmatic:Model {bad_model} not found, falling back to '
f'patternOmatic\'s default language model: en_core_web_sm',
cm.output[1])
def main(args: List) -> None:
"""
PatternOmatic's script main function wrapper
Args:
args: Command Line Input Arguments
Returns: None
"""
LOG.info('Parsing command line arguments...')
try:
cli = ArgumentParser(
description=
'Finds the Spacy\'s Matcher pattern for the given samples',
epilog='...using actual Artificial Intelligence')
# Samples
cli.add_argument('-s',
'--sample',
action='append',
required=True,
nargs='+',
type=str,
help='A sample phrase')
# Spacy Language Model
cli.add_argument('-l',
'--language',
nargs='?',
type=str,
default='en_core_web_sm',
help='Spacy language model to be used')
# Configuration file to be used
cli.add_argument(
'-c',
'--config',
nargs='?',
type=str,
help='Configuration file path to be used',
default=None,
)
# Parse command line input arguments/options
parsed_args = cli.parse_args(args)
# Join sample arguments
for index, item in enumerate(parsed_args.sample):
parsed_args.sample[index] = ' '.join(item)
#
# Find patterns
#
patterns_found, _ = find_patterns(
parsed_args.sample,
configuration=parsed_args.config,
spacy_language_model_name=parsed_args.language)
LOG.info(f'Patterns found: {patterns_found}')
except Exception as ex:
LOG.critical(f'Fatal error: {repr(ex)}')
raise ex
def test_find_patterns_when_config_instance_provided(self):
""" Checks when setting up a Config instance before find_patterns invocation works """
config = Config()
config.max_runs = 10
patterns, _ = find_patterns(self.my_samples)
super().assertEqual(10, len(patterns))
def test_find_patterns_when_only_samples_provided(self):
""" Tests that providing just samples makes the find_pattern keeps working """
patterns, _ = find_patterns(self.my_samples)
super().assertEqual(4, len(patterns))