def en_parsers_speed(self):
file_path = os.path.join(lexnlp_test_path, 'long_parsed_text.txt')
with codecs.open(file_path, 'r', encoding='utf-8') as fr:
text = fr.read()
ge_path = os.path.join(lexnlp_test_path, 'lexnlp/extract/en/tests/test_geoentities/')
entities_fn = ge_path + 'geoentities.csv'
aliases_fn = ge_path + 'geoaliases.csv'
geo_config = list(DictionaryEntry.load_entities_from_files(entities_fn, aliases_fn))
times = {} # type: Dict[str, float]
self.check_time(text, lambda s: list(get_amounts(s)), 'get_amounts', times)
self.check_time(text, lambda s: list(get_acts(s)), 'get_acts', times)
self.check_time(text, lambda s: list(get_citations(s)), 'get_citations', times)
self.check_time(text, lambda s: list(get_conditions(s)), 'get_conditions', times)
self.check_time(text, lambda s: list(get_constraints(s)), 'get_constraints', times)
self.check_time(text, lambda s: list(get_copyright(s)), 'get_copyright', times)
self.check_time(text, lambda s: list(_get_courts(s)), 'get_courts', times)
self.check_time(text, lambda s: list(get_cusip(s)), 'get_cusip', times)
self.check_time(text, lambda s: list(get_dates(s)), 'get_dates', times)
self.check_time(text, lambda s: list(get_definitions(s)), 'get_definitions', times)
self.check_time(text, lambda s: list(get_distances(s)), 'get_distances', times)
self.check_time(text, lambda s: list(get_durations(s)), 'get_durations', times)
self.check_time(text, lambda s: list(get_geoentities(s, geo_config)), 'get_geoentities', times)
self.check_time(text, lambda s: list(get_money(s)), 'get_money', times)
self.check_time(text, lambda s: list(get_percents(s)), 'get_percents', times)
self.check_time(text, lambda s: list(get_pii(s)), 'get_pii', times)
self.check_time(text, lambda s: list(get_ratios(s)), 'get_ratios', times)
self.check_time(text, lambda s: list(get_regulations(s)), 'get_regulations', times)
self.check_time(text, lambda s: list(get_trademarks(s)), 'get_trademarks', times)
self.check_time(text, lambda s: list(get_urls(s)), 'get_urls', times)
self.assertTrue('get_amounts' in times)
def test_percent(self):
text = "I'm 146.5 percent sure (or just 100%)"
ds = list(get_percents(text))
self.assertEqual(2, len(ds))
ants = list(get_percent_annotations(text))
self.assertEqual(2, len(ants))
self.assertEqual('en', ants[0].locale)
self.assertEqual('percent', ants[0].sign)
self.assertEqual(146.5, ants[0].amount)
self.assertEqual('%', ants[1].sign)
self.assertEqual(100.0, ants[1].amount)
def parse(self, text, text_unit_id, _text_unit_lang,
**kwargs) -> ParseResults:
found = list(percents.get_percents(text, return_sources=True))
if found:
unique = set(found)
return ParseResults({
PercentUsage: [
PercentUsage(text_unit_id=text_unit_id,
amount=item[1],
amount_str=item[3],
unit_type=item[0],
total=item[2],
count=found.count(item)) for item in unique
]
})
def _extract_variants_from_text(self, text: str, **kwargs):
percents = list(get_percents(text, float_digits=8))
if not percents:
return None
return [percent[2] * 100 for percent in percents]
def getPercentages(self):
mem = []
percentages = list(get_percents(self.bill_text))
for percent in percentages:
mem.append(str(percent[1]) + str(percent[0]))
self.bill.info['percentages'] = mem