def test_from_iso_date_string():
str_value = "2017-08-23"
datum = Datum()
datum.from_iso_date_string(str_value)
assert datum.value.day == 23
assert datum.value.month == 8
assert datum.value.year == 2017
def date_finder(text):
date =""
date_pattern = '%{YEAR:year}-%{MONTHNUM:month}-%{MONTHDAY:day}'
matches = list(datefinder.find_dates(s))
match_date = re.search('\d{4}-\d{2}-\d{2}', s)
try:
print "====using dateutil"
for i in s.splitlines():
d = parser.parse(i)
print(d.strftime("%Y-%m-%d"))
except Exception as e:
print e
try:
print "====pygrok==="
grok = Grok(date_pattern)
print(grok.match(s))
except Exception as e:
print e
try:
print "====using date==="
if len(matches) > 0:
date = matches[0]
print date
else:
print 'No dates found'
except Exception as e:
print e
try:
print "====using date==="
date = datetime.datetime.strptime(match_date.group(), '%Y-%m-%d').date()
print date
except Exception as e:
print e
try:
print "====using Chunkgrams==="
chunkGram = r"""NE:{<NNP>+<CD>}"""
chunkParser = nltk.RegexpParser(chunkGram)
sentences = nltk.sent_tokenize(text)
tokenized_sentences = [nltk.word_tokenize(sentence.strip()) for sentence in sentences]
tagged_sentences = [nltk.pos_tag(i) for i in tokenized_sentences]
chunked_sentences = [chunkParser.parse(i) for i in tagged_sentences]
entity_names = []
for tree in chunked_sentences:
entity_names.extend(extract_entity_names(tree))
print entity_names
except Exception as e:
print e
try:
print "===using pydatum=="
datum = Datum()
print (datum.from_iso_date_string(text))
except Exception as e:
print e
try:
print "===using dateparser=="
date = search_dates(text.decode('ascii','ignore'))
print date
except Exception as e:
print e
