def get_entities_from_sent(f1):
entities = {}
nlp = spacy.load('pt_core_news_sm')
doc = nlp(f1)
for ent in doc.ents:
list = []
if get(entities, str(ent.label_)) is not None:
list = get(entities, str(ent.label_))
list = push(list, str(ent))
set_(entities, str(ent.label_), list)
nlp = spacy.load('en_core_web_sm')
doc = nlp(f1)
for ent in doc.ents:
if str(ent.label_) == 'TIME' or str(ent.label_) == 'ORDINAL' or str(
ent.label_) == 'DATE' or str(ent.label_) == 'CARDINAL' or str(
ent.label_) == 'MONEY' or str(ent.label_) == 'QUANTITY':
list = []
if get(entities, str(ent.label_)) is not None:
list = get(entities, str(ent.label_))
list = push(list, str(ent))
set_(entities, str(ent.label_), list)
return entities
def get_word_entitie_from_sent(f1):
doc = nlp(f1)
#print(doc)
entities = []
for ent in doc.ents:
tuple = str(ent) + " " + str(ent.label_)
push(entities, tuple)
return entities
def get_syntactic_dependency_from_sent(f1):
doc = nlp(f1)
dependencies = []
for token in doc:
push(dependencies, (token.head.text, token.text, token.dep_))
return dependencies
def get_obj_pos_from_sent(f1):
doc = nlp(f1)
pos = {
"ADJ": [],
"ADP": [],
"ADV": [],
"AUX": [],
"CONJ": [],
"CCONJ": [],
"DET": [],
"INTJ": [],
"NOUN": [],
"NUM": [],
"PART": [],
"PRON": [],
"PROPN": [],
"PUNCT": [],
"SCONJ": [],
"SYM": [],
"VERB": [],
"X": []
}
for token in doc:
list = []
id = (str(token.pos_))
# if get(pos, id) is not None:
# list = get(pos, id)
list = push(list, (str(token.orth_)).lower())
set_(pos, id, list)
return pos
def process_adverb(adverbs, f1, f2):
f1 = f1.lower()
f2 = f2.lower()
doc = nlp(f1)
dependencies = []
len1 = 0
len2 = 0
adv1 = []
adv2 = []
for item in adverbs:
item = item.lower()
if item in f1:
len1 = len1 + 1
push(adv1, str(item))
if item in f2:
len2 = len2 + 1
push(adv2, str(item))
return [len1, len2, abs(len1 - len2), len_intersection(adv1, adv2)]
def chooseSong(song_data):
"""
chooses a song to tweet and updates recently choosen songs
"""
recent = song_data['recentSongIds']
songs = song_data['songs']
# filter out recently choosen songs and randomly choose a song
filtered_song_ids = pydash.filter_(songs.keys(), lambda x: x not in recent)
song_id = pydash.sample(filtered_song_ids)
# get chosen song and increment play count
song = songs[song_id]
song['playCount'] += 1
# pop least recently choosen song and push new one
if len(recent) == 7:
pydash.shift(recent)
pydash.push(recent, song_id)
return song
def get_word_pos_from_2_sents(f1, f2):
doc = nlp(f1)
dependencies = []
for token in doc:
if (token.pos_ == "PUNCT" or token.pos_ == "SPACE"):
continue
word_pos = deburr(token.text.lower()) + " " + token.pos_
push(dependencies, word_pos)
s1 = dependencies
dependencies = []
doc = nlp(f2)
for token in doc:
if (token.pos_ == "PUNCT" or token.pos_ == "SPACE"):
continue
#print(token.text, token.pos_)
word_pos = deburr(token.text.lower()) + " " + token.pos_
push(dependencies, word_pos)
s2 = dependencies
return s1, s2
def datetime_clusterer(tuplet: Tuple[List[Dict], int],
message: Dict) -> Tuple[List, int]:
message_anchored_time = round_to_nearest_interval_minutes(
message.get('timestamp'))
# No cluster: create a cluster
if not tuplet:
return [{'rate': 1, 'time': message_anchored_time}], 1
pulse_clusters, max_pulse_rate = tuplet
latest_pulse = pulse_clusters.pop()
# Message fits in cluster - cluster it up!
if latest_pulse.get('time') == message_anchored_time:
new_pulse_rate = latest_pulse.get('rate') + 1
new_pulse_clusters = _.push(
pulse_clusters, _.assign(latest_pulse,
{'rate': new_pulse_rate}))
if max_pulse_rate >= new_pulse_rate:
return new_pulse_clusters, max_pulse_rate
return new_pulse_clusters, new_pulse_rate
# Message doesn't fit in cluster
# lock in latest cluster, create new cluster but also fill
# in missing clusters in between
old_pulse_cluster = _.push(pulse_clusters, latest_pulse)
old_pulse_cluster = _.concat(
zero_pulses(start_time=latest_pulse.get('time'),
end_time=message_anchored_time,
interval=interval), old_pulse_cluster)
return _.push(old_pulse_cluster,
_.assign({}, {
'rate': 1,
'time': message_anchored_time
})), max_pulse_rate
def zero_pulses(start_time: datetime, end_time: datetime,
interval: int) -> List[Dict]:
minutes_difference = end_time - start_time
intervals_difference = \
((minutes_difference.seconds // 60) // interval) - 1
if intervals_difference > 0:
pulses = []
for x in range(0, intervals_difference):
pulses = _.push(
[], {
'rate': 0,
'time': start_time + timedelta(minutes=interval * (x + 1))
})
return pulses
return []
def test_push(case, expected):
assert _.push(*case) == expected
def remove_nones(pulses: List[Union[List[Dict], None]],
pulse: Union[List[Dict], None]) -> List[List[Dict]]:
if pulse is None:
return pulses
return _.push(pulses, pulse)
def get_sinonimos(palavra):
# print("-------------")
# print(palavra)
palavra = deburr(kebab_case(palavra))
base_url = "https://www.sinonimos.com.br/"
list_no_occur = get(nao_consta, "sinonimos")
if has(sinonimos, palavra):
#print("Palavra já consta no dicionário de sinônimos")
return get(sinonimos, palavra)
if palavra in list_no_occur:
#print("Não foram encontrados sinônimos para a palavra")
return []
else:
#print("\nNova Palavra: ", palavra)
request_fail = True
while request_fail:
try:
site = requests.get(base_url + palavra)
request_fail = False
except Exception as e:
print(e)
pass
data = BeautifulSoup(site.content, 'html.parser')
try:
h1 = data.find('h1').getText()
except Exception as e:
if palavra is not None:
list_no_occur = push(list_no_occur, palavra)
set_(nao_consta, "sinonimos", list_no_occur)
save_json("nao_consta", nao_consta)
return []
if (h1 == "Página Não Encontrada"):
#print(h1)
if palavra is not None:
list_no_occur = push(list_no_occur, palavra)
set_(nao_consta, "sinonimos", list_no_occur)
save_json("nao_consta", nao_consta)
return []
else:
content = data.find('div', attrs={'id': 'content'})
try:
div = content.findAll('div', attrs={'class': 's-wrapper'})
except Exception as e:
print(e)
if palavra is not None:
list_no_occur = push(list_no_occur, palavra)
set_(nao_consta, "sinonimos", list_no_occur)
save_json("nao_consta", nao_consta)
return []
aux = 0
for sentido in div:
aux = aux + 1
lista_sinonimos = []
try:
try:
key = lower_case(
sentido.find('div', attrs={
'class': 'sentido'
}).getText().strip(":"))
except Exception as e:
print(e)
key = "sinonimos" + str(aux)
pass
values = sentido.findAll('a',
attrs={'class': 'sinonimo'},
text=True)
values2 = sentido.findAll('span')
# print(values2)
all_values = union(values, values2)
#print(all_values)
for value in all_values:
lista_sinonimos.append(value.getText().strip(":"))
set_(sinonimos, palavra + "." + key, lista_sinonimos)
print("Sinônimo Salv@ no Dicionário")
except Exception as e:
print("\nError:\n" + str(e))
return []
save_json("sinonimos", sinonimos)
return get(sinonimos, palavra)
def get_antonimos(palavra):
# print("-------------")
# print(palavra)
palavra = deburr(kebab_case(palavra))
base_url = "https://www.antonimos.com.br/"
list_no_occur = get(nao_consta, "antonimos")
if has(antonimos, palavra):
#print("Palavra já consta no dicionário de antônimos")
return get(antonimos, palavra)
if palavra in list_no_occur:
#print("Não foram encontrados antônimos para a palavra")
return []
else:
request_fail = True
while request_fail:
try:
site = requests.get(base_url + palavra)
request_fail = False
except Exception as e:
print(e)
pass
data = BeautifulSoup(site.content, 'html.parser')
try:
h1 = data.find('h1').getText()
except Exception as e:
if palavra is not None:
list_no_occur = push(list_no_occur, palavra)
set_(nao_consta, "antonimos", list_no_occur)
save_json("nao_consta", nao_consta)
return []
if (h1 == "Página Não Encontrada"):
#print(h1)
if palavra is not None:
list_no_occur = push(list_no_occur, palavra)
set_(nao_consta, "antonimos", list_no_occur)
save_json("nao_consta", nao_consta)
return []
else:
content = data.find('div', attrs={'id': 'content'})
try:
div = content.findAll('div', attrs={'class': 's-wrapper'})
except Exception as e:
print(e)
if palavra is not None:
list_no_occur = push(list_no_occur, palavra)
set_(nao_consta, "antonimos", list_no_occur)
save_json("nao_consta", nao_consta)
aux = 0
for sentido in div:
aux = aux + 1
lista_antonimos = []
try:
try:
key = lower_case(
sentido.find('div', attrs={
'class': 'sentido'
}).getText().strip(":"))
except Exception as e:
key = lower_case(
sentido.find('strong').getText().strip("."))
pass
#print(sentido.find('p', attrs={'class':'antonimos'}))
p = sentido.find('p', attrs={
'class': 'antonimos'
}).getText()[3:]
try:
p = str(p.encode('raw_unicode_escape').decode('utf-8'))
except Exception as e:
print(e)
pass
#print(p)
#print(p.encode('utf-8'))
all_values = p.split(',')
#print(all_values)
for value in all_values:
lista_antonimos.append(value.strip(":").strip(' '))
set_(antonimos, palavra + "." + key, lista_antonimos)
print("Antônimo Salv@ no Dicionário")
except Exception as e:
print("\nError:\n" + str(e))
return []
save_json("antonimos", antonimos)
return get(antonimos, palavra)