def _get_time_line(self, t_user):
try:
cur = tweepy.Cursor(self.api.user_timeline,id=t_user.id)
t_timeline = []
for t_el in cur.items():
try:
log.info("appending user timeline")
t_timeline.append(t_el)
except Exception as e:
if 'Rate limit exceeded' in str(e):
log.info('oook wil be sleep...')
time.sleep(60)
self._count_requests += 1
log.debug("get timeline +1")
timeline = [
m_hash_dict({'text': element.text, 'retweets': element.retweet_count, 'initted': element.created_at})
for element in t_timeline
] #retieving user perls
return timeline
except Exception as e:
if 'Rate limit exceeded' in str(e):
log.info('oook wil be sleep...')
time.sleep(60)
return self._get_time_line(t_user)
else:
raise e
def __get_statistic_of_tweets(data):
"""
input some [] with strings
which return [] of {freq :<frequency_of_this_element_in_all_elements_in_input>, type:<type_name>, entity:<word>}
//also in db - will can create collection entities, which include some interests
"""
all_data = process_to_words(data)
types = {mention: [], word: [], hash_tag: [], url: []}
#separate and implying string types
for element in all_data:
__imply_string_obj(element, types)
#calculate all frequencies:
result = []
all_len = float(len(all_data))
for type, arr in types.items():
for el in arr:
result.append(m_hash_dict({'entity': el,
'freq': float(all_data.count(el)) / all_len,
'type': type}))
return {'statistic': list(set(result)),
hash_tag: list(set(types[hash_tag])),
mention: list(set(types[mention]))}
def diff_str(element1, element2):
difference_element(
difference_element.s_changed,
m_hash_dict({difference_element.s_a_old: element1, difference_element.s_a_new: element2}),
)
def _validate_dict_of_array(array):
if len(array) and isinstance(array[0], dict) and not isinstance(array[0], m_hash_dict):
return [m_hash_dict(el) for el in array]
return array
def _prep_fields(fields, exclude):
fields_ = [field for field in fields.items() if not exclude(field[0])]
fields_ = m_hash_dict(dict(fields_))
return fields_
@staticmethod
def diff_str(element1, element2):
difference_element(
difference_element.s_changed,
m_hash_dict({difference_element.s_a_old: element1, difference_element.s_a_new: element2}),
)
@staticmethod
def __print_difference(m_difference):
if not m_difference:
print None
return
dict = m_difference.__dict__
for item in dict.items():
print item
class h_set(set):
def __hash__(self):
hashes = []
for x in self:
hashes.append(hash(x))
return sum(hashes)
if __name__ == "__main__":
print set(m_hash_dict({1: 1, 2: 2, 3: 4}))
# diff = difference_factory.diff_arrays([[1, 1], 1, 1, 1, [1, 1]], [2, [2, 2], 2, [2, 2]])
# print diff
def serialise(self):
return m_hash_dict(self.__dict__)
