def add_emails(self, topic_ids):
'''
add all emails and email_blobs
'''
# loop over emails
for email in range(0, self.num_emails): # loop over emails
em = self.metadata[email]
dtime_orig = dateparse(em['Date'])
dtime_utc = dtime_orig.astimezone(pytz.utc)
values = nparray([em['Subject'], em['From'], em['To'], em['Cc'],
em['Bcc'], dtime_orig, dtime_utc])
values = nparray([value.replace("'", " ") if
(value and isinstance(value, str)) else value for value in values])
rows = nparray(['subject', 'sender', 'receiver', 'cc', 'bcc',
'send_time', 'send_time_utc'])
bool = nparray([True if a else False for a in values])
self.add_email('email', rows[bool], values[bool])
for idx2, t_id in enumerate(topic_ids): # loop over topics
rows = nparray(['topic_id', 'topic_probability'])
values = nparray([t_id, self.email_prob[idx2, email]])
self.add_blob('email_blob', rows, values)
评论列表
文章目录