async def show_message_edits(self, from_id, entity_id, message_id):
message_edits = self.dialog_stats.get_message_edits(
entity_id, message_id)
results = [
'Кажется, ты переслал сообщение из отслеживаемого диалога',
'Выведем доп. информацию по нему...',
'Число правок: {}'.format(len(message_edits) - 1)
]
if len(message_edits) > 0:
last_version = None
for message_edit in message_edits:
results.append('')
results.append('**Версия {} / {}**'.format(
message_edit['version'], message_edit['max_version']))
date = StatusController.datetime_from_str(
message_edit['taken_at'], '%Y-%m-%d %H:%M:%S%z')
results.append('Дата: {}'.format(
StatusController.datetime_to_str(date)))
results.append('Сообщение: \n[{}]\n'.format(
self.dialog_stats.remove_message_tags(
message_edit['message'])))
if last_version:
edit_ratio = self.dialog_stats.get_str_difference_ratio(
self.dialog_stats.remove_message_tags(
last_version['message']),
self.dialog_stats.remove_message_tags(
message_edit['message']))
results.append('Процент правок: {0:0.2f}%'.format(
100 * edit_ratio))
diff_counts = self.dialog_stats.get_str_difference_counts(
last_version['message'], message_edit['message'])
results.append('Число замен : {}'.format(
diff_counts['replaces_count_edit']))
results.append('Число вставок : {}'.format(
diff_counts['inserts_count_edit']))
results.append('Число удалений: {}'.format(
diff_counts['deletes_count_edit']))
last_version = message_edit
results = "\n".join(results)
await self.send_message_to_user(from_id, results)
async def get_me_dialog_statistics(self, user_id, date_from=None, title='за всё время', only_last_dialog=False, skip_vocab=False):
new_normal_form_cache = CacheHelper().get_from_cache('normal_forms', 'dialog_stats', False)
if new_normal_form_cache:
self.normal_form_cache = new_normal_form_cache
new_word_type_form_cache = CacheHelper().get_from_cache('word_type_forms', 'dialog_stats', False)
if new_word_type_form_cache:
self.word_type_form_cache = new_word_type_form_cache
days_a = '?'
if (not date_from) and (not only_last_dialog):
res = self.db_conn.execute(
"""
SELECT *
FROM `activity`
ORDER BY taken_at ASC
""",
[])
rows = list(res.fetchall())
days_a = 1
if len(rows) > 1:
date1 = StatusController.datetime_from_str(rows[0]['taken_at'])
date2 = StatusController.datetime_from_str(rows[len(rows) - 1]['taken_at'])
days_a = round((date2 - date1).total_seconds() / (24 * 60 * 60))
results = []
last_dialogue_date = None
try:
user_entity = await self.tg_client.get_entity(PeerUser(user_id))
except:
user_entity = None
if user_entity and (type(user_entity) == User):
if not date_from:
res = self.db_conn.execute(
"""
SELECT m.*,
(SELECT version FROM `messages` m1 WHERE m1.`entity_id` = m.`entity_id` AND m1.message_id = m.message_id AND m1.from_id = m.from_id ORDER BY version DESC LIMIT 1) as 'max_version',
(SELECT removed FROM `messages` m1 WHERE m1.`entity_id` = m.`entity_id` AND m1.message_id = m.message_id AND m1.from_id = m.from_id ORDER BY version DESC LIMIT 1) as 'is_removed'
FROM `messages` m
WHERE m.`entity_id` = ? OR m.`entity_id` = ?
ORDER BY m.`taken_at` ASC, m.`message_id` ASC, m.`version` ASC
""",
[str(user_id), str(self.tg_client.me_user_id)]
)
else:
date_from = StatusController.datetime_to_str(date_from, '%Y-%m-%d')
res = self.db_conn.execute(
"""
SELECT m.*,
(SELECT version FROM `messages` m1 WHERE m1.`entity_id` = m.`entity_id` AND m1.message_id = m.message_id AND m1.from_id = m.from_id ORDER BY version DESC LIMIT 1) as 'max_version',
(SELECT removed FROM `messages` m1 WHERE m1.`entity_id` = m.`entity_id` AND m1.message_id = m.message_id AND m1.from_id = m.from_id ORDER BY version DESC LIMIT 1) as 'is_removed'
FROM `messages` m
WHERE (m.`entity_id` = ? OR m.`entity_id` = ?) AND (m.`taken_at` > ?)
ORDER BY m.`taken_at` ASC, m.`message_id` ASC, m.`version` ASC
""",
[str(user_id), str(self.tg_client.me_user_id), date_from]
)
rows = list(res.fetchall())
me_name = self.tg_client.me_user_entity_name
another_name = await self.tg_client.get_entity_name(user_id, 'User')
dialog_name = me_name + ' <-> ' + another_name
results.append('**Диалог '+dialog_name+' ('+title+')'+':**')
results.append('')
results.append('Сообщений диалога в БД: ' + str(len(rows)))
if len(rows) > 0:
date_start = StatusController.datetime_from_str(rows[0]['taken_at'], '%Y-%m-%d %H:%M:%S%z')
if not only_last_dialog:
results.append('Самое раннее сообщение диалога в БД: ' + StatusController.datetime_to_str(date_start))
if len(rows) > 1:
date_end = StatusController.datetime_from_str(rows[len(rows) - 1]['taken_at'], '%Y-%m-%d %H:%M:%S%z')
seconds_count = (date_end - date_start).total_seconds()
days_count = seconds_count / (24 * 60 * 60)
messages_count = len(rows)
if (not date_from) and (not only_last_dialog):
results.append('Длительность общения: {0:0.2f} суток'.format(days_count))
if not only_last_dialog:
results.append('Средняя частота сообщений: {0:0.2f} в сутки'.format(messages_count / days_count))
max_dialog_question_interval = round((24 * 60 * 60) * 1.25)
max_dialog_non_question_interval = round((24 * 60 * 60) * 0.75)
max_dialog_hello_as_second_message_offset = round((24 * 60 * 60) * 0.25)
dialog_hello_words = ['привет', 'приветствую', 'здравствуй', 'здравствуйте']
dialog_hello_phrases = ['доброе утро', 'доброго утра', 'добрый день', 'доброго дня', 'добрый вечер', 'доброго вечера']
dialog_hello_stop_context = ['-привет', 'всем привет', 'привет»', 'привет"']
msg_len_me = 0
msg_me_cnt = 0
msg_me_max_len = 0
msg_len_another = 0
msg_another_cnt = 0
msg_another_max_len = 0
me_deletes = 0
another_deletes = 0
me_hello = 0
another_hello = 0
me_words = []
another_words = []
dialogues = []
active_dialog = []
last_msg_from_id = None
last_date = date_start
last_msg_is_question = False
edited_messages_me = {}
edited_messages_another = {}
edited_sequence_interrupted = {}
last_message_row = None
for row in rows:
if not row['message']:
row['message'] = ''
if int(row['removed']) == 1 or int(row['is_removed']) == 1:
if int(row['version']) == int(row['max_version']):
if int(row['from_id']) == self.tg_client.me_user_id:
me_deletes = me_deletes + 1
else:
another_deletes = another_deletes + 1
else:
if row['message_id'] not in edited_sequence_interrupted:
for k_int in edited_sequence_interrupted.keys():
if type(edited_sequence_interrupted[k_int]) == dict:
if edited_sequence_interrupted[k_int]['from_id'] != row['from_id']:
edited_sequence_interrupted[k_int]['interrupts'].append(row['message_id'])
if int(row['max_version']) > 1:
if int(row['version']) < int(row['max_version']):
if row['message_id'] not in edited_sequence_interrupted:
edited_sequence_interrupted[row['message_id']] = {
'from_id': row['from_id'],
'interrupts': []
}
else:
if row['message_id'] in edited_sequence_interrupted:
edited_sequence_interrupted[row['message_id']] = len(edited_sequence_interrupted[row['message_id']]['interrupts']) > 0
if last_message_row and (int(last_message_row['version']) < int(last_message_row['max_version'])) and (row['from_id'] != last_message_row['from_id']):
edited_sequence_interrupted[row['message_id']] = True
if int(row['version']) == int(row['max_version']):
last_message_row = row
if int(row['max_version']) > 1:
if int(row['from_id']) == self.tg_client.me_user_id:
if row['message_id'] not in edited_messages_me:
edited_messages_me[row['message_id']] = []
edited_messages_me[row['message_id']].append(row)
else:
if row['message_id'] not in edited_messages_another:
edited_messages_another[row['message_id']] = []
edited_messages_another[row['message_id']].append(row)
if int(row['version']) == int(row['max_version']):
message_lower = str(row['message']).lower()
message_words = re.sub("[^\w]", " ", message_lower).split()
message_words = list(filter(lambda x: x and self.is_valid_word(x, []), message_words))
message_hello_words = list(filter(lambda x: x in dialog_hello_words, message_words))
message_stop_contexts = list(filter(lambda x: message_lower.find(x) >= 0, dialog_hello_stop_context))
msg_from_id = int(row['from_id'])
msg_is_question = str(row['message']).find('?') >= 0
msg_is_hello = (len(message_hello_words) > 0) and (len(message_stop_contexts) == 0)
if not msg_is_hello:
for d_ph in dialog_hello_phrases:
if message_lower.find(d_ph) >= 0:
msg_is_hello = True
break
if not skip_vocab:
nform_list = [self.get_normal_form(x) for x in message_words]
if msg_from_id == self.tg_client.me_user_id:
me_words = me_words + nform_list
else:
another_words = another_words + nform_list
if msg_is_hello:
if msg_from_id == self.tg_client.me_user_id:
me_hello = me_hello + 1
else:
another_hello = another_hello + 1
msg_len = len(row['message'])
if msg_from_id == self.tg_client.me_user_id:
msg_len_me = msg_len_me + msg_len
msg_me_cnt = msg_me_cnt + 1
if msg_len > msg_me_max_len:
msg_me_max_len = msg_len
else:
msg_len_another = msg_len_another + msg_len
msg_another_cnt = msg_another_cnt + 1
if msg_len > msg_another_max_len:
msg_another_max_len = msg_len
msg_date = StatusController.datetime_from_str(row['taken_at'], '%Y-%m-%d %H:%M:%S%z')
if (
(len(active_dialog) == 0) or (
last_msg_is_question and
((msg_date - last_date).total_seconds() > max_dialog_question_interval)
) or (
not last_msg_is_question and
((msg_date - last_date).total_seconds() > max_dialog_non_question_interval)
)
):
if len(active_dialog) > 0:
dialogues.append(active_dialog)
active_dialog = []
active_dialog.append(row)
last_date = msg_date
if last_msg_from_id != msg_from_id:
last_msg_is_question = msg_is_question
else:
last_msg_is_question = last_msg_is_question or msg_is_question
last_msg_from_id = msg_from_id
for k_int in edited_sequence_interrupted.keys():
if type(edited_sequence_interrupted[k_int]) == dict:
edited_sequence_interrupted[k_int] = len(edited_sequence_interrupted[k_int]['interrupts']) > 0
me_edit_stats = self.get_edit_stats(edited_messages_me, edited_sequence_interrupted)
another_edit_stats = self.get_edit_stats(edited_messages_another, edited_sequence_interrupted)
if len(active_dialog) > 0:
dialogues.append(active_dialog)
active_dialog = []
if only_last_dialog:
dialogues = [dialogues[len(dialogues) - 1]]
else:
if len(dialogues) > 0:
last_dialogue_date = StatusController.datetime_from_str(dialogues[len(dialogues) - 1][0]['taken_at'], '%Y-%m-%d %H:%M:%S%z')
else:
last_dialogue_date = None
answers_me = 0
answers_wait_seconds_me = 0
answers_another = 0
answers_wait_seconds_another = 0
longest_len = 0
longest_dialog = None
shortest_len = 0
dia_me_start = 0
dia_another_start = 0
dia_me_finish = 0
dia_another_finish = 0
dia_between_seconds = 0
dia_between_max = 0
dia_between_max_from = None
dia_between_max_to = None
dia_between_cnt = 0
last_dia_end = None
for dial in dialogues:
dial_len = len(dial)
if (shortest_len == 0) or (dial_len < shortest_len):
shortest_len = dial_len
if (longest_len == 0) or (dial_len > longest_len):
longest_len = dial_len
longest_dialog = dial
if dial_len > 0:
first_dial = dial[0]
last_dial = dial[len(dial) - 1]
if int(first_dial['from_id']) == self.tg_client.me_user_id:
dia_me_start = dia_me_start + 1
else:
dia_another_start = dia_another_start + 1
if int(last_dial['from_id']) == self.tg_client.me_user_id:
dia_me_finish = dia_me_finish + 1
else:
dia_another_finish = dia_another_finish + 1
if last_dia_end:
curr_first_dia_begin = StatusController.datetime_from_str(first_dial['taken_at'], '%Y-%m-%d %H:%M:%S%z')
dia_between_seconds_curr = (curr_first_dia_begin - last_dia_end).total_seconds()
dia_between_seconds = dia_between_seconds + dia_between_seconds_curr
dia_between_cnt = dia_between_cnt + 1
if dia_between_seconds_curr > dia_between_max:
dia_between_max = dia_between_seconds_curr
dia_between_max_from = last_dia_end
dia_between_max_to = curr_first_dia_begin
last_dia_end = StatusController.datetime_from_str(last_dial['taken_at'], '%Y-%m-%d %H:%M:%S%z')
last_msg_id = None
last_msg_date = None
for dia in dial:
msg_date = StatusController.datetime_from_str(dia['taken_at'], '%Y-%m-%d %H:%M:%S%z')
if last_msg_id and (last_msg_id != int(dia['from_id'])):
seconds_between = (msg_date - last_msg_date).total_seconds()
if seconds_between > (60 * 60 * 4) and ((last_msg_date.time().hour < 6) or (last_msg_date.time().hour >= 22)):
# somebody just sleep
last_msg_date = msg_date
last_msg_id = int(dia['from_id'])
continue
if last_msg_id == self.tg_client.me_user_id:
answers_another = answers_another + 1
answers_wait_seconds_another = answers_wait_seconds_another + seconds_between
else:
answers_me = answers_me + 1
answers_wait_seconds_me = answers_wait_seconds_me + seconds_between
last_msg_date = msg_date
last_msg_id = int(dia['from_id'])
if dia_between_cnt > 0:
dia_between_time = dia_between_seconds / dia_between_cnt
dia_between_time = "{0:0.2f} сут.".format(dia_between_time / (60 * 60 * 24))
dia_between_time_max = "{0:0.2f} сут.".format(dia_between_max / (60 * 60 * 24))
dia_between_time_max = StatusController.datetime_to_str(dia_between_max_from) + ' --- ' + StatusController.datetime_to_str(dia_between_max_to) + ' ('+dia_between_time_max+')'
else:
dia_between_time = '?'
dia_between_time_max = '?'
if answers_another > 0:
answers_wait_seconds_another = answers_wait_seconds_another / answers_another
another_answer_time = "{0:0.2f} мин.".format(answers_wait_seconds_another / 60)
else:
another_answer_time = '?'
if answers_me > 0:
answers_wait_seconds_me = answers_wait_seconds_me / answers_me
me_answer_time = "{0:0.2f} мин.".format(answers_wait_seconds_me / 60)
else:
me_answer_time = '?'
if not date_from:
self.tg_client.entity_controller.set_entity_answer_sec(user_id, answers_wait_seconds_me, answers_wait_seconds_another)
longest_dates = ''
longest_hours = 0
if longest_len > 1:
msg_date1 = StatusController.datetime_from_str(longest_dialog[0]['taken_at'], '%Y-%m-%d %H:%M:%S%z')
msg_date2 = StatusController.datetime_from_str(longest_dialog[longest_len - 1]['taken_at'], '%Y-%m-%d %H:%M:%S%z')
longest_hours = (msg_date2 - msg_date1).total_seconds() / (24 * 60 * 60)
longest_dates = StatusController.datetime_to_str(msg_date1) + ' --- ' + StatusController.datetime_to_str(msg_date2)
valid_word_types = ['СУЩ', 'МЕЖД']
valid_word_types_str = (",".join(valid_word_types)).lower()
me_top_10 = None
me_last_cnt = None
me_words_count = 0
all_words_me = {}
if (not skip_vocab) and (len(me_words) > 0):
for word in me_words:
if word and (word not in all_words_me):
all_words_me[word] = True
me_words_count = len(all_words_me)
me_words = list(filter(lambda x: x and self.is_valid_word(x, valid_word_types), me_words))
wordlist = sorted(me_words)
wordfreq = [wordlist.count(p) for p in wordlist]
dic = dict(zip(wordlist, wordfreq))
me_words = list(sorted(dic.items(), key=lambda x: x[1], reverse=True))
words = me_words
half_words = round(len(words) / 2)
if half_words < 3:
half_words = 3
elif half_words > 15:
half_words = 15
top_10 = list(filter(lambda x: x[1] > 1, words[0:half_words]))
top_10 = list(map(lambda x: '**' + str(x[0]) + '** (' + str(x[1]) + ')', top_10))
last_word, last_cnt = words[len(words) - 1]
last_cnt_words = map(lambda x: x[0], filter(lambda s: s[1] == last_cnt, words))
last_cnt_words = list(sorted(last_cnt_words, key=lambda x: len(x), reverse=True))
last_cnt_cnt = len(last_cnt_words) - half_words
if last_cnt_cnt < 3:
last_cnt_cnt = 3
elif last_cnt_cnt > 10:
last_cnt_cnt = 10
last_cnt_words = last_cnt_words[0:last_cnt_cnt]
me_top_10 = top_10
me_last_cnt = last_cnt_words
another_top_10 = None
another_last_cnt = None
another_words_count = 0
all_words_another = {}
if (not skip_vocab) and (len(another_words) > 0):
for word in another_words:
if word and (word not in all_words_another):
all_words_another[word] = True
another_words_count = len(all_words_another)
another_words = list(filter(lambda x: x and self.is_valid_word(x, valid_word_types), another_words))
wordlist = sorted(another_words)
wordfreq = [wordlist.count(p) for p in wordlist]
dic = dict(zip(wordlist, wordfreq))
another_words = list(sorted(dic.items(), key=lambda x: x[1], reverse=True))
words = another_words
half_words = round(len(words) / 2)
if half_words < 3:
half_words = 3
elif half_words > 15:
half_words = 15
top_10 = list(filter(lambda x: x[1] > 1, words[0:half_words]))
top_10 = list(map(lambda x: '**' + str(x[0]) + '** (' + str(x[1]) + ')', top_10))
last_word, last_cnt = words[len(words) - 1]
last_cnt_words = map(lambda x: x[0], filter(lambda s: s[1] == last_cnt, words))
last_cnt_words = list(sorted(last_cnt_words, key=lambda x: len(x), reverse=True))
last_cnt_cnt = len(last_cnt_words) - half_words
if last_cnt_cnt < 3:
last_cnt_cnt = 3
elif last_cnt_cnt > 10:
last_cnt_cnt = 10
last_cnt_words = last_cnt_words[0:last_cnt_cnt]
another_top_10 = top_10
another_last_cnt = last_cnt_words
me_not_another_words_count = 0
me_not_another_top = None
another_not_me_words_count = 0
another_not_me_top = None
if (not skip_vocab) and (len(me_words) > 0) and (len(another_words) > 0):
me_not_another_top = []
for word in me_words:
if word[0] not in all_words_another:
me_not_another_top.append(word)
me_not_another_words_count = len(me_not_another_top)
me_not_another_top = me_not_another_top[:15]
me_not_another_top = list(map(lambda x: '**' + str(x[0]) + '** (' + str(x[1]) + ')', me_not_another_top))
another_not_me_top = []
for word in another_words:
if word[0] not in all_words_me:
another_not_me_top.append(word)
another_not_me_words_count = len(another_not_me_top)
another_not_me_top = another_not_me_top[:15]
another_not_me_top = list(map(lambda x: '**' + str(x[0]) + '** (' + str(x[1]) + ')', another_not_me_top))
results.append('Сообщений '+another_name+': {0} ({1:0.3f} Kb.)'.format(msg_another_cnt, msg_len_another/1024))
results.append('Сообщений '+me_name+': {0} ({1:0.3f} Kb.)'.format(msg_me_cnt, msg_len_me/1024))
if msg_another_cnt > 0:
results.append('Средняя длина сообщения '+another_name+': {0:0.2f} сим.'.format(msg_len_another / msg_another_cnt))
if msg_me_cnt > 0:
results.append('Средняя длина сообщения '+me_name+': {0:0.2f} сим.'.format(msg_len_me / msg_me_cnt))
results.append('Самое длинное сообщение '+another_name+': ' + str(msg_another_max_len) + ' сим.')
results.append('Самое длинное сообщение '+me_name+': ' + str(msg_me_max_len) + ' сим.')
results.append('Число приветствий от '+another_name+': ' + str(another_hello))
results.append('Число приветствий от '+me_name+': ' + str(me_hello))
if not only_last_dialog:
results.append('')
results.append('Число диалогов: ' + str(len(dialogues)))
results.append('Сообщений в самом коротком диалоге: ' + str(shortest_len))
results.append('Сообщений в самом длинном диалоге: ' + str(longest_len))
results.append('Самый длинный диалог: ' + longest_dates + ' ({0:0.3f} сут)'.format(longest_hours))
if len(dialogues) > 0:
if dia_between_time != '?':
results.append('Среднее время между диалогами: ' + str(dia_between_time))
if dia_between_time_max != '?':
results.append('Самое большое время между диалогами: ' + str(dia_between_time_max))
results.append('Инициатор диалога ' + another_name + ': {0:0.2f} %'.format(100 * dia_another_start / len(dialogues)))
results.append('Инициатор диалога ' + me_name + ': {0:0.2f} %'.format(100 * dia_me_start / len(dialogues)))
results.append('Завершитель диалога ' + another_name + ': {0:0.2f} %'.format(100 * dia_another_finish / len(dialogues)))
results.append('Завершитель диалога ' + me_name + ': {0:0.2f} %'.format(100 * dia_me_finish / len(dialogues)))
else:
results.append('Сообщений в диалоге: ' + str(longest_len))
results.append('Продолжительность диалога: ' + longest_dates + ' ({0:0.3f} сут)'.format(longest_hours))
if dia_me_start > 0:
results.append('Инициатор: ' + me_name)
elif dia_another_start > 0:
results.append('Инициатор: ' + another_name)
if dia_me_finish > 0:
results.append('Завершитель: ' + me_name)
elif dia_another_finish > 0:
results.append('Завершитель: ' + another_name)
if another_answer_time != '?':
results.append('В среднем ' + another_name + ' отвечает за: ' + another_answer_time)
if me_answer_time != '?':
results.append('В среднем ' + me_name + ' отвечает за: ' + me_answer_time)
results.append('')
if me_edit_stats['edited_messages_count'] > 0 or another_edit_stats['edited_messages_count'] > 0:
if (not date_from) and (not only_last_dialog):
results.append('За время активности скрипта ('+str(days_a)+' сут.):')
results.append('Отредактировано сообщений {}: {}'.format(another_name, another_edit_stats['edited_messages_count']))
results.append('Отредактировано сообщений {}: {}'.format(me_name, me_edit_stats['edited_messages_count']))
results.append('Отредактировано сообщений {} после ответа на него: {}'.format(another_name, another_edit_stats['sequence_interrupted_cnt']))
results.append('Отредактировано сообщений {} после ответа на него: {}'.format(me_name, me_edit_stats['sequence_interrupted_cnt']))
results.append('Процент редактируемых сообщений {0}: {1:0.2f}%'.format(another_name, 100 * another_edit_stats['edited_messages_count'] / msg_another_cnt))
results.append('Процент редактируемых сообщений {0}: {1:0.2f}%'.format(me_name, 100 * me_edit_stats['edited_messages_count'] / msg_me_cnt))
if another_edit_stats['max_1_message_edits'] > 0:
results.append('Макс. число правок одного сообщения {}: {} ("{}")'.format(another_name, another_edit_stats['max_1_message_edits'], self.cut_text(another_edit_stats['max_edits_message'])))
if me_edit_stats['max_1_message_edits'] > 0:
results.append('Макс. число правок одного сообщения {}: {} ("{}")'.format(me_name, me_edit_stats['max_1_message_edits'], self.cut_text(me_edit_stats['max_edits_message'])))
if another_edit_stats['max_1_message_diff_percent'] > 0:
results.append('Макс. процент правок одного сообщения {0}: {1:0.2f}% ("{2}")'.format(another_name, another_edit_stats['max_1_message_diff_percent'], self.cut_text(another_edit_stats['max_changed_message'])))
if me_edit_stats['max_1_message_diff_percent'] > 0:
results.append('Макс. процент правок одного сообщения {0}: {1:0.2f}% ("{2}")'.format(me_name, me_edit_stats['max_1_message_diff_percent'], self.cut_text(me_edit_stats['max_changed_message'])))
results.append('')
results.append('В среднем правок на 1 сообщение {0}: {1:0.2f}'.format(another_name, another_edit_stats['mid_1_message_edits']))
results.append('В среднем правок на 1 сообщение {0}: {1:0.2f}'.format(me_name, me_edit_stats['mid_1_message_edits']))
results.append('Средний суммарный процент изменений редактируемых сообщений {0}: {1:0.3f}%'.format(another_name, another_edit_stats['message_edit_mid_summ_percent']))
results.append('Средний суммарный процент изменений редактируемых сообщений {0}: {1:0.3f}%'.format(me_name, me_edit_stats['message_edit_mid_summ_percent']))
results.append('Среднее время между правками одного сообщения {0}: {1:0.2f} мин.'.format(another_name, another_edit_stats['message_edit_mid_time_sec'] / 60))
results.append('Среднее время между правками одного сообщения {0}: {1:0.2f} мин.'.format(me_name, me_edit_stats['message_edit_mid_time_sec'] / 60))
results.append('')
results.append('Средняя правка (число замен / вставок / удалений) {0}: {1:0.2f} / {2:0.2f} / {3:0.2f}'.format(another_name, another_edit_stats['replaces_count_avg'], another_edit_stats['inserts_count_avg'], another_edit_stats['deletes_count_avg']))
results.append('Средняя правка (число замен / вставок / удалений) {0}: {1:0.2f} / {2:0.2f} / {3:0.2f}'.format(me_name, me_edit_stats['replaces_count_avg'], me_edit_stats['inserts_count_avg'], me_edit_stats['deletes_count_avg']))
results.append('')
results.append('Удалений сообщений ' + another_name + ': ' + str(another_deletes))
results.append('Удалений сообщений ' + me_name + ': ' + str(me_deletes))
if me_top_10 or another_top_10:
results.append('')
results.append('Различных слов ' + another_name + ' в диалогах: ' + str(another_words_count))
results.append('Различных слов ' + me_name + ' в диалогах: ' + str(me_words_count))
results.append('')
results.append('Самые частые слова ('+valid_word_types_str+') ' + another_name + ' в диалогах: ' + (", ".join(another_top_10)) + '')
results.append('')
results.append('Самые частые слова ('+valid_word_types_str+') ' + me_name + ' в диалогах: ' + (", ".join(me_top_10)) + '')
results.append('')
results.append('Самые редкие слова ('+valid_word_types_str+') ' + another_name + ' в диалогах: ' + (", ".join(another_last_cnt)) + '')
results.append('')
results.append('Самые редкие слова ('+valid_word_types_str+') ' + me_name + ' в диалогах: ' + (", ".join(me_last_cnt)) + '')
if me_not_another_words_count > 0:
results.append('')
results.append('Слова ' + me_name + ' ('+valid_word_types_str+'), которые ' + another_name + ' ни разу не использовал: **' + str(me_not_another_words_count) +'** шт. Самые частые: ' + (", ".join(me_not_another_top)) + '')
if another_not_me_words_count > 0:
results.append('')
results.append('Слова ' + another_name + ' ('+valid_word_types_str+'), которые ' + me_name + ' ни разу не использовал: **' + str(another_not_me_words_count) +'** шт. Самые частые: ' + (", ".join(another_not_me_top)) + '')
CacheHelper().save_to_cache('normal_forms', 'dialog_stats', self.normal_form_cache)
CacheHelper().save_to_cache('word_type_forms', 'dialog_stats', self.word_type_form_cache)
return {
'results': results,
'last_dialogue_date': last_dialogue_date
}
def get_edit_stats(self, edited_messages: dict, edited_sequence_interrupted: dict):
edits_count = 0
max_1_message_edits = 0
max_1_message_diff = 0
mid_1_message_edits = 0.0
message_edit_mid_full_percent = 0
message_edit_mid_summ_percent = 0
message_edit_mid_time_sec = 0
edited_messages_count = len(edited_messages)
max_edits_message = ''
max_changed_message = ''
replaces_count_avg = 0
inserts_count_avg = 0
deletes_count_avg = 0
sequence_interrupted_cnt = 0
if edited_messages_count > 0:
for edit_rows in edited_messages.values():
# print([(x['taken_at'] + ': "' + x['message'] + '"') for x in edit_rows])
edit_ratio_summ = 0
last_version = None
summ_edits_time = 0
message_edit_cnt = len(edit_rows) - 1
mid_1_message_edits = mid_1_message_edits + message_edit_cnt
if message_edit_cnt >= max_1_message_edits:
max_1_message_edits = message_edit_cnt
max_edits_message = self.remove_message_tags(edit_rows[len(edit_rows) - 1]['message'])
replaces_count_edit = 0
inserts_count_edit = 0
deletes_count_edit = 0
for edit_row in edit_rows:
if last_version:
curr_time = StatusController.datetime_from_str(edit_row['taken_at'], '%Y-%m-%d %H:%M:%S%z')
last_time = StatusController.datetime_from_str(last_version['taken_at'], '%Y-%m-%d %H:%M:%S%z')
summ_edits_time = summ_edits_time + abs((curr_time - last_time).total_seconds())
edits_count = edits_count + 1
edit_ratio = self.get_str_difference_ratio(self.remove_message_tags(last_version['message']), self.remove_message_tags(edit_row['message']))
edit_ratio_summ = edit_ratio_summ + edit_ratio
diff_counts = self.get_str_difference_counts(last_version['message'], edit_row['message'])
inserts_count_edit = inserts_count_edit + diff_counts['inserts_count_edit']
replaces_count_edit = replaces_count_edit + diff_counts['replaces_count_edit']
deletes_count_edit = deletes_count_edit + diff_counts['deletes_count_edit']
last_version = edit_row
edit_ratio_full = self.get_str_difference_ratio(self.remove_message_tags(edit_rows[0]['message']), self.remove_message_tags(last_version['message']))
if edit_ratio_full >= max_1_message_diff:
max_1_message_diff = edit_ratio_full
max_changed_message = self.remove_message_tags(edit_rows[len(edit_rows) - 1]['message'])
message_edit_mid_full_percent = message_edit_mid_full_percent + edit_ratio_full
message_edit_mid_summ_percent = message_edit_mid_summ_percent + edit_ratio_summ
message_edit_mid_time_sec = message_edit_mid_time_sec + (summ_edits_time / message_edit_cnt)
replaces_count_avg = replaces_count_avg + replaces_count_edit
inserts_count_avg = inserts_count_avg + inserts_count_edit
deletes_count_avg = deletes_count_avg + deletes_count_edit
msg_id = edit_rows[0]['message_id']
if (msg_id in edited_sequence_interrupted) and edited_sequence_interrupted[msg_id]:
sequence_interrupted_cnt = sequence_interrupted_cnt + 1
mid_1_message_edits = mid_1_message_edits / edited_messages_count
message_edit_mid_full_percent = 100 * message_edit_mid_full_percent / edited_messages_count
message_edit_mid_summ_percent = 100 * message_edit_mid_summ_percent / edited_messages_count
message_edit_mid_time_sec = message_edit_mid_time_sec / edited_messages_count
replaces_count_avg = replaces_count_avg / edited_messages_count
inserts_count_avg = inserts_count_avg / edited_messages_count
deletes_count_avg = deletes_count_avg / edited_messages_count
max_1_message_diff = 100 * max_1_message_diff
return {
"edits_count": edits_count,
"sequence_interrupted_cnt": sequence_interrupted_cnt,
"edited_messages_count": edited_messages_count,
"max_1_message_edits": max_1_message_edits,
"max_1_message_diff_percent": max_1_message_diff,
"mid_1_message_edits": mid_1_message_edits,
"message_edit_mid_full_percent": message_edit_mid_full_percent,
"message_edit_mid_summ_percent": message_edit_mid_summ_percent,
"message_edit_mid_time_sec": message_edit_mid_time_sec,
"max_edits_message": max_edits_message,
"max_changed_message": max_changed_message,
"replaces_count_avg": replaces_count_avg,
"inserts_count_avg": inserts_count_avg,
"deletes_count_avg": deletes_count_avg,
}