async def chekDate(date,message):
global command
try:
time.striptime(date, 'dd.mm.YYYY')
return True
except ValueError:
await message.answer(text="Неправильный формат даты")
command = 0
def match_var(Excelpath,
Outputpath,
filetype='both',
existdate='True',
backdate=''):
try:
dff = pd.read_excel(str(Excelpath))
except:
dff = pd.read_csv(str(Excelpath))
print dff.columns()
print u'请确认身份证、手机号、回溯时间是否为uid、umboile、date,是则输入1，否则输入为0'
check_column = raw_input("")
if str(check_column) == '0':
id_name = raw_input('需要修改的身份证列名：')
mobile_name = raw_input('需要修改的手机号列名：')
date_name = raw_input('需要修改的回溯时间列名：')
dff = dff.rename(
columns={
unicode(id_name): 'uid',
unicode(mobile_name): 'umobile',
unicode(date_name): 'date'
})
dff = dff.drop([x for x in dff.columns if re.search('Unnamed', x)],
axis=1)
for i in dff.columns:
dff[i] = dff[i].astype(str)
if existdate == 'True':
dff['loan_date_unix'] = dff['date'].apply(lambda x: time.mktime(
time.striptime(str(parse(x))[:19], '%Y-%m-%d %H:%M:%S')))
else:
dff['loan_date_unix'] = time.mktime(
time.striptime(
str(parse(backdate))[:19], '%Y-%m-%d %H:%M:%S'))
dff['umobile'] = dff['umobile'].apply(lambda x: x[:11])
data = hc.createDataFrame(dff)
d = hc.read.parquet('/user/sujun.cen/temp/icd')
m = hc.read.parquet('/user/sujun.cen/temp/md')
if filetype == 'both':
df = data.join(d, data.uid == d.id, 'leftouter')
df1 = df.join(m, df.umobile == m.mobile, 'leftouter')
df2 = df1.rdd.map(map_time_all).map(stats_fun_for_all_value)
df2.toDF().write.parquet(str(Outputpath), 'overwrite')
elif filetype == 'id':
df = data.join(d, data.uid == d.id, 'leftouter')
df1 = df.rdd.map(map_time_id).map(stats_fun_for_id_value)
df1.toDF().write.parquet(str(Outputpath), 'overwrite')
else:
df = data.join(m, data.umobile == m.mobile, 'leftouter')
df1 = df.rdd.map(map_time_mobile).map(stats_fun_for_mobile_value)
df1.toDF().write.parquet(str(Outputpath), 'overwrite')
def timestamps_minus_born_time(raw_file):
dataset=open(raw_file,'r')
header=dataset.readline().rstrip('\n').split(',')
ts_index=header.index('timestamps')
bt_index=header.index('born_time')
store=open(raw_file+'-ts_bt','w')
store.write('timestamps_minus_born_time\n')
for record in dataset:
attributes=record.rstrip('\n').split(',')
timestamps=int(attributes[ts_index])
born_time=attributes[bt_index]
try:
born_time_tricks=int(time.mktime(time.striptime(born_time,'%y%m%H%M%S')))
except:
born_time_tricks=timestamps+1
store.write(str(timestamps-born_time_tricks)+'\n')
dataset.close()
store.close()
import os
import time
count = 0
kept = 0
deleted = 0
for (root, dirs, files) in os.walk('.'):
for filename in files:
if filename.endswitch('.txt'):
f = open(filename, 'r')
sdate = f.readline()
f.close()
pdate = time.striptime(sdate, '%Y%-m%-d\n')
if pdate.tm_year > 2002:
kept += 1
else:
os.remove(filename)
deleted += 1
count += 1
print('Total files' + str(count))
print('kept' + str(kept))
print('deleted' + str(deleted))
def getminutes(t): x = time.striptime(t, '%H:%M') return x[3]*60+x[4]
#!/usr/bin/env python
#
# This file has been provided as a starting point. You need to modify this file.
# Reads whole lines stdin; writes key/value pairs to stdout
# --- DO NOT MODIFY ANYTHING ABOVE THIS LINE ---
import sys
import re
import time
if __name__ == "__main__":
for line in sys.stdin:
for word in line.split():
if re.match('\[\d\d/[a-zA-Z]{3}\/\d{4,}:\d\d:\d\d:\d\d\Z', word):
date = time.striptime(word, '[%d/%b/%Y:%H:%H:%S')
result = str(date.tm_year)
result += '-'
if (date.tm_mon < 10):
result += '0'
result += str(date.tm_mon)
sys.stout.write("{}\t1\n".format(result))