Python核心编程习题之——正则表达式

[bh][aiu]t

[a-zA-Z]+\s[a-zA-Z]+

[a-zA-Z]+,\s[a-zA-Z]

[a-zA-Z]\w+

\d+\s([a-zA-Z]+\s*)+

^www[^\s]+com$

[-+]?\d+

[-+]?\d+[lL]

\d+\.\d*

\d*(-\d+j)?

\w+@[\w\.]+.com

http://[\w\.]+\.(com|cn)

patt = r"<\w+\s'(\w+)'>"
data = type(1)
m = re.search(patt,data)
if m:
print(m.group(1))

1[0-2]

\d{4}(-\d{6}-\d{5}|(-\d{4}){3})

from random import randrange, choice
from string import ascii_lowercase as lc
from time import ctime

tlds = ('com', 'edu', 'net', 'org', 'gov')

for i in range(randrange(5,11)):
dtint = randrange(10**10)
dtstr = ctime(dtint)
llen = randrange(4, 8)
c = []
for j in range(llen):
c.append(choice(lc))
login = ''.join(c)
dlen = randrange(llen, 13)
d = []
for j in range(dlen):
d.append(choice(lc))
dom = ''.join(d)
with open('redata.txt','a') as f:
line = '%s::%s@%s.%s::%d-%d-%d' %(dtstr, login, dom, choice(tlds), dtint, llen, dlen)
f.writelines(line+'\n')

import re

days = ['Mon','Tue','Wed','Thu','Fri','Sat','Sun']
days_count = {}
with open('redata.txt','r') as f:
patt = r'\w{3}'
for line in f:
m = re.match(patt,line)
day = m.group()
if day not in days:
continue
if day not in days_count:
days_count[day] = 1
else:
days_count[day] += 1
print(days_count)

import re
def find(line):
patt = r'(\w+)\s(\w+)\s+(\d+)\s(\d+:\d+:\d+)\s(\d+)::(\w+@\w+\.\w+)::(\d+-\d+-\d+)'
m = re.match(patt,line)
return m

def find_email(email):
patt = r'(\w+)@(\w+)\.(\w+)'
m = re.match(patt,email)
return m

with open('redata.txt','r') as f:
for line in f:
m = find(line)
#年
year = m.group(5)
#月
month = m.group(2)
#日
day = m.group(3)
#时间
time = m.group(4)
#星期
week = m.group(1)
#时间戳
timestamp = year+' '+month+' '+day+' '+time+' '+week
#地址
print(timestamp)
email = m.group(6)
m = find_email(email)
name = m.group(1)
main_host = m.group(2)
second_host = m.group(3)
print(name,main_host,second_host)

((\d+-)|(\(\d+\)\s))?\d+-\d+