<Automate the boring stuff with python> --- Chapter VII of the canonical example & regular & greedy matching phone number and email

Chapter VII to find a phone number by a string, compare the difference between regular expressions whether to use the program, significantly more concise wording regular, easy to expand.
Mode: 3 digits, a dash, three numbers, a dash, and then four digits. For example: 415-555-4242

. 1  Import Re
 2  '' ' 
. 3  do not find a regular pattern, matching three numbers, a dash, three numbers, a dash, 4 digits
 . 4  EX. 111-222-3334
 . 5  ' '' 
. 6  
. 7  DEF isPhoneNo (text):
 . 8      IF len (text) = 12 is! :
 . 9          return False
 10      for I in Range (0,3 ):
 . 11          IF  Not [I] .isdecimal text ():
 12 is              return False
 13 is      IF text [ . 3] =! ' - ' :
 14          return False
 15     for I in Range (4,7 ):
 16          IF  Not text [I] .isdecimal ():
 . 17              return False
 18 is      IF text [. 7] =! ' - ' :
 . 19          return False
 20 is      for I in Range (8,12 ) :
 21 is          IF  Not text [I] .isdecimal ():
 22 is              return False
 23 is      return True
 24  
25  '' ' 
26 is  a regular expression matching the above pattern
 27  ' '' 
28  DEF regPhoneNo (text):
29     phoneNoReg=re.compile(r'\d\d\d-\d\d\d-\d\d\d\d')
30     res=phoneNoReg.search(text)
31     if res != None:
32         print('phone No find by reg: '+ res.group())
33 
34 print(isPhoneNo('123-122-9090'))
35 print(isPhoneNo('1234123321'))
36 msg = 'call me at 415-443-1111 tomorrow. 415-443-2222 is my office'
37 for i in range(len(msg)):
38     tmp = msg[i:i+12]
39     if isPhoneNo(tmp):
40         print('phone No find: ' + tmp)
41     regPhoneNo(tmp)
42 print('msg find end')

View Code

 

Python regular expression default is "greedy", which means that under ambiguous circumstances, they will match the longest string possible.

Braces "non-greedy" version match the shortest possible string, followed by a question mark at the end of braces

Example:

'' ' 
Examples and non-greedy greedy Python matches 
'' ' 
DEF showGreedReg (): 
    greedReg = the re.compile (R & lt ' (HA) {3,5} ' ) 
    nonGreedReg = the re.compile (R & lt ' (HA) ? {3,5} ' ) 
    InP = ' hahahahahah ' 
    R1 = greedReg.search (InP) 
    R2 = nonGreedReg.search (InP)
     Print ( ' Greed RES REG: ' + r1.group ())
     Print ( ' nongreed REG RES : ' + r2.group ())

showGreedReg()

View Code

 

Chapter 7 project for the phone number and email regular extraction, the clipboard section omitted here.

 1 import pyperclip, re
 2 phoneReg=re.compile(r'''(
 3 (\d{3}|\(\d{3}\))?   #area code
 4 (\s|-|\.)?     #separator
 5 (\d{3})    #first 3 digits
 6 (\s|-|\.)?     #separator
 7 (\d{4})    #last 4 digits
 8 (\s*(ext|x|ext.)\s*(\d{2,5}))?
 9 )''', re.VERBOSE
10     )
11 
12 emailReg=re.compile(r'''(
13 [a-zA-Z0-9_-]+    #username
14 @    #@
15 [a-zA-Z0-9_-]+    #domain name
16 (\.[a-zA-Z]{2,4})
17 )''', re.VERBOSE
18     )

View Code

A phone number from "optional" area code beginning, the area code followed by a question mark packets.

Because the code may only three digits (i.e., \ d {3}), or three figures in parentheses (i.e., \ (\ d {3} \)), so that the two parts should be connected with the pipe symbol.

This part of multi-line strings can be combined with regular expression comment # Area code, to help you remember (\ d {3} | \ (\ d {3} \))? What is to be matched Yes.
Telephone numbers may be divided character spaces (\ S), a dash (-) (.) Or periods, these portions should be connected by piping.

The following regular expression is simple parts: 3 digits, followed by another delimiter, followed by four digits.

The last part is optional extension, including any number of spaces,
then ext, x or EXT., Then followed by 2-5 digits.

 

E-mail user name part address one or more characters, which may include: lower and upper case letters, numbers, dot, underscore, percent sign, plus or dashes.

All of these can be classified into a character: [a-zA-Z0-9 ._ % + -].
Splitting name and username @ symbol domain to allow less classified characters, only letters, numbers, and dashes periods: [a-zA-Z0-9.-] .

The last is the "dot-com" part (technically called "top-level domain"), it can actually be "dot-anything". It has 2-4 characters.

re.VERBOSE, ignoring the regular expression string whitespace and comments

At this point, the end of the contents of Chapter VII, practical projects strong passwords to detect the next issue of blog