Regex.py

import re
message = 'Call me at the nuber 435-333-3333 tomorrow, or at 432-231-3433'

phoneNumRegex = re.compile(r'\d\d\d-\d\d\d-d\d\d\d')
mo = phoneNumRegex.search(message)   #stops at the first occurence. Returns some find regarding the found match
mo = phoneNumRegex.findAll(message)  #finds all of them

#mo is None if no maches have been found
print(mo.group())  #search and return the matches


#use the paranthesis to group the different parts; if the paranthesis are part of the path we are looking for they must be back-slashed
message = 'Call me at the nuber 435-333-3333 tomorrow'
phoneNumRegex = re.compile(r'(\d\d\d)-(\d\d\d-d\d\d\d)')
mo = phoneNumRegex.search(message)   #stops at the first occurence. Returns some find regarding the found match
print(mo.group())   #search and return the matches
mo.group(1)
mo.group(2)


#search using the pipe for
batRegex re.compile(r'Bat(man|mobile|copter')
mo.group(1)     #to know which one among the different keywords match the string

#####################
####### Repetion ####
#####################

#match a specific number of repetions with ? * +. All these special characther should be escaped if we need them as part of the path
batRegex = re.compile(r'Bat(wo)?man')  #wo can appear 0 or 1 time
phoneNumRegex = re.compile(r'(\d\d\d)?-\d\d\d-d\d\d\d')

batRegex = re.compile(r'Bat(wo)*man')  #wo can appear 0 or more time
batRegex = re.compile(r'Bat(wo)+man')  #wo can appear 1 or more time

batRegex = re.compile(r'Bat(wo){3}man')  #wo can appear 1 or more time
message = 'My nubers are 435-333-3333,432-231-3433,432-231-3433'
phoneNumRegex = re.compile(r'((\d\d\d-)?\d\d\d-d\d\d\d(,)?){3}')   #match 3 phone numbers in a row, optionally separated with a comma, optionally having an area code

batRegex = re.compile(r'Bat(wo){3,5}man')  #wo can appear 3 to 4 times
batRegex = re.compile(r'Bat(wo){3,}man')   #wo can appear 3 or more times


sMessage = '123456789'
digitRegex = re.compile(r'(\d){3,5})')   #greedy match by default. group() will match 12345
digitRegex = re.compile(r'(\d){3,5}?)')  #non greedy match. It will return 123 as a match


#############################
####### Search VS FinaAll ###
#############################

mo = phoneNumRegex.search(message)   #returns a search object
mo = phoneNumRegex.findAll(message)  #if there are no groups returns a list with the matched patterns
                                     #if there are groups then it returns a list of tuples, each tuple contains what is matched per each group;
                                     #it doesn't return the matches


#############################
####### Character classes ###
#############################

#\d, \D, \w, \W, \s, \S -> character classes
#search for all the words which are folloing a number
re.compile(r'\d+\s\w+')

#[aiou] -> custom character class
mo = re.compile(r'[aeiou]')  #match whatever of this letters. Same as r('a|e|i|o|u')
mo = mo.findAll(message)     #returns list of matching vowels

re.compile(r'[a-fA-F]')
re.compile(r'[ˆaeiou]')  #match whatever is not one of these letters

re.compile(r'ˆHello')  # match if the string begins with hello
re.compile(r'World$')  # match if the string begins with hello

#############################
####### Match whater char ###
#############################

#whatever character but a new line
re.compile(r'.at')  #matches lat, rat, fat...

message = r'<To serve humans> for dinner.>'
mo = re.compile(r'<(.*)>')  #the match is To serve humans> for dinner.
mo = re.compile(r'<(.*?)>') #to have the non-greedy match, which will return To server humans

re.compile(r'.*',re.DOTALL) #with this option it will be matched whatever char, includig the new line
mo = re.compile(r'[aeiou]', re.I)  #match whatever of this letters. Low or upper case because of the flag case insensitive

#############################
####### Find and replace  ###
#############################

sMessage = 'Agent Fabio and Agent Elisa'
mo = re.compile(r'Agent \w+')
mo.sub('Replacement', sMessage)  #automatically findall the matches of the pattern (Agent Fabio, Agent Elisa) and replacing them with the string "Replacement"

#matches in these case will be still (Agent Fabio, Agent Elisa) but the findall returns ony (F, E).
#We return as match only the content of the group!!
mo = re.compile(r'Agent (\w)\w*')
#replace Agent Fabio with Agent F*****, and Agent Elisa with Agent E*****
#\1 indicates the match we have in the group1
mo.sub(r'Agent \1****', sMessage)

#to add comments or split the regular exp we can use verborse flag
#with the verbose flag the spaces are not considered as part of the pattern
#and we can also add comments
re.compile(r'''
    (\d\d\d-) |   #area code
    (\d(\d\d\d\d) #comment1
    \d\d\d''', re.VEROBSE)

#to pass multiple options we can use the bitwise OR operator