Python uses for re module for regular expressions. It usually starts out like prog = re.compile(pattern), and then running a method like match() or search() on prog.
The following is a complete example which finds a phone number in a larger string.
prog = re.compile(r'(\d{3})[-).\s][\s]?(\d{3}[-.\s]?\d{4})')
number = prog.search('This is my phone number: 123-4567-8901. Call me.')
print(number) # 123-4567-8901Very handy regex playground: https://pythex.org/.
findall() Looks for all occurances of a patterh and returns a list. If returned list contains groups, those groups will be in tuples. E.g. re.compile(r'a').findall('a dog is a type of animal') would return ['a', 'a', 'a', 'a'] as there are four "a's" in the searched string.
search() Looks for the first occurance of a pattern and returns a Match object. E.g. re.compile(r'a').search('a dog is a type of animal') would return a Match object with property match='a'.
match() Works like search() but checks only at the beginning of a string. (search() checks everywhere.)
{x,y} matches a number of occurances of the preceeding expression at a minimum of x and a maximum of y. If only one value is given, the match will operate on that number of occurances. {x,} will match any number of occurances equal to or greater than x, and {,y} will match 0 to the value of y occurances. The returned value is the longest possible match by default; when followed by a ?, as in {x,y}?, it returns the shortest.
[] matches characters inside the brackets. Emphasis on "characters", not "string". [abc] would look for a, b, and c, not abc. Special characters (?, *, etc.) are taken literally inside [] and do not need to be escaped. It can also look for ranges of whole numbers and letters. [0-5] matches numbers between and including 0 and 5; [a-z] matches all lowercase letters; [a-zA-Z] matches all lowercase and uppercase letters.
? Preceeding part is optional. E.g. wolf(-)?dog will match both wolf-dog and wolfdog.
* Preceeding part repeats zero or multiple times. E.g. English(Pera)?Language will match EnglishLanguage, EnglishPeraLanguage, and EnglishPeraPeraPeraLanguage.
+ It's like * but requires at least one occurance.
^ When placed at the beginning of a regex, indicates that match must begin with the following pattern. E.g. r'^hello' will return True when searching the string hello world but not greetings and hello.
$ When placed at the end of a regex, indicates that match much end with the following pattern. E.g. r'$world' will return True when searching the string hello world but not hello world goodbye.
. Matches any character except for newline.
.* Matches everything except for newline, returning the largest thing found. To return smallest thing found, use .*?. To include newline, use re.compile('.*', re.DOTALL).
There are a number of built-in character classes:
\d Any number from 0 to 9
\D Any character that is not a number from 0 to 9
\w Any letter, number, or the underscore character
\W Any character that is not a letter, number, or the underscore character
\s Any space, tab, or newline character
\S Any character that is not a space, tab, or newline