Switch branches/tags
Nothing to show
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Type Name Latest commit message Commit time
Failed to load latest commit information.


rere: regex redone

from rere import *

money_regex = Exactly('$') + Digit*2 + (Exactly('.') + Digit*2).zero_or_one

money_regex.match('$23.95') # ==> MatchObject(...)

Isn't this better than re.compile('\\$\\d\\d(\\.\\d\\d)?')?


Run the following command to install:

pip install rere

This may require root (sudo).

Python 2.7+ and 3.3+ are supported.


To get started using rere, you need to know the logic of the regular expression pattern that you wish to build. To learn more about regular expressions and their usage, please visit Wikipedia: Regular Expression.

Once you know what sort of pattern you wish to match strings against, you can use rere to automatically generate the string patterns that you wish to use. Additionally, there is functionality built in to rere to call Python's built-in re library to do the matching for you (match() or match_prefix).

See above for the example.


Regex Components

The following components can be used individually, or added together (with +) create compound regexes.


  • string: the string that is exactly what you want to match against

Use exactly to describe a part of a regex that you wish to be the exact string of your choosing.

For example, if you want to match for the exact string, 'cat',

regex = Exactly('cat')
regex.match('cat') # ==> MatchObject(...)
regex.match('Cat') # ==> None
regex.prefix_match('catapult') # ==> MatchObject(...)
regex.prefix_match('bobcat') # ==> None

Exactly takes care of any required escaping, so you can do things like:

regex = Exactly('$2.00\n')
regex.match('$2.00\n') # ==> MatchObject(...)

(If you had to write a raw regex for the above, it might look something like re.compile('\\$2\\.00\\\n'). Ew.)



Use AnyChar when you want to match any single character (special or otherwise, including newlines).

regex = Exactly('hello') + AnyChar
regex.match('hello!') # ==> MatchObject(...)
regex.match('hello1') # ==> MatchObject(...)
regex.match('hello!!') # ==> None
regex.match('hello\n') # ==> MatchObject(...)



Use Digit when you want to match any single digit (from 0 to 9).

regex = Exactly('hello') + Digit
regex.match('hello!') # ==> None
regex.match('hello1') # ==> MatchObject(...)
regex.match('hello09') # ==> None



Use Letter when you want to match any English letter (case insensitive).

regex = Exactly('hello') + Letter
regex.match('helloB') # ==> MatchObject(...)
regex.match('hellob') # ==> MatchObject(...)
regex.match('hello9') # ==> None
regex.match('hello\n') # ==> None
regex.match('helloBb') # ==> None



Use Whitespace when you want to match whitespace ([ \t\n\r\f\v]).

regex = Exactly('hi') + Whitespace
regex.match('hi ') # ==> MatchObject(...)
regex.match('hi\n') # ==> MatchObject(...)
regex.match('hi b') # ==> None



Use Anything when you want to match absolutely anything (special or otherwise, including newlines). The empty string will also be matched.

regex = Exactly('hello') + Anything
regex.match('hello!') # ==> MatchObject(...)
regex.match('hello!!') # ==> MatchObject(...)
regex.match('hello\n') # ==> MatchObject(...)
regex.match('Hellohello') #==> None


  • pattern: a string containing a raw regex (using the syntax from re)

Simply match the provided regular expression. This allows you to use legacy regexes within rere expressions.

For example, if you have an existing regex for phone numbers (like r"\(\d\d\d\) \d\d\d-\d\d\d\d"), and you want to match one or more of them:

regex = RawRegex(r"\(\d\d\d\) \d\d\d-\d\d\d\d").one_or_more

Combining Components

All regex components implement several common functions. They can be combined and nested in many ways, such as:

regex = (Exactly('cat') + Exactly('dog').zero_or_one).one_or_more
regex.match('catcatdogcatdogcatdog') # ==> MatchObject(...)
regex.match('catdogdog') # ==> None


Use the zero_or_one property to describe how many repetitions of a string are required to match the pattern, in this case, only zero or one.

regex = Exactly('ab').zero_or_one
regex.match('aba') # ==> None
regex.match('ab') # ==> MatchObject(...)
regex.match('') # ==> MatchObject(...)


Use the zero_or_more property to describe how many repetitions of a string are required to match the pattern, in this case, any number (zero or more).

regex = Exactly('ab').zero_or_more
regex.match('ababab') # ==> MatchObject(...)
regex.match('ab') # ==> MatchObject(...)
regex.match('') # ==> MatchObject(...)
regex.match('aba') # ==> None


Use the one_or_more function to describe how many repetitions of a string are required to match the pattern, in this case, at least one.

regex = Exactly('ab').one_or_more
regex.match('ababab') # ==> MatchObject(...)
regex.match('ab') # ==> MatchObject(...)
regex.match('') # ==> None
regex.match('aba') # ==> None


  • name: the name of your group

You can assign a your regex part to a group. This allows those who want to use re's group functionality an easy way of working with it.

For example, say you want to group dollars and cents separately for a money regex.

regex = (Exactly('$') + Digit.one_or_more.as_group('dollars') +
         Exactly('.') + (Digit * 2).as_group('cents'))
match = regex.match('$24.13')
match.groupdict() # ==> {'dollars': '24', 'cents': '13'}

Addition (+)

You can form a regex from separate parts and combine them together with the + sign.

regex = Exactly('cat') + Exactly('dog')
regex.match('catdog') # ==> MatchObject(...)

Multiplication (*)

If you want a part (or a full) regex to be repeated a specified number of times, use the * sign.

regex = Exactly('cat') * 2
regex.match('catcat') # ==> MatchObject(...)

Or (|)

If need "Either or" logic for your regex, use |.

regex = Exactly('cat') | Exactly('dog')
regex.match('cat') # ==> MatchObject(...)
regex.match('dog') # ==> MatchObject(...)
regex.match('fish') # ==> None