Specifying a Regular Expression for the Shortest Match

Contributed by: David Beazley and Brian K. Jones

Difficulty level: Easy

Visit the Python Cookbook: Strings and Text Lab to test out this Shortcut in a live practice environment.

Problem

You’re trying to match a text pattern using regular expressions, but the regular expression is identifying the longest possible matches of a pattern. Instead, you would like to change it to find the shortest possible match.

Solution

This problem often arises in patterns that try to match text enclosed inside a pair of starting and ending delimiters (e.g., a quoted string). To illustrate, consider this example:

>>> str_pat = re.compile(r'\"(.*)\"')
>>> text1 = 'Computer says "no."'
>>> str_pat.findall(text1)
['no.']
>>> text2 = 'Computer says "no." Phone says "yes."'
>>> str_pat.findall(text2)
['no." Phone says "yes.']
>>>

In this example, the pattern r'\"(.*)\"' is attempting to match text enclosed inside double quotes. However, the * operator in a regular expression is greedy, so matching is based on finding the longest possible match. Thus, in the second example involving text2, it incorrectly matches the two quoted strings.

To fix this, add the ? modifier after the * operator in the pattern, like this:

>>> str_pat = re.compile(r'\"(.*?)\"')
>>> str_pat.findall(text2)
['no.', 'yes.']
>>>

This makes the matching nongreedy and instead produces the shortest match.

Discussion

This recipe addresses one of the more common problems encountered ...