Substitute multiple whitespace with single whitespace in Python [duplicate]

Solution 1:

A simple possibility (if you'd rather avoid REs) is

' '.join(mystring.split())

The split and join perform the task you're explicitly asking about -- plus, they also do the extra one that you don't talk about but is seen in your example, removing trailing spaces;-).

Solution 2:

A regular expression can be used to offer more control over the whitespace characters that are combined.

To match unicode whitespace:

import re

_RE_COMBINE_WHITESPACE = re.compile(r"\s+")

my_str = _RE_COMBINE_WHITESPACE.sub(" ", my_str).strip()

To match ASCII whitespace only:

import re

_RE_COMBINE_WHITESPACE = re.compile(r"(?a:\s+)")
_RE_STRIP_WHITESPACE = re.compile(r"(?a:^\s+|\s+$)")

my_str = _RE_COMBINE_WHITESPACE.sub(" ", my_str)
my_str = _RE_STRIP_WHITESPACE.sub("", my_str)

Matching only ASCII whitespace is sometimes essential for keeping control characters such as x0b, x0c, x1c, x1d, x1e, x1f.

Reference:

About \s:

For Unicode (str) patterns: Matches Unicode whitespace characters (which includes [ \t\n\r\f\v], and also many other characters, for example the non-breaking spaces mandated by typography rules in many languages). If the ASCII flag is used, only [ \t\n\r\f\v] is matched.

About re.ASCII:

Make \w, \W, \b, \B, \d, \D, \s and \S perform ASCII-only matching instead of full Unicode matching. This is only meaningful for Unicode patterns, and is ignored for byte patterns. Corresponds to the inline flag (?a).

strip() will remote any leading and trailing whitespaces.

Solution 3:

For completeness, you can also use:

mystring = mystring.strip()  # the while loop will leave a trailing space, 
                  # so the trailing whitespace must be dealt with
                  # before or after the while loop
while '  ' in mystring:
    mystring = mystring.replace('  ', ' ')

which will work quickly on strings with relatively few spaces (faster than re in these situations).

In any scenario, Alex Martelli's split/join solution performs at least as quickly (usually significantly more so).

In your example, using the default values of timeit.Timer.repeat(), I get the following times:

str.replace: [1.4317800167340238, 1.4174888149192384, 1.4163512401715934]
re.sub:      [3.741931446594549,  3.8389395858970374, 3.973777672860706]
split/join:  [0.6530919432498195, 0.6252146571700905, 0.6346594329726258]

EDIT:

Just came across this post which provides a rather long comparison of the speeds of these methods.