Thread by @reuvenmlerner

Many #Python beginners are confused by space characters, empty strings, and the term "whitespace." After all, how can nothing be something? (Or: How can something be nothing?)

A thread about nothing!

(Cue the Seinfeld theme, I guess...)

Given:

s = 'a b'

This string contains three characters. As humans, we only see the "a" and "b", two characters separated by a space.

But computers don't work that way: The space character is a character. It takes up just as much space in memory as either "a" or "b".

We can see this if we iterate over the characters in s, printing their Unicode numbers:

>>> s = 'a b'
>>> for c in s:
... print(ord(c))
...
97
32
98

The empty string, by contrast, is ... empty, with zero length. It is not the same as the space character, as we can see here:

>>> s = ''
>>> s = ' '
>>> len(s)
1
>>> s = ''
>>> len(s)
0
>>> '' == ' '
False

Another example: Empty strings are False in a boolean context. All other strings, including space, are True:

>>> for s in ['', ' ']:
... if s:
... print(f'"{s}" is True-ish')
... else:
... print(f'"{s}" is False-ish')
...
"" is False-ish
" " is True-ish

You can use ' ' when splitting strings, but not '':

>>> s = 'ab cd ef'
>>> s.split(' ')
['ab', 'cd', 'ef']
>>> s.split('')
Traceback (most recent call last):
File "<stdin>", line 1, in <module>
ValueError: empty separator

(We'll return to str.split in a bit.)

Space is just one of several characters that Python calls "whitespace." These characters affect how things are printed, but we only see them as empty space. The others are:

\n - newline, aka line feed
\r - carriage return
\t - tab
\v - vertical tab

When computers output to printers, "going down one line" meant two separate actions: (1) Returning the print head to and (2) Descend one line. Thus, two separate characters, carriage return + line feed.

Windows still uses CR+LF for "end of line." Unix uses just LF.

Also: Typewriters had "tabs," indicating commonly used columns. Pressing the "tab" key jumped to the next tab.

Today, "tab" (\t) moves you to the next column that's a multiple of 8; run this code to see:

>>> print('\t'.join('abcdefg'))
>>> print('0123456789' * 5)

Each of these whitespace characters has its own Unicode code:

>>> for c in ' \r\n\t\v':
... print(f'{ord(c)}')
...
32
13
10
9
11

None of these is the same as the empty string. And all are True in a boolean context (e.g., "if" or "while").

BTW, I've never used \v.

By default, the str.strip method removes all whitespace (any combination of these characters) from the start and end of a string:

>>> s = '\r\n\v a b c\t \r\n'
>>> s.strip()
'a b c'

Note: (1) Strings are immutable, so s is unaffected, and (2) It ignores internal whitespace.

Another place we see whitespace is in the "str.split" method, which returns a list of strings. For example:

>>> s = 'a b c'
>>> s.split(' ')
['a', 'b', 'c']

Here, ' ' was used as the separator.

But what about this:

>>> s = 'a b c'

>>> s.split(' ')
['a', '', 'b', '', 'c']

We told Python that wherever it sees a space character, ' ', it should cut and create a new list element. So it did.

Solve this by calling str.split with no arguments:

>>> s.split()
['a', 'b', 'c']

This uses all whitespace — any length, any combination — as a field separator.

If you use regular expressions, you can describe "any one whitespace character" with the \s metacharacter. Similarly, \S means "anything *but* a whitespace character."

What?!? You don't know regular expressions? Learn them for free: RegexpCrashCourse.com

This thread is the result of a question someone asked in this week's "Python for non-programmers" corporate training, wondering what I meant by "whitespace," and how it's different from space and the empty string.

Anything about whitespace I didn't cover here? Ask away!

Oh, I forgot to mention the str.isspace method. It returns True if the string contains only whitespace characters:

>>> s = ' \n\r\v\t'
>>> s.isspace()
True
>>> s = 'a \n\r\v\tb'
>>> s.isspace()
False