Question

I am just looking for some advice on best practice. What is the best way to check for strings that are empty or include nothing but space character in them?

I generally use regex like this $string =~/^\s*$/ to see if the variable is blank. Is there a more accepted way or this is as good as any other? Is it better to use regex or to chomp and then check for empty string?

Thanks.

Answer 1

Personally, I generally invert the sense, and search for a single non-whitespace character. So I’d say something like:

next unless $string =~ /\S/;

As ikegami mentioned, \s (and \S) can be a bit idiosyncratic about exactly what characters are considered whitespace, and you can use \p{Whitespace} (or in my case \P{Whitespace}) for more consistent results.

Out of curiosity, I benchmarked not /\S/ versus /^\s*\z/. On an 80 character string (with Perl 5.14.2), I found not /\S/ was about 20% faster for strings of spaces, and /^\s*\z/ was about 30% faster for strings of non-spaces. It’s unlikely that checking for blank lines is the bottleneck in your program, but it certainly doesn’t hurt to use a simpler regexp.

use Benchmark 'cmpthese';

our $text = ' ' x 80;
#our $text = 'abcd' x 20;

cmpthese(-3, {
    'pos'  => '$text =~ /^\s*\z/',
    'neg'  => 'not $text =~ /\S/',
    'negU' => 'not $text =~ /\P{Whitespace}/',
});

Example results for spaces:

          Rate  pos negU  neg
pos  3562012/s   --  -2% -15%
negU 3651619/s   3%   -- -13%
neg  4194275/s  18%  15%   --

Example results for non-spaces:

          Rate negU  neg  pos
negU 4951234/s   --  -3% -24%
neg  5112271/s   3%   -- -22%
pos  6555251/s  32%  28%   --