On Fri, May 28, 2010 at 12:44, Uri Guttman <[email protected]> wrote:
>>>>>> "CO" == Chas Owens <[email protected]> writes:
>
> CO> On Fri, May 28, 2010 at 01:05, Uri Guttman <[email protected]> wrote:
> CO> snip
> >> $foo = $string =~ /^([^-])+-/ ? $1 : '' ;
> >>
> >> that will grab something from the start to the first - and grab it. if
> >> it matched it will assign it to $foo, otherwise assign ''. (and '' is
> >> called the null string, not null. perl has no null things unlike
> >> databases).
> CO> snip
>
> CO> This seems like an overcomplicated regex to me, what is the benefit of
> CO> doing this over
>
> CO> my ($foo) = $string =~ /(.*?)-/; # $foo will be undef if there is
> CO> no match
>
> the negated char class is usually faster than most similar methods. i
> just like it as it says what i really want - a string without any - chars.
> also anchoring helps too in saying this string must be at the beginning
> (he wants the first field).
Really? I have never found it so. Could you explain what is wrong
with my benchmark (I get similar results on 5.12)?
Perl version 5.008008
anchored_nongreedy => foo
charclass => foo
nongreedy => foo
foo:
Rate charclass anchored_nongreedy nongreedy
charclass 748127/s -- -5% -7%
anchored_nongreedy 783752/s 5% -- -3%
nongreedy 805306/s 8% 3% --
foo-:
Rate charclass anchored_nongreedy nongreedy
charclass 750868/s -- -3% -10%
anchored_nongreedy 778073/s 4% -- -6%
nongreedy 829906/s 11% 7% --
xxxx:
Rate charclass anchored_nongreedy nongreedy
charclass 753628/s -- -4% -8%
anchored_nongreedy 783752/s 4% -- -5%
nongreedy 822940/s 9% 5% --
xxx-:
Rate charclass anchored_nongreedy nongreedy
charclass 759211/s -- -2% -7%
anchored_nongreedy 778073/s 2% -- -4%
nongreedy 813440/s 7% 5% --
#!/usr/bin/perl
use strict;
use warnings;
use Benchmark;
print "Perl version $]\n";
my $s = "foo-";
my %subs = (
nongreedy => sub {
my ($x) = $s =~ /(.*?)-/;
return $x;
},
anchored_nongreedy => sub {
my ($x) = $s =~ /^(.+?)-/;
return $x;
},
charclass => sub {
my ($x) = $s =~ /^([^-]+)-/;
return $x;
},
);
for my $k (keys %subs) {
print "$k => ", $subs{$k}(), "\n";
}
for $s ( "foo", "foo-", "x" x 10_000, ("x" x 10_000) . "-") {
print "\n", substr($s, -4, 4), ":\n";
Benchmark::cmpthese -1, \%subs;
}
--
Chas. Owens
wonkden.net
The most important skill a programmer can have is the ability to read.
--
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
http://learn.perl.org/
