Weird hash access problem

tehjord has asked for the wisdom of the Perl Monks concerning the following question:

Hi, im making some cheap chess game, but I got a very weird problem when accessing an hash using $1 following a regex match. The regex used works perfectly, but the problem is, when I try to access my hash table using the $1 var, it simply does not work. Ive tried first copying $1 to a temp var and then using that temp var to access my hash but with no success. Here is my code which should explain everything:

use strict;

my %h = (a1 => 1);

my $test = "a1 to a2";

if ($test =~ /([a-hA-H][1-8]\s*)to(\s*[a-hA-H][1-8])/)
{
    print "test work\n"; 
}

print $h{$1}; #print nothing
print $h{a1}; #print 1

print $1; #print "a1"
[download]

THX FOR THE REPLIES ! works !

Comment on Weird hash access problem Download Code

Replies are listed 'Best First'.
Re: Weird hash access problem by toolic (Bishop) on May 26, 2011 at 18:31 UTC
Try this (Basic debugging checklist, Tip #3): `print ">>>$1<<<"; #print "a1"` [download] You will see that there is a space after a1: `>>>a1 <<<` [download] 'a1 ' is not the same as 'a1' as a hash key. You should also use warnings because it tells you which line is fishy. See also: chess	[reply] [d/l] [select]
Re: Weird hash access problem by wind (Priest) on May 26, 2011 at 18:30 UTC
Change your regex, you're capturing the spacing: `if ($test =~ /([a-hA-H][1-8])\sto\s([a-hA-H][1-8])/) {` [download]	[reply] [d/l]
Re: Weird hash access problem by kennethk (Abbot) on May 26, 2011 at 18:30 UTC
`$1` does not contain what you think it does. If you modify your code to wrap your output in a delimiter: `use strict; my %h = (a1 => 1); my $test = "a1 to a2"; if ($test =~ /([a-hA-H][1-8]\s)to(\s[a-hA-H][1-8])/) { print "test work\n"; } print "x$h{$1}x\n"; #print nothing print "x$h{a1}x\n"; #print 1 print "x$1x\n"; #print "a1"` [download] you get: `test work xx x1x xa1 x` [download] You are grabbing whitespace in addition to your square id. You'll get your expected result if you modify your regular expression to: `/([a-hA-H][1-8])\sto\s([a-hA-H][1-8])/`	[reply] [d/l] [select]
Re: Weird hash access problem by ikegami (Patriarch) on May 26, 2011 at 18:32 UTC
`print $1; #print "a1"` No, it prints "a1 ". Your parens are misplaces in your regex pattern. ... Too slow. But wow, 4 replies in a span that could be as little as 61 seconds!	[reply] [d/l]
Re: Weird hash access problem by jfroebe (Parson) on May 26, 2011 at 18:32 UTC
The scope of "$1" with respect to the regular expression is within the if block IIRC. Who knows what "$1" is outside of if block. Btw, you forgot "use warnings;" after "use strict;" Jason L. Froebe Blog, Tech Blog	[reply]
Re^2: Weird hash access problem by ikegami (Patriarch) on May 26, 2011 at 18:33 UTC
It's being used in the same block as the one in which it's set.	[reply]
Re^3: Weird hash access problem by jfroebe (Parson) on May 26, 2011 at 19:07 UTC
Interesting. I would have thought that the scope of what is in the if (...) would be only within the if (...) { } code block. The fact that it works doesn't mean the code should be written that way IMHO. Jason L. Froebe Blog, Tech Blog	[reply]