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

Hi,

It's my first time playing with Text::Balanced module. From the code below (a quick test script I wrote..), shouldn't $next contain '#name#' after the call to extract_delimited() sub?

As of now, the subroutine simply sets $@ indicating this: "Not a delimited pattern, detected at offset 0".

My thinking was that the extract_delimited() sub would pull out 'first' delimited substring inside a given larger string. This is not the case, however ;-). So, for that matter, how do I do it using this module? I'm not sure which one of the imported methods I should use to achive the desired result (say, pull out next '#' delimited substring). 
use Text::Balanced qw (
                        extract_delimited
                        extract_bracketed
                        extract_quotelike
                        extract_codeblock
                        extract_variable
                        extract_tagged
                        extract_multiple
                        gen_delimited_pat
                        gen_extract_tagged
                       );

my $source = 'Hello World, #name#!';
my ($cfml_expr_match, $next);

#eval {
    $cfml_expr_match = gen_delimited_pat(q{#},q{#});
    $next = extract_delimited($source, '#', '', '#');
#};

if ($@) {
    print "Hmm.. got this: $@\n";
}

# some code here...

print "\n";

[download]


Note: I've uncommented eval{...} since it would clear the $@ variable, therefore, not letting me to check on the status of a call to a Text::Balanced method. Any idea why this is a case? Sorry for yet another loosely related question ;-).

"There is no system but GNU, and Linux is one of its kernels." -- Confession of Faith

Edit kudra, 2002-01-10 Changed Delimited to Balanced in title per ntc request

Replies are listed 'Best First'.
(crazyinsomniac) Re: using Text::Delimited...
by crazyinsomniac (Prior) on Jan 05, 2002 at 11:33 UTC
    I think you misunderstood the pod. From what I understand, the snippet above would extract stuff like ##. I think this is the behaviour your want (even note the eval dealy. Since there is only one $@ variable, and all eval calls set it, when you got nested ones, you gotta check $@ inside your eval). Enjoy. 
    #!/usr/bin/perl -w
use strict;
use Text::Balanced qw (
                        extract_delimited
                        extract_bracketed
                        extract_quotelike
                        extract_codeblock
                        extract_variable
                        extract_tagged
                        extract_multiple
                        gen_delimited_pat
                        gen_extract_tagged
                       );

my $source = 'Hello World, #name#!';
my ($cfml_expr_match, $next);

eval {
$cfml_expr_match = gen_delimited_pat(q{#},q{#});
$next = extract_delimited($source, '#','[^#]+','#');
print "There is only one \$@, so Inside my eval, .. got this: $@\n" if
+ $@;
};

if ($@) {
    print "Hmm.. got this: $@\n";
}
else
{
    print "CFML: $cfml_expr_match\n";
    print "next: $next\n";
}


# some code here...

print "\n";
__END__
CFML: (?:\#(?:[^\#]*(?:(?:\#\#)[^\#]*)*)\#)
next: #name#
## If I use your '#','','#' params, the error i'd get is
inside my eval, Hmm.. got this: Not a delimited pattern, detected at o
+ffset 0

    [download]

     
    ______crazyinsomniac_____________________________
    Of all the things I've lost, I miss my mind the most.
    perl -e "$q=$_;map({chr unpack qq;H*;,$_}split(q;;,q*H*));print;$q/$q;"

[reply]
[d/l]

Back to Seekers of Perl Wisdom