taking white space out between closing and opening tags

[chuleto1]

Problem:
I would like to delete all tab space "\t" new lines "\n " and more than one continuos space " " inside opening and closing tags and replacing them with ONE space per instance.
$text = "<tag>
      The purpose of the applicant rating session is for you,
the applicant, to provide a sample of your       
effective teaching skills.</tag>"


The desired result would be:

$text = "<tag>The purpose of the applicant rating session is for you, the applicant, to provide a sample of your effective teaching skills.</tag>"

[Ovid]

Here's a quick, untested, stab at it. Let's assume for this example that you are talking about <p> tags.

use HTML::TokeParser::Simple;

# assumes that $text is a scalar containing the actual HTML
my $p = HTML::TokeParser::Simple->new( \$text );

my $token;
do { $token = $p->get_token } until $token->is_start_tag('p');

my $new_text = $token->return_text;

do ( $token = $p->get_token ) {
    my $temp = $token->return_text;
    if ( $token->is_text ) {
        $temp =~ s/\s+/ /g;  # collapse whitespace
        $temp =~ s/^\s//;    # remove initial whitespace
        $temp =~ s/\s$//;    # remove trailing whitespace
    }
    $new_text .= $temp;
} until $token->is_end_tag('p');

$new_text .= $token->return_text;

This is a much cleaner method (and accurate) method of accomplishing this task than most regex solutions. I also happen to think that HTML::TokeParser::Simple is easier to use than many other HTML parsing modules. Of course, I may be biased as I wrote that module :)

Cheers,
Ovid

Join the Perlmonks Setiathome Group or just click on the the link and check out our stats.

[dpuu]

split the problem: first get the string, then condense it. Assuming you can't use any of te std XML/HTML modules to get the text, you could try:

sub condense { $_[0] =~ s/\s+/ /g }
$in =~ s/(<tag>)(.*?)(<\/tag>)/ $1 . condense($2) . $3 /ge;

--Dave

[Mr. Muskrat]

I'll help with the regex requirements.

#/usr/bin/perl -w
use strict;
my $text = "<tag>\n\tThe purpose of the applicant rating session is fo
+r you,\nthe applicant, to provide a sample of your\t\neffective teach
+ing skills.</tag>";
$text =~ s/\s+/ /g; # convert white space to a single space
print $text;

The print statement is just there to show what's taken place.
edited to match the given example...
Of course, this still leaves a space between the tag and the text... dpuu and Ovid both give better ways of doing it... but I wasn't paying attention.