/ 
HTML-Truncate-0.12
River stage one • 1 direct dependent • 1 total dependent
/ HTML::Truncate

NAME

HTML::Truncate - (beta software) truncate HTML by percentage or character count while preserving well-formedness.

VERSION

0.12

ABSTRACT

When working with text it is convenient and common to want to truncate strings to make them fit a desired context. E.g., you might have a menu that is only 100px wide and prefer text doesn't wrap so you'd truncate it around 15-30 characters, depending on preference and typeface size. This is trivial with plain text using substr but with HTML it is somewhat difficult because whitespace has fluid significance and open tags that are not properly closed destroy well-formedness and can wreck an entire layout.

HTML::Truncate attempts to account for those two problems by padding truncation for spacing and entities and closing any tags that remain open at the point of truncation.

NOTA BENE

Previous versions did not truncate characters to any good degree of accuracy so the internals were rewritten a bit starting with 0.12. It might still not be ideal but it is much closer than previously.

Backwards compatibility from 0.11 and earlier is broken in renaming the method "utf8" to "utf8_mode".

SYNOPSIS

use strict;
use HTML::Truncate;

my $html = '<p><i>We</i> have to test <b>something</b>.</p>';
my $readmore = '... <a href="/full-article">[readmore]</a>';

my $html_truncate = HTML::Truncate->new();
$html_truncate->chars(20);
$html_truncate->ellipsis($readmore);
print $html_truncate->truncate($html);

# or

use Encode;
my $ht = HTML::Truncate->new( utf8_mode => 1,
                              chars => 1_000,
                             );
print Encode::encode_utf8( $ht->truncate($html) );

XHTML

This module is designed to work with XHTML-style nested tags. More below.

WHITESPACE AND ENTITIES

Repeated natural whitespace (i.e., "\s+" and not " &nbsp; ") in HTML -- with rare exception (pre tags or user defined styles) -- is not meaningful. Therefore it is normalized when truncating. Entities are also normalized. The following is only counted 14 chars long.

\n<p>\nthis     is   &#8216;text&#8217;\n\n</p>
^^^^^^^12345----678--9------01234------^^^^^^^^

METHODS

  • HTML::Truncate->new

    Can take all the methods as hash style args. "percent" and "chars" are incompatible so don't use them both. Whichever is set most recently will erase the other.

    my $ht = HTML::Truncate->new(utf8 => 1,
                             chars => 500, # default is 100
                             );

  • $ht->utf8_mode

    Set/get, true/false. If utf8_mode is set, utf8_mode(1) is also set in the underlying HTML::Parser, entities will be transformed with decode and the default ellipsis will be a literal ellipsis and not the default of &#8230;.

  • $ht->chars

    Set/get. The number of characters remaining after truncation, excluding the "ellipsis".

    Entities are counted as single characters. E.g., &copy; is one character for truncation counts.

    Default is "100." Side-effect: clears any "percent" that has been set.

  • $ht->percent

    Set/get. A percentage to keep while truncating the rest. For a document of 1,000 chars, percent('15%') and chars(150) would be equivalent. The actual amount of character that the percent represents cannot be known until the given HTML is parsed.

    Side-effect: clears any "chars" that has been set.

  • $ht->ellipsis

    Set/get. Ellipsis in this case means --

    The omission of a word or phrase necessary for a complete
syntactical construction but not necessary for understanding.

    What it will probably mean in most real applications is "read more." The default is &#8230; which if the utf8 flag is true will render as a literal ellipsis, chr(8230).

    The reason the default is &#8230; and not "..." is this is meant for use in HTML environments, not plain text, and "..." (dot-dot-dot) is not typographically correct or equivalent to a real horizontal ellipsis character.

  • $ht->truncate($html)

    Also can be called with arguments--

    $ht->truncate( $html, $chars_or_percent, $ellipsis );

    No arguments are strictly required. Without HTML to operate upon it returns undef. The two optional arguments may be preset with the methods "chars" (or "percent") and "ellipsis".

    Valid nesting of tags is required (alla XHTML). Therefore some old HTML habits like <p> without a </p> are not supported and will cause a fatal error.

    Certain tags are omitted by default from the truncated output. There will be a mechanism to custom tailor these--

    skipped tags

    These will note be included in truncated output by default.

    <head>...</head> <script>...</script> <form>...</form>
<iframe></iframe> <title>...</title> <style>...</style>
<base/> <link/> <meta/>
    tags allowed to self-close

    See emptyElement in HTML::Tagset.

  • $ht->add_skip_tags( qw( tag list ) )

    Put one or more new tags into the list of those to be omitted from truncated output. An example of when you might like to use this is if you're thumbnailing articles and they start with <h1>title</h1> or such before the article body. The heading level would be absurd with a list of excerpts so you could drop it completely this way--

    $ht->add_skip_tags( 'h1' );

  • $ht->dont_skip_tags( qw( tag list ) )

    Takes tags out of the current list to be omitted from truncated output.

  • $ht->repair

    Set/get, true/false. If true, will attempt to repair unclosed HTML tags by adding close-tags as late as possible (eg. <i><b>foobar</i> becomes <i><b>foobar</b></i>). Unmatched close tags are dropped (foobar</b> becomes foobar).

  • $ht->on_space(1)

    This will make the truncation back up to the first space it finds so it doesn't truncate in the the middle of a word.

  • $ht->cleanly(qr/[\s[:punct:]]+\z/)

    This is on by default and the default cleaning regular expression is shown. It will make the truncation strip any trailing spacing and punctuation so you don't get things like "The End...." or "What? ..." You can cancel it with <$ht-cleanly(undef)>> or provide your own regular expression.

COOKBOOK (well, a recipe)

Template Toolkit filter

For excerpting HTML in your Templates. Note the "add_skip_tags" which is set to drop any images from the truncated output.

use Template;
use HTML::Truncate;

my %config =
   (
    FILTERS => {
        truncate_html => [ \&truncate_html_filter_factory, 1 ],
    },
    );

my $tt = Template->new(\%config) or die $Template::ERROR;

# ... etc ...

sub truncate_html_filter_factory {
    my ( $context, $len, $ellipsis ) = @_;
    $len = 32 unless $len;
    $ellipsis = chr(8230) unless defined $ellipsis;
    my $html = shift || return '';
    my $ht = HTML::Truncate->new();
    return sub {
        $ht->add_skip_tags(qw( img ));
        return $ht->truncate( $html, $len, $ellipsis );
    }
}

Then in your templates you can do things like this:

[% FOR item IN search_results %]
  <div class="searchResult">
    <a href="[% item.uri %]">[% item.title %]</a><br />
    [% item.body | truncate_html(200) %]
  </div>
[% END %]

See also Template::Filters.

AUTHOR

Ashley Pond V, <ashley@cpan.org>.

LIMITATIONS

There may be places where this will break down right now. I'll pad out possible edge cases as I find them or they are sent to me via the CPAN bug ticket system.

This is not an HTML filter

Although this happens to do some crude HTML filtering to achieve its end, it is not a fully featured filter. If you are looking for one, check out HTML::Scrubber and HTML::Sanitizer.

BUGS, FEEDBACK, PATCHES

Please report any bugs or feature requests to bug-html-truncate@rt.cpan.org, or through the web interface at http://rt.cpan.org/NoAuth/ReportBug.html?Queue=HTML-Truncate. I will get the ticket, and then you'll automatically be notified of progress as I make changes.

THANKS TO

Kevin Riggle for the "repair" functionality; patch, Pod, and tests.

Lorenzo Iannuzzi for the "on_space" functionality.

SEE ALSO

HTML::Entities, HTML::TokeParser, the "truncate" filter in Template, and Text::Truncate.

HTML::Scrubber and HTML::Sanitizer.

COPYRIGHT & LICENSE

Copyright 2005-2008 Ashley Pond V.

This program is free software; you can redistribute it or modify it or both under the same terms as Perl itself.

DISCLAIMER OF WARRANTY

Because this software is licensed free of charge, there is no warranty for the software, to the extent permitted by applicable law. Except when otherwise stated in writing the copyright holders or other parties provide the software "as is" without warranty of any kind, either expressed or implied, including, but not limited to, the implied warranties of merchantability and fitness for a particular purpose. The entire risk as to the quality and performance of the software is with you. Should the software prove defective, you assume the cost of all necessary servicing, repair, or correction.

In no event unless required by applicable law or agreed to in writing will any copyright holder, or any other party who may modify and/or redistribute the software as permitted by the above licence, be liable to you for damages, including any general, special, incidental, or consequential damages arising out of the use or inability to use the software (including but not limited to loss of data or data being rendered inaccurate or losses sustained by you or third parties or a failure of the software to operate with any other software), even if such holder or other party has been advised of the possibility of such damages.