Sean Allen wrote:
> anyone done anything about getting html tidy to play nice with
> template toolkit syntax?
>
> don't want to redo what is already done, google isn't helping then again
>
> template toolkit tidy html
>
> in any combo bring up a ton of stuff isn't relevant so maybe its out
> there just lost in the noise.
>
If I understand your question right I've been doing something like this
in an application of mine:
sub tidy_html_string {
my $content=shift @_;
use Regexp::Common qw( balanced );
use MIME::Base64 ();
use Encode ();
use HTML::Tidy ();
my $tidy=HTML::Tidy->new();
# Obfuscate TT markup
# MIME::Base64 must be done on bytes, not perl character strings
$content=~s/$RE{balanced}{-begin => "[%"}{-end => "%]"}{-keep}/'tt_begin:'
. MIME::Base64::encode($1,'')
. ':tt_end'/ge;
# HTML::Tidy wants to work on perl character data
$content=Encode::decode_utf8($content);
$content=$tidy->clean($content);
$content=Encode::encode_utf8($content);
# Get back obfuscated TT markup
$content=~s/$RE{balanced}{-begin => "tt_begin:"}{-end =>
":tt_end"}{-keep}/
my $var=$1;
my $len=length($var);
MIME::Base64::decode(substr($var, 9, $len-16 ));
/ge if $content;
return $content;
}
If your input string is not binary UTF-8 text, some twiddling with the
encode/decode functions will be needed. Just remember that MIME::Base64
works on bytes, but HTML::Tidy likes to work on perl character strings.
If your input is plain 7-bit ASCII the encoding/decoding is a non-issue
and can be ignored.
Regards,
Robin Smidsrød
_______________________________________________
templates mailing list
[email protected]
http://mail.template-toolkit.org/mailman/listinfo/templates
