Re: [GeSHi-users] Getting Geshi to recognize multi-character SYMBOLS for perl.

[Sir R. B. <rbu...@ci...>]

Benny Baumann wrote:
> There actually are two ways to do this with GeSHi.
>
> The first way is using COMMENT_REGEXP to match those magics (only those
> $# given they are in front of a keyword) and thus hide them from further
> processing. This can be done by something like this:
>
> 42 => '/(?<!\w)\$\#(?=\w)/',
>
> where 42 is any still free index in the comment_regexp array of the
> perl.php language file.
>
> Second version, you define a symbol group for multi-char symbols (like
> $_, $#, ...) and place its definition BEFORE the existing single char
> symbol groups. This is to ensure GeSHi uses the multi-char symbols with
> higher priority than single-char ones, as the processing of symbols is
> ordered by definition of the groups they belong to.
>
> Could you provide me with a list of these additional multi-char symbols
> (as complete as possible), so I can do a fix in the release?
>
> Regards,
> BenBE.
>
> Sir Robert Burbridge schrieb:
>   
>> Hey all,
>>
>> I've just started using Geshi -- it was a dependency of the
>> "codecolorer" plugin for WordPress -- and I'm impressed so far.  I
>> have a fledgling code blog for which I was trying out some syntax
>> highlighters and I came across this apparent limitation in Geshi:  in
>> highlighting perl code, it seems to recognize the basic "symbols" in
>> perl ($, @, %, etc.) but I couldn't get it to recognize perl's "$#"
>> symbol; the "last index" symbol.
>>
>> You can see an example of the (apparent) failure here:
>>     http://tech.ification.org/?p=12
>>
>> Notice that in the snippet the variables are all highlighted except
>> for occurrences of $#arr:
>>
>>     ### Create an array.
>>     my @arr = ('a', 'b', 'c', 'd', 'e');
>>     ### $length is now 5
>>     my $length = scalar(@arr);
>>     ### $lastIndex is now 4
>>     my $lastIndex = $#arr;
>>     ### Get the last element of the array:  'e'
>>     my $el = $arr[$#arr];
>>     ### Get the fourth-to-last element:  'b'
>>     my $el = $arr[$#arr - 4];
>>
>> For some reason those are missed.  I peeked into the code and have
>> already opened the perl.php language file and modified it thus:
>>
>>     131     'SYMBOLS' => array(
>>     132         '$#', '<', '>', '=',
>>     133         '!', '@', '~', '&', '|',
>>     134         '+','-', '*', '/', '%',
>>     135         ',', ';', '?', ':'
>>     136         ),
>>
>> (notice the new first element of the array).  This wasn't sufficient
>> to implement it.  I'm no php whiz, but I can get by in it.  Can
>> someone offer me some advice about how to get perl's extra magic
>> recognized?
>>
>> Thanks!
>>
>> -Sir Robert
>>
>>     
Thanks, Benny. 

I went ahead and put together a complete (I hope!) expo of perl syntax 
that is likely to be relevant here.  It shows Geshi in action against a 
slew of perl black magic.  You can see it here 
(http://tech.ification.org/?p=63).  I didn't include it in this message, 
as it would be pretty useless.

Also, I took the liberty of addressing some of the things I noticed at 
the top of the post.  I'm not familiar with Geshi's inner workings, so 
some of these might be desired behavior, or not addressable here.  Also, 
I pointed out some possible regexp patterns that could be used to 
alleviate a few of the less straightforward issues I noticed.

Let me know if there's anything more I can do!

Thanks again,

-Sir Robert