boardom

[imeridian]

Hah! $15 is a bargain... I charge $65 an hour with a minimum of a half hour for *Anything*

[Gadget Girl]

Hi Indi I remember you from the GM and weren't you on the MT boards~good to see you over here.

Yeah, $15 is a bargin, it's hard when you are first starting out. I guess I can't use that line any more eh?

Sara

[imeridian]

I only go to the MT boards to complain about how their script wrecks servers by hanging up... or causing dependencies on perl module versions... and otherwise creating administrative havoc for me... *Grumbles*

[annia]

i think the hack is very easy to install.. :? where is the problem?

[Gadget Girl]

*LOL* Indi!

Thanks Annia, I thought it was too~

Sara

[annia]

eeewww.. i just found a bug in this hack.. if you write a link containing your friend's name like

[Aryana]

Wow, that's really cool annia. Thanks =)

Hm... would that really be a problem? I mean, why would you even have your friend's name in a link other than the auto-link?

[michel v]

Annia, that's because your code (and the original smilies code) doesn't check that we're not between < and > before it converts stuff.
That would require regular expressions, and I'm planning on adding that transparently in later versions.

In fact, I'm coding a little script to convert stuff into tags, and that should be a generic script to handle smilies, friends' names, acronyms etc. It's still in the debug stage, as the regex seems to be a bit too greedy at the moment. It does, however, check that we're not between < and > when converting, to avoid disasters.
It might be available as a "filter" for b2 0.x or built-in for 1.x.

It handles data in this form (could read it from a file or a string):

[Aryana]

[Gadget Girl]

Aryana, I'm sorry you had a problem with the tutorial! I thought I wrote it so it was easy to paste and figure out what went where!

Sara

[Aryana]

It's probably a problem with my main index.

[allusion]

Michael that would be extremely powerful. Is any of the code in the CVS yet? How far along are you?

[michel v]

Not very far.
In fact it works as expected on text that doesn't have html tags in it. Once I put html, it start fucking up, putting little <=""></> strings around.
I'll probably ask help for it on sitepointforums.com, if I don't have it working tomorrow.

[michel v]

OK, the code is there:
[php:1:f43f48134a]<?php

$tagifylist = trim('

"AOL" acronym title "America Online"
"API" acronym title "Application Interface"
"CGI" acronym title "Common Gateway Interface"
"CMS" acronym title "Content Management System"
"CSS" acronym title "Cascading Style Sheets"
"DNS" acronym title "Domain Name System"
"DOM" acronym title "Document Object Model"
"DTD" acronym title "Document Type Definition"
"DVD" acronym title "Digital Video Disc"
"FAQ" acronym title "Frequently Asked Questions"
"FOAF" acronym title "Friend of a Friend vocabulary"
"FSF" acronym title "Free Software Foundation"
"GFDL" acronym title "GNU Free Documentation License"
"GPL" acronym title "GNU General Public License"
"HDTV" acronym title "High Definition TeleVision"
"IE" acronym title "Internet Explorer"
"IIRC" acronym title "if I remember correctly"
"IIS" acronym title "Internet Infomation Server"
"JSP" acronym title "Java Server Pages"
"KB" acronym title "kilobytes"
"KDE" acronym title "K Desktop Environment"
"HTML" acronym title "HyperText Markup Language"
"LGPL" acronym title "GNU Lesser General Public License"
"MB" acronym title "megabyte"
"MSDN" acronym title "Microsoft Developer Network"
"MSIE" acronym title "Microsoft Internet Explorer"
"MSN" acronym title "Microsoft Network"
"OPML" acronym title "Outline Processor Markup Language"
"P2P" acronym title "Peer To Peer"
"PBS" acronym title "Public Broadcasting System"
"PDF" acronym title "Portable Document Format"
"PHP" acronym title "PHP Hypertext Processor"
"RDF" acronym title "Resource Description Framework"
"RSS" acronym title "Rich Site Summary"
"SOAP" acronym title "Simple Object Access Protocol"
"SSN" acronym title "Social Security Number"
"URL" acronym title "Uniform Resource Locator"
"VNC" acronym title "Virtual Network Computing"
"W3C" acronym title "World Wide Web Consortium"
"WCAG" acronym title "Web Content Accessibility Guidelines"
"WYSIWYG" acronym title "what you see is what you get"
"XHTML" acronym title "eXtensible HyperText Markup Language"
"XML" acronym title "eXtensible Markup Language"
"XSL" acronym title "eXtensible Stylesheet Language"
"XSLT" acronym title "eXtensible Stylesheet Language Transformation"

');

function tagifythis($text) {
global $tagifylist, $cache_tagifylist;

/* let's build an array of searches and replaces from
$tagifylist so that we don't have to regex match
everything every time we call this function */
if (empty($cache_tagifylist)) {
$tagifylist = explode("\n", $tagifylist);
foreach ($tagifylist as $tag_item) {
preg_match('#^"(.+?)"\s+(\w+)(.*)#', $tag_item, $blah);
$tag_match = $blah[1];
$tag_tag = $blah[2];
$tag_item = $blah[3];
preg_match_all('#\s(\w+)\s+"(.+?)"#', $tag_item, $blah, PREG_SET_ORDER);
$tag_attributes = '';
foreach($blah as $attribute) {
$tag_attributes .= ' '.$attribute[1].'="'.$attribute[2].'"';
}
$tag_match = '#>([^<]{0,})([^a-zA-Z0-9]{1})('
.preg_quote($tag_match)
.')([^a-zA-Z0-9]{1})([^>]{0,})<#';
$tag_replace = '>$1$2<'.$tag_tag.$tag_attributes.'>$3</'.$tag_tag.'>$4$5<';
if (strlen($tag_match)) {
$cache_tagifylist[] = array($tag_match, $tag_replace);
}
}
}

/* let's dance! */
foreach ($cache_tagifylist as $tag_item) {
$match = $tag_item[0];
$replace = $tag_item[1];
$text = '> '.$text.' <';
$text = preg_replace($match, $replace, $text, 1);
$text = substr($text, 1, -1);
}

return($text);
}

?>[/php:1:f43f48134a]

There are three (ok, two and a half) outstanding issues remaining.
1. It won't convert a string that is right between tags, like <i>CSS</i>, but it will convert <i> CSS </i> just fine.
2. On my server at least, it converts only the last match, instead of the first match.
3. I got lazy and forgot to put support for self-closing tags.

Other than that, it just works. You can see it in action on my blog.

The TODO list:
1. Fix the above three issues (number 3 should be quick).
2. Make it so that it won't convert a string into a tag if it's already surrounded by that tag (to avoid double <acronym><acronym> stuff for example, or double <a><a> links).

--
You can apply this function as a filter on your blog, if you are running the CVS version of b2.
It is quite simple, just open your template and paste this anywhere before the loop:
[php:1:f43f48134a]<?php
include('tagifythis.php');
add_filter('the_content', 'tagifythis');
?>[/php:1:f43f48134a]

[sidcarter]

Hi michel,

Have the bugs been fixed ? is the tagifythis.php file in CVS now ?

TIA