Discussion on 1.4 (Page 2 of 2)

Personally I would be tempted to keep the old parser for 1.4, this does sound like feature-creep to me, but since I have been absent for pretty much all of the development of 1.4 and don't aim to use it myself I'm really not too fussed either way.

@ridgerunner: You do make some good points, but keep in mind a bigger filesize doesn't always equate to worse performance (and I'm sure no-one is worried about 14kb space on their server), in fact the main reason for the size increase is because it now does both pre and post parsing - meaning the post parsing (that is done on each view) does less work and should be quicker. Also if there are errors in the regex' used in 1.3, please point them out.

ridgerunner: the problem with the alt attribute is an accessibility issue. Specifically 1.2 does not allow the poster to set the alt attribute which of course is a major issue for anybody who cannot see the image. Whether that is a problem in the real world can be debated but it does mean that forum owners who want to make their forums accessible are unable to do so. The current default of using the url is no benefit to anyone. The paragraph issue is more just a question of doing things right. A post which looks like it has 10 paragraphs should actually have 10 paragraphs rather than one block of text divided by br's.

I too, am a "standardista" and am all for semantic, accessible markup (and separation of markup, style and behavior - this is one of the things that led me to PunBB in the first place.) I agree that having an ALT attribute would be a very good thing. As for the <br />, I would say that two newlines in a row '\n\n' should be converted to a paragraph, but one by itself converted to a <br />. At first glance at the code, this would appear to be a pretty minor change to the 1.2 parser.

I'm working on a detailed write up as we speak. (Unfortunately I won't be able to post it here because the regexs themselves have BBcode tags embedded inside them which confuses the forum parser even when placed inside a CODE tag. D'oh!)

If it sounded as though I was attacking you personally, I apologize. I never intended that tone at all. I was just giving my critical input which was requested by the topic of this thread. I know that it is one thing to complain about how bad something is, and quite another to instead offer solutions to the problems. In that spirit, I'll be offering some positive contributions shortly...

Great, that's the way I like it

That was an excellent read, thanks for that! I'll see about implementing and testing it tomorrow.

And done in [1080].

I proofread the article and made a few changes/corrections. Check the Revision History page for details.

FluxBB Parser Regular Expressions

Note: For some reason, this article does not yet print correctly under Firefox2 or IE6. It does print correctly from Opera9 (not sure about any other browsers/versions).

Fixed in [1081].

That looks wicked! Regex mastery.