So as none of you may now, at the core of my little framework here is a pretty sweet home grown parser, that basically spits out a DOM tree. It takes all sorts of input and turns it into pretty smileys, auto-paragraphs and more coolness. There's a few things it doesn't do too well though, so I rewrite a new parser from scratch over the past 2 days.

One of the main selling points for the old parser was the fact that it was very input error resistant. You could basically throw any crappy string at it, and it would translate it correctly into XML the best it could. With that error tolerance I sacrificed some flexibility. One of the sweet features is cleaning up people's new lines and auto creating paragraphs. Which I realized shouldn't really be a core parser thing, and is better left for a second pass. Hence my next parser. There are a few ways I could've gone about starting this project, but I figured taking a real XML parser as a base would work pretty well.

I basically extended the PHP built-in DomDocument, and added the auto-paragraphing, smiley parsing etc. on top of that, as a second pass over the tree. This makes parsing a lot easier, but a lot more rigid since DomDocument doesn't tolerate any errors.

I might revisit the old parser, but for now the extra features and stability outweigh the flexibility and fault tolerance in my mind.

What does this mean for the blog? It means I can now do code snippets that are highlighted correctly