JTextile 1.0 is now out. Thanks to Gareth Simpson for his hard work whilst I was out sunning myself.

There are a few bugs to be cleared, but if there are any I haven’t spotted, just leave a comment.

Fixed some bugs, introduced a new rule, as proposed by Scott in the comments.

I give up. I give in. I concede defeat. Regular expressions have been for a long time been my weakest field, and my attempts to make a Java version of Textile have ended in ignominy.

Has anyone already performed this regex-a-plenty task? Or is anyone else willing to give it a go? Go on Lazyweb, be my hero.

(Mark Pilgrim also wrote a version in Python, which you can see here)

Okay, I ‘fess up. I don’t get it.

Why should I bother using xhtml:body when I could just as well be using content:encoded ? What’s the inherent benefit? I don’t buy the bandwidth cost answer, either.

There’s a nice short overview which probably summarises all the main points, and back when it was all new I didn’t get it either, but I wasn’t publishing RSS then, so thought I was just missing something. Now I am publishing, and I still don’t get it.

Sure you get a namespace, and it’s probably nicer, in a web egalitarian way, but is it actually better?

So what’s the deal? Why should I change my RSS feed to support xhtml:body ?

Thanks to a suggestion from Phil Ringnalda, my RSS feed has now moved. Hopefully this could mean better HTML parsing and RSS 2.0 production. xhtml:body here we come.

My new RSS feed

→ Something is broken though. My RSS description contents are much shorter than they should be, although links to comments and real permalinks now appear correctly. But I rather think the contents of the posts are more important…

→ It stops after the first <span> tag. Most of them are unnecessary additions from a Textile clone, so just taking them out should provide the full post in the <description> section.

I don’t have an MT weblog (‘cause I’m “powered by Blogger”), but even I understand the principles behind trackback.

It seems that Dave’s been having trouble, but I don’t see what there is to get (although admittedly, this is after I read a very good explanation), I have a short html form for when I want to send a trackback from my Blogger blog, which takes the TB URL (which is the trackback:ping=”xx” bit of the embedded RDF snippet), then enter my entry title and an excerpt and hit submit. It takes about a minute, which is probably about 59 seconds longer than a native MT user, but it works.

Dave may also like to know that both Sam Ruby and Mark Pilgrim have the embedded RDF in their weblog posts, uncommented and commented respectively, I suspect the commented version is more correct, but I could easily be wrong.

A quick (and imperfect) workaround for the lack of Blogger RSS feeds is the service provided by Voidstar, which necessitates you add a <span> into your main template, and then generates your RSS feed for you.

Not a bloody bad idea given that a number of “A-list” people have said they won’t read a website which doesn’t provide an RSS feed (here is where I’m supposed to link to Phil Ringnalda and Sam Ruby saying exactly this, but I can’t find their comments offhand).

Anyway, all this makes my new RSS feed, which astonishingly, even validates

All would be perfect if it didn’t seem to insert random exclamation marks and break the anchor link on occassion, replacing it with a link I’ve made in the text. But I can live with this, at least temporarily.


More importantly than anything else at the moment though (translate: bugging me the most) is that Blogger doesn’t generate RSS feeds.

Blogger Pro users seem to be able to do this, but with the widespread proliferation of free and simple blogging tools which all provide this, this is something Blogger really needs to sort out.

Suffice to say that as soon as I have a Joggle server set up, this site will be available in RSS 0.91 and 2.0.

And about bloody time.

I think I’ve discovered my favourite diveintomark post ever. And who said programmers couldn’t be funny?