Doing Things at the Wrong Time of Day

hex on 2002-08-24T09:22:39

...Or, another all-nighter.

Haven't done much on wikibot recently, barring getting it functioning to a basic extent on my new server, which required some changes to cope with an unexpected change from HTML to XHTML in UseMod's output. I've run into a strange bug causing some page updates over HTTP to write over each other instead of follow each other, which I have to look more closely into, and I'm taking a break due to gumption erosion (yes, the other day I read Zen and the Art of Motorcycle Maintenance).

Anyway, it's now coming up to 10am and I've been here at the EasyInternetCafe, Tottenham Court Road, since about 9pm last night. I spent the first couple of hours fiddling - whilst browsing and IRCing, of course - on and off with RSS.

LiveJournal has a feature that paid users have their own RSS feed (e.g. mine). I decided to do something with it, and set out with these two objectives:

  1. Get the ten most recent items.
  2. Display them on a webpage, which does the above whenever loaded.

With a bit of help from webreference and surprisingly little pain, I was soon able to cobble together the new front page of my site, which now displays the last ten of my journal entries. If anything, I spent equal the amount of time that it took me to do the code on getting the damn Apache configuration right so that a) the script would run and b) everything else wouldn't.

Now that it's working, it makes me happy, because although it's hardly anything new it's another small step for me on the perl ladder.

Some hours of doing other interweb things later, at around 7:00am, I decided to try and get the Daily Chump Bot working. Step one, of getting the actual (Python) script running wasn't a problem. However, it produces XML files. Which meant I needed to get XSLT working.

Now, in an article on webreview.com, Edd Dumbill, co-author of the chumpbot, says the following:

Use mod_xslt, or another XSLT-aware Web serving environment, such as Cocoon, to transform the XML dynamically from the Web server. This is the method the Daily Chump uses, and sample XSLT sheets are included in the Chump program distribution.

Oh, that should be easy, then. Right?

Of course not.

I decided to try getting mod_xslt to work. First step: find the homepage. Then find the appropriate debian packages. apt-get install libapache-mod-xslt. OK. Add the following files to httpd.conf with the appropriate paths:

LoadFile                    libexec/libsablot.so
LoadFile                    libexec/libxmlparse.so
LoadFile                    libexec/libxmltok.so
LoadModule mod_xslt_module  libexec/mod_xslt.so

Oh, wait... those top three files didn't come in the package. Great. Lucky for me blech was around to tell me about libwww0 and libxmltok1.

I installed those, and then blech told me about the extreme hairiness of the code running scriborg's front page XSLT. At which point the last small shreds of my gumption evaporated in little curly wisps.

Well, maybe next time.