22 Sep 2000 jgg   » (Master)

It has been brought to my attention that people have written processors to convert docbook reference pages into nroff suitable for man. I am overjoyed.

Years ago I sought this exact thing and came up empty so I decided to use yodl. Now I'm converting 1600 lines of yodl into docbook SGML - and overall the output result is much nicer but there are still problems - I will have to adjust the stylesheets and things.

I think I will also tank debiandoc-sgml and put those texts into a docbook format as well. Wee, unified documentation, I only need to remember one markup! What could be better?

So, I'm really doing all this so I can sit down and write an absolutely huge guide on how to use libapt. I'm doing that because I wrote some Python bidings (and bod is writing perl ones..). I can expect C++ programmers to grok the source, but not Ptyhon folks - so a big ole doc is the only answer.

I love writing documents, I expect this will take a good week to finish, have copious diagrams and things. I'm targeting something between 50 and 100 printed pages so we will see how that goes.

Xfree 4 won't stay up on my box. It hangs hard occasionaly, usually during screen bitmap copies. I guess the Matrox Mysitque support is slightly untested. That said, X4 is ungodly fast on this card. It is easially faster than some fast PIII's at pure graphic ops and it is just a P166!

Still have to add some framework for abstracting the things Conectiva added, the differences are quite small, but I'm going to clean some crap up in the process. rpm has some weird notions, like versioning order differs for depends and 'newness' , but otherwise it is an amazingly clean fit - one new dependency type, Obsoletes, which is really just Conflicts+Replaces rolled into one. The fact that APT can represent both systems fully (it actually will represent RPM more fully that most RPM tools!) is fairly interesting. I wonder what other packaging systems out there have the required structure. If RPM wasn't so ungodly expensive to generate all the meta data it might be fairly nice.

Latest blog entries     Older blog entries

New Advogato Features

New HTML Parser: The long-awaited libxml2 based HTML parser code is live. It needs further work but already handles most markup better than the original parser.

Keep up with the latest Advogato features by reading the Advogato status blog.

If you're a C programmer with some spare time, take a look at the mod_virgule project page and help us with one of the tasks on the ToDo list!