30 Jan 2004 shlomif   » (Master)

Version Control Systems Overview Article

An article that I wrote with an overview of Version Control Systems got published on O'Reilly's ONLamp.com. I wrote a suggestions to it (with a summary, IIRC) to the editors of O'ReillyNet some monthes ago. chromatic answered me, said he would publish it and asked me to write a draft. I wrote a draft and sent it to a couple of people I know from the version control business, and they helped me revise and correct it. Tom Lord of Arch fame was especially helpful with his commentary, but I received a lot of useful commentary from other people.

The published article is not identical to the final draft I have. Some things were clearly modified, and I believe that it detracted a lot from the article's colour. But whatever.


I have no more lectures and tutorials for this semester. My first test is in two weeks, and the next one a few days afterwards, which gives me some time to study. I have a lot of material to cover as far as Thermodynamics is concerned, and also need to prepare some material summary pages for Statistics. (and naturally need to prepare for that too)

Work on WWW::Form

It turns out we forgot to include the README in the MANIFEST of WWW::Form 1.13. I fixed it along with a test to make sure it won't be neglected. Now, the question is what will make sure the test file will be included there. (Who will watch the watcher?).

I also did some refactoring of the code. The original code had a function to set all the fields parameters as inputted from the user. I converted it to one small loop that calls a function that sets each parameter individually. I thought that was enough to add new parameters for every form field, but then thought of a better idea, and so created a function to return them as a hash.

Now, the user can sub-class this method alone, call the original and add more parameters of his own to the hash reference that is returned.

I'm doing test-driven development now, writing tests before writing the code.

Hebrew XML with Perl

I wrote an article about processing Hebrew with Perl (it's in Hebrew). I received some useful commentary on it and revised it accordingly. Now I hope it will be published in the Perl and Hebrew page.

PCLinuxOS Live CD

I downloaded the PCLinuxOS LiveCD which is a Mandrake-based Linux Live CD. The first burning produced a CD that booted but got stuck with some errors in the middle of the initialization. It was probably because I did too many things at once. So I re-verified the MD5 sum and did another copy. This time it worked perfectly.

Well, I only played with it a little. I found out it did not have gvim, or joe, (guess that if you want to have both GNOME and KDE, you have to give up something), and so it wasn't ideal as a rescue CD. Still, it seems nice. Probably a nice way to introduce Linux for newbies, or install Mandrake for them.


I stambled across a bug in the GIMP Curves dialog when I was looking for interesting GIMP bugs to solve. So, I tried to reproduce it on my system, but could not. One thing I was able to do, even with fewer clicks than were described was in the bug report, is to grab the pointer for the window with a few rapid clicks. It is so bad, I have to kill the GIMP to release it.

However, Sven Neumann could not reproduce it on his machine, and he claims it is a problem in my X server. I tried to disable the Nvidia acceleration, but this did not help either. Even a bug fix from Sven to disable the explicit pointer grab did not help. What could be the problem?


I ended up using the XSL stylesheets of DocBook for processing "The Eternal Jew" into XHTML. They are quite nice and I was able to customize them to include a CSS stylesheet and other stuff. Now, I'm stuck as I want to make sure a specific itemizedlist will be CSS-classed according to its role attribute. This turned out to require a hairy XSL customization, and I guess I'll have to learn XSLT better to do it by myself.

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!