Hey folks, I want to give a final SoC Report as well. Sorry for being a bit late. First a short one: * Last bug (nearly) closed, so official project finished * anyway: new code is underway * new ideas as well * so I want to keep contributing to Evolution and GNOME in general. That's it, you're done :P If you still have free time, you may read further on: First of all, thanks a lot Google, thanks a lot GNOME and thanks a lot Herzi for making this Summer of Code happen. As announced last week I wanted to fix the last open bug (new threads appear expanded, although the user selected to "collapse all"). It turned out, that it's not reproducible anymore anyway. But as I found that out, I have rewritten state-storage engine to use GConf instead of writing to disk. So I could see which nodes are expanded and which are collapsed. This is not (yet) for production use as it's very dirty right now (debug code.. you know..). But why do I say that?! Well my point is, that I burned a lot of time with things turned out to be useless or just not worth the time. For example dealing with autotools or compilation of Evolution in general. I feel, that I spent at least a third of my time, maybe even half of it, just to keep (either Evolution or it's dependencies, but actually most time both...) compiling. Is that really everything what 35 years of C programming has to offer? I actually wanted to code and not to deal with such things. To not scare new potential contributors away, I patched the Evolution wiki a bit so that there are desperate people can lookup whether they missed something. Another issue was, that I was looking for race conditions in that state-saving magic, but it turned out not to be a problem. It took me a huge amount of time and even more discipline to get through the codebase and see how things play together. Actually even the evolution-data-server is involved in all this threading thing and I didn't even manage to look through it yet. My bottomline is, that I feel my Summer of Code went worse than others. Spoken in terms of written LoC or implementing new shiny things. And although times were quite tough and the pressure was quite high sometimes, I am happy to participated the SoC. As said in the short version and suggested above, new code is underway. Namely a GConf rewrite of at least the state-saving and account things. Right now GConf is abused storing full XML strings. If things are rewritten using GConf you are, at least in theory, able to lock things down as much as you like. Other ideas, of which I think will make Evolution rock, are ignorable threads or taggable emails. Who knows, maybe one of this will be my next SoC project ;-) Well then guys, again: thanks a lot! see ya all at next GUADEC! :) Cheers, Tobi
Attachment:
signature.asc
Description: This is a digitally signed message part