[12:07] <ev> so dh v9 tries to execute .maintscript files if they're +x. Bug or the packager should know better?
[12:23] <xnox> ev: packager should know better, it's documented behaviour.
[12:25] <ev> this whole needing a separate degree to distribute an application thing will never catch on :)
[12:26] <antarus> thats just what developers who don't understand how complex packaging is would say ;p
[12:26] <ev> it doesn't need to be complex
[12:26] <ev> it shouldn't be complex
[12:26] <antarus> heh
[12:26] <antarus> I agree the interfaces are poor
[12:26] <antarus> but it is inherently complex
[12:26] <ev> but yes, the complexity is baked in. "Hey, here's a root shell. Have fun!"
[12:26] <ev> indeed
[12:27] <antarus> you could say the same thing about most programming though ;p
[12:27] <ev> sure - but programming is hard enough. Lets not layer complexity on top of that :)
[12:27] <antarus> (if you are distributing a simple applicatoin I would argue that packaging is fairly trivial)
[12:28] <antarus> at work you can basically in your 'makefile' just write "builddeb(name, version, release, owner, depends=[foo, bar,baz])
[12:28] <antarus> and the system will build you a deb of your code
[12:28] <antarus> now you can't do anything complex with it (conffiles, alternatives, etc..)
[12:28] <antarus> because its hard to make a simple interface for that ;p
[12:32] <xnox> ev: imho it's better than a single monolithic file with pseudo shell functions and requirement to explicitely list every single file & folder that will be 'owned' by the package... (rpm)
[12:33] <antarus> real packages use ebuilds anyway
[12:33] <antarus> ;p
[12:42] <silvos> hi, are you talking about how simple (or not) building ubiquity or just packaging? (having problem to compile and run it here) http://paste.ubuntu.com/1009633/
[12:45] <silvos> http://paste.ubuntu.com/1009648/  config.log
[12:55] <xnox> silvos: ./debian/rules update-local
[12:55] <xnox> silvos: fakeroot ./debian/rules binary
[12:55] <xnox> done
[12:56]  * xnox ubiquity is a non-standard package, because it includes bits and pieces of other packages
[13:05] <silvos> thanks xnox
[13:06] <silvos> I thinks its fails again, I changed value in .po to be sure, and when I run ubiquity its runs ubiquity from the apt-get install one with old .po value.
[13:06] <silvos> I feel I don't understannd every thing
[13:24] <cm-t> Hi
[13:26] <cm-t> I followed same as silvos but I fail to run ubiquity, and I don't find many information on compiling it.
[13:27] <silvos> I might forgot some step, is there a wiki page or something like that explainning the compilation ?
[13:41] <xnox> silvos: after fakeroot ./debian/rules binary
[13:41] <xnox> $ sudo dpkg -i ../*.deb
[13:41] <xnox> to install newly compiled ubiquity
[13:41] <xnox> then run it.
[14:08] <silvos> well, I don't have .deb because of error, I'm trying to understand them
[14:32] <silvos> http://paste.ubuntu.com/1009761/  if ever
[14:33] <xnox> silvos: did you run: fakeroot ./debian/rules binary
[14:33] <xnox> or just ./debia/rules build
[14:33] <xnox> ?
[14:33] <xnox> binary will give you actuall debs to install
[14:34] <silvos> http://paste.ubuntu.com/1009761/  is the result of    fakeroot ./debian/rules binary >> bin.txt
[14:36] <silvos> hum
[14:36] <silvos> i should use the error output
[14:36] <xnox> yes, please. that does look incomplete
[14:37] <xnox> lp:ubiquity currently can only be build on quantal, because it has been ported to python3 and a couple of dependenies are not in precise
[14:37] <xnox> if you want precise's ubiquity get it with $ apt-get source ubiquity
[14:37] <xnox> or there should be a precise branch.
[14:38] <cm-t> oh
[14:46] <silvos> http://paste.ubuntu.com/1009777/
[14:47] <silvos> ( fakeroot ./debian/rules binary &> bin.txt )
[22:23] <CIA-62> debian-installer: cjwatson * r1698 ubuntu/ (build/pkg-lists/gtk-common debian/changelog): ttf-freefont-udeb was renamed to fonts-freefont-udeb.