[02:14] Hi Laney, if you're about, can you review https://lists.ubuntu.com/archives/devel-permissions/2014-September/000717.html ? === SpamapS_ is now known as SpamapS [02:35] slangasek, the local-finished is just to say "cloud-init-local already ran". which it could have run to completion before cloud-init-blocknet started. [02:35] i'm out til monday, just figured i'd respond. [02:35] later all. === mfisch` is now known as mfisch === mfisch is now known as Guest17883 === Guest17883 is now known as mfishc === mfishc is now known as mfisch === mfisch is now known as Guest41508 === Guest41508 is now known as mfisch === arrrghhhAWAY is now known as arrrghhh [06:47] What do you guys think about new Poettering idea about changing the way OS is being packaged: http://0pointer.net/blog/revisiting-how-we-put-together-linux-systems.html === udevbot_ is now known as udevbot [07:45] derek-g: vommit on that and anything systemd related === debfx_ is now known as debfx === soren_ is now known as soren === Zic_ is now known as Zic [10:29] bluesabre_: should get to it tomorrow [10:29] sorry nobody else did === bluesabre_ is now known as bluesabre === ttx` is now known as ttx === Ursinha-afk is now known as Ursinha === asac` is now known as asac === Quintasan_ is now known as Quintasan === tarpman_ is now known as tarpman === GridCube_ is now known as GridCube === plars_ is now known as plars [18:06] mitya57: How come you didn't just sync pyqt5? [18:12] Is it possible to write Unity lenses in C? (instead of C++ or Python) === _morphis is now known as morphis === tlyu_ is now known as tlyu [19:11] smoser`: ah. so the idiomatic way of doing that with wait-for-state would be to make sure cloud-init-local is in state 'started' once it finishes running, so that dependent jobs can key on the state [20:48] umh.. I'm unable to find any documentation about the differences between the amd64 and am64+mac isos. Any hint? [21:29] mapreri: http://askubuntu.com/a/40480 [21:32] Logan_: thanks [21:33] np [22:05] I'm a newbie following the Ubuntu packaging guide and got the following error message: [22:05] dpkg-genchanges: error: badly formed line in files list file, line 1 [22:07] The git repository I started with compiled and installed w/o incident. All I'm trying to do is wrap it in Debian frosting. [22:07] (Specifically, I'm following http://packaging.ubuntu.com/html/packaging-new-software.html) [22:09] That sounds like the "debian/files" file (which is supposed to be autogenerated) has somehow had bad data written to it [22:12] maxb: I forgot to mention, upon receiving the error I did "find . -name "*files*" and got no hits (other than two files in the source repository: files.html and files.js) [22:16] Well that's odd. [22:16] Is your packaging already in a git repository somewhere that I could clone and look at? [22:17] maxb: https://github.com/w3c/tidy-html5 [22:18] I don't see a debian/ directory in there [22:21] max: according to the documentation "bzr dh-make" should generate it, and it did. I don't use pastebin often. Is it just http://pastebin.com/ or is there a preferred other pastebin? I can paste the console log/ [22:21] . [22:21] maxb: according to the documentation "bzr dh-make" should generate it, and it did. I don't use pastebin often. Is it just http://pastebin.com/ or is there a preferred other pastebin? I can paste the console log. [22:23] pastebin.com is pretty heavy with ads [22:23] paste.ubuntu.com is less so [22:23] http://paste.ubuntu.com/8346124/ [22:24] The thing you really need to bear in mind about dh-make is it just generates a starting point which will have *way* too much distracting boilerplate [22:24] Expect to need to delete lots [22:25] From all that "warning: ignoring deletion" stuff, I infer you've ended up with an .orig.tar.gz which contains the results of a previous build [22:25] It would be a good idea to clean that up [22:26] maxb: Ah. The documentation only mentioned getting rid of the *.ex and *.EX example files. [22:26] That's certainly part of it. A good principle to follow is the end result shouldn't contain anything you don't understand what it's there for :-) [22:27] maxb: That would leave me with an empty directory. ;-) I will pursue the purging. [22:27] I am still somewhat confused by the ultimate error - I can't see what would be causing that [22:28] There are four files in debian/ which are truly core: changelog, compat, control and rules [22:28] You'll probably also have one named install by the time you're done. [22:29] But limiting yourself to that subset to start with should help you cut through the distractions whilst you work on initial packaging [22:31] I have copyright, which was boilerplate but I copied the text of the distributed git into it. [22:32] copyright is certainly required if you're aiming for official Debian/Ubuntu eventually, but doesn't have any functional influence on the package build [22:33] maxb: I'll get rid of the docs, README.*, source/format (and source/) and try again. [22:33] maxb: Other than those, everything else is in your list of core files. [22:34] This package has a really quite weird Makefile install target [22:45] ubuntourist: https://github.com/maxb/tidy-html5/commit/2ce2511c25a0dc6e017fca389654df4ff060f7b9 [22:46] maxb: Thanks. I'll have a look and compare to what I've ended up with, to learn. [22:48] I still don't understand how the "files list" error you had would have been caused - so all I can do is tell you that that commit results in a reasonable looking .deb for me