[02:50] morning [03:10] /c/c [03:10] heya TheMuso :) [03:10] Hey mgolisch. [03:10] gah typing [03:10] hey emgent [03:10] :) [03:22] codehosting is down to fix a regression cuased by the rollout === asac_ is now known as asac [05:04] afternoon all [05:08] morning Hobbsee_ :) [05:08] Good night [05:09] lol [05:10] It’s 7:10 here, a good time to go to sleep. [05:10] am ? [05:11] I rather avoid the AM/PM notation. I have nothing against using a 12-hour clock, but having the time of each half go from 12 to 1 to 11 is just too bereft of logic. :-) [05:12] So yes, 7, not 19. :-) [05:13] yeah, i think so. good night ion_ :) [05:13] :-) [05:14] here 6.14 am [05:32] ion_: to disambiguate, use a leading 0 [05:35] hello. does anyone know where i would submit code for newspost? the newspost webpage hasnt updated since 2003 [05:45] ooook so much for that [06:33] Good morning [06:34] pitti: Guten Morgen [06:34] elmo: it's off by default in hardy-updates; if you manually turned it on in /etc/default/apport, you can disable it there again [06:35] ion_: "prove pitti right" -> I made some joke about "go, keybuk, go! push it to -ubuntu20" :) [06:35] pitti: can you give the MIR for libpar-dist-perl a quick look if something is missing? bug #236298 [06:36] https://bugs.edge.launchpad.net/ubuntu/+source/libpar-dist-perl/+bug/236298 [06:36] calc: ok, thanks; I was just wondering why both tasks are open; if you are working on the bug, all is fine [06:36] pitti1 [06:37] bryce, cjwatson: I agree; it's generally better to have public bugs for SRUs; a second public one WFM if the primary bug has a good justification for being private [06:37] arthur-: thanks; did you see the mail I wrote to the Debian BTS? [06:37] good morning [06:37] A question about merges: the wiki on merges says to file a bug and assign it to yourself -- is it not okay to file a bug requesting a merge but not work on it? [06:38] pwnguin: no, that's not ok [06:38] pwnguin: there's not reason to file bugs for "plzmergekthxbye" [06:38] pwnguin: since we already have MoM to remind us about it [06:38] how does MoM get updated? [06:38] s/not/no/ [06:38] morning people :) [06:38] pwnguin: it automatically updates daily (or even more often, not sure) [06:38] hey emgent [06:39] hey dholbach [06:39] hi pitti [06:39] pitti: well hrm. i dont see wacom-tools on it =( [06:41] pwnguin: that's because Ubuntu's version is newer [06:41] 1:0.7.9.8-0ubuntu3 >> 0.8.0 [06:41] huh [06:41] no idea why we have an epoch, thouch [06:41] though [06:41] thats no good [06:41] so right, in this case a merge bug would make sense [06:42] wacom-tools (1:0.7.9.3-2ubuntu1) hardy; urgency=low [06:42] - Bump the epoch because of old Ubuntu packaging. [06:43] -- Timo Aaltonen Sun, 30 Dec 2007 18:59:29 +0200 [06:43] tjaalton: ^ why did you do this? that'll stick forever, and we'll never have the chance to properly merge/sync with Debian? [06:44] that second part doesnt really sound like a question ;) [06:45] pitti: It's been there since Dapper... [06:45] * Bump epoch to make xserver-xorg-driver-wacom newer than the one in breezy. [06:46] ah, then that bit got dropped from teh changelog; sorry, tjaalton [06:46] wgrant: ah, ok; bah, there are less sticky ways to do that... [06:47] pitti: It does say 'Remaining changes [06:47] right [06:48] i wonder what can be done about that [06:48] Nothing. [06:48] the only way would be to beg Debian to bump it [06:48] but they most likely won't [06:49] so we're stuck with it [06:49] Upstream could change the versioning scheme [06:49] I had a nice DD that bumped an epoch for us because we had another use for the name first. [06:49] Guten Morgen dholbach [06:49] wgrant: right, but TBH, if I were only a DD, I wouldn't bump the epoch "just because" [06:49] Ron's usually pretty cool [06:50] im not sure if he'd do something quite that silly [06:50] given the "destructive" nature of epoch's, I'ld say packages that gain an epoch would be ideal candidates for the NEW queue (or similar) [06:50] what about renaming the package? [06:51] pwnguin: that only makes it worse [06:51] hi geser [06:51] Chipzz: yeah, indeed [06:51] pitti: how so? [06:51] pwnguin: can't sync from debian then [06:51] pwnguin: then merging becomes even more difficult, and you have to additionally keep transitional packages for a while [06:51] pitti: and by gain I mean gain or increase an existing epoch [06:51] given that the software is "linux wacom" upstream and "wacom-tools" in debian/ubuntu [06:52] s/gain or/acquire or/ [06:52] Chipzz: I've always thought that would be a very good idea. [06:52] there MIGHT be a chance at that [06:53] clearly ubuntu wacom is in a bad spot at this point, just trying to brainstorm anything that might be better than the status quo [06:54] pitti: is subscribing fixed you in the newest bzr revision of py-lp-bugs? [06:54] dholbach: in bzr main head, yes [06:55] is MoM amenable to some sort of special casing wacom-tools? [06:55] pitti: I still get http://paste.ubuntu.com/16348 [06:55] pitti: nevermind [06:55] pitti: it uses the system pylpbugs [06:55] yes, it's not yet fixed in any ubuntu release [07:03] i love it when people close bugs because i accidentally use the wrong terminology [07:03] status invalid [07:04] fortunately they went and did it three days later [07:07] pitti: yes, I've asked the debian developer to bump it but got no response [07:07] it's not maintained by the XSF [07:07] tjaalton: right; I wouldn't put a lot of hope into Debian bumping the epoch; thanks for asking, though! [07:09] tjaalton: well, in the mean time, is there something we can do to make wacom-tools updates more visible to Ubuntu developers? [07:11] i have an RSS feed of versions moving around in debian [07:13] Maybe MoM could automagically drop epochs if they're higher in Ubuntu? [07:13] Or maybe not automagically. [07:14] how many broken epochs does ubuntu have? [07:15] wgrant: That strikes me as prone to breakage [07:15] pwnguin: we use this http://people.ubuntu.com/~bryce/Xorg/versions_current.html [07:15] okie [07:16] StevenK: More breakage than having a differing epoch in the first place? [07:16] wgrant: It isn't breakage, it's just ... different [07:18] pwnguin: there were quite a few in the X packages, but most of them are now in debian too. most of the unsyncable packages are x11proto's which have a different tarball (and no hope of a new release :) [07:18] (anytime soon) [07:18] \o/ === jussio1 is now known as jussi01 [08:24] pitti, hi [08:24] hi tkamppeter [08:24] itti, it is about the blueprint of the PDF printing workflow [08:25] pitti, you tell I did not mention what needs to be changed in CUPS. [08:26] pitti, In the second paragraph of the "Implementation" part I told that nothing needs to be changed in the CUPS daemon itself, only filters need to be added. [08:27] tkamppeter: right, I meant the filters [08:27] they are shipped in the cups package [08:28] pitti, the extra filters can be shipped in the CUPS backage but they do not need to be shipped there necessarily. [08:28] tkamppeter: right, but that doesn't matter much [08:28] tkamppeter: so which filters do we still need? pstopdf? [08:28] pitti, probably as long as they are not official upstream part of the CUPS package, they will come in an extra package. [08:29] tkamppeter: didn't Mike intend to switch to PDF internally for 1.4? [08:30] The filters are the following, AFAIK: pdftopdf, imagetopdf, texttopdf, pdftoraster, pstopdf, pdftoijs, pdftoopvp [08:31] I do not know whether Mike has this intention, this would be great. If he does it, he will grab the filters from OpenPrinting at SourceForge.jp, as these filters are not yet in the 1.4 SVN. [08:54] pitti, I have updated the spec now. [08:55] tkamppeter: danke === fta_ is now known as fta [09:03] seb128: would you have some time to verify bug 152246? it's my upload, so I can't officially verify [09:03] Launchpad bug 152246 in gthumb "gThumb change date to exif date" [Undecided,Fix committed] https://launchpad.net/bugs/152246 [09:03] pitti: sure [09:05] seb128: urgh, that bug isn't SRUish at all; fixing [09:08] seb128: bug is ok now [09:09] pitti: ok, going to try that in a few minutes [09:09] pitti: I just have to upgrade before ;-) [09:09] seb128: thanks [09:17] Does anyone know why myspell-en-au is in universe, while the other three myspell-en-* are in main? [09:17] Hewus: language-support-writing-en depends on the others, but not myspell-en-au [09:17] it's probably a bug [09:18] cjwatson: thank you, I'll try and take care of it [09:22] I am trying to get a consistent Intrepid installation onto a 32-bit PC and have a problem: [09:22] I cannot install KDE apps because of [09:22] The following packages have unmet dependencies: [09:22] kdebase-kio-plugins: Depends: kdelibs4c2a (>= 4:3.5.8-1) but it is not going to be installed [09:22] Depends: kdesktop but it is not going to be installed [09:23] but I have kdelibs4c2a 4:3.5.9.dfsg.1 on this box, so the dependency should be fulfilled. [09:26] tkamppeter: while apt's output is often suboptimal when this sort of thing happens, http://people.ubuntu.com/~ubuntu-archive/testing/intrepid_probs.html indicates that kdebase-kio-plugins is entirely uninstallable in intrepid [09:26] so you probably won't be able to get far until that's sorted out [09:27] cjwatson, thank you. I will keep Hardy on at least one box for now so that I can use digikam there. [09:28] it's pretty early in the cycle to start running the development version, isn't it? [09:28] liw: why? [09:28] without people running it we'll never know what stuff to fix [09:28] Probably because stuff like this happens all the time? :) [09:29] pitti, yeah, but things break so much that one runs the risk of not getting anything else done than bug reporting :) [09:29] tkamppeter: what was the problem: I got disconnected, so I just got your "I have a problem" and then liw's comment [09:29] pitti: uninstallable kdebase-kio-plugins [09:29] hm, traditionally I upgraded to the dev release on the day we opened it [09:30] I don't report all those bugs yet, since they usually shake out themselves without bug spam [09:30] Keybuk: valgrind: I think you're right, looks like one of Kees' changes was responsible [09:30] RAOF: uninstallable packages usually don't break your box; the upgrade should just hold them back [09:30] I'll talk with him about it later [09:31] pitti: Yeah. FWIW, *I'm* running intrepid. [09:31] pitti: I think we should proceed with bug 199190 [09:31] Launchpad bug 199190 in libitpp "libittp fortran transition" [Undecided,Fix committed] https://launchpad.net/bugs/199190 [09:32] norsetto: ah, did anyone test it? [09:33] pitti: no, it runs a self test at the end [09:33] that doesn't check if the package actually installs and works [09:34] pitti: installs it installs, wether it works or not, nobody has any idea. The package as is now doesn't install [09:34] tkamppeter: KDE is stuck half way between KDE 3 and 4 and will not be installable until the rest of the main inclusion reports get reviewed [09:34] norsetto: if installation works, I'm good with it [09:37] pitti: bug 221664 looks nasty [09:37] Launchpad bug 221664 in lilo "lilo loads wrong data for big initramfs (?)" [Undecided,New] https://launchpad.net/bugs/221664 [09:40] cjwatson: eww, indeed; the documentation of "large-memory" seems to indicate that this would break on 'older' systems, but not how old 'old' is here [09:42] nothing that immediately jumps out on google [09:47] pitti, it was about trying to install a KDE app in Intrepid. KDE is a komplete mess currently, we need for the KDE 4 introduction to complete. [09:47] tkamppeter: right; see Riddell's explanation above [09:49] sbeattie: Ping [09:50] pitti: dankeschoen! [09:57] norsetto: what for? :-) I didn't do anything yet [09:58] * norsetto hugs pitti [09:58] * pitti hugs norsetto back [10:13] hi all [10:14] hi RicardoPerez [10:14] pitti: hi! one question... [10:14] pitti: in which packages are the firefox translations in hardy-proposed? [10:14] RicardoPerez: language-pack-XX-base [10:14] RicardoPerez: I answered your mail, too [10:15] pitti: yes, I saw it. I updated firefox, xulrunner & all the langpacks, but firefox is still in English... :( [10:16] Riddell: dpkg -l language-pack-es-base|grep es-base [10:18] pitti: there are directories inside language-pack-es-base, called /usr/lib/firefox-addons/extensions/langpack-es-ES@firefox-3.0.ubuntu.com and /usr/lib/xulrunner-addons/extensions/langpack-es-ES@xulrunner-1.9.ubuntu.com, but they're empty... [10:19] asac: ^ any idea? [10:19] * pitti takes a look at the source package [10:19] hmm [10:19] i tested es.tar.gz at least [10:19] pitti: correction: they contains a directory called chrome, which is empty, too [10:20] hm, the mozilla.tar.gz looks fine here [10:20] source package is fine here too [10:20] * asac installs langpacks [10:21] * pitti downloads .deb [10:21] pitti: ....langpack-en-GB@xulrunner-1.9.ubuntu.com/chrome CONTAINS a file called en-GB.jar [10:22] looks good here: [10:22] http://paste.ubuntu.com/16368/ [10:22] -rw-r--r-- root/root 157031 2008-05-29 10:52 ./usr/lib/xulrunner-addons/extensions/langpack-es-AR@xulrunner-1.9.ubuntu.com/chrome/es-AR.jar [10:23] -rw-r--r-- root/root 57414 2008-05-29 10:49 ./usr/lib/firefox-addons/extensions/langpack-es-AR@firefox-3.0.ubuntu.com/chrome/es-AR.jar [10:23] .deb looks fine here [10:23] mmmmm.... I don't have too many files... [10:23] asac: same here [10:23] maybe something went wrong with the Replaces: [10:24] http://paste.ubuntu.com/16369/ [10:25] ricardo@kadath:~$ dpkg -l language-pack-es-base | grep ^ii [10:25] ii language-pack-es-base 1:8.04+20080527 translations for language Spanish; Castilian [10:25] RicardoPerez: try dpkg -S usr/lib/xulrunner-addons [10:25] err [10:25] :) [10:25] RicardoPerez: I guess if you sudo apt-get install --reinstall language-pack-es-base, it'll work [10:26] pitti: how to figure what happened? [10:26] pitti: mmmmmm.... strange.... the .deb file contains more files than I have installed... I'll try --reinstalling it [10:26] sudo apt-get --reinstall install language-pack-es-base [10:26] RicardoPerez: can you keep the current state for a second? [10:26] asac: sure, I wait [10:27] RicardoPerez: dpkg -S langpack-es-ES@xulrunner-1.9.ubuntu.com [10:27] ? [10:27] what does that give you? [10:27] ricardo@kadath:~$ dpkg -S langpack-es-ES@xulrunner-1.9.ubuntu.com [10:27] language-pack-es-base: /usr/lib/xulrunner-addons/extensions/langpack-es-ES@xulrunner-1.9.ubuntu.com [10:27] language-pack-es-base: /usr/lib/xulrunner-addons/extensions/langpack-es-ES@xulrunner-1.9.ubuntu.com/chrome [10:27] ricardo@kadath:~$ [10:27] that looks correct [10:28] pitti: where is the replaces state tracked? [10:28] e.g. in /var/ ... ? [10:28] the es-AR.jar & es-ES.jar files are missing [10:28] I think what happened is that the upgrade order was done in a way that the Replaces: wasn't done properly [10:28] asac: should be /var/lib/dpkg/info/.list [10:29] ... are missing in the installed, but not in the .deb... [10:29] RicardoPerez: grep langpack-es-ES@xulrunner-1.9.ubuntu.com /var/lib/dpkg/info/* [10:30] err, [10:30] RicardoPerez: grep langpack-es-ES@xulrunner-1.9.ubuntu.com /var/lib/dpkg/info/*.list [10:30] ricardo@kadath:~$ grep langpack-es-ES@xulrunner-1.9.ubuntu.com /var/lib/dpkg/info/*.list [10:30] /var/lib/dpkg/info/language-pack-es-base.list:/usr/lib/xulrunner-addons/extensions/langpack-es-ES@xulrunner-1.9.ubuntu.com [10:30] /var/lib/dpkg/info/language-pack-es-base.list:/usr/lib/xulrunner-addons/extensions/langpack-es-ES@xulrunner-1.9.ubuntu.com/chrome [10:30] ricardo@kadath:~$ [10:31] RicardoPerez: did you ever had the language-pack-gnome-es-base installed? [10:31] asac: -es was one of the packages where the gnome update packages had a newer version than -base, right? [10:31] ricardo@kadath:~$ dpkg -s language-pack-gnome-es-base | grep Status [10:31] Status: install ok installed [10:31] ricardo@kadath:~$ [10:31] asac: that just affected three languages AFAIK (es, pt, and zh?) [10:32] asac: so for those we have three different versions of translations [10:32] pitti: we have, but the packages are similar to the other packages [10:32] it was basically just a manual update that shipped a better chrome.manifest [10:33] dont see why it would make any difference over lets say -de [10:33] pitti: ah ... you mean because those were in language-pack-gnome-es [10:33] yeah. maybe dpkg has problems to deal with three replaces ;) [10:34] right [10:34] I didn't get reports for other languages about this kind of problem [10:35] yes, but -es is a popular language ... so maybe a quick catch [10:35] asac: so maybe; (1) update language-pack-es-base (files), then (2) update language-pack-gnome-es{,-base} [10:36] lp-es-base replaced lp-gnome-es-base, and that conflicted with lp-gnome-es (still in the old version) [10:37] yeah [10:38] so we have to put them into language-pack-es as a first measure? or are we brave enough to fix conflicts/replaces? [10:39] pitti: ? i could upload language-pack-es -zh -pt if thats what you like [10:41] I can downgrade my langpacks and wait for new langpacks updates uploaded into hardy-proposed, if you want... [10:41] asac: no idea how to solve it, I'm afraid, as long as we don't understand what's happening in the guts of dpkg [10:42] might mvo help? [10:42] asac: we might need to add a Conflicts: to force upgrade ordering [10:42] i. e. ensure that we don't install the new language-pack-es-base before upgrading language-pack-gnome-es{,-base} [10:43] RicardoPerez: can you purge langpacks, install the ones from hardy, and then manually upgrade first -gnome-* and then language-pack-es{,-base}? that should work [10:43] pitti: ok, i'll do that in a moment [10:43] RicardoPerez: gracias [10:49] pitti: works! [10:49] RicardoPerez: great [10:49] cjwatson: when is supposed to be the accepted time that ppas will be able to build for intrepid? [10:49] Hobbsee: I didn't know it was delayed [10:50] RicardoPerez: just for confirmation, if you do the same, but first upgrade language-pack-es{,-base} and then *gnome*, it fails? [10:50] should at most be a matter of chroots being set up? (i.e. -> infinity) [10:50] cjwatson: apparently it's delayed, because it's waiting on word from the distro team. [10:50] infinity: poke [10:50] gar [10:50] pitti: I'll do that in a moment, again :) [10:50] ok, I'll go talk with them [10:50] cjwatson: cool, thanks [10:50] RicardoPerez: thanks a lot; if that is the reason, we can add a Conflicts: [10:53] pitti: as you say, if I do the upgrade in the reverse order, it don't work :) [10:54] superm1, tseliot: I just tried to DKMSify http://linuxtv.org/hg/v4l-dvb (I want to provide a driver backport for hardy), but I have some trouble with upstream's Makefile; it doesn't build any modules when being called from /lib/modules/2.6.24-17-generic/build; do you have some experience with that and could help me? [10:54] RicardoPerez: awesome; thanks a lot for testing! [10:54] pitti: thanks to you :) [10:55] asac: ok, so AFAICS we should add "Conflicts: language-pack-gnome-es (<< 8.04+20080527), language-pack-gnome-es-base (<< 8.04+20080527)" to language-pack-es-base? likewise for pt and zh? [10:55] Keybuk: around? I think enabling PPAs for Intrepid requires a techboard member [10:56] cjwatson: err, why? [10:57] pitti: I have no experience with that driver however if you upload your code somewhere I think I can give you hand with DKMS [10:57] is there a button that needs to be pushed? [10:57] Keybuk: apparently it requires turning on 'support-virtualization' in each distroarchrelease [10:57] ? [10:57] now I must leave. see you! [10:57] https://launchpad.net/ubuntu/intrepid/amd64/+admin, https://launchpad.net/ubuntu/intrepid/i386/+admin, https://launchpad.net/ubuntu/intrepid/lpia/+admin [10:57] tseliot: that's the (essential part of) the Makefile: http://paste.ubuntu.com/16376/ [10:57] ENOPERM [10:58] Keybuk: seriously? argh [10:58] duckie required, I suspect [10:58] or maybe infinity [10:58] or my tech board membership expired already [10:58] your techboard membership is still active [10:58] though five days - shouldn't that be extended? [10:58] dunno, ask sabdfl ;) [10:59] tseliot: running "make" in /var/lib/dkms/v4l-dvb/0.20080328/build/ works, but running make KERNELRELEASE=2.6.24-17-generic -C /lib/modules/2.6.24-17-generic/build M=/var/lib/dkms/v4l-dvb/0.20080328/build (as dkms does) doesn't [10:59] tseliot: dkms passes M=/var/lib/dkms/v4l-dvb/0.20080328/build to make [10:59] pitti: yes that makes sense [10:59] pitti: actually, shouldn't the language pack conflict all other lang packs with lower version anyway? [10:59] tseliot: apparently the kernel upstream Makefile is supposed to handle M=, but that seems to need some support in the target Makefile, too [11:00] asac: AFAIR we used Replaces: instead of a million Conflicts: to avoid confusing apt-get [11:00] asac: and I have never seen it fail for the 'usual' case (just two versions) [11:01] asac: this is a special kind of a three-way replaces:, which apparenlty has issues [11:01] pitti: yes, i think its a rare thing that translations migrate from gnome to normal and vv. [11:01] pitti: what does your dkms.conf look like? [11:01] asac: so, it's certainly worth testing to generally replace some Replaces: with Conflicts:, but we shouldn't overdo it [11:01] yep. lets please keep the solution as simple as possible for hardy [11:02] and review the conflicts/replaces in intrepid ;) [11:02] pitti: btw, could langpack-o-matic pass the lsb release version the package is targetted for? [11:02] tseliot: http://paste.ubuntu.com/16378/ (just the first 5; there are 209 modules built from that) [11:03] i need to maintain different blacklists for different releases for instance [11:03] asac: you mean the target? like gutsy-proposed, intrepid, etc? yes, that's no problem [11:03] yes [11:03] for me the lsb version would be better though ;) [11:03] but well. [11:03] asac: like 8.04? [11:03] yes [11:03] asac: no problem, you can have that as well [11:03] thats what i current use locally. but i can adapt my transformer if its easier for you [11:03] I have a code name -> version mapping anyway [11:04] ah good [11:04] asac: however, that will loose the distinction between release and -proposed; do you need that? [11:04] maybe => 8.04 updates (if things go to proposed) [11:04] but yeah. at best just pass the whole target ... ill take care of the rest [11:04] asac: I don't have a meaningful lsb_release there, just 'hardy-proposed' and a mapping to '8.04' [11:05] ok ... Ill let you know when i have setup the code to deal with that. but after this update ;) [11:06] asac: I think manually adding the three conflicts: is minimally invasive and the easiest solution I can think of; WDYT? [11:06] pitti did you add a line like the following to your dkms.conf? MAKE[0]="make module KERNDIR=/lib/modules/$kernelver SYSSRC=$kernel_source_dir" [11:06] asac: I'll do that in langpack-o-matic, I have everything there already [11:07] i think they should be fine. [11:07] tseliot: no, I didn't; I need to? [11:07] we should test the upgrade path for those three afterwards. [11:07] yes, absolutely [11:07] let me know when they are in so i can test all combinations in a chroot [11:07] well ... all combinations i have in mind ;) [11:07] pitti: yes, you should, so that DKMS can build the module [11:08] tseliot: and that won't run make 209 times? [11:08] tseliot: it just needs to run 'make' in the target dir, that's all [11:08] s/target/dkms source directory/ [11:09] pitti: why should it run make 209??? DKMS is a sensible system [11:09] pitti: will you reupload all? if so we have to take care the the right fi.tar.gz is used ;) [11:09] pitti: yes, I picked that up [11:09] rookery should be fixed in case you want to recreate them for real [11:09] asac: no, I was just going to upload -zh, -es, and -pt [11:09] ok good [11:09] asac: I already put your -fi to rookery [11:10] pitti: what do you mean by "put my -fi" ? do you preserve the tar.gz ? [11:10] asac: yes, I unpacked your sources to rookery [11:10] ah ok. didn't know that you use some kind of source caching. grewat [11:10] it shouldn't actually matter, but just to be on the safe side [11:10] yep [11:12] pitti: let me know how it goes ;) [11:13] tseliot: I will, thanks for the hint! just fixing the langpack mess first [11:15] pitti: is it possible to a) add a new binary package to hardy-proposed, b) add a new source package? [11:16] it is technically possible and sometimes also allowed by policy ;-) [11:16] doko: technically yes, practically we did that only once or so [11:16] it needs a really really good justification anyway [11:17] pitti: noticed that libmudflap0-4.1-dev is not built anymore, so -fmduflap doesn't work with gcc-4.1 [11:18] doko: do we care enough about 4.1 to fix that in hardy? [11:18] pitti: the other package would be ca-certificates-java (ca-certificates built for java); cannot be built from ca-certificates, because this is in main, and requires at least openjdk as a b-d [11:19] pitti: well, I don't care that much about 4.1 [11:19] java can't use /etc/ssl/certs/*? [11:20] no, has it's own format [11:20] yay [11:21] lib/security/cacerts or so is the file in the java installation directory [11:21] mgolisch: well, a mere path could be worked around with a symlink [11:22] but if it actually uses its own encoding of SSL certificates, that's more tricky [11:22] yeah the certs are saved in a so called keystore file [11:23] . o O { just gotta love the Sun guys for not using already existing platfom independent standard formats ... } [11:23] mgolisch: do you know if it is possible to generate them on the client side? [11:23] mgolisch: then we could just add that call to ca-certificate's postinst === persia_ is now known as persia [11:24] since ca-certificates itself can change, and the user can reconfigure which certificates he wants, that would then automatically apply to Java as well [11:24] woot woot [11:24] I isolated the Xubuntu black screen of death bug :] [11:24] pitti: do you have powers to proactively prevent xulrunner source package from being synched? [11:24] asac: yes [11:24] asac: we can add it to the sync blacklist [11:25] the new package is still in experimental, but once it reaches unstable the xulrunner source will ship the same package names we have in xulrunner-1.9 [11:25] (+ some more) [11:25] not sure what would happen :/ [11:25] asac: but as long as our's has 'ubuntu' in the version, it won't get autosynced [11:25] asac: and in principle, if Debian adopts our changes, we could sync; or is there a general reason why we will never ever do that? [11:26] pitti: well. they still have a package split we dont want [11:26] they ship a -common package and a libmozjs for which mike added so-versioning [11:27] (not sure why he kept that as afaik there is no standalone rdepends on libmozjs) [11:27] pitti: and the bad is that debian doesnt use a patch-system ... so no sync possible [11:27] eww [11:28] yeah ... glandium rolled-back his opinion that it is sane to have a patch-system for mozillas [11:28] not sure why [11:28] now he has a private git archive [11:28]