| apostols__ | Hi | 00:09 |
|---|---|---|
| apostols__ | I have problem with upgrade timezone in Ubuntu Dapper | 00:09 |
| apostols__ | This dont have tzdata package and i need upgrade Venezuelan timezone (like http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=443202) | 00:12 |
| ubotu | Debian bug 443202 in tzdata "Venezuelan time zone change, September 2007" [Normal,Fixed] | 00:12 |
| Kmos | apostols__: do you have -proposed active on dapper ? | 00:13 |
| apostols__ | Kmos, No | 00:13 |
| Kmos | 2007j-0ubuntu0.7.10 | 00:13 |
| Kmos | this is the lat one in -updates and -propopsed | 00:14 |
| Kmos | -proposed | 00:14 |
| Kmos | it has the changes for venezuela | 00:15 |
| Kmos | * Replace tzdata2007i.tar.gz with new version tzdata2007j: - Updates DST rules for Venezuela. | 00:15 |
| Kmos | https://edge.launchpad.net/ubuntu/+source/tzdata | 00:15 |
| Kmos | as you can see here | 00:15 |
| apostols__ | Kmos, This dont workfor Dapper | 00:16 |
| apostols__ | Kmos, tzdata package dont exist in Dapper | 00:16 |
| apostols__ | Kmos, Just exist locales package | 00:16 |
| Kmos | apostols__: it must exist | 00:19 |
| apostols__ | Kmos, dont exist in Dapper | 00:20 |
| Kmos | you have all repositories active? | 00:20 |
| RAOF | Kmos: He's right. tzdata was introduced in Edgy. | 00:20 |
| StevenK | Kmos: tzdata was split out from locales in Edgy | 00:20 |
| apostols__ | Kmos, http://packages.ubuntu.com/cgi-bin/search_packages.pl?keywords=tzdata&searchon=names&subword=1&version=dapper&release=all | 00:20 |
| Kmos | ups.. he's right.. | 00:21 |
| Kmos | StevenK: you're right | 00:21 |
| Kmos | as always =) | 00:21 |
| apostols__ | How to fixed its? | 00:21 |
| StevenK | apostols__: Make sure a bug is filed in Launchpad and has a task open against locales in Ubuntu Dappper | 00:23 |
| apostols__ | Oh | 00:23 |
| apostols__ | :S | 00:23 |
| Kmos | !info locales dapper | 00:23 |
| ubotu | locales: common files for locale support. In component main, is required. Version 2.3.18.7 (dapper), package size 3208 kB, installed size 12788 kB | 00:23 |
| slangasek | apostols__: have you checked whether version 2.3.18.7 of locales in dapper-updates includes this fix? | 00:24 |
| Kmos | https://edge.launchpad.net/ubuntu/+source/langpack-locales | 00:24 |
| Kmos | slangasek: that version has the fix.. | 00:24 |
| Kmos | and it's at -updates | 00:24 |
| apostols__ | slangasek, I have installed Version: 2.3.18.7 | 00:29 |
| apostols__ | slangasek, in my source.lists have deb http://ve.archive.ubuntu.com/ubuntu/ dapper-updates main restricted | 00:30 |
| apostols__ | slangasek, this package is in main tree | 00:30 |
| slangasek | apostols__: and this version has the wrong timezone data for Venezuela? | 00:30 |
| apostols__ | slangasek, yes | 00:30 |
| apostols__ | slangasek, one second. I try test the timezone | 00:31 |
| apostols__ | slangasek, This work | 00:34 |
| apostols__ | root@arismendi:~# date | 00:35 |
| apostols__ | Sun Dec 9 02:59:53 VET 2007 | 00:35 |
| apostols__ | root@arismendi:~# date | 00:35 |
| apostols__ | Sun Dec 9 02:30:01 VET 2007 | 00:35 |
| apostols__ | slangasek, Thanks for you help | 00:37 |
| slangasek | apostols__: are you saying that it works now? I don't understand what you're showing, all I see is that your clock jumped back in time by a half hour. :) | 00:37 |
| apostols__ | sladen, Yeah, this change of timezone back in time half hour (the change is -04:00 to -04:30) | 00:38 |
| === dendrobates is now known as dendro-away | ||
| apostols__ | slangasek, Yeah, this change of timezone back in time half hour (the change is -04:00 to -04:30) | 00:38 |
| apostols__ | sladen, excuseme | 00:38 |
| slangasek | ok | 00:40 |
| stratus | slangasek, please come back w/ my vorlon! | 01:04 |
| === tonyy is now known as tonyyarusso | ||
| infinity | geser: Have you filed a bug about the versioned build-dep thing? | 02:01 |
| infinity | geser: Would be nice to have a way to track it. | 02:01 |
| === tonyy is now known as tonyyarusso | ||
| * Hobbsee waves | 02:11 | |
| pwnguin | which is the best way to measure a program's memory usage in GNOME? | 04:00 |
| === asac_ is now known as asac | ||
| === jamesh_ is now known as jamesh | ||
| === RAOF_ is now known as RAOF | ||
| dholbach | good morning | 06:11 |
| === Shely is now known as iE18 | ||
| === LongPointyStick is now known as Hobbsee | ||
| === \sh_away is now known as \sh | ||
| pitti | Good morning | 08:09 |
| dholbach | hey pitti | 08:11 |
| \sh | moins | 08:15 |
| dholbach | hey seb128 | 08:18 |
| seb128 | hello dholbach | 08:18 |
| seb128 | pitti: can you give a build retry to evolution? | 09:31 |
| pitti | seb128: done | 09:31 |
| seb128 | pitti: danke | 09:31 |
| seb128 | pitti: ditto for nautilus? ;-) | 09:39 |
| pitti | done | 09:39 |
| seb128 | thanks! | 09:39 |
| TheMuso | If an archive admin is around and doing duties, could they please promote python-pyatspi from universe? Its source package, at-spi is in main. Python-atspi is needed by gnome-orca to function. Otherwise, I assume I need to file a bug to get this taken care of? | 09:51 |
| pitti | TheMuso: no, that's fine | 09:51 |
| pitti | TheMuso: done | 09:52 |
| TheMuso | pitti: Thanks. | 09:55 |
| StevenK | pitti: It seems gcrypt is still causing havoc | 09:58 |
| StevenK | creating libqgis_core.la | 09:58 |
| StevenK | /bin/sed: can't read /lib/libgcrypt.la: No such file or directory | 09:58 |
| pitti | meh; can libtool finally decide where it wants to look for the .la? | 09:58 |
| StevenK | It seems not | 09:59 |
| StevenK | Should we put the .la in /lib, and have a symlink from /usr/lib? | 09:59 |
| pitti | StevenK: I thought Riddell already did something like that? | 10:01 |
| StevenK | He did? | 10:01 |
| pitti | ah, that was in libgpg-error | 10:01 |
| pitti | so, yes | 10:02 |
| pitti | (bwah) | 10:02 |
| StevenK | So, should I do the same for gcrypt? | 10:03 |
| pitti | StevenK: either that, or we give up and revert all our patches *sigh* | 10:03 |
| StevenK | pitti: I daresay adding a symlink should be simple-ish | 10:04 |
| pitti | it is | 10:04 |
| TheMuso | Is libtool just not realizing that .la files it needs to fnd or packages have moved or something? | 10:04 |
| TheMuso | find | 10:04 |
| StevenK | pitti: I think we need to ask for battle pay, and go on a rampage hurting the libtool authors | 10:04 |
| pitti | TheMuso: it's behaving bizarre | 10:04 |
| pitti | TheMuso: the basic problem is that libtool seems to have no way to install a lib into /lib | 10:05 |
| TheMuso | pitti: No kidding. | 10:05 |
| pitti | so, if it's there, it sometimes looks for the .la in /usr/lib, and sometimes in /lib | 10:05 |
| TheMuso | Right. | 10:05 |
| === seb128_ is now known as seb128 | ||
| * StevenK hacks libgcrypt11 | 10:06 | |
| StevenK | pitti: Real file in /lib or /usr/lib? | 10:06 |
| pitti | StevenK: /usr/lib, I'd say, but it shouldn't matter much | 10:07 |
| pitti | just for consistency with libgpg-error | 10:07 |
| soren | dpkg uploaded. | 10:13 |
| * soren starts to sweat uncontrollably | 10:13 | |
| pitti | wooo! | 10:13 |
| StevenK | Hah | 10:14 |
| * pitti congratulates soren for being TIL now | 10:14 | |
| * seb128 hugs soren | 10:14 | |
| soren | pitti: :p | 10:15 |
| StevenK | pitti: Giving libgcrypt11 a nice hard test build before uploading | 10:17 |
| soren | I've been looking into kvm and qemu a lot lately, and I've stumbled upon bug 64501. | 10:20 |
| ubotu | Launchpad bug 64501 in openhackware "FTBFS" [Critical,Triaged] https://launchpad.net/bugs/64501 | 10:20 |
| soren | It's an arch: all package that needs to be built on powerpc. | 10:20 |
| soren | Would be acceptable to change that package to be arch: powerpc, build it, and make an arch: all package that ships the resulting files (kind of like ia32-libs for amd64)? | 10:20 |
| pitti | eww | 10:22 |
| soren | I know :) | 10:22 |
| pitti | soren: why not just keep it as Arch: powerpc? | 10:22 |
| soren | Better suggestions? | 10:22 |
| soren | Because it's a BIOS image that qemu needs on other archs as well. | 10:22 |
| pitti | ah | 10:22 |
| soren | ..to be able to emulate powerpc. | 10:22 |
| soren | And it only builds on powerpc. | 10:23 |
| * pitti rolls back to "Eww" | 10:23 | |
| StevenK | Hah | 10:23 |
| pitti | soren: do you know why it can only be built on powerpc? | 10:23 |
| pitti | does it actually compile the bios? | 10:23 |
| soren | AFAIR, yes. | 10:23 |
| TheMuso | soren: I can test build on PowerPC if you need it tested before you upload. | 10:24 |
| soren | TheMuso: I have access to a powerpc machine, but thanks for the offer! | 10:24 |
| TheMuso | soren: Ok no problem then. | 10:24 |
| soren | pitti: It might just be the case that my crossbuild-fu isn't strong enough. | 10:25 |
| StevenK | Or that dpkg stole it | 10:25 |
| * StevenK sniggers, and feeds soren's complex | 10:25 | |
| * soren whimpers | 10:26 | |
| StevenK | pitti: libgcrypt11 uploaded, and I've upped my bounty on the libtool developers. | 10:28 |
| pitti | great, thanks | 10:28 |
| pitti | /msg StevenK let's have a good beating at it in the afternoon | 10:28 |
| StevenK | The bounty, the developers or libtool? :-) | 10:29 |
| Treenaks | StevenK: hmm.. are those devs in paris? | 10:29 |
| pitti | libtool; I'm not *that* cruel :) | 10:29 |
| StevenK | Treenaks: I'm only naming the guilty party. | 10:29 |
| StevenK | :-P | 10:29 |
| pitti | StevenK: for the developers, my evil plan is to lock them into a room with libgpg-error and libtool, and not allow them to come out again until they made it work right | 10:30 |
| StevenK | pitti: Add libgcrypt11 and cryptsetup with /usr on a seperate partition, and you have my vote. | 10:30 |
| Hobbsee | soren: how the *hell* did you manage to bork dpkg like that? | 10:33 |
| soren | Hobbsee: :p | 10:33 |
| Hobbsee | YOU BROKE IT! BAD SOREN! | 10:33 |
| soren | Hobbsee: You should at least until the thing has actually built and been published before trying that :) | 10:34 |
| Hobbsee | soren: haven't you heard that buildd admins can see, even before others can, and can smell breakage almost immediately? | 10:34 |
| soren | Hobbsee: No, I must have missed the memo. | 10:35 |
| Hobbsee | soren: now you know. | 10:35 |
| soren | Noted :) | 10:35 |
| soren | You know... One day, a long time from now, I'll stumble upon that note, and wonder why "buildd admins have superhuman sense of smell". | 10:36 |
| soren | I could make it even more cryptic and make it say "buildd admins smell really well". | 10:37 |
| Hobbsee | haha | 10:37 |
| * Hobbsee has uploaded dpkg before, and only just survived to tell the tale | 10:37 | |
| StevenK | soren: You could make it an in-joke and s/ really well// | 10:38 |
| * StevenK hides from Hobbsee, pitti and Mithrandir | 10:38 | |
| * Hobbsee beats StevenK | 10:38 | |
| StevenK | Ouch! | 10:38 |
| soren | Hobbsee: That's odd. I don't see your name in the changelog? | 10:38 |
| Hobbsee | soren: i sponsored a change | 10:39 |
| soren | Hobbsee: Oh. | 10:39 |
| Hobbsee | then had iwwj attempting to eat me | 10:39 |
| StevenK | Who's change did you sponsor? | 10:40 |
| StevenK | Oh damn | 10:40 |
| StevenK | pitti: Can you reject libgcrypt11? | 10:40 |
| pitti | StevenK: no | 10:41 |
| StevenK | Drat | 10:41 |
| * StevenK quickly uploads a new version | 10:41 | |
| pitti | StevenK: nowadays sources wander straight into DONE | 10:41 |
| pitti | no reject-from-accepted any more | 10:41 |
| StevenK | Bad Soyuz | 10:42 |
| StevenK | There we go, uploaded | 10:42 |
| Hobbsee | StevenK: no, it's just more efficient now :P | 10:42 |
| * StevenK raises an eyebrow. | 10:43 | |
| Hobbsee | you're complaining about it being more efficient? | 10:43 |
| StevenK | It introduces that you can't reject uploads, and besides, it seems to wait until sources are published before creating build records | 10:43 |
| * StevenK idly wonders if Soyuz deals with Pending -> Superseded | 10:45 | |
| StevenK | Hrm. So far both -2ubuntu6 and -2ubuntu7 are Pending | 10:46 |
| ogra | urgh, somewhitng really weird happened to my fonts since the last update | 10:50 |
| ogra | ah, restarting FF helps :) | 10:51 |
| pitti | StevenK: no, it doesn't; build records can be created immediately after upload | 10:51 |
| pitti | of course it actually takes eons, but it's not tied to publishing | 10:52 |
| StevenK | Oh right, so it the build machinery just taking that long, and not waiting for the publisher | 10:52 |
| pitti | right | 10:54 |
| pitti | StevenK: ideally build records would be created immediately on accepting, and that's in fact the plan | 10:55 |
| lool | Is it easy to run my own local CD builds, perhaps with locally modified packages or package lists, to have ISOs to run in emulators? | 11:07 |
| lool | Or are you people using daily images and pushing changes via hardy to see how they affect the live CD? :) | 11:08 |
| soren | Hmm... I'm trying to determine what changed in debian-policy between 3.7.2.2 and 3.7.3... The changelog lists only bugfixes afaics, but that would usually cause the version to be bumped to 3.7.2.3 rather than 3.7.3.0 wouldn't it? | 11:09 |
| lool | soren: Well there are some changes in upgrading-checklist such as using ${binary:Version} instead of ${Source-Version} which are required by the new version | 11:10 |
| lool | soren: Or the recommendation to link to the GFDL in /usr/share/common-licenses instead of copying it | 11:11 |
| soren | lool: Hm.. Ok. That's it? | 11:12 |
| lool | soren: There are other things as well; upgrade-checklist lists some stuff which you probably wouldn't want to care about, but it has some useful bits I think | 11:13 |
| lool | I don't think it's very useful to mention "may do" stuff when it's already widely done for example :) | 11:14 |
| soren | lool: I'm just looking for a list of functional changes (not just fixed typos or whatever), so that I can tell if I can update my package to "Standards-Version: 3.7.3". :/ | 11:14 |
| lool | soren: Well upgrade-checklist is meant to be exactly this | 11:14 |
| soren | lool: Ah... | 11:15 |
| soren | lool: Ok, now I get it. :) Thanks. | 11:15 |
| lool | It's a bit too long to my taste, but it's supposed to list effective things you should check when upgrading to a newer standard-versions | 11:15 |
| lool | On my packages, I'd probably double-check binary:Version, check copyright for GPLv3 or GFDL, check menu sections (but probably lintian would tell me anyway), and that's probably all I'm using | 11:16 |
| soren | * The Source field in a .changes file may contain a version number | 11:16 |
| soren | in parentheses. | 11:16 |
| lool | Yeah, that's really a lot of chatter for something probably two people care about | 11:17 |
| soren | That sounds odd. | 11:17 |
| soren | lool: Ok. Thanks for pointing out upgrade-checklist a sufficient number of times for me to actually get it. :) | 11:19 |
| lool | soren: Ah you didn't know about the fiel? | 11:19 |
| soren | lool: Nope. | 11:19 |
| lool | I discovered it after a lot of time too; but it's mentionned in the changelog this time around :) | 11:20 |
| === cprov-out is now known as cprov | ||
| soren | Oh, really? | 11:20 |
| TomaszD | who can I speak with about whitelisting a laptop for acpi? | 11:21 |
| cjwatson | lool: it's unfortunately tedious to run one's own CD builds | 11:21 |
| cjwatson | the code's all there, but it's something of a pain to set up | 11:21 |
| cjwatson | if you need it, I can help you | 11:21 |
| cjwatson | you also need a full local mirror, at least of main and restricted | 11:22 |
| soren | lool: Well, even if it did, I wouldn't have considered looking at it. I think I expected it to be a checklist for when upgrading packages, like: "Make sure if files move, you put conflicts: and replaces:" or something. I didn't put much thought into the matter, clearly. | 11:22 |
| cjwatson | the Source thing is support for dpkg experiments I think | 11:22 |
| TomaszD | someone resigned from triaging my bug and I've been waiting for nine months | 11:23 |
| TomaszD | https://bugs.launchpad.net/ubuntu/+source/acpi-support/+bug/92806 | 11:23 |
| ubotu | Launchpad bug 92806 in linux-source-2.6.22 "N340S8 laptop needs to have ACPI enabled" [Undecided,Confirmed] | 11:23 |
| lool | cjwatson: Is it ok to have a http_proxy like squid? | 11:23 |
| cjwatson | lool: no, it really needs a mirror on disk | 11:23 |
| lool | Ah | 11:23 |
| lool | That sounds like many intermediate steps to reach the point where I can do actual tests; I think I'll defer that a little then | 11:23 |
| cjwatson | I suppose you could mount one with something to do with fuse over a squid proxy but argh | 11:24 |
| lool | cjwatson: Is it easy to override packages or package lists -- bit still keeping the mirrors a clean true mirror? | 11:24 |
| lool | s/bit/but | 11:24 |
| cjwatson | lool: the cdimage code has some (unused for some time) support for local packages | 11:25 |
| cjwatson | back when we were doing warty it took a while to get Ubuntu itself into shape, so I had local overrides of a few packages in cdimage in order to be able to build installable CDs | 11:25 |
| cjwatson | that would be easier than changing the mirror | 11:25 |
| lool | Ok; perhaps I'll just boot a CD and use casper USB persistence if that's still available to install my experiments and see how to do that properly later | 11:26 |
| cjwatson | typically I just modify things on the fly if I need to do this sort of test, I must admit | 11:26 |
| lool | Ok; thanks for the info | 11:26 |
| cjwatson | I do think it would be useful to have a facility for this eventually, but no capacity for it for hardy ... | 11:26 |
| cjwatson | (as in a proper user-oriented facility for customising CD images; we talked about it at UDS) | 11:27 |
| lool | cjwatson: If I wanted to look into building hardy daily images, where should I start? I think it's "ubuntu-cd"? | 11:28 |
| cjwatson | http://people.ubuntu.com/~cjwatson/bzr/cdimage/mainline/ and also check out the bits listed in configs/devel | 11:28 |
| soren | cjwatson: How do you usually test stuff like gfxboot and isolinux then? Extract an existing iso, replace the relevant files, re-mkisofs, test? | 11:28 |
| cjwatson | soren: I have a little manually-driven test suite for the gfxboot theme | 11:29 |
| lool | cjwatson: Noted, thanks | 11:29 |
| cjwatson | it's just a thing where I can do 'make -C test' and it fishes stuff out of known places in my local directory layout | 11:29 |
| cjwatson | I don't use a full tree for it, just a few files that are enough to test the theme | 11:30 |
| soren | cjwatson: Alright. | 11:30 |
| cjwatson | soren: http://people.ubuntu.com/~cjwatson/tmp/gfxboot-test.tar.gz | 11:30 |
| cjwatson | bit dated but hasn't changed much either | 11:31 |
| === Hobbsee is now known as LongPointyStick | ||
| cjwatson | pitti: I assume you'll want another d-i for dapper-proposed? | 11:31 |
| * soren looks | 11:31 | |
| pitti | cjwatson: yes, please; if that lbm ever condescends to build | 11:32 |
| pitti | but yeah, having d-i in the queue would be great | 11:32 |
| LongPointyStick | where's my supersonic jet? | 11:32 |
| soren | cjwatson: Ah, clever. | 11:33 |
| cjwatson | pitti: heading queuewards now | 11:33 |
| soren | LongPointyStick: Where are you going in that? | 11:33 |
| === dholbach_ is now known as dholbach | ||
| pitti | Riddell: do you know about libkarma? the MIR (https://wiki.ubuntu.com/MainInclusionReportlibkarma) is almost useless | 11:50 |
| pitti | Riddell: does it actually work on a standard Ubuntu kernel? the cited pmount bug is long obsolete | 11:50 |
| === jpatrick_ is now known as jpatrick | ||
| Riddell | pitti: I'm told it does yes | 11:50 |
| cjwatson | TomaszD: the bug was confirmed so there was no longer a need for a triager anyway | 11:55 |
| cjwatson | TomaszD: odd though, the patch for your system still seems to be there | 11:55 |
| TomaszD | cjwatson, indeed, but it doesn't work :( | 11:56 |
| TomaszD | after dapper or edgy I have to acpi=force in menu.lst | 11:56 |
| cjwatson | TomaszD: yeah, it's just that you explicitly said the patch was removed and I don't see that it was | 11:57 |
| cjwatson | so trying to work out what happened | 11:57 |
| TomaszD | cjwatson, ok. You know IANAE, but I thought that once I don't have the functionality the patch must have been dropped by mistake | 11:58 |
| cjwatson | TomaszD: to be clear, we're talking about ACPI sleep, right? | 11:58 |
| pitti | asac: xulrunner MIR> when we will switch firefox to 3 by default? | 11:58 |
| cjwatson | (you just said "ACPI" in the bug) | 11:58 |
| TomaszD | cjwatson, we're talking about turning on ACPI at all | 11:58 |
| TomaszD | cjwatson, because I get the message about the bios being to old and going under the bios cut-off age | 11:59 |
| cjwatson | err, acpi-support doesn't have one of those | 11:59 |
| TomaszD | cjwatson, so it's not whitelisted then? | 11:59 |
| cjwatson | perhaps it would help if you copied the *exact* message you're seeing into the bug | 11:59 |
| asac | pitti: in january i guess ... there is still too much work going on upstream to not cause any confusion ... for now i need xul 1.9 in main to update all the other rdepends. | 12:00 |
| seb128 | mdke: any news about the yelp layout patch update? | 12:00 |
| asac | pitti: maybe b2 will be good enough though (which will be next week i guess) | 12:01 |
| * pitti switches his desktop to ffox 3 for testing love | 12:01 | |
| TomaszD | it appears just for a second, it says your bios falls under the bios cut-off age, because it's older than 1998 (which isn't true, it's 2002), so acpi functions will have to be forced to be enabled cjwatson | 12:01 |
| pitti | asac: ah, I see; but we'll definitively drop 2 for hardy | 12:01 |
| cjwatson | TomaszD: I've commented on the bug; it's not an acpi-support problem, it's a kernel problem | 12:01 |
| asac | pitti: yes. we cannot support ffox 2 for such a long time | 12:01 |
| TomaszD | whenever I install a new version of ubuntu cjwatson I just go menu.lst and acpi=force there, I got used to it, but acpi was working in dapper | 12:02 |
| asac | pitti: as proposed in https://wiki.ubuntu.com/XulrunnerGecko ... we will allow universe rdepends to switch to xulrunner 1.8 ... in case they don't get ready in time. | 12:02 |
| cjwatson | I imagine the kernel got stricter | 12:03 |
| pitti | asac: ah, so in ffox 3 I don't have those ugly typewriter fonts (as broken by recent fontconfig), but I don't have the smooth fonts either; but slightly better, I'd say :) | 12:03 |
| === xerakko_ is now known as xerakko | ||
| asac | pitti: yeah ... there is a request to patch xulrunner for subpixel stuff (bug 164640) ... maybe thats the problem? | 12:04 |
| ubotu | Launchpad bug 164640 in xulrunner-1.9 "Build Firefox 3 against a subpixel-patched cairo" [Undecided,Confirmed] https://launchpad.net/bugs/164640 | 12:04 |
| pitti | asac: I guess so | 12:04 |
| pitti | asac: ok, promoted | 12:05 |
| asac | pitti: rock! | 12:05 |
| pitti | asac: but what's wrong with switching to 3 early? after all, testing is why we all run hardy now? | 12:06 |
| seb128 | xulrunner1.9 promoted? | 12:07 |
| seb128 | which means we can build the GNOME packages using it now? | 12:07 |
| asac | seb128: yeah ... if you apply the patches i have ;) | 12:07 |
| asac | seb128: https://code.edge.launchpad.net/~mozillateam/xulrunner/xulrunner-porting | 12:08 |
| asac | there is yelp, devhelp and epiphany ... though epiphany is against current trunk | 12:08 |
| asac | i will add the python gtkmozembed today i hope (after cleaning things up) | 12:09 |
| asac | seb128: ah ... totem is in that xulrunner-porting thing as well | 12:09 |
| pitti | cjwatson: WDYT about libssh? There's a MIR for it, but I have some concern about having two ssh implementations in main | 12:10 |
| asac | pitti: i wanted to fix xulrunner to consider all the needed plugins/extensions directories before making firefox-3.0 default ... otherwise we will end up installing plugins in a place that is not ment to be the final one. | 12:10 |
| pitti | asac: ah, that's a good point, yes | 12:11 |
| seb128 | asac: do you want to do the upload or should have a look to that? I've nothing special to do this afternoon so I'm fine looking at the changes and the wiki page I said I would read the other day and upload the xulrunner versions ;-) | 12:11 |
| asac | seb128: i would prefer if you could take the uploading part and maybe even upstream submission | 12:12 |
| asac | seb128: before you touch a package better ask me for any regression i know of ... so you can decide if its good enough for initial upload | 12:13 |
| asac | to start with: yelp patch has no known regression for me | 12:14 |
| === dendro-away is now known as dendrobates | ||
| seb128 | asac: ok, will do that after lunch then, thanks | 12:17 |
| asac | seb128: nothing to hurry ... you need to wait till the xulrunner-1.9 package i uploaded today is build because it changed the names of the .pc files to the final ones | 12:18 |
| seb128 | asac: ok | 12:22 |
| cjwatson | pitti: since openssh is never likely to provide a library, it doesn't hugely bother me | 12:24 |
| pitti | cjwatson: yeah, but it just feels...wrong | 12:24 |
| cjwatson | your call | 12:25 |
| * gicmo yawns | 12:26 | |
| LongPointyStick | for the love of the great green arklesneezer, this is absolutley ridiculous! | 12:46 |
| LongPointyStick | grr! | 12:46 |
| Nafallo | hmmm | 12:46 |
| Nafallo | LongPointyStick: what now? | 12:47 |
| LongPointyStick | my system has frozen 4 times in the past 10 minutes! | 12:47 |
| LongPointyStick | that's a new low for hardy. | 12:47 |
| LongPointyStick | tjaalton: where should i start looking as to why it's freezing? | 12:48 |
| dholbach | MOTU Q&A session in 11 minutes in #ubuntu-classroom | 12:48 |
| tjaalton | LongPointyStick: well, there have been no updated x-packages lately, to my knowledge :) | 12:52 |
| LongPointyStick | tjaalton: this has happened for a while, but seems worse atm. | 12:53 |
| LongPointyStick | oh, interesting. compiz is showing 100% cpu | 12:53 |
| ogra | compiz was updated today :) | 12:53 |
| ogra | and yesterday | 12:54 |
| tjaalton | well, there you go :) | 12:54 |
| LongPointyStick | and now it behaves, with another machine ssh'd in | 12:58 |
| Hobbsee | or not. | 13:00 |
| StevenK | Heh | 13:00 |
| asac | hmm on lpia i get a strange build failure for xulrunner-1.9 : http://paste.ubuntu.com/2524/ | 13:01 |
| Nafallo | compiz doesn't do the newer Intels yet, right? | 13:01 |
| Hobbsee | Nafallo: this isnt' really a newer intel | 13:03 |
| Nafallo | Hobbsee: was more of a general question dear :-) | 13:03 |
| Nafallo | GM965 / X3100 or so. | 13:04 |
| Hobbsee | oh | 13:04 |
| asac | pitti: maybe the promotion in the mids of a xul build caused these "failed to upload" errors? https://edge.launchpad.net/ubuntu/+source/xulrunner-1.9/1.9~b1+nobinonly-0ubuntu2 | 13:06 |
| Hobbsee | tjaalton: actually, i don't think you get blamed for this. it appears that compiz shows 100% CPU and locks up (while the programs all still run), when more than 1 active "notify me" window is there. | 13:06 |
| pitti | asac: right; let me give them back | 13:06 |
| Hobbsee | asac: yes, that's correct | 13:06 |
| asac | pitti: thanks! | 13:06 |
| torkel | Nafallo: they removed the blacklistning of 965 in -0ubuntu3 | 13:07 |
| * asac lunch | 13:07 | |
| Nafallo | torkel: cheers. sounds scary 'nough to update according to Hobbsee though ;-) | 13:07 |
| Nafallo | ...and Fujitsu | 13:08 |
| Hobbsee | Fujitsu: you still around? | 13:17 |
| === cassidy_ is now known as cassidy | ||
| pitti | asac: is there a way to tell gnome to use ffox-3 as my prefered application? it opens ffox 2 on clicking on a link ATM, but ffox 2 doesn't appear in gnome-default-applications-properties | 13:23 |
| lool | pitti: You can set a manual command? | 13:23 |
| lool | pitti: %s for the URL | 13:24 |
| pitti | yeah, of course | 13:24 |
| pitti | but it shuold still appear there | 13:24 |
| lool | It needs to be patched in control-center IIRC | 13:24 |
| seb128 | pitti: how is the binary called? | 13:24 |
| pitti | firefox-3.0 | 13:24 |
| pitti | oh, it's hardcoded there? | 13:24 |
| lool | Yeah | 13:24 |
| pitti | I thought it was some .desktop file somewhere | 13:24 |
| pitti | after all, even lynx appears there :) | 13:25 |
| lool | I think it has a mapping | 13:25 |
| seb128 | pitti: /usr/share/gnome-control-center/gnome-default-applications.xml has the list of known applications | 13:25 |
| pitti | seb128: aah, I understand; thanks | 13:25 |
| seb128 | need to patch gnome-control-center | 13:25 |
| seb128 | I'll do it | 13:25 |
| pitti | seb128: well, shouldn't be necessary in the end | 13:25 |
| pitti | seb128: once we ship 3 by default, the binary will be 'firefox' (hopefully :) ) | 13:26 |
| seb128 | if the binary is renamed firefox no need to rename | 13:26 |
| seb128 | right | 13:26 |
| pitti | I just wondered by which mechanism a package appears there | 13:26 |
| pitti | so I was concerned that ffox-3 needs to ship a .desktop file for that, or so | 13:26 |
| seb128 | that would be a better way to do things | 13:27 |
| seb128 | but at the moment the list is in gnome-control-center | 13:27 |
| pitti | seb128: ok; nevermind then | 13:27 |
| pitti | and thanks for the explanation | 13:27 |
| seb128 | you are welcome | 13:27 |
| === cjwatson_ is now known as cjwatson | ||
| === Igorots is now known as Knightlust | ||
| === Hobbsee_ is now known as Hobbsee | ||
| === doko__ is now known as doko | ||
| === \sh is now known as \sh_away | ||
| xhaker | seb128, can i discuss default torrent client for Ubuntu with you? | 14:30 |
| seb128 | xhaker: if you want too, I'm not sure I'm the best person for that discussion though since I don't know the alternatives nor use bittorrent | 14:32 |
| xhaker | seb128, just thought you were appropriate since you are a "gnome" guy. My take is that gnome-btdownload + bittorrent should step back | 14:33 |
| xhaker | I've been thinking to myself, that Transmission is very much the best for a default torrent client | 14:34 |
| xhaker | While trying out fedora8 live through an usbkey, i noticed they have transmission there. :) I wish we could do the same. | 14:36 |
| seb128 | xhaker: maybe mail ubuntu-desktop or ubuntu-devel-discuss list about some rational on why you think transmission is better | 14:36 |
| seb128 | I think some people did in the past but other users replied that transmission was not stable and crash all the time, etc | 14:36 |
| xhaker | seb128, it might be true.. but i've been following the development.. they're really active.. and they're running up for the 1.0 release.. last version on the repositories is 0.95 | 14:37 |
| xhaker | Will send my rationale to ubuntu-desktop ml | 14:37 |
| seb128 | xhaker: is there anybody looking at the launchpad bugs for transmission? like an upstream guy? | 14:37 |
| azeem | just because they call it 1.0 doesn't mean it's bug-free | 14:37 |
| seb128 | that would show they have interest to have it used in ubuntu ;-) | 14:38 |
| seb128 | xhaker: thanks | 14:38 |
| xhaker | azeem, it means they're pretty confident it'll be good.. check their trac page.. they look into HIG, and there are lots of worrying with memory leaks :D | 14:39 |
| * ogra wonders if there is a way to compile dillo with xulrunner now :P | 14:44 | |
| * xhaker is not smart enough to understand why debian and ubuntu build-depend differently in respect of xulrunner-dev and firefox-dev | 14:47 | |
| poningru | xhaker please also add deluge to the possibility of default torrenting clients | 14:47 |
| xhaker | poningru, don't get me wrong.. I'm using deluge right now.. I have both clients installed.. but Transmission is leaps ahead in ease of use | 14:48 |
| ogra | xhaker, lawyers will uderstand that :) | 14:48 |
| xhaker | poningru, it just downloads, and does a good job at that. :) hence the proposal for shipping in the default install | 14:49 |
| poningru | xhaker, ah k | 14:49 |
| === cprov is now known as cprov-lunch | ||
| poningru | someone also needs to solve the natting problem with those things | 14:50 |
| poningru | as in some client program that works with routers that automatically adds a port | 14:50 |
| poningru | forwarding | 14:50 |
| xhaker | Transmission will use http://miniupnp.free.fr/ projects.. provides upnp and natpmp | 14:52 |
| xhaker | It will maybe handle more routers than their solution now | 14:52 |
| pitti | seb128: I assume totem-pl-parser should go into main? | 15:33 |
| seb128 | pitti: yes | 15:33 |
| seb128 | pitti: totem will depends on it when it's accepted | 15:33 |
| seb128 | pitti: danke ;-) | 15:34 |
| pitti | seb128: looks good to me, accepted | 15:34 |
| pitti | *whistle* NEW is empty again | 15:35 |
| * Hobbsee uploads more | 15:37 | |
| ScottK | pitti: Did you figure out how to remove crack from partner yet? | 15:37 |
| pitti | ScottK: yes, it's actually gone; I just didn't close the bug yet | 15:38 |
| ScottK | pitti: Great. Thanks for that/ | 15:38 |
| ScottK | /. | 15:38 |
| === cprov-lunch is now known as cprov | ||
| Nafallo | ehrm | 16:00 |
| Nafallo | "remove crack for partner". that can be quite misunderstood ;-) | 16:00 |
| pitti | BenC: re lbm patch> I sent a mail about that this morning with a pointer to the patch | 16:03 |
| BenC | pitti: oops, must have missed that, sorry | 16:03 |
| pitti | BenC: no problem; it's attached to bug 164449, so you can grab it from there | 16:04 |
| ubotu | Launchpad bug 164449 in linux-backports-modules-2.6.15 "undefined symbols in mptspi driver" [High,Fix released] https://launchpad.net/bugs/164449 | 16:04 |
| Hobbsee | BenC: can i ask when we get l-u-m for .24? | 16:04 |
| BenC | Hobbsee: before monday | 16:04 |
| Hobbsee | BenC: \o/ | 16:04 |
| pitti | cjwatson: wrt. getting the noptrace group into base-passwd; do you think we should discuss abusing an existing group and put that into a meeting (TB or whatever), or shall I file a bug about adding 'noptrace' to base-passwd? | 16:30 |
| cjwatson | I'd like it to be discussed on debian-devel | 16:32 |
| cjwatson | base-passwd is one package I really don't want to be out of sync | 16:32 |
| cjwatson | and I want complete propriety in how changes to the master files happen | 16:33 |
| pitti | ok, I'll send a mail there | 16:33 |
| cjwatson | thanks | 16:34 |
| ScottK | Nafallo: Pick a better word that characterizes uploading software to a public repository with known remote code exploits unpatched? | 16:37 |
| keescook | mornin' | 16:40 |
| Nafallo | ScottK: foolness. what did I do now? | 16:41 |
| ScottK | Nafallo: Not you. I was responding to your comment about crack and partner. | 16:45 |
| Nafallo | ScottK: ah :-) | 16:45 |
| lool | I don't understande why some uploads to hardy-changes aren't signed | 16:47 |
| cjwatson | lool: syncs are unsigned | 16:47 |
| lool | 3.16.0-4ubuntu1 => probably not a sync; I think the body of the mail has in fact a signature, but Mutt doesn't recognize it as such | 16:48 |
| cjwatson | we just copy those from Debian, and they're done on a trusted system and inserted through basically a back door in the queue system rather than inventing a fake key to sign them with | 16:48 |
| cjwatson | ah | 16:48 |
| lool | Either an encoding issue or a Mutt bug | 16:48 |
| cjwatson | ScottK: not that it was the right thing to do; but we did know that the problems in openssl097 did not affect vmware | 16:49 |
| ScottK | cjwatson: I understand that. That was no where publically documented and certainly doesn't help a bit when people (our competition) says ubuntu-server isn't secure enough to take seriously. The percpetion (if not the fact) was very poor. | 16:50 |
| ScottK | It also pained me personally after investing a lot of time in getting it removed from Universe. | 16:50 |
| lool | Ah it's UTF-8 | 16:51 |
| lool | That's the data confusing Mutt | 16:51 |
| cjwatson | ScottK: I know, we still have work to do on a proper response that includes documentation on what will be allowed in partner; it's in progress ... | 16:52 |
| ScottK | cjwatson: Good to hear. Thanks. | 16:52 |
| * ScottK grumbles again that it'd be better if Partner were more clearly a Canonical production than an Ubuntu one. | 16:52 | |
| cjwatson | what can we do to make that clearer? | 16:53 |
| cjwatson | or, perhaps a fairer question, in what places is it currently unclear? | 16:54 |
| ScottK | In LP it's not clear at all. | 16:54 |
| ScottK | For example, bugs against things in Partner are described as "in Ubuntu" | 16:54 |
| cjwatson | hmm, I didn't realise that packages in partner showed up under Ubuntu | 16:54 |
| cjwatson | I agree that that's unfortunate | 16:54 |
| === dendrobates is now known as dendro-away | ||
| cjwatson | (or perhaps I did know and forgot) | 16:55 |
| ScottK | If you go to the front page for the 7.10 release, Partner is listed as just another repository. | 16:55 |
| ScottK | cjwatson: There's an LP bug. Let me find it. | 16:55 |
| cjwatson | I don't see partner mentioned on /ubuntu/gutsy | 16:56 |
| ScottK | cjwatson: Bug #153798 | 16:56 |
| ubotu | Launchpad bug 153798 in soyuz "canonical partner repo packages showing as "in ubuntu"" [Undecided,Confirmed] https://launchpad.net/bugs/153798 | 16:56 |
| * ScottK looks again | 16:56 | |
| cjwatson | ah, it's on /ubuntu | 16:57 |
| ScottK | Yes. Sorry about that. | 16:57 |
| ScottK | That's actually worse. | 16:57 |
| cjwatson | followed up to the bug | 16:58 |
| ScottK | cjwatson: Thanks. | 16:58 |
| cjwatson | (not with good news, I'm afraid) | 16:58 |
| ScottK | cjwatson: I understand your answer, but as a non-Canonical developer, it's very frustrating. | 17:00 |
| blueyed | mjg59`: are you planning to merge powernowd? I've collected a whole bunch of fixes in the last days, which are attached to bug 67341. If you like, I could do the merge and include those - then you could just sponsor it. | 17:50 |
| ubotu | Launchpad bug 67341 in powernowd "powernowd doesn't use /etc/default/powernowd anymore" [Low,Triaged] https://launchpad.net/bugs/67341 | 17:50 |
| === cprov is now known as cprov-out | ||
| geser | pitti: please give-back: compiz-fusion-plugins-extra | 18:03 |
| === dendro-away is now known as dendrobates | ||
| === cjwatson_ is now known as cjwatson | ||
| cjwatson | ScottK: some things are frustrating both inside and outside Canonical ... | 18:11 |
| ScottK | ;-) | 18:14 |
| pitti | cjwatson: d-devel@ mail about ptrace() sent; let the flamefest begin :) | 18:18 |
| pitti | keescook: ^ I BCCed you to that FYI | 18:23 |
| keescook | pitti: cool; thanks. | 18:23 |
| === mekius_ is now known as mekius | ||
| * pitti blinks | 18:28 | |
| pitti | cjwatson: I just wanted to add a gutsy task to your SRU bug, but guess which release is missing on https://bugs.edge.launchpad.net/ubuntu/+source/finish-install/+bug/174689/+nominate | 18:29 |
| ubotu | Launchpad bug 174689 in finish-install "hvc/hvsi consoles not handled" [Undecided,New] | 18:29 |
| pitti | cjwatson: any idea why? | 18:29 |
| cjwatson | because I targeted it myself 10 minutes ago or so | 18:29 |
| pitti | ah, heh; sweet race | 18:30 |
| pitti | thanks | 18:30 |
| pitti | I stared at the check boxes for at least 10 seconds until I realized :) | 18:30 |
| cjwatson | might be worth a Malone bug | 18:30 |
| === dendrobates is now known as dendro-away | ||
| === dendro-away is now known as dendrobates | ||
| === pedro__ is now known as pedro_ | ||
| geser | pitti: please give-back: gdome2-xslt | 18:44 |
| geser | pitti: please give-back: ocamlgraph | 18:50 |
| pitti | geser: both done | 18:50 |
| geser | pitti: please give-back: compiz-fusion-plugins-extra | 18:55 |
| pitti | done | 18:57 |
| geser | thanks | 18:58 |
| Eckzillor | Hello | 19:06 |
| === jdong is now known as jdong2 | ||
| === jdong2 is now known as jdong | ||
| === mathiaz_ is now known as mathiaz | ||
| === ember_ is now known as ember | ||
| sistpoty | hi folks | 20:01 |
| pochu | hiya sistpoty | 20:03 |
| sistpoty | hi pochu | 20:03 |
| geser | should build-dependencies on gs-common be replaced with ghostscript or is build-depending on gs-common still ok (especially for a package in main)? | 20:14 |
| tjaalton | any hal experts around? | 20:40 |
| Burgundavia | tjaalton: you might be better to ask in #hal or #gnome-hackers on gimpnet | 20:42 |
| tjaalton | Burgundavia: right, I'll do that if I can't figure it out myself eventually | 20:44 |
| tjaalton | the problem, that is | 20:44 |
| === jdong is now known as jdong2 | ||
| === jdong2 is now known as jdong | ||
| === nixternal_ is now known as nixternal | ||
| === Adri20001 is now known as Adri2000 | ||
| AgentHeX | i've got a bug in the gnome power manager where the display brightness will "dim" to a value higher than what i manually set with the keyboard shortcut when it thinks i'm idle. i would like to fix this bug. what should i download to take a look? | 22:29 |
| ScottK | AgentHeX: I have a vague recollection that there was a bug on this already. Dunno if it's been fixed in Hardy or not. | 22:29 |
| AgentHeX | ScottK: i'm on gutsy right now. i was kind of looking for an introduction to developing in ubuntu. i'm a C++ coder but i have little experience coding in linux, and i'm looking for a way i can explore. any hints? | 22:30 |
| ScottK | Sorry. I'm a KDE person. | 22:31 |
| AgentHeX | ScottK: i'm pulling eclipse atm, so i'll have that soon, and i figured the gpm bug might be a good start | 22:31 |
| AgentHeX | ScottK: how might i go about getting the source for the gnome power manager? | 22:31 |
| ScottK | I'd suggest looking for the gdm bug and seeing what's already there (it had a proposed patch, dunno how good it is). | 22:32 |
| slangasek | "apt-get source gnome-power-manager" | 22:32 |
| AgentHeX | ah | 22:32 |
| AgentHeX | slangasek: i was looking in synaptic. | 22:32 |
| ScottK | AgentHeX: First lesson is developers will taunt you if you insist on using GUI tools and not command line. | 22:33 |
| slangasek | they will? | 22:33 |
| * slangasek covers up his update-manager | 22:34 | |
| Nafallo | hmm. I use sudo update-manager from a terminal ;-) | 22:34 |
| AgentHeX | ScottK: i would *rather* use a GUI IDE, but i *could* use a CLI | 22:36 |
| ScottK | You'll get over it ;-) | 22:37 |
| AgentHeX | ScottK: i'm familiar enough with vim that i prefer it to a generic text editor (gedit, notepad, et al), and it's proably a good thing since i'm on a laptop and the touchpad is kinda annoying | 22:37 |
| ScottK | Excellent. | 22:37 |
| ScottK | I think your basic strategy of find stuff that annoys you and try to fix it is a good one. | 22:38 |
| AgentHeX | ScottK: but i'd still rather use a gui. if there was a gui version of vim, i'd be in heaven. | 22:38 |
| ScottK | I think there is, but I don't recall for sure. | 22:38 |
| tkamppeter | pitti. hi | 22:38 |
| AgentHeX | ScottK: i'm kinda kidding. what's the point of having a gui version of vim anyway? it's designed to be lightweight. that gui adds tons of cruft | 22:39 |
| AgentHeX | ScottK: now once i actually have the source for the package, how do i go about compiling it? will i need source for all the dependencies as well? | 22:39 |
| * ScottK has no idea, but it's FOSS, so all it takes is one coder who feels liek it. | 22:39 | |
| AgentHeX | i'm pretty sure apt-get source will drop the source in my /usr/src path. can you confirm? | 22:42 |
| ScottK | No, it'll drop it into the current working directory. | 22:43 |
| AgentHeX | ah... do not want in home | 22:43 |
| ScottK | https://wiki.ubuntu.com/PackagingGuide/Howtos/BuildingTheSourcePackage?highlight=%28build%29 is a start on how to compile it. It'll lead you to other quesitons, but I have to run. | 22:43 |
| Kopfgeldjaeger | n8 | 22:44 |
| tophat | has anyone used git instead of svn? | 22:48 |
| tkamppeter | doko, did you upload hplip and hal-cups-utils for me? | 22:51 |
| geser | tkamppeter: is the new dpkg already built? | 22:53 |
| geser | tkamppeter: the current version of hplip needs the newer dpkg for building | 22:53 |
| AgentHeX | ScottK: how might i go about running a modified gpm? do i need to replace the original and hope i don't bork my system? how might i debug something like gpm? | 22:53 |
| slangasek | AgentHeX: anyway, "vim-gnome" :) | 22:55 |
| doko_ | tkamppeter: yes | 22:56 |
| AgentHeX | slangasek: hah. nice | 22:58 |
| tkamppeter | doko_: Thank you very much | 23:00 |
| AgentHeX | oh snap. i can't see my appearances window to turn off compiz. | 23:06 |
| AgentHeX | HAH! i did it blind! | 23:06 |
| tkamppeter_ | geser, I have taken HPLIP from the Debian RPM as I am working together with Mark Purcell from Debian on the maintenance and so our packages and Debian's are the same. | 23:07 |
| tkamppeter_ | On this build I had a problem and I had to remove "--dpkg-shlibdeps-params=--ignore-missing-info" from dh_shlibdeps in debian/rules. | 23:07 |
| tkamppeter_ | My system is a Hardy which I have updated today. | 23:07 |
| tkamppeter_ | geser, would this mean that I can reintroduce this "--dpkg-shlibdeps-params=--ignore-missing-info" on dh_shlibdeps as soon as the newest dpkg hits Hardy? | 23:07 |
| tkamppeter_ | dpkg --version | 23:07 |
| tkamppeter_ | Debian `dpkg' package management program version 1.14.12ubuntu1 (i386) | 23:07 |
| tkamppeter_ | s/Debian RPM/Debian SVN/ | 23:07 |
| === Traxer is now known as Traxer|on | ||
| === Traxer|on is now known as Traxer | ||
| geser | tkamppeter_: if I understand it all correctly, yes | 23:12 |
| tkamppeter_ | geser, so which version dpkg must be so that I can re-introduce "--dpkg-shlibdeps-params=--ignore-missing-info" on dh_shlibdeps? | 23:12 |
| geser | --ignore-missing-info was introduced around .8 or .9 and .12 is getting build now | 23:12 |
| slangasek | tkamppeter_: if you're working with hplip, how about fixing it so that it has a proper shlibs file and doesn't *need* the --ignore-missing-info? | 23:13 |
| geser | according to the changelog --ignore-missing-info works since dpkg 1.14.8 | 23:13 |
| tkamppeter_ | So strange that it did not work with 1.14.12ubuntu1 | 23:14 |
| slangasek | I don't know what possessed the maintainer to do that instead of looking at the reason why dpkg-dev thought the package was wrong :P | 23:14 |
| tkamppeter_ | Mark Purcell has introduced --ignore-missing-info. I do not even know for what that is good for. | 23:15 |
| slangasek | it's to stop dpkg-shlibdeps erroring out with a message about missing shlibs for a library dependency | 23:15 |
| slangasek | and the reason dpkg-shlibdeps was erroring out was because he has a library that's missing a shlibs file (required by policy), which means the package interdependencies are wrong | 23:16 |
| tkamppeter_ | HPLIP seems to build and install fine without --ignore-missing-info. | 23:16 |
| slangasek | not with the new dpkg | 23:17 |
| geser | should build-dependencies on gs-common be replaced with ghostscript or is build-depending on gs-common still ok (especially for a package in main)? | 23:17 |
| tkamppeter_ | geser, I think this can be moved to ghostscript. ghostscript has a transitional package gs-common, but things should depend on the main package. | 23:18 |
| slangasek | note that the "ghostscript" package doesn't exist in dapper, so this will affect ease of backporting | 23:20 |
| geser | slangasek: it's a change for graphviz, see bug #174749 | 23:22 |
| ubotu | Launchpad bug 174749 in graphviz "[hardy] Drop libttf-dev from Build-Depends" [Undecided,New] https://launchpad.net/bugs/174749 | 23:22 |
| geser | gs-common is in universe but ghostscript provides gs-common so build-depending on it should work, but to be on the safe side I changed it | 23:22 |
| slangasek | there's no need here to be "on the safe side". It's perfectly valid to build-depend on virtual packages | 23:24 |
| slangasek | and as I said, build-depending on "ghostscript" is an additional change that has to be made for backports (as well as increasing the size of the Ubuntu delta here, of course) | 23:24 |
| ScottK | slangasek: If it doesn't exist in Dapper, then I'd just backport ghostscript first (no regression risk). | 23:25 |
| geser | so I should revert it back to gs-common? | 23:25 |
| slangasek | ScottK: uh? in what sense is there no risk of regression? it's a complete reorganization of the source packages | 23:26 |
| geser | ScottK: I guess ghostscript was called gs-somehting in dapper | 23:26 |
| ScottK | slangasek: I haven't actually looked at it, but generally if a package doesn't exist, it's good for backporting (I do check before I actually approve these things). Nevermind then. | 23:26 |
| slangasek | ScottK: right; this isn't a "ghostscript doesn't exist in dapper", it's a "someone thought it would be clever to rename all the ghostscript-related packages in Debian because renaming is FUN" | 23:28 |
| ScottK | slangasek: Ah. Yes. That'd be totally different. | 23:28 |
| ScottK | slangasek: I'd say you got the first two letters right. | 23:29 |
| slangasek | sorry for my hash collision | 23:30 |
| slangasek | :) | 23:30 |
| tkamppeter_ | geser, gs-common is in Universe? I thought we have taken it from the distro? It does not make sense any more with the new ghostscript. ghostscript contains everything now. So please remove gs-common from the Universe. | 23:32 |
| tkamppeter_ | geser, in dapper the default ghostscript was gs-esp, as alternative there were also gs-gpl and gs-afpl. | 23:33 |
| ScottK | tkamppeter_: Does it have any rdepends left? | 23:34 |
| tkamppeter_ | slangasek, we decided on renaming it /the Debian maintainer and me) to make it searchable more easily. Searching for ghostscript gives better results than for gs. | 23:35 |
| tkamppeter_ | and with the merger of ESP and GPL Ghostscript (http://www.cups.org/espgs/) which I have done this year all gs-... got obsolete. | 23:36 |
| === tkamppeter_ is now known as tkamppeter | ||
| tkamppeter | ScottK, what are rdepends? | 23:37 |
| slangasek | tkamppeter_: a) gs-common is being built from the ghostscript source as a dummy package, to facilitate smooth upgrades; I'm not sure why it needs to be a transitional package instead of just the gs-gpl and gs-esp packages, but that's the rationale given. b) "apt-cache search ghostscript" already worked fine? | 23:37 |
| tkamppeter | slangasek, is it not the case that for every package which my new package replaces I have to provide a transitional package? And ghostscript replaces gs-common. | 23:40 |
| slangasek | tkamppeter: you just said "please remove gs-common from the Universe", how exactly do you expect it to be removed and still be a transitional package? | 23:40 |
| slangasek | the gs-common package in universe /is/ the transitional package | 23:41 |
| slangasek | as for whether every package needs a transitional package, no; you generally only want transitional packages for things that an end-user is likely to install directly | 23:41 |
| tkamppeter_ | slangasek, so "apt-get dist-upgrade" would work also without transitional packages? Only "apt-get install gs-common" needs them? | 23:52 |
| tkamppeter_ | slangasek, then with the gs-common in Universe being the transitional package everything is OK. Nothing needs to be removed. | 23:53 |
| slangasek | tkamppeter_: "apt-get dist-upgrade" would not do what you want without /some/ transitional packages. But I think users are unlikely to have gs-common alone installed, right? They're likely to have gs-esp or gs-gpl installed? | 23:54 |
| slangasek | tkamppeter_: in which case, you only need to provide transitional packages for gs-esp and gs-gpl, since gs-common would be sorted out according to dependencies | 23:54 |
| tkamppeter_ | slangasek, I understand, gs-common is like a lib... package which does not make sense alone. So can I take it out on the next ghostscript packaging then? | 23:57 |
| slangasek | tkamppeter_: since ghostscript also has a Provides: gs-common that will meet the needs of reverse-deps, that's what I would suggest, yes | 23:58 |
| tkamppeter_ | slangasek, OK and thanks for the help. | 23:58 |
| slangasek | sure | 23:59 |
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!