/srv/irclogs.ubuntu.com/2009/06/14/#ubuntu-mozillateam.txt

RainCTfta: Hey, are you around?00:12
ftaRainCT, I am00:13
RainCTfta: I'm looking at Prism. Where do those .webapp things come from?00:15
ftathe ones in the package? from upstream, but they are completely obsolete and should be removed00:16
ftamy debian/rules is dirty because of all the workarounds to fix their syntax00:17
RainCTYes, I'm wondering if we need such .webapp things or just a .desktop file is enough00:17
ftawebapps contain several files. icons are a problem for us but json files are good to have00:20
RainCTfta: ok, so if we have a .desktop and optionally a .js and we  fetch an icon from the web, we can zip that together and get a .webapp?00:22
RainCT(and can those .webapp things be installed in ~/ too?)00:22
ftathey are in ~/.webapps00:23
ftafor each webapp, we need webapp.ini (created by prism from the UI), optionally localstore.json (to predefine the window size, etc), the icon (we fetch from the web), all those in a .webapp file or unpacked in ~/.webapp, plus a desktop file (that we must create)00:25
RainCTfta: can you send me a sample ~/.webapp dir with some webapps?00:28
ftaforgot, and a .css (optional)00:29
ftahttp://paste.ubuntu.com/195295/00:30
ftathe icon must be both inside the webapp dir (or file) and in the gnome dir for the desktop file00:31
RainCTor using an absolute path in the .desktop file00:32
RainCTfta: would you mind dropping me those files in a tarball so that I can see what they actually look like?00:34
ftahttp://www.sofaraway.org/ubuntu/tmp/google-maps.tar.gz00:36
ftahere is it00:36
RainCTthanks00:36
ftait was part of my initial webapps meta package00:37
ftai had some fun doing the css, it's no longer up to date though :(00:38
RainCTfta: ok so in the repository we would have, for each web application: 1. <webapp>.desktop containing a [Desktop Entry] section with the actual content for the .desktop, a [Parameters] section for the webapp.ini and an [Icon] section with info from where to get the icon; 2. Optionally, <webapp>.json, 3. Optionally, <webapp>.css.    How does this sound?00:41
ftasounds ok00:42
asacfta: how are you trying to build it?00:43
asaci dont think we should use moz build system for that00:43
ftaasac, sure, i'm using the original stuff00:44
asacwhat is the original stuff? makefile.gen00:44
ftaa raw configure and a template called makefile.cfg00:47
ftano .in/.ac00:47
asacfta: isnt makefile.gen the right one?00:47
ftanope00:48
fta# Generated automatically from makefile.cfg by configure.00:48
ftamoz changed a lot of int to int16, it's probably what is making the tests fail00:49
ftafor me, a int is 32 most of the time, not 1600:49
ftaasac_, gone?01:10
ftaasac, asac_: https://code.edge.launchpad.net/~fta/+junk/jpeg-6b+moz.upstream01:17
=== asac_ is now known as asac
ftaasac, and https://code.edge.launchpad.net/~fta/+junk/libjpeg6b-moz.dev02:17
ftaavailable here: https://edge.launchpad.net/~fta/+archive/staging02:17
ftai need to check if the mmx/sse2 things work.. not sure with those nspr ifdefs02:18
rippsfta: how do you intiate a manual update for ppabot?04:34
rippsupdate-pkg -n isn't working04:56
=== micahg1 is now known as micahg
rippsfta: I have an idea for adding changelogs to ppabot, I think it would work if it was added to the ppabot-pkgs-$pocket.conf the same way local_init works. Have the user supply a command to gather changelog info, and then let ppabot inject into the changelogs06:36
micahgping asac06:41
micahgnevermind06:43
micahgnot a regression06:44
rippsfta: if you or anybody else here knows how to increment the version using ppabot, I'd very much like to know07:35
rippsfta: ping09:33
ftaripps, pong11:28
ftaripps, what's wrong with -n?11:30
rippsfta: yo, I'm still having trouble get another set of packages to work with the ppa-scripts. I've done practically everything, including completely deleting all the branches locally and remotely and starting fresh, but I still can't get it to upload.11:30
BUGabundohi11:30
ftaripps, show me11:30
rippsfta: Also, I can't get packages to increment to ~ppa2.11:30
BUGabundofta: nice discussion going on on identica about #chrome vs #chomium licence11:30
rippsThe method you told me about the other day doesn't work anymore.11:31
ftahm, it should. please pastebin your logs11:31
rippsfta: it's sync, update-pkg -n, then sync -p. correct?11:32
ripps*-P11:32
rippsOkay, let me start fresh with the mpd packages and I'll post the logs, give me a sec11:33
ftanope, no need for -n in update-pkgs, it's just to create a new entry in changelog if the previous one was closed (= released instead of UNRELEASED)11:34
ftain rev 105, i've added a -n | --no-tarball-update to daily directly11:35
ftahttps://code.edge.launchpad.net/~fta/+junk/ppa-scripts11:36
rippsfta: what I really need is a simple force dput option. because while I was uploading some new gmpc packages, I experienced a timeout error while dputting gmpc-plugins, and now it won't let me upload them again. I only got the karmic distro through11:36
ftayou should have all the files in the PPA dir, just go there and dput manually11:38
rippsOkay, I've just re-deleted all the mpd branches locally and on launchpad, and I've deleted any existing mpd tarballs. I'm rerunning ./daily.sh mpd and pastebining it.11:38
ftaby default, it's in ../ppa11:38
rippsfta: except if I run ./daily.sh again, it deletes ppa/, I didn't know It hadn't uploaded until I was already trying to work another set of packages.11:39
ftaoh ok. deleting ppa/ should not be a problem as the last tarball is still in tarballs/ and the rest is in the branches11:41
rippsfta: http://pastebin.com/f4f74aaa911:43
ftasomething is wrong in mpdscribble's get-orig-source11:45
fta#11:46
fta# Done (created mpdscribble_0.17+git20090608.b5ed2a0.orig.tar.gz)11:46
fta#11:46
fta-rw-r--r-- 1 ripps ripps 56530 2009-06-14 05:42 mpdscribble_0.17+git20090608.b5ed2a0.orig.tar.gz11:46
fta#11:46
ftaTAG = -r11:46
fta#11:46
ftamake: TAG: Command not found11:46
fta#11:46
ftamake: *** [get-orig-source] Error 12711:46
rippsfta: okay, I fixed that just now, and I ran daily.sh mpd again, and I uploaded something, but only mpdscribble this time. It didn't even try the mpd and mpc packages that go with it.11:47
fta?11:48
rippsthat's my response11:48
ftabtw, you should never have to delete anything, especially branches, it's causing even more problems.11:49
rippswell, i figure deleting everything locally and remotely allows me to start fresh, without worrying about something I did that screws the whole system up.11:49
rippsThe biggest problem with this system (as far as I can tell) is that it's not very happy about any kind of errors. Stupid errors I admit, but they're just making me confused and upset.11:51
ftayep, i agree. there's room for improvement there but most of the time, you really need to fix the errors before doing the rest11:52
rippsIt'd be nice if it would tell me the errors. Sometimes it leaves me scratching my head.11:53
rippsSo, is daily.sh -n supposed to force a dput update?11:53
ftaonly if there's something new (tarball or packaging changes)11:54
ftawell, no11:54
ftadaily -n is just for packaging changes11:54
ftait will not run update-pkgs11:54
ftait's the same as running sync with all the parameters11:55
rippsfta: okay, walk me through, step by step how to force a dput update when there's no changes to the packaging or the tarballs11:56
ftahm, the bot is not supposed to do anything in that case.11:59
ftawhy would you want to do that btw?12:00
rippsfta: because, it only uploaded mpdscribble.12:00
rippsI now it won't upload anything12:00
rippsppa/ only has the changes files for mpdscribble too12:00
fta? it stopped because of an error in mpdscribble, if you fixed that, restart the full thing and it will continue with the other packages12:01
rippsfta: I DID!12:01
ftaif it doesn't, there's something else => pastebin12:01
ftai thought my logs were already too verbose, apparently not enough12:02
rippsfta: http://pastebin.com/f7177eed312:04
rippsI updated mpdscribble and it uploaded, but mpd and mpd did nothing12:04
ripps*mpc12:04
ftaweird. what was the command you used?12:09
rippsfta: ./daily mpd 2>&1 |pastebinit12:10
ripps^./daily.sh12:10
rippsIt seems to me, it only dputs when it's been flagged as changed. Should I make some kind of appetrary change to mpd and mpc's packaging and see if it does something?12:12
rippsfta: Okay, I've got an even stranger one. It didn't upload the orig.tar.gz. The mpdscribble that uploaded failed to build because it only included the debian/ directory12:16
ftayep, isnative=1, that's wrong12:17
ftai see12:18
fta#12:18
fta  - compare 0.16 (UNRELEASED) and 0.17~git20090525.ba2cda7-0ubuntu112:18
fta#12:18
fta     => isclosed=0 isnative=112:18
fta#12:18
fta0.16 is clearly recognized as a native version12:18
ftait should have been 0.16-x or 0.16-xubuntuy12:19
ftai guess that if you fix your .head branches to start with a non native version, it should work12:20
rippsfta: okay, I'll give it a shot12:20
rippsso, replace 0.16 with 0.16-a?12:21
ftayep12:21
ftanative packages are special, they require no merge12:21
ftaripps, use 0.16-112:22
rippsfta: woohoo, they're uploading!12:38
ftaasac, my libjpeg doesn't work at all. it makes apps crash, same error as with make test12:38
ftaripps, excellent.12:38
rippsNow, how do I correct the botched gmpc-plugins upload?12:38
ftaripps, what's wrong with it?12:39
rippsEverything worked correctly, but dput experienced a network timeout and canceled the rest of the uploads. I just need to reupload them.12:39
rippsThe change files in ppa/ are long gone now12:40
ftain the logs, you see the dput command, just copy paste it12:40
ftait's at the end12:40
rippsnothing to repload, there's no more changes files anymore12:40
ftahm, the easiest way is to commit something to the packaging branch and resync12:42
rippshmm... perhaps, in the future, I'll make an override file that I just write random crap to when I need to make an a fixed upload. Of course, theirs usually small aesthetic changes I can make to the control file or something to do that.12:48
ftaripps, once you get used to working with the bot, you will probably no longer need to do that at all.12:55
rippsfta: hmm... something's wrong, it's not building the orig.tar.gz (and the one tarballs/ is old)...12:56
BUGabundono sound on flash! bummer13:00
rippsfta: http://pastebin.com/f4a7552e813:01
ftawhich one?13:02
rippswhich what? it seems it's not putting the orig.tar.gz properly... I'm not sure what it's doing...13:05
ftaripps, is gmpc-plugins_0.18.95~bzr11.orig.tar.gz still in tarballs/ ?13:10
rippsfta: no13:10
ftadid you drop it at some point?13:10
rippsfta: didn't touch it13:10
rippsthere's *bzr8* in there though?13:10
ftaer..13:11
rippsbizarre.. because earlier today I uploaded a bzr11 to karmic...13:11
rippsand the day before that I did bzr10 for all the the distros succesfully...13:11
ftaok, got it, it's a bug in daily.sh13:13
ftalast line :P13:14
ftai lazily rely on sort to do the clean-up13:15
ftabut bzr11 < bzr8 according to sort13:15
rippsBut that shouldn't take place until after the dputting is done. The bzr11 that's made during this run should still exist long enough to upload13:16
ftayep but it's ripps3 so you probably run daily at least twice on that tarball13:17
ftafor now, just recreate it using debian/rules get-orig-source, move it to tarballs/ and drop the old one (bzr8). I'll fix it on my side13:18
fta(add -L ... and whatever flag you usually need)13:22
ftaripps, fixed in the branch. http://paste.ubuntu.com/195639/13:38
alexbodnhello friends13:39
alexbodnhi fta, acas13:39
alexbodnhi asac13:44
rippsfta: there's a flaw somewhere, I tried to initiate an upload of mpdscribble but it won't upload the orig.tar.bz, even though the changes depend on it. It inevitably leads to launchpad emailing me about a missing orig.tar.gz13:45
ftaripps, it will only upload the orig if it's the 1st iteration, i.e. ~ppa1, not ~ppa2/3/... and the tarball has to be there when you push ppa113:46
rippsfta: unfortunatly, it wasn't, and since all of the changes in ppa/ are setup not to include orig.tar. I'm stuck until upstream changes the source.13:47
ftaripps, you can work around it. either manually, or by cheating in sync-ppa13:50
rippsI think I can just manuall run bzr bd in a the karmic directory so it will create a changes with orig.tar13:50
ftayou can edit sync-ppa, and in bzr_cmds, change ppa2 to use -sa instead of -sd, then change it back13:51
ftabut i don't not really see this as a bug in the bot, or a limitation. it's a workaround for the bug i just fixed (and probably some destructive trial/errors on your side too)13:54
ftahopefully, you'll have everything in place soon13:54
ftai need to figure out what i should do with those logs.. create a log file somewhere, send a real email, or something13:56
ftaideas welcome13:56
ftaasac, ^^, you too :)13:56
rippsfta: did you read my idea for including changelog support?13:56
ftaripps, yes, i have to think about it, but a generic hook would sure be nice to have13:57
ftai need to think about which arguments it should receive13:57
ftaprobably the previous version number13:57
ftaso you can extract the previous rev number13:57
ftaor the prev date, depending on the vcs, and versioning scheme you used13:58
rippsI have to leave for a while, bbl14:05
asacfta: can you put you jpeg work somewhere?15:36
ftahttps://edge.launchpad.net/~fta/+archive/staging15:37
ftahttps://code.edge.launchpad.net/~fta/+junk/libjpeg6b-moz.dev15:37
asacoh found it15:37
ftahttps://code.edge.launchpad.net/~fta/+junk/jpeg-6b+moz.upstream15:37
asacthx15:37
ftabut it's obviously not ready15:37
ftaasac, so, what do you think?15:52
ftaBUGabundo, how?16:08
BUGabundoasking me?16:08
BUGabundoI know nothing about repositories or bzr16:08
ftalol16:08
BUGabundoa bit OT but if anyone needs it: http://ppa-search.appspot.com/16:17
ftalp already has something to search inside ppas16:25
BUGabundodid not know that16:26
BUGabundothis one looks nice16:26
BUGabundoit as a cload16:26
ftahttps://edge.launchpad.net/ubuntu/+ppas16:28
BUGabundothanks16:29
BUGabundospamming^W µblogging it now16:29
=== ejat is now known as e-jat
ftaripps, you're ppa is green. does this mean you're all set now?17:56
rippsfta: yep, i hope so. Copied the good packgages to their proper ppa's in gmpc-trunk team. I'm going to run daily for a few days to see if everything works the way it should, if it works for a while without outside assistance, I'll move move it from my staging ppa to the proper one's on gmpc-trunk17:57
ftasounds great17:58
BUGabundo!ping17:58
ubottuping yourself ;-) really the diodes all down my left side are sore17:58
=== BUGabundo1 is now known as BUGabundo
micahgis there a special place to put firefox extension packaging requests?18:55
ftathe wiki19:07
micahgnevermind..found the project19:07
micahgthere's a FF extensions project19:08
micahgI added it there19:08
ftai'm not sure anyone would notice19:14
micahgpeople use it19:16
micahglast post was 6/119:16
ftai mean, i'm not sure people packaging addons will notice, but maybe i'm wrong19:18
micahgok, well next time I talk to some of the other packagers, I'll check with them if they use it19:19
ftagood :)19:20
micahgjust trying to clean up the old firefox package in Ubuntu19:21
micahgWhere do language requests go?19:22
micahgupstream?19:23
ftanot sure19:23
micahgBUGabundo any ideas?19:23
BUGabundo1don't ask me !19:23
ftaBUGabundo is clueless, but he dents a lot ;)19:23
BUGabundo1that's me19:24
BUGabundo1hey I get offended by that19:24
BUGabundo1I'm not clueless19:24
BUGabundo1I just know nothing about packaging or mozilla19:24
BUGabundo1or code19:24
BUGabundo1or the lottery number19:24
=== BUGabundo1 is now known as BUGabundo
lupine_86asac: ping ?21:17
BUGabundo!ask | lupine_8621:18
ubottulupine_86: Please don't ask to ask a question, simply ask the question (all on ONE line, so others can read and follow it easily). If anyone knows the answer they will most likely reply. :-)21:18
lupine_86asac: BUGabundo is shepherding me to bug you about, um, a bug ;) - I can reproduce 384493 (wpa-psk with NM and the kde4,3 plasmoid not working)21:20
BUGabundohumm KDE?21:21
BUGabundoyou never mention KDE before21:21
asaclupine_86: plasmoid issues are usually because the plasm nm widget has bugs21:24
asacso asign it there21:24
lupine_86mm, there seems to be an open bug or ten already21:26
BUGabundoif they are the same, I guess  you could dupe them21:27
asacq21:27
asaclupine_86: if there is a triaged bug already use that as the dupe target21:27
ftaasac, so, jpeg?21:34

Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!