[00:00] <rickspencer3> https://wiki.ubuntu.com/DesktopTeam/Meeting/2010-08-24
[00:00] <rickspencer3> I didn't have time to pull out the irc logs :(
[00:00] <rickspencer3> or update the wiki
[00:00] <rickspencer3> hold on
[00:02] <robert_ancell> rickspencer3, i'm doing that now
=== bjf is now known as bjf[afk]
[00:02] <rickspencer3> robert_ancell, too late
[00:04] <robert_ancell> rickspencer3, just beaten!  BTW, put a space before the <<BR>> because it screws up links that people post
[00:04] <rickspencer3> *sigh*
[00:04] <rickspencer3> robert_ancell, feel free to fix that up ;)
[00:05] <rickspencer3> nm, doing it
[00:06] <rickspencer3> oh dang it
[00:06] <RAOF> Right.  I caught much of that log care of getting highlighted about the apport item that looks like it'll slip.
[00:07] <rickspencer3> wtf
[00:07] <rickspencer3> robert_ancell, looks like we both edited it, sorry, my bad
[00:07] <rickspencer3> RAOF, what item?
[00:08] <robert_ancell> rickspencer3, heh, ok, you fix it.  otherwise we'll delete it twice ;)
[00:08] <RAOF> pitti's “pick up gpu hangs, turn on drm.debug, and ask users to submit next time” item.
[00:08] <rickspencer3> so, no apport for gpus?
[00:08] <RAOF> No _improvement_ on apport for gpus.
[00:08] <rickspencer3> RAOF, well, that's rally too bad
[00:08] <rickspencer3> RAOF, is the existing system at all useful?
[00:08] <RAOF> Yes.
[00:08] <rickspencer3> heading into Beta, they'll be lots of good bug reports
[00:08] <RAOF> The current system is pretty kick arse.
[00:08] <RAOF> Well, for intel.
[00:09] <RAOF> That puts me in mind of another goal :)
[00:09] <rickspencer3> ok
[00:09] <rickspencer3> RAOF, so you don't seem to dissapointed
[00:10] <RAOF> There's no support for -ati or -nouveau, but they need kernel support before the gpu hang stuff can work for them anyway.
[00:10] <rickspencer3> ack
[00:10] <RAOF> I'm not terribly disappointed, no.
[00:10] <rickspencer3> so move any related work items out of maverick then
[00:11] <RAOF> Will do.
[00:11] <rickspencer3> so, in terms of U1 ribbon
[00:11] <rickspencer3> the U1 team addressed the bug report
[00:11] <rickspencer3> we should see tomorrow if the desktop team still thinks it's a problem
[00:12] <rickspencer3> this was the bug if anyone is interested
[00:12] <rickspencer3> wendar, still around at all?
[00:13] <rickspencer3> we started the meeting by introducing wendar :)
[00:14] <RAOF> Our new Technical Architect?
[00:14] <rickspencer3> RAOF, robert_ancell everything else is status updatey stuff
[00:14] <rickspencer3> RAOF, correct
[00:14] <rickspencer3> except 2 things
[00:15] <rickspencer3> 1. everyone needs to get their review goals in!
[00:16]  * rickspencer3 whip cracking noises
[00:16] <robert_ancell> don
[00:16] <robert_ancell> e
[00:16] <rickspencer3> yah
[00:16] <rickspencer3> same with Luke
[00:16] <rickspencer3> TheMuso, that is
[00:16] <rickspencer3> RAOF, so that leaves you
[00:16] <RAOF> Yup.
[00:16] <rickspencer3> (and pretty much everybody else on the desktop team)
[00:16] <chrisccoulson> ;)
[00:17] <kenvandine> hey folks
[00:17]  * kenvandine waves
[00:17] <rickspencer3> chrisccoulson, kenvandine ... go put in your goals!!!!
[00:17] <RAOF> Howdie!
[00:17] <kenvandine> rickspencer3, i will... first thing in the morning :)
[00:17] <kenvandine> or maybe later tonight
[00:17] <chrisccoulson> it's morning here already ;)
[00:17] <kenvandine> hehe
[00:17]  * kenvandine needs to get these kids to bed, it's been a long day
[00:18] <rickspencer3> tomorrow is fine, just please!
[00:18] <rickspencer3> good night kenvandine
[00:18] <kenvandine> i will :)
[00:18] <rickspencer3> chrisccoulson you too!
[00:18] <chrisccoulson> rickspencer3, yeah, i'll look at that later too ;)
[00:19] <rickspencer3> chrisccoulson sounds good
[00:19] <rickspencer3> tomorrow morning is fine :)
[00:19] <kenvandine> everytime i tell myself i am going to look at those goals, someone pings me and gets me going on some bug
[00:19] <kenvandine> :)
[00:19] <rickspencer3> I am going to get so busted
[00:19] <rickspencer3> I have like 20 direct reports right now
[00:19] <kenvandine> sorry
[00:19] <kenvandine> :)
[00:19] <rickspencer3> and only 4 people have put in goals so far
[00:19] <kenvandine> yeah... sucks to be you :)
[00:19] <kenvandine> yikes
[00:20] <kenvandine> your going to have a busy thursday night :)
[00:20]  * rickspencer3 wipes sweat from brow
[00:20] <rickspencer3> ok, robert_ancell, RAOF anything to add for eastern edition?
[00:20] <RAOF> Soluble bugs are fun!
[00:20] <rickspencer3> ?
[00:21] <robert_ancell> RAOF, soluble in alcohol?
[00:21] <kenvandine> :)
[00:21] <rickspencer3> is that like putting ants in an acid bath or something?
[00:21] <RAOF> Sorry; badly timed.  That was in response to kenvandine :)
[00:21] <kenvandine> hehe
[00:21] <rickspencer3> heh
[00:21]  * RAOF prefers electroplating ants to disolving them in acid
[00:21] <rickspencer3> robert_ancell, mdz had an idea for cheese
[00:21]  * bcurtiswx_ is SRU empathy
[00:22]  * bryceh waves
[00:22] <robert_ancell> nothing to add for me.  Maverick seems pretty stable from here.  The Cheese performance is awful but not necessarily worse than before?
[00:22] <rickspencer3> hiya bryceh!
[00:22] <rickspencer3> long time no see
[00:22] <robert_ancell> rickspencer3, yeah?
[00:22] <rickspencer3> robert_ancell, yeah so his idea was ...
[00:22] <rickspencer3> look for a fast encoder, even if it's in bad
[00:22] <rickspencer3> if it's there, just use it
[00:22] <rickspencer3> then fall back to theora and low compression, low resolution if not
[00:22] <robert_ancell> rickspencer3, I tried mpeg but it seemed even worse
[00:23] <rickspencer3> dang it!
[00:23] <rickspencer3> so basically, we can't encode on the fly?
[00:23] <RAOF> Didn't you find that performance increased with a different muxer?
[00:24] <rickspencer3> RAOF, if you are addressing me, then "sort of"
[00:24] <rickspencer3> I can use an avimuxer with no encoding
[00:24] <rickspencer3> so I can quickly create crystal clear videos, but they are HUGE
[00:24] <robert_ancell> It appears that the encoders are not fast enough.  There is an element in the pipeline which drops frames that are too late (?) and removing that seems to make the video have more frames.  But I'm not sure if that's safe to remove
[00:24] <RAOF> Ah.  That's not quite what I was thinking :)
[00:24] <RAOF> The other option would be to have a huge queue in there to buffer, but that doesn't work for really long videos.
[00:25] <robert_ancell> RAOF, really, does the muxer add much delay?  I thought it was more or less just interleaving the video and audio and adding a header?
[00:25] <rickspencer3> robert_ancell, how does record my desktop work?
[00:25] <RAOF> robert_ancell: Depends on the container format; some want to have a pretty big buffer because they need to know a bunch of future data.
[00:26] <rickspencer3> because if I run that while I have a web cam at 30 fps, record my desktop makes videos that work, and the desktop works
[00:26] <robert_ancell> rickspencer3, don't know. I'll have a look at it
[00:27] <rickspencer3> no easy answers, huh?
[00:27] <robert_ancell> looks like it uses libtheora directly
[00:27]  * rickspencer3 wants easy answers
[00:27] <robert_ancell> I'd like that too :)
[00:27] <rickspencer3> so just completely rewrite cheese so it's not gstreamer based?
[00:27] <rickspencer3> anyway, that is good data
[00:28] <rickspencer3> it implies that the bug is in gstreamer, right?
[00:28]  * RAOF wouldn't think that gstreamer would have _too_ large a performance impact.
[00:28] <RAOF> Yeah.
[00:28] <robert_ancell> RAOF, but the buffer should not make any difference right?  Essential you are taking a stream of frames from the video and encoding them - you should only drop frames if you get behind in CPU
[00:29] <RAOF> Right, but you detect getting behind the CPU by watching how many frames you've got queued up in some buffer somewhere.
[00:30] <RAOF> Have you tried futzing with the various speed/quality knobs on theoraenc?
[00:31] <RAOF> Maybe record my desktop sets some really conservative values for the tunables?
[00:31] <robert_ancell> RAOF, yeah, dropped the quality to zero, it had an effect on the output quality but not significantly so.  Reducing resolution helps but you can go low enough to get smooth video
[00:31] <robert_ancell> can't go low enough
[00:32] <RAOF> There's also the “speed” tunable, which is searching for motion vectors.
[00:32] <RAOF> I think that's likely to be processor-intensive; does setting it to “0” speed things up?
[00:33] <RAOF> (Or maybe 2 - the property description seems to conflict with the property name)
[00:33] <robert_ancell> RAOF, speed-level?
[00:33] <RAOF> YEah.
[00:33] <RAOF> That's what I meant.
[00:40] <rickspencer3> okay, anyone else here waiting to see what robert_ancell says?
[00:40]  * rickspencer3 munches popcorn
[00:40] <bryceh> mmm
[00:42] <robert_ancell> rickspencer3, :P
[00:44]  * robert_ancell compiling...
[00:45] <robert_ancell> no real change.   Trying with quality=0 as well
[00:46] <robert_ancell> RAOF, I assume speed-level=0 is the fastest?
[00:46] <RAOF> robert_ancell: That's how I read the description, but you'd think speed-level=2 would be faster :)
[00:48] <robert_ancell> hmm, with speed-level=0 and quality=0 it looks crap but stabilises after 13s and runs perfectly smooth
[00:57] <rickspencer3> robert_ancell, what settings does record-my-desktop use?
[00:57] <rickspencer3> or does libtheora just work totally differently?
[00:58] <robert_ancell> rickspencer3, gstreamer is using libtheora so it should be similar
[01:00] <robert_ancell> I can read the settings but I don't know what most of them do
[01:01] <robert_ancell> Hmm, if I take videoscale out of the pipeline it goes smooth in 9s.  I don't know why there is a videoscale element as I don't think it needs to do any scaling
[01:02] <rickspencer3> I've got to go
[01:02] <rickspencer3> g'night all
[01:02] <robert_ancell> rickspencer3, bye
[02:20] <RAOF> robert_ancell: If you're touching gnome-session in the near future could you pick up https://bugs.edge.launchpad.net/gnome-session/+bug/623700 ?  If not, I could prepare a new package for sponsoring.
[02:20] <ubot2> Launchpad bug 623700 in gnome-session (Ubuntu) (and 1 other project) "Edge condition in idle monitor (affects: 1) (heat: 6)" [Undecided,New]
[02:21] <robert_ancell> RAOF, there's no plan to touch it anytime soon - you can upload anyway as it's in the desktop set
[02:22] <RAOF> I don't think I can.
[02:22] <RAOF> I don't _think_ I'm in ~ubuntu-desktop
[02:26] <RAOF> I'll prepare an update anyway, then.
[02:26] <robert_ancell> RAOF, oh, you should be!
[02:26] <RAOF> I guess so :)
[02:27] <RAOF> Almost all of what I want to touch isn't in that set, though ;)
[02:27] <robert_ancell> I think we can trust you...
[02:33] <RAOF> I'd need to front the TB to become a member of ubuntu-desktop, right?  I may as well go the whole hog for core-dev if I'm doing that.
[02:34] <robert_ancell> RAOF, I don't think so, you just need to request membership on LP and the team owner (seb128) just needs to decide if you've made sufficient contributions to the desktop team
[02:35] <RAOF> Oh, well, then.  I may just do that.
[02:54] <bcurtiswx_> SRU sponsor bug #623657
[02:54] <ubot2> Launchpad bug 623657 in empathy (Ubuntu) "New Upstream Release 2.30.3 (affects: 1) (heat: 8)" [Wishlist,New] https://launchpad.net/bugs/623657
[03:58] <bcurtiswx_> g'nite all
[04:12] <RAOF> Anyone feel like sponsoring gnome-session?  http://cooperteam.net/Packages/gnome-session_2.31.6-0ubuntu2_source.changes and the related merge request https://code.edge.launchpad.net/~raof/gnome-session/fix-bug-623700/+merge/33605
=== almaisan-away is now known as al-maisan
[07:29] <pitti> Good morning
[07:33] <intelliant> Hello!
[07:33] <intelliant> Is this a good place to discuss ldap client auth issues/
[07:33] <intelliant> ?
[07:35] <intelliant> I am authenticating an ubuntu LTS 10.04 desktop client against a centos based openLDAP server following this guide - https://help.ubuntu.com/community/LDAPClientAuthentication#Notes%20for%207.10%20and%20later
[07:38] <and471> intelliant, better to try #ubuntu
[07:38] <and471> intelliant, this channel is more for development, whereas #ubuntu is for support :)
[07:38] <intelliant> and471: thanks
[07:38] <and471> intelliant, no problem :)
[07:42] <and471> vish, rock on! :)
[07:54] <and471> mvo, hey there, how do you get the display name of a ppa in SC, and should I use the same method in software-properties-gtk?
[08:04] <mvo> good morning and471
[08:06] <mvo> and471: we use the "label" property of the pkg.origins part
[08:06] <mvo> and471: i.e. python -c 'import apt; cache=apt.Cache(); print cache["apt"].candidate.origins[0].label'
[08:07] <and471> mvo, ok I'll have a go :)
[08:12] <mvo> :)
[08:16] <and471> mvo, can I ask why the source-code check button in software-properties is weird?
[08:18] <vish> and471: :)
[08:20] <vish> and471: did the gdebi 'remember checkbox' fix get released?
[08:21] <mvo> and471: its probably in "half-pressed" state because some of your source does not have a deb-src line
[08:22] <vish> and471: hmm, maybe not looks like the update was before the lp change.
[08:28] <and471> mvo, ah ok thanks :)
[08:29] <and471> vish, no, it is in the gdebi branch but mvo just needs to release it, but I don't know whether he will becuase gdebi is being replaced by software-center
[08:29] <vish> k..
[08:35] <mvo> and471: I will upload it still, gdebi is not obsolete as it has features like inspecting the content of debs that are interessting to a lot of users
[08:35] <and471> vish, ^
[08:36] <vish> and471: thanks non-contributing dev! ;p
[08:36] <and471> hehe
[08:45] <glatzor> morning mvo
[08:47] <mvo> hey glatzor, good monring
[08:50] <glatzor> mvo: i just realized that we don't have got a sperate privilege to install unauthenticated packages in aptdaemon
[09:01] <mvo> glatzor: oh, right. I think that should be there
[09:04] <glatzor> mvo, it would just be a "yes" "no" privilege -  I would grant it to every active user by default
[09:05] <mvo> glatzor: hm, shouldn't it be "no" by default so that there is a extra confirmation step for unauthorized packages?
[09:06] <glatzor> mvo, but policykit isn't about confirmation - it is about authenticating the user
[09:06] <glatzor> but I want to add an unauthenticated property to the transaction
[09:07] <mvo> glatzor: aha, I see what you mean. ok, so its up to the softwrae to decide what to do (that makes perfect sense)
[09:07] <glatzor> it should contain data after the simulate call
[09:07] <RAOF> Is there any particular reason to have a separate polkit authorisation for unauthenticated packages unless you'd like to have different permissions on unauth packages?
[09:10] <glatzor> mvo, may I point you to  #587004
[09:11] <glatzor> lp:#587004
[09:11] <mvo> bug #587004
[09:11] <ubot2> Launchpad bug 587004 in update-notifier (Ubuntu) (and 1 other project) "Update-notifier triggers an aptdaemon launch on start up (affects: 1) (heat: 6)" [Undecided,Confirmed] https://launchpad.net/bugs/587004
[09:11]  * mvo looks
[09:12] <glatzor> RAOF, I could imagine  an organizsation that would like to allow users to install additional software but only authenticated one.
[09:12] <glatzor> RAOF, so you could grant install-or-remove by auth_admin and set install-unauthenticated to no
[09:13] <desrt> RAOF: we discussed in some BoF at some point that installing additional software from authenticated repositories is almost a non-op
[09:13] <RAOF> glatzor: Ah, right.  I wasn't thinking of the more-restrictive direction.  That makes sense.
[09:15] <glatzor> mvo, why doesn't software-center use the simulate method to get the to be removed packages?
[09:16] <glatzor> mvo, SoftwareCenterApp.on_application_request_action just checks the installed rdepends
[09:17] <mvo> glatzor: that is a missing feature :/ I just didn't get around to it, but I can do that today
[09:17] <mvo> glatzor: it definitely should use it
[09:19] <glatzor> mvo, how do you want to handle the installation? shouldn't there be a confirmation if the installation of an app would remove a conflicting one
[09:20] <and471> mvo, am I correct in assuming softwareproperties/gtk/DialogAptKey.py is obselete?
[09:20] <mvo> glatzor: yes, I think that should be there
[09:21] <mvo> and471: yes
[09:22] <and471> mvo, I shall delete it then
[09:26] <mvo> ok
[09:27] <and471> mvo, why does gtk/CdromProgress use update-managers' translation domain? and can I remove this?
[09:29] <pitti> seb128: I put back french langpacks, FYI :)
[09:29] <pitti> OO.o fix landed, and it seems that my recent debhelper "drop upstream changelogs" fix now causes CDs to get smaller and smaller over time \o/
[09:30] <mvo> and471: hm, hm, I'm pretty sure that this is a bug
[09:31] <and471> mvo, okay that is fine, I shall fix it, just wanted to make sure it wasn't something intentionally special :)
[09:37] <and471> be back after a restart
[09:37] <glatzor> mvo,  to set the priority of aptdaemon bugs in ubuntu I have get ubuntu-core developer?
[09:38] <mvo> glatzor: bug control irrc
[09:39] <mvo> glatzor: bdmurray should be able to grant that priv
[09:39] <mvo> glatzor: I asked him on #ubuntu-devel
[09:40] <glatzor> thanks
[09:40] <mvo> glatzor: he is in the us, so it may take a bit
[09:42] <asac> who to blame for pushing ooo to proposed in the pre-beta-freeze days? this could have waited three days when all builders are idle ;)
[09:42] <asac> ah doko
[09:42] <asac> kk
[09:42] <mvo> glatzor: https://edge.launchpad.net/~ubuntu-bugcontrol <- that is the team
[09:44] <glatzor> mvo, I would like to add the app name to metadata of transactions in software-center. this way we could provide better error messages
[09:44] <glatzor> mvo, e.g. "Failed to install APPNAME"
[09:45] <mvo> glatzor: I do this already as "sc_appname" iirc, but just to the "free" metadata, if you make it a "official" propety, even better
[09:46] <mvo> glatzor: what is the best place to check for the simulate method integraton? in the gtk examples?
[09:46] <glatzor> mvo, I just haven't seen this ok
[09:48] <glatzor> mvo, it is just a call to client.Transaction.simulate(). afterwards you can check the client.Transaction.dependencies property
[09:49] <glatzor> e.g. client.Transaction.dependencies[enums.PKGS_REMOVE] gives you the to be removed packages
[09:51] <glatzor> mvo, since the dependency calculation takes some time (nearly a second on my old computer) you should perhaps use inline defers
[09:51]  * mpt cheers glatzor and mvo
[09:52]  * glatzor hugs mpt
[09:52] <glatzor> servus!
[09:53] <and471> mvo, mpt, in software-properties you have the 'add cd-rom' button, but it can be a cd-rom or dvd, would 'Add disk' be a good alternative?
[09:54] <glatzor> mpt, I know that I am quite late in the current development cycle, but I would like to address the error messages in aptdaemon againg
[09:54] <glatzor> mpt, I have created a new wiki page http://wiki.debian.org/aptdaemon/error-messages
[09:55] <mvo> glatzor: thanks
[09:55] <mvo> hey mpt :)
[09:56] <mpt> glatzor, I was just about to say "didn't you already have a wiki page on that?", but I see this is the same one updated :-)
[09:56] <glatzor> mpt, would this be a good way to discuss the messages? I could also refractor the code in aptdaemon/enums.py so that error enums, strings and descriptions are more closer organized and easier to edit
[10:00] <glatzor> mpt, we could also overwrite the messages in software-center  and make use of the application name which is stored in the transaction by software-center. So "Failed to resolve dependencies" could be "Failed to install APPNAME since required software isn't available"
[10:00] <mpt> glatzor, yes, that's why I was cheering :-)
[10:04] <glatzor> mpt, would you only like to overwrite the header or the text or both?
[10:05] <mpt> glatzor, does "Title" here mean the text that appears in the title bar, or the text that appears in bold inside in the window?
[10:05] <glatzor> mpt, the <big><b> in the dialog
[10:06] <glatzor> perhaps I should rename it to short and long description :)
[10:06] <glatzor> or summary and description
[10:11] <mpt> glatzor, how about we use "Primary text" and "Secondary text", like the HIG
[10:12] <mpt> http://library.gnome.org/devel/hig-book/stable/windows-alert.html.en#alert-text
[10:13] <and471> mpt, <ping> (sorry to be impatient but I have to go soon)
[10:14] <glatzor> mpt, do you want the primary text only to be of the kind "ACTION APPNAME failed"?
[10:14] <mpt> and471, oh, sorry, didn't see you there. I think mvo could confirm that it could even be a USB key, too.
[10:14] <and471> mpt, in that case 'Add volume'?
[10:15] <mpt> and471, yes, I think so
[10:15] <and471> k
[10:15] <and471> thanks mpt
[10:15] <mpt> glatzor, give me a minute, I'll see
[10:16] <and471> mpt, is it with a capital 'V', I never know when to capitalise in GUIs
[10:16] <and471> ?
[10:16] <mpt> and471, yes, buttons use Title Case
[10:16] <and471> ok
[10:17] <mpt> so do menus, tabs, window titles, and table headers
[10:17] <mpt> Most other things use Sentence case
[10:20]  * mpt is pleased to discover that Opera has fixed its menus to use Title Case
[10:23] <mpt> glatzor, so for all of these can I now assume that aptdaemon knows the name of the thing it was trying to install or remove?
[10:24] <huats> morning
[10:25] <glatzor> mpt, software-center stores the application name in the meta data of the transaction and we know if they are getting installed, removed or upgraded.
[10:26] <mpt> ok, thanks
[10:27] <mpt> glatzor, for ERROR_INCOMPLETE_INSTALL, is it the package you're trying to install that didn't install correctly, or is it some other previous package that messed up the database or something?
[10:28] <glatzor> mpt, right it's a previous installation that failed. the correct behavior would be to fire the "FixIncompleteInstallation" transaction.
[10:28] <glatzor> this is basically a call of "dpkg --configure -a"
[10:29] <mpt> glatzor, so normally, ERROR_INCOMPLETE_INSTALL shouldn't appear at all?
[10:29] <glatzor> mpt, you cannot perform a new action if an old one failed to complete
[10:29] <mpt> glatzor, why not? Because it might involve re-authenticating?
[10:30] <glatzor> mpt, that is an apt policy thing. apt wants a clean dpkg status.
[10:31] <glatzor> incomplete install means that a previous installation was cancelled in a very unfriendly way
[10:31] <mpt> glatzor, oh, sorry, I thought you meant aptdaemon in particular, rather than apt in general
[10:31] <mpt> ok
[10:31] <glatzor> mpt, think of a post install maintainer script that tried to download some files from the internet or restart a daemon
[10:32] <mpt> glatzor, so is it practical for aptdaemon to automatically run "dpkg --configure -a" instead of showing this error? What would be the benefits and disadvantages of that?
[10:32] <glatzor> mpt, or a computer crash
[10:33] <glatzor> mpt, aptdaemon only tries to recover a failed aptdaemon transaction by running dpkg --configure -a afterwards internally
[10:33] <glatzor> from the outside this looks like the transaction just takes a little bit longer
[10:33] <mpt> sure
[10:34] <mpt> But if the computer crashed, so aptdaemon never had the chance to do that (and if there wasn't a crash it wouldn't have needed to anyway), and then you restart and try to install or remove something...
[10:35] <chrisccoulson> RAOF - re bug 623700 - we had a similar issue in gnome-power-manager, and i actually found another issue entirely (not sure if it affects gnome-session too)
[10:35] <ubot2> Launchpad bug 623700 in gnome-session (Ubuntu) (and 1 other project) "Edge condition in idle monitor (affects: 1) (heat: 6)" [Undecided,New] https://launchpad.net/bugs/623700
[10:35] <glatzor> mpt, but it could also make sense to run "dpkg --configure -a" in the transaction when encountering a broken installation
[10:35] <chrisccoulson> what was happening in gnome-power-manager is that the idletime counter was being reset even before g-p-m had registered the negative alarm
[10:35] <chrisccoulson> (which can happen if you wiggle the mouse quickly)
[10:35] <chrisccoulson> so it was missing the alarm entirely
[10:36] <glatzor> so I could add a new status "Recovering from previously failed installation"
[10:36] <mpt> glatzor, right. So we wouldn't show that error message any more.
[10:36] <glatzor> mpt, the recovery can also fail :)
[10:36] <mpt> We might show an even worse one if the "dpkg --configure -a" fails, but much less often. :-)
[10:38] <mpt> glatzor, for ERROR_PACKAGE_NOT_INSTALLED and ERROR_PACKAGE_ALREADY_INSTALLED, what benefit is there to showing the error at all?
[10:39] <glatzor> mvo, what do you think about running "dpkg --configure -a" automatically inside a transaction if we encounter a dirty dpkg journal?
[10:39] <mvo> glatzor: sounds fine to me
[10:39] <mvo> glatzor: our users will be happy about it I'm sure
[10:39] <mvo> I mean, you don't want to run it automatically on server, but for us it should be fine
[10:40] <glatzor> mpt, aptdaemon doesn't want to make any decisions for the client application. if you don't care about this error, you can just ignore it in software-center
[10:40] <and471> mvo, https://code.launchpad.net/~and471/software-properties/gtkbuilder-and-gui-polish/+merge/33618
[10:41] <mvo> and471: nice!
[10:41] <and471> mvo, converts to gtkbuilder and there are a few other tweaks
[10:41] <and471> mvo, (such as using buttonboxes instead of hboxes)
[10:41] <and471> mvo, could you test before merging tho
[10:41] <mvo> and471: cool, I'm in the process of finishing a small update-manager UI change, once that is uploaded I have a look
[10:41] <and471> mvo, I haven't been able to test, for example the add sources list dialog
[10:42] <and471> mvo, sure, I am leaving now so you can take your time, just post any comments on the merge request and I'll look at them this evening
[10:42] <and471> see ya everyone
[10:42] <mpt> tchau and471
[10:42] <mpt> glatzor, ok, I'll report a bug for USC to do that
[10:43] <mvo> bye and471
[10:43] <glatzor> bye and471
[10:53] <seb128> pitti, http://people.canonical.com/~platform/desktop/versions.html
[10:53] <seb128> \o/
[10:55] <pitti> yippiie
[10:55] <pitti> I'd also like to move the WI tracker there
[10:55] <seb128> ;-)
[10:55] <seb128> pitti, go for it, you should have access as well
[10:55]  * pitti hugs seb128
[10:55]  * seb128 hugs pitti back
[10:58] <mpt> glatzor, for ERROR_PACKAGE_DOWNLOAD_FAILED and ERROR_REPO_DOWNLOAD_FAILED, does it make sense to have a "Retry" button?
[11:11] <mpt> mvo, is tremolux in charge of getting the add-ons branch in? I see he's made some tweaks
[11:12] <mvo> mpt: he was kind enough to review/test, its ulimately on me to do the real merge, still on my radar
[11:13] <mpt> glatzor, similarly, would it make sense for the ERROR_CACHE_BROKEN alert to have a "Repair" button that runs the "apt-get install -f" itself?
[11:32] <glatzor> mvo, is there a clean way to detect half-configured|installed packages?
[11:33] <glatzor> would be nice to call dpkg --configure -a too. otherwise the installation of an app fails because the installation of another package cannot be completed
[11:38] <mvo> glatzor: dpkg --configure -a should deal with that automatically
[11:39] <mvo> glatzor: I guess you ask how to figure out if a pkg is in that state ? I need to check, you can iterate over the pkg and check the state, but there might be a quicker way. I can check after lunch
[11:39] <vish> mpt: jupiter colors, ftw !
[11:40] <glatzor> mvo, I would like to separate the errors for the user
[11:40] <kklimonda> mvo: do you have any plans for what to do with a really broken dpkg database? something that completely borks dpkg - like bogus packages.files or errors in info file.
[11:41] <glatzor> kklimonda, install a backup
[11:41] <kklimonda> glatzor: right - because we know that all users do those :)
[11:41] <mvo> kklimonda: no good plan, its just hard to know what is right in a automatic way. the best is probably what glatzor said (unless you have specific examples where its possilbe to figure it out)
[11:41]  * mvo really needs to go for lunch now
[11:42] <glatzor> bye mvo!
[11:43] <kklimonda> mvo: in case of broken .files (or .md5sum) a forced package reinstall may probably do the trick (and dpkg can obviously detect that because it dies crying like a little girl ;))
[11:45] <seb128> chrisccoulson: do you know if anybody is tracking chromium an appmenu?
[11:46] <chrisccoulson> seb128 - not that i'm aware of
[11:46] <seb128> ok
[11:46] <seb128> do you know who was supposed to do that?
[11:46] <seb128> there is an item for it on the chromium spec
[11:46] <seb128> or rather a bug linked to the spec
[11:46] <chrisccoulson> i'm not sure if anyone was ever assigned to work on that
[11:46] <seb128> there is also one about stock ids for icons
[11:47] <seb128> I guess we don't need to track the stock ids one
[11:47] <seb128> but having chromium working with appmenu would be nice
[11:47] <seb128> davidbarth, ^
[11:47] <chrisccoulson> yeah, it would
[11:47] <seb128> is anybody from your team tracking that? bratsche?
[11:47] <chrisccoulson> having firefox working with it would be nice too ;)
[11:47] <seb128> yeah, that seems harder to get though
[11:48] <seb128> chrisccoulson: btw you still have work items on that spec you should probably postpone, ie the mir one
[11:48] <chrisccoulson> oh, i thought i'd already postponed that?
[11:49] <seb128> let me check if that's the tracker being buggy
[11:49] <seb128> chrisccoulson: https://blueprints.edge.launchpad.net/ubuntu/+spec/desktop-maverick-chromium
[11:49] <chrisccoulson> no, you're right
[11:49] <seb128> no you didn't ;-)
[11:49] <seb128> you can probably postpone the xulrunner dependency one as well
[11:50] <chrisccoulson> yeah, i must have been thinking of another WI :0
[11:50] <chrisccoulson> urrm, ;)
[11:50] <chrisccoulson> i got the shift key the wrong way round there
[11:50] <seb128> chrisccoulson: you updated the pyxpcom yesterday
[11:50] <seb128> so maybe you were thinking about this one ;-)
[11:50] <seb128> chrisccoulson: thanks
[11:51] <and471> seb128, is UI freeze start of tomorrow or end of tomorrow?
[11:51] <seb128> and471, somewhat between those
[11:51] <seb128> usually start of u.s day
[11:51] <seb128> ie mid european day
[11:52] <seb128> well it's not an hard freeze so uploads will still go through
[11:52] <seb128> often people get late updates during the day still
[11:53] <chrisccoulson> are people going to be rushing to get last minute builds in before the freeze?
[11:53] <and471> seb128, ok thanks
[11:53] <chrisccoulson> i have the first builds for firefox 3.6.9 to do today ;)
[11:53] <seb128> lol
[11:53] <seb128> excellent timing ;-)
[11:53] <chrisccoulson> perfect timing as always ;)
[11:53] <chrisccoulson> heh
[11:53] <seb128> don't worry about buildds
[11:54] <seb128> upload time count, not build time
[11:54] <seb128> so if things build over night that's ok
[11:54] <seb128> dpm, hi
[11:56] <desrt> anyone seen didrocks
[11:56] <desrt> ?
[11:57] <chrisccoulson> he's on vacation i think
[11:57] <desrt> oof
[11:58] <seb128> desrt, hey
[11:58] <desrt> seb128: hey
[11:58]  * desrt is in paris
[11:58] <seb128> do you need him for something? can we help maybe since he's not there?
[11:58] <seb128> oh ok
[11:59] <seb128> he will be back to work next week but I'm not sure he will be on paris
[11:59] <seb128> they are moving next month and he's visiting flats this week in Lyon
[12:00] <desrt> i figured that would be the case.  he mentioned that he was likely going to be involved with moving-related things
=== MacSlow is now known as MacSlow|lunch
[12:18] <dpm> hey seb128
[12:19] <seb128> dpm, how are you?
[12:19] <dpm> seb128, fine, thanks :), how's the day so far?
[12:19] <seb128> busy as usual ;-)
[12:20] <dpm> I knew the answer :)
[12:20] <seb128> dpm, I wanted to ask if we need to do anything for the unity places translations
[12:20] <seb128> is there a place where translators can see what is new and should be translated?
[12:21] <seb128> or do they need "those new components need translations, that's default UNE"
[12:21] <seb128> ie some sort of note dropped somewhere
[12:21] <seb128> in which case where is the somewhere ;-)
[12:23] <dpm> seb128, I don't think you guys need to do anything as long as translations work. I approved the templates and set the priority high so that they appear on the first page along with the unity template: https://translations.launchpad.net/ubuntu/maverick. The only thing missing is testing that the translations actually work. We've got the Global Jam coming in this weekend, so I'd expect people to start translating and seeing the translations with the nex
[12:23] <dpm> t language pack. What I should do is to perhaps send a heads up e-mail to the translators list, with the message you are suggesting.
[12:25] <seb128> dpm, ok thanks
[12:25] <seb128> dpm, it should work I've tested that before upload
[12:25] <dpm> seb128, awesome
[12:25] <seb128> I did set some strings in the fr.po, built a .mo and installed it
[12:25] <seb128> unity showed those correctly in ui
[12:25] <seb128> unity showed those correctly in the ui
[12:25] <seb128> dpm, ok, so seems we are on shape, thanks
[12:26] <dpm> seb128,  I've actually got a question related to that. Should I disable the window-picker-applet and the go-home-applet? They were used in UNR - UNE, and last time I looked they were in main, but I'm not sure if unity uses them at all
[12:27] <dpm> I meant disable the templates in LP
[12:27] <seb128> unity doesn't use them
[12:27] <seb128> I'm not sure if linaro or arm use those though
[12:27] <seb128> you can lower the priority for sure
[12:27] <seb128> but maybe check with asac or ogra if they use them before turning off the template
[12:27] <vish> could someone open the Lucid task for Bug 587853 ? seems like a good fix for an LTS
[12:27] <ubot2> Launchpad bug 587853 in cheese (Ubuntu) (and 1 other project) "After video recording, record/take photo button becomes disabled (affects: 2) (dups: 1) (heat: 22)" [Low,Fix released] https://launchpad.net/bugs/587853
[12:28] <ogra> dpm, ubuntu-arm uses them
[12:29] <ogra> since we have no choice
[12:29] <seb128> well you could use gnome-panel
[12:29] <ogra> we do
[12:30] <seb128> or a normal desktop GNOME session
[12:30] <ogra> we cant, to heavy
[12:30] <seb128> or xfce ;-)
[12:30] <ogra> even gnome-panel is to heavy for use but we didnt get the alternative in time
[12:30] <seb128> anyway doesn't cost a lot to keep those around
[12:30] <ogra> right
[12:31] <ogra> please keep them at least until N
[12:31] <seb128> dpm, so maybe just set the priority a bit lower for translations
[12:31] <seb128> they should be mostly translated anyway
[12:31] <seb128> they didn't change a lot
[12:31] <ogra> we might steal the new ubiquity panel by then
[12:31] <ogra> which will use indicators all over
[12:31] <dpm> seb128, ogra, yeah, I'll do that: keep them and lower the priority. thanks
[12:31] <seb128> thanks
[12:31] <ogra> thanks :)
[12:32] <ogra> seb128, did dyfet talk to you about telepathy-glib any time recently ?
[12:32] <ogra> (the last upload FTBFS on armel failing two tests)
[12:44] <seb128> ogra, no, I noticed that somebody did a build retry and the testsuite failed the same way
[12:44] <ogra> yeah, that was me after dyfet told me it would build fine and he had tested
[12:45] <ogra> i would expect him to contact the maintainer first since he was tasked with fixing it since a while ... intresting he didnt
[12:46] <ogra> one failure is a segfault, the other seems to be invalid utf-8, not sure why the latter only affects armel
[12:56] <chrisccoulson> mvo - the policykit agent blocks logout now
[12:58] <mvo> chrisccoulson: oh? because of the session registering change?
[12:59] <chrisccoulson> mvo - probably ;)
[12:59] <chrisccoulson> presmably it's not responding to the session saving handshaking magic when i log out ;)
[13:00] <mvo> chrisccoulson: could you file a bug and assign to me ? I will fix it tomorrow, but today I need to get some UI freeze work done
[13:00] <chrisccoulson> mvo - sure, no problem
[13:01] <kiwinote> mvo: just fyi I'm working on the addons branch atm
[13:01] <chrisccoulson> hmmm, my connection is going super-slow atm :/
[13:01] <mvo> kiwinote: thanks, could you check with the stuff that gary did too?
[13:02] <bcurtiswx_> good morning
[13:03] <kiwinote> mvo: yeah, I took a look at what gary did, but it seems he just took trunk and merged the addons branch and did a little tweak to make it work. Could this be the case, or did you have the impression there was slightly more?
[13:04] <mvo> kiwinote: that is very likely
[13:04] <mvo> kiwinote: I did not look myself yet
[13:04] <kiwinote> ok, thanks
[13:05] <mvo> kiwinote: please let me know when you are done
[13:05] <mvo> kiwinote: I check the merge then :)
[13:06] <kiwinote> mvo: will do, I'm doing some biggish refactoring of the code, so it may only be towards dinner time. If you want it before then, just let me know and then I'll make less changes ;)
[13:07] <mvo> kiwinote: heh :) just refactoring or ui changes as well?
[13:07] <mvo> kiwinote: it has to go in now, but dinner time should still be good, I will just attack other stuff then in the meantime
[13:08] <kiwinote> mvo: just refactoring atm, to make it nice and easy to review for merging into trunk
[13:11] <mvo> kiwinote: great, I will just wait for your ack then
[13:22] <davidbarth> seb128: bratsche was in contact with upstream directly, we haven't assigned people to do that patch for maverick
[13:26] <glatzor> mvo, is there a plan to use the aptdaemon plugins for language-selector?
[13:26] <mvo> glatzor: yes, but language-selector is currently unmaintained
[13:27] <glatzor> mvo, i was just aksing out of curiosity
[13:28] <mvo> glatzor: right, it was on the radar and discussed a couple of times, but did not quite get done
[13:30] <glatzor> mvo, I pushed some more fixes to aptdaemon's 0.3 branch. would be nice if you could find some time to upload before beta freeze. I had to re-fractor the locking a little bit
[13:30] <tjaalton> hum, changing the path of user dirs in /etc/xdg/user-dirs.dirs to something else than the default means they don't have emblems in nautilus
[13:32] <glatzor> mvo: for 11.04 i want to split the deferreds into a separate package. perhaps somebody else finds them useful
[13:32] <mvo> glatzor: sure, will do
=== MacSlow|lunch is now known as MacSlow
=== oubiwann is now known as oubiwann-away
=== oubiwann-away is now known as oubiwann
[13:43] <seb128> davidbarth, right, I don't think we should do the changes, I'm just wondering if somebody is still in contact with the upstream team to make that happen maybe next cycle
[13:43] <seb128> hey nessita
[13:43] <nessita> good morning!
[13:43] <nessita> hey seb128, how is it going?
[13:43] <seb128> I'm fine thanks
[13:44] <seb128> how are you?
[13:44] <nessita> seb128: pretty good, though I'm in mute mode since my throat hurts a lot
[13:44] <nessita> seb128: so, paradise for the people that is with me :-P
[13:45] <davidbarth> seb128: yep, we're still in touch with them
[13:45] <seb128> nessita, lol, hope you get better then ;-)
[13:46] <seb128> davidbarth, ok, but I guess that's not likely to happen this cycle now?
[13:46] <seb128> davidbarth, ie nobody upstream picked up the task to work on that?
[13:46] <seb128> brb
[13:50] <davidbarth> seb128: nope
[13:50] <seb128> davidbarth, ok
[13:50] <seb128> davidbarth, thanks
[14:00] <chrisccoulson> pitti - if you have a few minutes, could i get an ack from you on bug 623962?
[14:00] <ubot2> Launchpad bug 623962 in firefox (Ubuntu Lucid) (and 1 other project) "Enabling the crash reporter for Lucid (affects: 1) (heat: 8)" [Undecided,New] https://launchpad.net/bugs/623962
[14:05] <ogra> seb128, seems telepathy-glib builds fine with -O0, mind if i set that for armel builds
[14:06] <seb128> we are on sync with debian right now
[14:06] <seb128> I would like to have it stay this way
[14:06] <ogra> are we? i thought we have vala changes anyway
[14:06] <seb128> can you try to talk to the debian guys about it? or upstream?
[14:06] <ogra> hmm
[14:06] <seb128> just because the new vala was waiting in debian NEW
[14:06] <ogra> i'm a bit short on time for that
[14:06] <seb128> it has been newed yesterday
[14:07] <seb128> we will be in sync again with next upload
[14:07] <ogra> it breaks image builds and we need to prepare for beta, i'm happy to file upstream and debian bugs for it
[14:07] <seb128> ok, do your change, I will try to talk to them
[14:07] <seb128> or better send the change to the debian bts
[14:07] <seb128> with the url of the build log
[14:07] <ogra> will do, thanks :)
[14:07] <seb128> thank you
[14:09] <pitti> chrisccoulson: answered
[14:19] <chrisccoulson> pitti, thanks. i've just responded too
[14:30] <pitti> chrisccoulson: also answered
[14:30] <chrisccoulson> pitti - cool, thanks!
[14:33] <mpt> glatzor, still around?
[14:34] <glatzor> mpt, right
[14:38] <mpt> glatzor, is ERROR_NO_PACKAGE really ever used for removals? I assume you can successfully use aptdaemon to remove a package that is no longer in your sources
=== zyga is now known as zyga-us-visa-stu
[15:10] <glatzor> mpt, NO_PACKAGE is raised if the requested package cannot be found in the cache
[15:10] <mpt> glatzor, ah, so if I did the aptdaemon equivalent of "apt-get remove fqwhgads"
[15:10] <glatzor> mpt, so you could ignore it in s-c
[15:11] <glatzor> mpt, right
[15:11] <mpt> Well, *in theory*, s-c should never trigger it. ;-)
[15:11] <mpt> as opposed to ERROR_PACKAGE_NOT_INSTALLED and ERROR_PACKAGE_ALREADY_INSTALLED, which it occasionally triggers
[15:12] <glatzor> right
[15:17] <vish> mpt: one odd name though is "Software Sources" ... maybe we need a good name for that, or just move it out of the System > Preferences, into Software Updater > Preferences ;)
[15:21] <vish> ie. the "Settings" button alone..
[15:22] <mpt> vish, yes, the awkward thing is that Software Sources is invoked by three separate programs in the default seed (USC, Update Manager, Synaptic)
[15:26] <mpt> glatzor, ok, I've now added suggestions for almost every error in <http://wiki.debian.org/aptdaemon/error-messages>. I've left out the ones I don't understand yet, and I'm sorry I don't have time to grok them, but that should be plenty enough to keep you busy. :-)
[15:32] <vish> mpt: yeah, but not sure why it would be awkward that they all call upon the same program.. just wanted to mention that it probably need not exist in the menu.
=== JanC_ is now known as JanC
[15:33] <vish> any changes in the 'Software Sources' is not final , it ends up triggering synaptic or similar to complete anyway.
[15:37] <lucidfox> I wonder if software-sources will ever be modified to use PolicyKit instead of global root access
[15:40] <rickspencer3> dang
[15:40] <rickspencer3> evolution is taking up 100% of my CPU :/
[15:40] <mpt> vish, that's an excellent point. Let's remove it. And if it's no longer a branded "program", its item no longer needs an odd-one-out icon in the USC "Edit" menu. :-)
[15:41] <vish> hehe.. :)
[15:43] <devildante> mvo: what's the status about the addons branch?
[15:44] <mvo> devildante: kiwinote is working on it, then I shall merge
[15:44] <mvo> devildante: sorry that its taking so long, but we will get there :)
[15:44] <devildante> mvo: np, and thanks a lot :)
[15:45] <vish> mvo: what do you think about hiding software sources from the menu?  [pls see above ^^conversation with mpt ]
[15:46] <mvo> vish: no strong opinion either way
[15:46] <kiwinote> mvo: if you can add a 'self.' in line 1143 of appdetailsview_gtk.py in trunk, that would be sweet ;)
[15:47] <vish> mvo: cool! /me files bug ;)
[15:47] <mvo> kiwinote: just commiting that, sorry
[15:48] <kiwinote> devildante: hi! I am indeed working on it, and I hope to be able to hand it over to mvo within an hour or two. We're getting there :)
[15:48] <kiwinote> mvo: np
[15:48] <devildante> yay!
[15:48] <mvo> tremolux was faster with the fix ;) thanks dude!
[15:49] <kiwinote> hehe, thanks tremolux ;)
[15:49] <devildante> vish, mvo, mpt: about hiding "software sources"... I don't know if you remember, but I had proposed on u-d-d that we separate software-properties into update-manager and software-center. What do you think?
[15:50] <tremolux> kiwinote, mvo: hehe, crazy!
[15:50] <mpt> devildante, possibly, but I haven't seen a complete wireframe to demonstrate that it's practical
[15:51] <vish> devildante: why do we need to separate? when were are trying to unite everything together..
[15:51] <devildante> "complete wireframe"?
[15:51] <devildante> mpt ^
[15:52] <mpt> devildante, as in, showing where everything that used to be in "Software Sources" would go
[15:53] <devildante> vish: By separating, we could offer settings appropriate to both the 2 apps without all the options available
[15:53] <devildante> mpt: hmm
[15:53] <devildante> mpt, I'll think of some mock-ups later
[15:57] <pitti> dpm, Riddell: does KDE use gettext or .ts files these days? Is .ts supported by LP?
[15:58] <dpm> pitti, KDE uses gettext (if I'm not mistaken they use qt but they overload their translate calls with gettext calls). .ts is not supported by LP
=== bjf[afk] is now known as bjf
[15:59] <pitti> dpm: ack, thanks
[16:02] <Riddell> pitti: KDE has always used gettext
[16:03] <vish> devildante: mpt: hmm , probably folks not interested in using synaptic/USC, and want to do it using cmdline might be interested in having that menu item..
[16:03]  * vish postpones filing bug..  :( 
[16:06] <rodrigo_> anyone knows about glibc brokenness on latest maverick packages? I seem to be having weird memory corruption bugs, and after 1 hour looking at my code, it doesn't seem to be my fault
[16:07] <and471> mvo, hey, I can't figure out how to get the display name of a ppa, from the address of the ppa. Can you recommend how to do this? (for software-properties-gtk)
[16:09] <mvo> and471: I'm in a meeting right now, can you ping me in some minutes (30 or so)? its a bit tricky, but not impossible (requires mucking around with the sourcelist object
[16:10] <and471> mvo, sure
[16:12] <mpt> virtuald, so they want to install and remove stuff with the command line, but not edit their sources.list with the command line? hmph :-)
[16:13] <vish> mpt: heh, i thought the same , but i was worried if i would mostly get yelled at for that bug ;p
[16:16] <mvo> hey nessita - a quick question. does the ubuntu-sso-login code detect revoked oauth tokens and re-asks for the password in this case (acquires a new one automatically)?
[16:17] <Laney> rodrigo_: Are you aware that the mono u1 bindings got broken?
[16:17] <mvo> glatzor: what do you think about automatically moving keys in a deb line to apts auth.conf when they are added via adptdaemon ? for repos that require a password
[16:17] <rodrigo_> Laney, yes, fixing them in a minute, as soon as I finish other stuff
[16:17] <Laney> rodrigo_: ah cool, you know the fix?
[16:17] <Laney> ajmitch: ^^^
[16:18] <rodrigo_> Laney, not yet
[16:18] <Laney> k
[16:18] <nessita> mvo: not at the moment
[16:18] <mvo> nessita: is that planed for -final ?
[16:18] <nessita> mvo: we haven't planned it, could you please file a bug?
[16:18] <mvo> nessita: I'm updating my work items currently and want to reflect the correct status
[16:18] <mvo> nessita: sure
=== zyga-us-visa-stu is now known as zyga
[16:19] <nessita> mvo: thank you, please assign to me
[16:20] <and471> glatzor, since michael is in a meeting, do you know how to do this? "I can't figure out how to get the display name of a ppa, from the address of the ppa. Can you recommend how to do this? (for software-properties-gtk)"
[16:23] <mvo> nessita: bug #624065 (hope its explained well enough?)
[16:23] <ubot2> Launchpad bug 624065 in ubuntu-sso-client "Please detect revoked SSO tokens (affects: 1) (heat: 6)" [Undecided,New] https://launchpad.net/bugs/624065
[16:24]  * nessita checks
[16:27] <nessita> mvo: do you know if current REST api allows to detect revoked tokens?
[16:27] <nessita> SSO api, I mean
[16:28] <mvo> nessita: not sure, my (naive) approach would be to use it against the sso api to get the "what-is-my-name", if that returns non-40x it should be ok
[16:29] <nessita> argh! :-)
[16:29] <nessita> that's a bit hacky. I'll check with achuni about this
[16:31] <Chipaca> mpt: re ubuntu sso client gui: can we *not* display the helper text (the descriptive text under the "Create <b>Appname</b> account" header) on the t&c page of the process? Currently it's shown there, so the text has to work on both registration form and the t&c page, and it's a lot harder to do right.
[16:32] <mpt> Chipaca, screenshot?
[16:32] <mvo> and471: I just checked the gtkbuilder branch, I can't access "Automatic updates" anymore, the box there, is that working for you?
[16:33] <Chipaca> mpt: https://docs.google.com/document/edit?id=1LCPnN4JfY6B1PFBnnHyM_ucmnimaLgZ4W_RcWsP4kYk
[16:33] <and471> mvo, sorry, forgot to push that fix
[16:34] <and471> mvo, ok fixed in rev 624
[16:34] <mvo> thanks and471
[16:36] <mpt> mvo, nessita: When I run "software-center --enable-buy" and choose "File" > "Reinstall Previous Purchases...", the dialog that comes up invites me "To connect this computer to ubuntu-software-center". Should I report that bug on software-center, ubuntu-sso-client, or both?
[16:37] <mpt> Chipaca, I need to sign in to see that document but my Canonical account doesn't work
[16:37] <Chipaca> mpt: weird, I set it to 'anyone with the link'
[16:37] <Chipaca> mpt: give me a minute and I'll put an actual screenshot
[16:37] <Chipaca> mpt: you want it as it is now, the proposed change, or both?
[16:38] <mpt> Chipaca, both if possible. :-)
[16:38] <Chipaca> ok. give me *two* minutes then :)
[16:38] <mvo> and471: thanks, one more. I get warnings on the commandline http://paste.ubuntu.com/483505/
[16:39] <vish> mpt: bug 624072 , more comments welcome :D
[16:39] <ubot2> Launchpad bug 624072 in software-properties (Ubuntu) "Hide 'Software Sources' from the System > Administration menu (affects: 1) (heat: 6)" [Undecided,New] https://launchpad.net/bugs/624072
[16:39] <mvo> and471: and the close button seems to be no longer connected :)
[16:40] <and471> mvo, I'll take a look
[16:40] <mpt> vish, no time, sorry
[16:40] <vish> ;p
[16:40] <mpt> but thank you
[16:41] <mvo> thanks and471!
[16:41] <and471> np :)
[16:43] <mvo> and471: same for revert, probably simply a changed singal name or something
[16:44] <and471> mvo, yeah, just realised glade doesn't copy the signal names over :/
[16:44] <nessita> mpt: that is the name that the sso service is getting as the app_name
[16:44] <nessita> mvo: what are you passing as "app_name"?
[16:45] <mpt> nessita, there's supposed to be explanatory text: "To reinstall previous purchases, sign in to the Ubuntu Single Sign-On account you used to pay for them." Without that, I think people will be pretty lost
[16:45] <and471> mvo, do we need that help button anymore?
[16:46] <nessita> mpt: the ubuntu sso service doesn't make any choice about what to show, is just using string passed as parameters
[16:47] <mvo> and471: no
[16:47] <and471> goodie
[16:47] <nessita> mvo: how are you calling the login_or_register... call?
[16:47] <mpt> nessita, ok, are there other applications that use the dialog, where it makes more sense?
[16:48] <mvo> nessita: its currently just using ubuntu software center but I can pass any string here, hold on a sec
[16:49] <nessita> mpt: sorry, but I'm not sure what you're asking. The SSO GUI has 3 pseudo-customizable parts: the "header", which is a fixed string provided by the design where the application name is given by the caller, the "description" or "help" text, which is a smaller text under the header, and the url to the terms
[16:49] <nessita> mpt: what part are you referring to? do you have a screenshot?
[16:50] <nessita> (the "description" or "help" text  is 100% given by the caller)
[16:50] <mvo> nessita: I have version 0.99.2 currently and I use self.proxy.login_to_get_credentials)
[16:50] <mvo> nessita: it seems to only pick up the first argument (appname) and display that in 0.99.2 - should I update?
[16:51] <and471> mvo, can you try again with the latest revision? this fixes the buttons, and should fix the errors
[16:51] <nessita> mvo: can you please show me the whole call you're making to self.proxy.login_to_get_credentials ?
[16:51] <mvo> and471: thanks
[16:52] <mvo> nessita: http://paste.ubuntu.com/483510/
[16:52]  * and471 can feel the pressure of UI freeze
[16:52] <Chipaca> mpt: http://ubuntuone.com/p/Ds8/ vs http://ubuntuone.com/p/Ds9/
[16:52] <mpt> nessita, http://imgur.com/c2W5W
[16:53] <mpt> and471, looking for something to do? ;-)
[16:53] <mpt> Chipaca, "Create Ubuntu account" should be "Create Ubuntu One account", right?
[16:54] <and471> mpt. hehe, well if mvo says that the gtkbuilder stuff is okay, then yeah, until he can help me with the software-properties ppa view :)
[16:54] <asac> pitti: are you doing proposed approval?
[16:54] <asac> ;)
[16:54] <Chipaca> mpt: yeah, this is just a test thrown together, not an actual app using the service
[16:54] <mpt> oh, right
[16:54] <nessita> mpt: as app_name you (or mvo in this case) should pass whatever you wanna show as the application name. So 'ubuntu-software-center' should be "Ubuntu Software Center"
[16:54] <nessita> mvo: you're passing the parameters sightly wrong
[16:54] <asac> pitti: maybe we could not let stuff like openoffice/linux and other big proposed uploads that are not super urgent in a few days before beta freeze ;)
[16:55] <nessita> mvo: the first parameter is the app_name as the user will read it
[16:55] <asac> rather during freeze where everything is idle ;)
[16:55] <nessita> mvo: and the second parameter is what it will be show below the big title when registering
[16:55] <mpt> nessita, unfortunately that wouldn't solve the problem. Ubuntu Software Center is already running perfectly well on my computer. :-) So it doesn't make sense to say "connect this computer to Ubuntu Software Center".
[16:55] <mvo> nessita: right, I will fix that, but it would be nice to have a help text for this dialog as well (as mpt pointed out)
[16:55] <asac> pitti: not really a complain. just a "bringing to attention" ;)
[16:55] <Chipaca> mpt: the T&C page shown is very ugly, and that's being fixed with @media selectors on the server side :) (yeah, your poor eyes still get to see it ugly for now)
[16:56] <mpt> Chipaca, I guess there's a previous step here that also shows the intro text?
[16:57] <Chipaca> mpt: yes
[16:58] <nessita> mpt: I see. I understand your need and you can please file a bug for ussoc, but this GUI is matching the design you and john lea agreed on. As far as I know we can't change that at this point
[16:58] <mpt> Chipaca, then as best I can understand it without seeing the previous step: Sure, no need to repeat the intro text.
[16:58] <nessita> mpt: translator are already translating these strings and we're on interface freeze
[16:58] <pitti> asac: well, I already delayed it for some days to not accept it together with the countless new SRU kernels
[16:58] <pitti> asac: if you need some other packages bumped, please shout my way, I'll rescore :)
[16:58] <Chipaca> nessita: we're *in* api freeze, *against* ui freeze :)
[16:59] <and471> nessita, still got tomorrow
[16:59] <mpt> nessita, I'm sorry I didn't see that. A lack of communication, not your fault.
[17:01] <mvo> and471: branch looks good now, seems to be working fine afaict, good work!
[17:01] <and471> mvo, thanks :)
[17:02] <and471> mpt, ok I can do something small now if you want
[17:02] <mpt> and471, the dialog we're talking about displays "This field is required." text if you click "Connect" without filling in both fields. A simple improvement would be to instead make the button insensitive whenever either field is empty.
[17:03] <mpt> and471, https://code.launchpad.net/ubuntu-sso-client
[17:03] <and471> mpt, sure
[17:03] <and471> nessita, should I then write a test for this?
[17:03] <and471> nessita, and is it the login dialog?
[17:04] <nessita> and471: one second, argeeng with the boss
[17:04] <and471> ok
[17:04] <mpt> Chipaca, also, that T&C frame needs a 1px border around it. (Common GTK mistake.)
[17:06] <Chipaca> mpt: you mean around the webkit, so it doesn't smush up with the scrollbar?
[17:07] <mpt> Chipaca, yes, though it would need it even if there wasn't a scrollbar (it visually "explains" the change from grey to white)
[17:07] <asac> pitti: hmm. the whole unity stack on armel ;)
[17:08] <asac> pitti: let me see what wasnt built
[17:08] <devildante> mvo: we want to rename update-manager to software-updater, see bug 618723
[17:08] <ubot2> Launchpad bug 618723 in update-manager (Ubuntu) (and 1 other project) "Rename Update Manager to Software Updater (affects: 3) (heat: 20)" [Wishlist,In progress] https://launchpad.net/bugs/618723
[17:08] <asac> pitti: oh. wait a bit. i think its now building ;) ... stay tuned
[17:08] <nessita> mpt: about the 'Connect' button being enabled. We discussedthis with JOhn lea, and we agreed we'll have the button enabled
[17:08] <Chipaca> mpt: ah, so a coloured border?
[17:08] <nessita> and471: we're not changing that
[17:09] <nessita> and471: we're only in bug fix mode, we need to keep the diffs to a minimun
[17:09] <mpt> devildante, mvo, if that rename takes mvo time, I don't mind at all if it gets delayed to Natty. It's not that important.
[17:09] <nessita> and471: and this behavior was agreed on with the design person at that point
[17:09] <devildante> mpt: okay, but I'll wait for his answer
[17:09] <and471> nessita, no problem, mpt told me to :)
[17:10] <mvo> devildante: I saw it, thanks for the work, but I don't feel comforatble at this point
[17:10] <nessita> and471, mpt: I know, we're having some trade off because all our design discussions were made with a different design person :-/
[17:10] <Chipaca> about validation, our assumption was that as-you-type validation was incredibly better, but we were overruled :(
[17:10] <devildante> mvo: we can delay it for narwhal
[17:10] <mvo> heh :)
[17:10] <mvo> nutty narwhal
[17:10]  * mvo really needs to go for dinner
[17:10]  * devildante will be afk for a while, can you discuss with mpt?
[17:10] <devildante> mvo: okay :)
=== devildante is now known as devildante-afk
[17:11] <and471> Chipaca, I think you just need to set the gtk_shadow type on the scrolledwindow to IN
[17:12] <Chipaca> mpt: and471: if you could run changes to ubuntu-sso-client past us, we might avoid work duplication (or changes that don't work for all use cases)
[17:12] <Chipaca> nessita in particular has the whole thing in her head, because she rocks
[17:12] <and471> Chipaca, sure, but I am not making any changes
[17:13] <mpt> and471, yeah, this counts as a change for Chipaca in the sense that something nessita needs to know about
[17:13] <and471> mpt, I am confused
[17:14] <mpt> and471, oh, are you talking about the button sensitivity, or the scrolledwindow style?
[17:14] <and471> mpt, I don't really know
[17:14] <nessita> and471, mpt: what could help is to have bug reports for all these requests, so we can schedule them for this cycle or the next, and we can follow the discussion in the report. That way we can point to others to that disussion if needed
[17:14] <and471> Chipaca, what were you talking about?
[17:15] <mpt> nessita, yes, the only reason I asked about it on IRC was because I didn't know what it needed to be filed against. I've reported bug 624097 now.
[17:15] <ubot2> Launchpad bug 624097 in software-center (Ubuntu) (and 1 other project) ""Reinstall Previous Purchases" sign-in dialog doesn't explain itself (affects: 1) (heat: 6)" [Undecided,New] https://launchpad.net/bugs/624097
[17:15] <Chipaca> and471: button sensitivity. But anything related to changes in ubuntu sso: interface, logic, api, the pizza topping, anything
[17:16] <nessita> mpt: you can always file it agains SSO and I can assign it to other project if needed :-)
[17:16] <and471> Chipaca, ah ok, well I thought mpt might have filed it, as he asked me to do it
[17:16] <nessita> Chipaca: speaking of pizza, shouldn't we having lunch right now?
[17:16] <mpt> and471, no, I just happened to notice it when I was looking at the dialog for this other bug
[17:16] <nessita> :-)
[17:16] <mpt> oh, pizza
[17:17] <and471> Chipaca, FYI just margherita is fine, no
[17:17] <Chipaca> lucnh!
[17:17] <and471> other toppings
[17:17] <Chipaca> lunch is now
[17:17] <nessita> and471, mpt: I'm sorry if I'm a little bitch about this, but we have tons of bugs to resolve and we're trying to respect freezes a lot
=== fta_ is now known as fta
[17:17] <Chipaca> except for rodrigo_, he stays here hacking
[17:17] <and471> nessita, hehe, no problem :)
[17:18]  * rodrigo_ wants lunch also
[17:18] <nessita> and471, mpt: so all the bug reports are welcomed, I'll review them :-)
[17:20] <and471> mpt, so anything else?
[17:24] <asac> so in the past i was able to start X :0; then gnome-terminal ... then mutter --replace (or any other window manager)
[17:24] <asac> this seems to have regressed in maverick for me
[17:24] <asac> also if i just do X:0; gnome-terminal and go to console with alt-1 again, what happens is that gnome-terminal crashes
=== nessita is now known as nessitalunch
[17:24] <asac> but i can switch to console without crashes if i use a full desktop session
[17:24] <asac> whats going on?
[17:25] <asac> RAOF: ^^
[17:25] <asac> wake up. a really unimportant, but urgent help request ;)
[17:25] <mpt> and471, I'm reporting a few other bugs in that dialog right now, but they're off-limits, so ...
[17:25] <and471> hehe
[17:25] <mpt> and471, maybe bug 623460?
[17:25] <ubot2> Launchpad bug 623460 in software-center (Ubuntu) "Pathbutton element that doesn't do anything is strongly highlighted (affects: 1) (heat: 6)" [Undecided,New] https://launchpad.net/bugs/623460
[17:27] <and471> mpt, meh, that is more for nzmm, he is still working on that other light themes bug, why don't you mention it on there?
[17:27] <and471> mpt, it is fine, I shall do something else
[17:41] <glatzor> and471, do you still need some assistance?
[17:43] <and471> glatzor, yes please :)
[17:43] <and471> glatzor, should I tell you the full scenario?
[17:47] <mpt> nessitalunch, ok, reported bug 624101 and bug 624117
[17:47] <ubot2> Launchpad bug 624101 in ubuntu-sso-client "Sign-on dialog is maximizable (affects: 1) (heat: 6)" [Undecided,New] https://launchpad.net/bugs/624101
[17:47] <ubot2> Launchpad bug 624117 in ubuntu-sso-client "Sign-on dialog default button is sensitive when unusable (affects: 1) (heat: 6)" [Undecided,New] https://launchpad.net/bugs/624117
[17:47] <mpt> nessitalunch, but the most important are bug 624097 and bug 624127
[17:47] <ubot2> Launchpad bug 624097 in software-center (Ubuntu) (and 1 other project) ""Reinstall Previous Purchases" sign-in dialog doesn't explain itself (affects: 1) (heat: 6)" [Undecided,New] https://launchpad.net/bugs/624097
[17:48] <ubot2> Launchpad bug 624127 in software-center (Ubuntu) (and 1 other project) "Sign-on dialog won't open a second time (affects: 1) (heat: 6)" [Undecided,New] https://launchpad.net/bugs/624127
[17:51] <mpt> tremolux, hi, how's hacking?
[17:52] <tremolux> mpt: ha, it's the day for it isn't it?
[17:53] <mpt> tremolux, how's my favorite pathbutton bug? :-)
[17:55] <tremolux> mpt: sorry, I haven't had time enough to get a fix for today's upload I fear
[17:56] <tremolux> mpt: do you think it's a UI freeze bug?
[17:57] <tremolux> mpt: seems just a regular bug, can still be fixed after freeze, no?
[17:57] <mpt> tremolux, sure, it doesn't change the UI in ways that would befuddle screenshots or help
[17:57] <mpt> that's a good point
[17:58] <tremolux> mpt: it seems that way to me too
[18:00] <mpt> tremolux, what are you working on at the moment?
[18:01] <tremolux> mpt: so I will fix this one, but I just can't get it in for today's upload
[18:01] <mpt> sure
=== al-maisan is now known as almaisan-away
=== MacSlow is now known as MacSlow|break
[18:02] <tremolux> currently reviewing addons, kiwinote did some additional work on it and we want this to go in today
[18:02] <tremolux> mpt: it's really cool stuff
[18:03] <tremolux> mpt: at the same time, I'm working on verifying the fix on LP for custom metadata for new-apps packages
[18:03] <mpt> excellent
[18:09] <and471> glatzor, should I tell you the full scenario?
=== devildante-afk is now known as devildante
[18:11] <mvo> kiwinote: is the addons ready ? or still being worked on (just curious :)
[18:12] <and471> mvo, have you ate?
[18:12] <mvo> and471: yes
[18:12] <and471> mvo, sorry that is terrible english, have you eaten?
[18:12] <and471> ah
[18:13] <and471> mvo, could you give me a hint of how to do this then?
[18:13] <and471> (the software-properties thing)
[18:13] <tremolux> mvo: there's a small bug when adding/removing addons, I'm tracking it now
[18:13] <and471> mvo, I think I have a way to do it, but it is a bit hacky
[18:18] <mvo> and471: ok, just use it for now and we can fixup stuff as we go along
[18:18] <mvo> and471: make sure its isolated in a function so that we can easily replace it
[18:18] <and471> mvo, what way would you have done it? (Could you just outline it?)
[18:22] <mvo> and471: hold on a sec, I write a small function
[18:22] <and471> mvo, oh okay then
=== nessitalunch is now known as nessita
[18:36] <mvo> and471: hm, my approach via sources = apt_pkg.SourceList(), metaindex iterating does not work, so I need to try something else. sorry
[18:38] <and471> mvo, my (quite slow) idea is to filter the apt.Cache based on a parsed version of pkg.candidate.uri
[18:38] <and471> then get the display_name from that
[18:38] <and471> would that work?
[18:38] <devildante> mpt: what is the reason for adding a background in software-center?
[18:38] <and471> devildante, prettyfication
[18:38] <and471> :)
[18:39] <mpt> devildante, it helps distinguish between "Get Software" screens and "Installed Software" screens. And, prettification. :-)
[18:39] <devildante> haha
[18:39] <and471> mpt, can we get that as a tag for bug reports? XD
[18:39] <devildante> I understand, but why clouds?
[18:40] <mpt> because the software is installed from the Internet
[18:40] <mpt> I'd like to get custom art for each department later on
[18:41] <devildante> thank you, mpt, and471 :)
[18:41]  * and471 finds it funny he got thanked for making up a word
[18:42] <devildante> heh ;)
[18:42] <and471> :)
[18:42] <devildante> mpt, vish, mvo: I just made a branch for hiding the menu entry: https://code.edge.launchpad.net/~ilidrissi.amine/software-properties/hide-desktop-file/+merge/33676
[18:43]  * devildante thinks we should have an ubot for code branches
[18:43]  * and471 agrees
[18:44]  * vish tickles ubot ! ;p
[18:47] <and471> mvo, how do you create an apt_pkg.SourceList object?
[18:47] <and471> mvo, (one that uses/etc/apt/sources.list)
[18:48] <kiwinote> hi mvo, tremolux! back from dinner now :)
[18:49] <kiwinote> tremolux: how's it getting on? managed or to find the bug, or is it something less obvious?
=== MacSlow|break is now known as MacSlow
[18:52] <tremolux> kiwinote: just looking at it, the problem is always the second time through mark_changes(), it does the wrong action
[18:55] <tremolux> kiwinote: btw, too many "hm, updated"s, eh?  ;)
[18:56] <kiwinote> tremolux: yeah, that was me checking that it got updated after transactions ;)
[18:57] <kiwinote> tremolux: although I think I removed most of the other random print statements I had inserted during testing ;)
[18:58] <tremolux> kiwinote: yeah, I wondered if you were checking if we are getting extra calls to that
[19:07] <kiwinote> tremolux: ok, found the bug, it's the restore call in about line 1479
[20:17] <asac> pitti: can you rescore https://edge.launchpad.net/ubuntu/+source/unity/0.2.28-0ubuntu3/+build/1933160 ?
[20:51] <devildante> vish: how will we deal with Scott Ritchie's comment?
[20:56] <vish> devildante: refresh :)
[20:57] <devildante> vish: thanks :)
[20:57] <vish> devildante: np.. thanks for a quick fix  ;)
[20:57] <devildante> vish: you're welcome :)
[21:17] <vish> devildante: if you add the wine ppa does it show up in SC ?
[21:17] <vish> it seems to hide for some reason! but is visible in synaptic
[21:19] <devildante> vish: will try
[21:20] <mvo> vish: it may take a bit until its showing
[21:22] <vish> mvo: oh! but still it seems odd, i'v been waiting for more than 10mins and tried re-opening SC a few times , but it is still not visible.
[21:23] <mvo> vish: *weeh* ok, that should definitely show up in that timeframe, looks like a bug then :/
[21:23] <vish> in Synaptic wine shows up as "/main (ppa.launchpad.net)"
[21:23] <devildante> mvo, vish: Failed to fetch http://ppa.launchpad.net/ubuntu-wine/ppa/ubuntu/dists/maverick/main/binary-i386/Packages.gz  404  Not Found
[21:23] <vish> mvo: i suspect it might be something wrong with wine ;)
[21:24] <vish> ah devildante !
[21:24] <vish> devildante: but how does it show up in synaptic?
[21:25] <vish> devildante: do you see it in Synaptic? or even missing there?
[21:25] <devildante> vish: I don't see it in synaptic nor USC
[21:25] <vish> devildante: well , i'm on lucid..
[21:26] <devildante> maverick here
[21:27] <devildante> vish: wait, ubuntu-wine is not available on Maverick :p
[21:27] <vish> devildante: lol!
[21:29] <devildante> vish: just reverted to lucid repo, it doesn't show up in usc but shows up in synaptic
[21:29] <vish> devildante: phew! not just me!
[21:30] <devildante> vish: go file a bug and I'll confirm it
[21:30] <vish> YokoZar: any idea, ^^ why wine ppa doesnt show up in Software Center ?
[21:31] <devildante> wait! it does show in USC!
[21:31] <devildante> I didn't see it, my bad :p
[21:31] <devildante> vish ^
[21:31] <vish> hrmmff! then why wont it show up for me :(
[21:33] <vish> devildante: ah! did you add the key?
[21:33] <devildante> vish: yep :)
[21:34] <devildante> So unauthenticated sources won't show?
[21:34] <vish> devildante: seems so!
[21:34]  * devildante is happy that his addons branch got merged :)
[21:34] <vish> now , is that a feature or a bug? :D
[21:35] <devildante> Maybe we should show them but with a BIG warning :p
[21:36] <vish> shouldnt adding "ppa:ubuntu-wine/ppa" also download the key?!
[21:36] <devildante> vish: it *does* download the key, but maybe not for you :p
[21:37] <vish> hrm!  lets try again! and i'll blame someone ;p
[21:45] <mpt> oh, add-ons merged?
[21:45]  * mpt updates trunk
[21:45] <tremolux> devildante: yes, great work on addons!!
[21:45] <devildante> tremolux: thx :)
[21:46] <tremolux> devildante: thanks to you!  it's a really nice feature and it's great to be able to have it for maverick
[21:47] <mpt> Oh! Who did the "Find it in the menu at"?
[21:47] <devildante> mpt: ?
[21:47] <devildante> mpt: (kiwinote did some tweaking)
[21:47] <tremolux> mpt: mvo did that this morning  ;)
[21:47] <mpt> brilliant, thanks mvo
[21:48] <mvo> cheers mpt
[21:49] <vish> mpt: unauthenticated are not supposed to show up in USC, is that by design.. or a bug?
[21:49] <vish> *ppa
[21:49] <mpt> vish, I don't understand the pros and cons well enough to have an opinion on that
[21:49] <mvo> vish: oh, that is a technical limitation of the current way apt works
[21:50] <mvo> vish: its a mix between bug and missing feature and oversight
[21:50] <vish> mvo: hehe :)
[21:50] <mvo> but I guess we should just call it a feature for now ;)
[21:50] <nessita> mvo: I closed this as an invalid for SSO, but I think is valid for software center https://bugs.edge.launchpad.net/ubuntu-sso-client/+bug/624127
[21:50] <ubot2> Launchpad bug 624127 in software-center (Ubuntu) (and 1 other project) "Sign-on dialog won't open a second time (affects: 1) (heat: 6)" [Undecided,New]
[21:51] <mpt> devildante, is now a good time for a few tweaks to the add-ons presentation?
[21:51] <mvo> nessita: isn't that just because the new ubuntu-sso-client has not landed yet?
[21:51] <mvo> nessita: or is the new signal api (with the app-name as first argument) in now?
[21:53] <devildante> mpt: what do you have in mind?
[21:53] <mpt> devildante, color and padding fixes
[21:54] <devildante> mpt: tomorrow is UIF, so we gotta be quick
[21:54] <mpt> yep
[21:54] <nessita> mvo: oh, right. The new API has landed yesterday, sorry for not being explicit about this
[21:54] <nessita> mvo: version 0.99.2 will have the latests API (ande definitive for maverick)
[21:55] <nessita> and*
[21:55] <mpt> devildante, first, can we make the add-on package name, and the brackets, the same light grey as "Version:", "License:", etc?
[21:55] <mpt> so that it's not cluttering the description so much
[21:58] <mvo> nessita: hm, so … according to https://wiki.ubuntu.com/SingleSignOn/UbuntuSsoClient the CredentialsError signal has 3 arguments, but I get only one and I'm pretty sure I have the latest client, let me kill it to ensure I really run with the update-to-version
[21:58] <devildante> mpt: good point
[21:59] <mvo> nessita: aha, all cool - it appears a old version was still around
[22:00] <devildante> mpt: the addons state bar ("Apply changes") is broken
[22:00] <devildante> mpt: try to choose an add-on to see what I mean
[22:00] <nessita> mvo: antoher thing about https://bugs.edge.launchpad.net/ubuntu-sso-client/+bug/624097
[22:00] <ubot2> Launchpad bug 624097 in software-center (Ubuntu) (and 2 other projects) ""Reinstall Previous Purchases" sign-in dialog doesn't explain itself (affects: 1) (heat: 6)" [High,Triaged]
[22:00] <mpt> devildante, the color and the padding? or something else?
[22:00] <nessita> mvo: we can add a new method call to the API, we're past feature freeze
[22:01] <devildante> mpt: when selecting an add-on, a green bar appears behind the description and shows nothing
[22:02] <nessita> mvo: I mean we can't :-/
[22:02] <mpt> devildante, I don't see that, either for an application that's installed already or one we can't
[22:02] <mvo> nessita: well, it requires a feature-freeze-exception. I changed the string now so that it reads "Ubuntu Software Center Store". that make it look ok-ish with the dialog
[22:02] <mvo> ^--- mpt
[22:03] <mpt> nessita, do I need to bribe you with chocolates?
[22:03] <mpt> devildante, sorry, "or one we can't" -> "or one that isn't"
[22:04]  * devildante is happy that aptdaemon's commit_packages() now has its own polkit auth dialog
[22:04] <mpt> devildante, oh dear, COLOR_GREEN_FILL etc
[22:05] <devildante> mpt: http://imgur.com/LANvS.png
[22:06] <tremolux> devildante: hmm, it seems to be working fine for me
[22:06] <mpt> devildante, I definitely don't get that in trunk.
[22:06] <devildante> hmm
[22:07] <devildante> I'll just re-update or something
[22:07] <nessita> mpt: we're talking about that, one minute
[22:07] <tremolux> devildante: I've seen that in kiwinote's branch, are you running that my chance?
[22:07] <tremolux> by chance?
[22:08] <devildante> tremolux: was running trunk
[22:08] <devildante> hmm, now it's okay
[22:08] <devildante> don't know what happened
[22:08] <devildante> but the green bar must be blue, I'll fix that
[22:08] <nessita> mpt, mvo: so, what we've agreed on is that we don't have resources to do it. But, what can be done is:
[22:09] <devildante> I lol'd at "Hello X Adventure" :p
[22:10] <nessita> mpt, mvo: we'll happily review patches and/or merge proposals to solve this, though we think that a new dbus call is a nonsense, and since you're asking for a freeze exception anyways, the current login_only call should be receiving an extra parameter "header_text"
[22:11] <nessita> mvo: please note that this change requires changes to both the DBus backend (file ubuntu_sso/main.py) and the GUI code (ubuntu_sso/gui.py)
[22:12] <nessita> mvo: we're also submitting only branches that have tests for the changes, both for the dbus layer and the gui layer. You can see tons of examples of tests within ubuntu_sso/tests/test_main.py and ubuntu_sso/tests/test_gui.py
[22:13] <mvo> nessita: right, thanks for this info
[22:17] <nessita> mvo: thank you
[22:19] <devildante> mpt: the "apply changes" bar is now blue :)
[22:19] <mpt> devildante, http://paste.ubuntu.com/483638/
[22:19] <mpt> devildante, so have you nuked all occurrences of the string "COLOR_GREEN"? :-)
[22:19] <mvo> nessita: I updated the bug #624127
[22:20] <ubot2> Launchpad bug 624127 in software-center (Ubuntu) (and 1 other project) "Sign-on dialog won't open a second time (affects: 1) (heat: 6)" [Medium,Confirmed] https://launchpad.net/bugs/624127
[22:20] <nessita> mvo: oh, me too! :-)
[22:20] <nessita> let 's see what resulted from that
[22:21] <devildante> mpt: yeah... at least, nothing is green now ;)
[22:21] <mvo> nessita: I attached a small script that seems to trigger a crash
[22:21]  * mpt is impressed that a comment posted 6 seconds ago appears before one posted 18 seconds ago
[22:22] <mpt> devildante, and is it grey in "Installed Software"?
[22:23] <devildante> mpt: yes, it's grey
[22:23] <mpt> devildante, excellent
[22:23] <nessita> mvo: ok, thanks!
[22:23] <devildante> mpt: weird, I thought it was blue :p
[22:24] <mpt> devildante, it should always be the same color as the installed state bar on the same screen
[22:24] <devildante> mpt: oh okay
[22:24] <devildante> (thank you view.section_color :p)
[22:26] <mpt> devildante, is section_color set by the "color =" in app.py?
[22:27] <mpt> (four of them)
[22:28] <mpt> I was trying to figure out why those colors are there while the rest of the colors are in appdetailsview_gtk.py
[22:28] <devildante> yes
[22:28] <devildante> because view.section_color is defined by app.py
[22:29] <devildante> fyi, I just copied from the installed state bar code
[22:30] <mpt> devildante, ok, next, the padding should be exactly the same as the installed state bar. It looks like that's defined in appdetailsview_gtk.py line 80.
[22:31]  * mvo is off to bed
[22:32] <mpt> goodnight mvo
[22:32] <devildante> mvo, bye :)
[22:35] <mvo> thanks
[22:35]  * mvo waves
[22:35] <devildante> mpt: done!
[22:36] <mpt> devildante, brilliant. Finally, the space between the left+right edges of the bar and the left+right edges of the pane should be the same as for the installed state bar.
[22:41] <vish> mpt: btw, Software sources has been removed from menu, you can go ahead and remove icon from SC's Edit too  ;)
[22:41] <rickspencer3> bug #613468
[22:41] <mpt> thanks vish
[22:41] <ubot2> Launchpad bug 613468 in soyuz "debian/control XB- fields not exported in PPAs (affects: 1) (heat: 12)" [Low,Fix committed] https://launchpad.net/bugs/613468
[22:42] <devildante> mpt: http://imgur.com/gI2Jd.png
[22:42] <rickspencer3> tremolux, hey
[22:42] <tremolux> rickspencer3: hiya!
[22:43] <rickspencer3> tremolux, so, should I be concerned that you set a work item to blocked due to the above bug?
[22:44] <rickspencer3> oh, I see, you simply moved it
[22:44] <tremolux> rickspencer3: oh, no, that was already set to blocked, I just moved it...the bug is fixed in LP but I haven't been able to do that exact test yet
[22:44] <rickspencer3> ok
[22:44] <mpt> devildante, excellent. Somehow though the "Cancel" and "Apply Changes" buttons in that screenshot have gotten around the wrong way. They're correct in trunk.
[22:45] <devildante> mpt: fixed :)
[22:45] <tremolux> rickspencer3: in summary, I have to test on dogfood.lp.net until the next LP release in Sep. 8th, and there's a problem getting my test package to build in the PPA there
[22:46] <mpt> devildante, great
[22:46] <devildante> mpt: anything else?
[22:46] <mpt> devildante, do you remember us talking about showing the "Total size:" row all the time, not just when you change add-ons?
[22:46] <tremolux> rickspencer3: was working with jelmer earlier to resolve it, and once it the test package builds, I can do the test and remove the block
[22:47] <rickspencer3> thanks tremolux
[22:47] <devildante> mpt, it's already there all the time: http://imgur.com/S2hf9.png
[22:48] <mpt> devildante, not in trunk: http://imgur.com/lt7vd
[22:50] <devildante> mpt: we show it even if it's already installed? that doesn't make sense to me...
[22:50] <mpt> devildante, only the "on disk" part
[22:51] <devildante> mpt: you mean, we should show "Total size: Already on disk"?
[22:52] <mpt> devildante, no, I mean like "apt-cache show software-center | grep Installed-Size"
[22:53] <mpt> devildante, search for the string "text describing the" in https://wiki.ubuntu.com/SoftwareCenter#software-item-screen :-)
[22:54] <devildante> mpt: okay, I'll do it ;)
[22:55] <mpt> thanks devildante
[22:57] <devildante> mpt: should we include the size of dependencies even when the pkg is installed?
[22:58] <mpt> devildante, no, just the individual package -- otherwise every KDE app would look enormous. :-)
[22:58] <devildante> haha, okay :)
[23:03]  * Riddell wonders why KDE apps get singled out there
[23:03] <mpt> Riddell, it would be the same for installing Gnome apps if USC was a KDE application
[23:04] <devildante> mpt: done!
[23:06] <mpt> devildante, and how about greying out the add-on package names?
[23:06] <devildante> mpt: I'll do it right now
[23:06] <mpt> thanks so much
[23:19] <mpt> grrrr
[23:20] <devildante> mpt: ???
[23:21] <devildante> mpt: oh well... http://imgur.com/uoJc1.png
[23:21] <mpt> devildante, please laugh at my diff <http://paste.ubuntu.com/483653/> then tell me what I'm doing wrong. :-)
[23:22] <devildante> you didn't do it in on_realize
[23:22] <devildante> see the PackageInfo code
[23:22] <mpt> oh
[23:24] <devildante> what do you think of http://imgur.com/uoJc1.png ?
[23:24] <mpt> devildante, sorry, got distracted. :-) That's great, except for the bold
[23:25] <mpt> so just minus the <b> and </b> I guess
[23:25] <devildante> mpt: done!
[23:26] <mpt> hurrah
[23:26] <devildante> do you want me to do the "choose add-ons" label bit?
[23:27] <mpt> devildante, yes please, I'm not smart enough
[23:27] <devildante> implying I'm smart :p
[23:27] <mpt> Just "Add-ons", no colon, same color as the "Total size:" etc
[23:27] <mpt> This is to make it seem less like they're something you *need* to twiddle
[23:29] <devildante> mpt: done! (and it looks nice)
[23:30] <mpt> devildante, ok, propose for merging :-)
[23:30] <devildante> mpt: you've run out of ideas, I guess :p
[23:34] <mpt> devildante, ha, I have some ideas for Natty ... For now, though, I just want it to look beautiful
[23:34] <mpt> There are all sorts of nice things we could do with animations, to make the screen changes less jarring
[23:35] <devildante> bling-bling :p
[23:35] <devildante> mpt: https://code.edge.launchpad.net/~ilidrissi.amine/software-center/addons-tweaking/+merge/33706
[23:35] <mpt> devildante, ah, one more I forgot: there's horizontal scrolling when the add-on labels are wider than the window.
[23:36] <devildante> mpt: any examples?
[23:36]  * devildante facepalm, just unmaximize it :p
[23:36] <mpt> devildante, Firefox
[23:36] <mpt> yeah
[23:38] <mpt> devildante, probably the simplest way to fix it is to allow the labels to wrap to multiple lines.
[23:39] <devildante> mpt: line wrap won't work
[23:39] <mpt> A more complicated way is to ellipsize the end of the add-on title, *without* altering the appearance of the package name after it.
[23:39] <devildante> mpt, because I separated the summary text from the pkgname into their own labels to apply the coloring only for the pkgname
[23:43] <mpt> devildante, so, let me amuse you with some pseudocode
[23:43] <mpt> if width(title) + width(package name) <= width available for label:
[23:43] <mpt>     lay out as normal
[23:43] <mpt> else:
[23:43] <mpt>     lay out package name up against the trailing edge of the pane
[23:43] <mpt>     ellipsize the package name in the remaining space
[23:44] <mpt> ("trailing" = "right" in LTR languages, "left" in RTL)
[23:45] <mpt> It's nearly midnight, so I'm going home, and I'll leave you with that puzzle :-)
[23:45] <mpt> thanks again for your work
[23:45] <devildante> mpt: wait a sec
[23:45] <mpt> hm?
[23:45] <devildante> mpt: I found a solution , lemme take a screenshot
[23:46] <devildante> mpt: is this acceptable: http://imgur.com/lqhrv.png
[23:47] <mpt> devildante, it's not ideal, but it's better than what we have now
[23:48] <devildante> mpt: I'll try to work out another solution
[23:48] <devildante> mpt: but thanks for your patience :)
[23:50] <mpt> Left edge of pane -> | ☑ 〠 Here's a really long add-on name that goes on and o... (really-long-addon) | <- Right edge of pane
[23:50]  * mpt really going home now
[23:51] <devildante> bye :)
[23:57] <devildante> bye?