/srv/irclogs.ubuntu.com/2013/01/29/#ubuntu-unity.txt

hyperairhmm, compiz needs the equivalent of emacs's C-h k01:01
=== Chucrute301 is now known as ChucruteROOT
=== ChucruteROOT is now known as Chucrute301
rperierHi all, bug 1019457 is still pending for PS Jenkins Bot, it seems not uploaded yet in ubuntu. That's normal ?08:43
ubot5bug 1019457 in unity (Ubuntu) "The Dash closes when trying to switch to the Command lens (Alt+F2)" [Medium,In progress] https://launchpad.net/bugs/101945708:43
sil2100rperier: hi! Let me check that08:49
sil2100rperier: I think it's already in Ubuntu08:50
sil2100rperier: since the version that includes your fix is 6.12.0daily13.01.25.1-0ubuntu1, and from what I remember that's the one in raring08:50
rperierI did not see the changelog in raring, I probably missed it :\08:51
jibelrperier, it will be in unity 6.12.0daily13.01.29-0ubuntu1 but automated tests failed and it didn't land into the distro08:57
rperierjibel: automated tests failed ? it has something to do with the test case I added for my fix or not ?08:59
sil2100rperier: I think it was due to a doc-string tab-character instead of space09:00
sil2100rperier: in the autopilot test - but we fixed that09:00
rperiersil2100: oh, I saw the commit yeah !09:01
sil2100jibel: could you also make sure that https://code.launchpad.net/~sil2100/unity/bump_compiz_dep_abi_break/+merge/145330 gets reviewed and included before the release?09:01
jibelrperier, no it's unrelated to your fix. autopilot run failed this morning with a "unity not installed" error, someone from the unity team have to look into this09:01
sil2100jibel: it might be due to the ABI bump09:01
rperierno because I don't like commit crap or broken tests :)09:02
sil2100jibel: since there was an ABI break in compiz, we bumped the ABI version for compiz, and now I also bumped the compiz-dev build-dep in unity09:02
sil2100jibel: so, unity might have failed installing because compiz was offering a new ABI version, while unity was still build with the old ABI - therefore installing compiz removed unity09:03
sil2100jibel: but if we get this merge in, we should be fine09:03
sil2100Just compiz needs a release as well09:04
jibelsil2100, yes, very likely, I think mterry does this kind of review while didrocks is away, or is there anyone else?09:04
sil2100jibel: yes, mterry usually does it, but it'll be some time till he appears - maybe Mirv could take a look right now instead? ^09:05
sil2100Mirv: ^09:05
rperierbtw, yesterday it was impossible for me to activate the warnings for unity.dash.view even if I export UNITY_LOG_SEVERITY="unity=warning;unity.dash.view=warning"  (I just try to understand what I do, so I add logs locally into my code). This is for unity-R on raring09:08
Mirvsure I can check and approve that one09:09
rperierhowever there's no hurry, I will retry today in the afternoon09:10
Mirvapproved09:12
Mirv(the sil2100's branch, that is)09:12
sil2100Mirv: thanks!09:22
=== dandrader is now known as dandrader|afk
=== dandrader|afk is now known as dandrader
ryeuhm, i think something went wrong after fixing the maximize/restore button - now one can't click on indicators when dash overlay is open10:37
ryebug #1101310 i mean10:37
ubot5bug 1101310 in unity (Ubuntu) "[staging][raring] Maximise/Restore button depends on the launcher icon size." [High,In progress] https://launchpad.net/bugs/110131010:37
luvrye: hi, I got raring installed in VM, I will make sure it compiles fine tonight - but I can't test because compiz is unusably slow in my VirtualBox - so I will send you a link to the patch/lp-branch tomorrow10:37
ryeluv: sure, compiz is slow because you are using llvmpipe rendering in virtualbox - you will want to install guest additions to get hw acceleration working10:38
luvcan i get guest additions for raring?10:43
luvoh and can i install them without gui? it's that slow ....10:44
ryeluv: looking for info..10:49
ryeluv: http://askubuntu.com/questions/207813/why-does-an-ubuntu-12-10-guest-in-virtualbox-run-very-very-slowly/214968#214968 ?10:50
* rye does not use virtualbox10:50
ryefor now10:50
luvthanks, i will have a look tonight10:52
popeysil2100: i have no desktop appearing on my desktop bare metal running staging..10:53
popeysil2100: http://paste.ubuntu.com/158546010:54
sil2100It doesn't look like staging to me10:55
sil2100It looks more like the daily PPA10:56
popeyoh balls10:56
sil2100popey: but anyway, let's maybe wait for one merge to go into unity10:56
* popey downgrades and starts again10:57
sil2100https://code.launchpad.net/~sil2100/unity/bump_compiz_dep_abi_break/+merge/145330 <- this one, I hope it'll get merged soon10:57
sil2100A bit strange, since it's waiting for merge since an hour already10:57
popeyok, staging in bare metal and staging in vm are both okay.. phew!11:03
sil2100Phew! So no VM problems with compiz?11:06
sil2100mmrazik: is everything ok with the merger right now?11:08
mmraziksil2100: didn't see any issues so far11:09
sil2100mmrazik: since I didn't see any new merges getting in and https://code.launchpad.net/~sil2100/unity/bump_compiz_dep_abi_break/+merge/145330 is approved for 1 hour already11:09
sil2100Is it busy?11:09
mmrazikno, not really. looking into ti.11:09
sil2100Thank you :)11:10
mmraziksil2100: its landing11:10
mmrazikas far as I can see the tests already finished11:10
mmrazikits just being merged11:10
mmraziksil2100: well.. it is merged11:10
sil2100mmrazik: excellent, phew! Thought something was wrong - thanks11:11
=== _salem is now known as salem_
=== mmrazik is now known as mmrazik|lunch
=== mmrazik|lunch is now known as mmrazik
rperierwell, in fact logs work fine :)12:24
rperierI probably did some craps yesterday ;)12:24
=== mpt_ is now known as mpt
rperierI found a fix for switching from command.lens to home.lens. Basically, when the DashView receive an ACTIVATE_REQUEST event (dash/DashView.cpp:OnActivateRequest() ), if this one is of type NOT_HANDLED AND the dash is visible AND the new len != the current one, instead of closing the dash we need to switch to this new len using lens_bar->Activate(id)12:43
rperierand I don't know if you're agree for remember the current len...12:44
rperierI probably need to open a bug first, then I explain the fix and then I attach a patch or I link a branch... this is probably a better approach12:47
bregmarperier, yes, that's the preferred approach13:04
mterryfginther, today's daily build of unity says "UNITY_NOT_INSTALLED"13:11
mterry(or rather, it has a file in the artifacts called that)13:11
jibelmterry, it was due to the ABI change that sil2100 fixed this morning. unity was not installable13:14
mterryjibel, ah makes sense13:14
sil2100mterry: hi, yes, I fixed it more or less I think ;)13:15
=== dandrader is now known as dandrader|afk
mterryOK.  Let me try rebuilding unity then13:16
sil2100We need new compiz and unity released at the same time13:19
rperierbregma: I will open a bug then :)13:20
jibelthey will because they are part of the same stack13:21
MCR1...volunteer for approving MPs needed...13:28
MCR1very easy (1 minute): https://code.launchpad.net/~mc-return/compiz/compiz.merge-fix1101641-use-snprintf-instead-of-sprintf/+merge/14522813:29
MCR1easy (1 minute): https://code.launchpad.net/~mc-return/compiz/compiz.merge-fix1101549-and-fix1101561-missing-break/+merge/14516113:29
MCR1advanced (5 minutes): https://code.launchpad.net/~mc-return/compiz/compiz.merge-fix1106270-showmouse-plugin-needs-port-to-GLES/+merge/14506913:30
MCR1advanced (5 minutes): https://code.launchpad.net/~mc-return/compiz/compiz.merge-fix1099100-thumbnail-title-text-issues.0/+merge/14495413:30
MCR1 all of those are already approved and tested, but still need a final green light...13:33
GrottenolmHi there, I lost something on my unity desktop: If I minimize XChat to the tray ... where can I find it afterwards? Any helpful hint will be appreciated.13:44
bregmaGrottenolm, is the xchat icon still in the Launcher on the left?14:12
Grottenolmbregma, no, it is not14:13
bregmaurgh14:14
Grottenolmbut the program is running14:14
GrottenolmI can't log in again with the same nickname14:14
MCR1JohnLea: Are you here ?14:16
ryebregma: i suppose xchat is using old style status icon and you may want to whitelist it14:22
MirvMCR1: went through 3/4, leaving the last one for someone else14:26
MCR1Mirv: Oh, thx a lot :)14:28
MCR1Mirv: Once they are merged I can start to fix conflicts on their follow-ups ;)14:28
=== francisco is now known as Guest38765
Mirvnp :)14:29
=== dandrader|afk is now known as dandrader
rperierbregma: bug opened, this is the bug 110895615:06
ubot5bug 1108956 in Unity "The dash closes when trying to switch from Command lens to home lens" [Undecided,New] https://launchpad.net/bugs/110895615:06
bregmarperier, thanks15:06
sil2100oh, the other way around15:07
sil2100heh ;)15:07
davmor2hey guys is there a way round a system not sleeping or locking if the dash or an indicator are in use ie select one don't move the mouse away system does sleep/lock till the mouse if indicator is closed15:20
XetiusIs there a way to set an application to override system key bindings when it is active, and have the default system keybindings active when it's not?15:23
=== Jim_Lahey is now known as bobweaver
mterryfginther, I started a build 52 for unity-head earlier today from the web interface.  Now it's gone and not on the jenkins dash anymore.  :(15:32
fginthermterry, looking15:33
fginthermterry, something weird happend. I received email that the ps-unity-autopilot-release-testing job finished (unstable) but it's not on the server15:43
mterryfginther, I started this job from the web UI.  That might be an unusual path?  I think didrocks usually uses his script15:43
sil2100I sometimes use the web panel as well, so hm15:44
fginthermterry, that shouldn't be a problem.15:44
mterrynm then  :-)15:44
mterryfginther, shall I start a new one, and see what number it gives it / if the same thing happens?15:45
fginthermterry, I saw some other odd behavior, which looked like jenkins was restarting. But I can't find evidence either way15:45
mterryOK, well.  I guess I'll try again15:46
fginthermterry, yes, might as well get some good test results :)15:46
mterryfginther, yup, gave it #5315:46
fginther#52 must be a ghost15:47
sil2100Ghost builds, brr15:48
rperieris it an acceptable proposal to put LensBar::GetActiveLensId() public ? I need it for my fix (see the previous bug that I pasted below)15:54
rperiers/below/above/15:54
JohnLeaMCR1; hyia, I'm here now ;-)16:04
MCR1JohnLea: Hey, just wanted to ask you if you could once again okay this: bug 106916516:07
ubot5bug 1069165 in Unity "Unity Standard Shortcuts: Make it easier for Windows users" [Undecided,Incomplete] https://launchpad.net/bugs/106916516:07
MCR1I have solutions for both in mind, but have to discuss it with other devs, because it needs a few changes16:08
MCR1JohnLea: For the Alt+Enter Fullscreen/Unfullscreen keys we have a lightweight Compiz plugin to do it16:09
JohnLeaMCR1; also when thinking about keyboard shortcuts take a look at https://docs.google.com/a/canonical.com/document/d/1jqeKtIJwqLtl58Wk_fqjr9Rrgxn9zsouCYOo-cZsLSE/edit16:10
MCR1JohnLea: Very interesting.16:12
MCR1JohnLea: One big problem is the "Restores or minimises" shortcut -> this is not an easy thing to accomplish16:13
JohnLeaMCR1; if  1) a Windows key combination does not clash with an existing Ubuntu key combination and 2) adding the key combination does not remove any existing Ubuntu keyboard shortcut ...   then I think we should add it.  I don't think we should add all of these to the keyboard shortcut overlay because we have only limited space there.  If you could open up bugs to for each of the new key combos you would like to change, and in the bug16:14
JohnLea report say "This keyboard shortcut does not clash with any existing Ubuntu shortcut, and is being added to help windows users transition to Ubuntu.  Also adding this shortcut does not removing any existing shortcut"  (or something along those lines)  I'll ack the bug reports.16:14
JohnLeaMCR1; yes, I know that one is some work to do...16:14
MCR1JohnLea: I am for removing "minimises" and make a own shortcut to "minimise" until the groundstructure to fix it properly is laid16:15
JohnLeaMCR1; also see all the bugs listed at the bottom of the doc, several are still open and would be good ones to fix16:15
MCR1JohnLea: because otherwise people see it as a bug16:15
MCR1JohnLea: Yes, thanks a lot for the document. I have saved it ;)16:15
JohnLeaMCR1; yes, also good luck fixing the restore/minimise bug! ;-)16:16
MCR1JohnLea: So we should either just remove the words "and minimises" or remove those words and add another shortcut to minimise a window16:16
JohnLeaMCR1; let's not add another shortcut atm, and focus on fixing the bug16:17
JohnLeabut we can change the text until it is fixed16:17
MCR1JohnLea: ok, that is what I meant16:17
JohnLeaMCR1; cool ;-)16:17
MCR1JohnLea: because the fix is really complicated16:17
JohnLeaMCR1; I know, we have fixed many of the easier bug already, a lot of the bugs we have now are the hard ones.  If this was easy to do it would prob. be fixed by now!16:19
MCR1(I do not want to nerve you with details)16:19
JohnLeaMCR1; I'll let you bother Sam about that ;-)16:19
MCR1ok ;)16:19
=== dandrader is now known as dandrader|afk
ricotzTrevinho, hi16:52
Trevinhoricotz: hey16:56
ricotzTrevinho, please let me know if you figured out the "Unable to find default local directory monitor type" problem ;)16:57
ricotzmeaning this one https://jenkins.qa.ubuntu.com/job/bamf-mbs-ci/build=pbuilder,distribution=quantal,flavor=amd64/37/console16:57
Trevinhoricotz: I've noticed that...16:57
Trevinhoricotz: but I don't know what happens, it seems pretty random16:57
ricotzTrevinho, i got reports for other apps even a mono-app failing on that one16:58
Trevinhoricotz: I didn't go into the glib internals, however16:58
ricotzTrevinho, weirdly this happened on precise too16:58
Trevinhoricotz: of coure I can calm-down the builder (by hiding the warning), but don't see why this happens...16:59
ricotzTrevinho, yeah, but it seems to be some real problem17:00
Trevinhoricotz: all happens here http://codesearch.debian.net/show?file=glib2.0_2.33.12%2Breally2.32.4-3%2Fgio%2Fglocaldirectorymonitor.c&line=287&numfiles=3182#L28717:00
ricotzcan't reproduce it myself though17:00
=== dandrader|afk is now known as dandrader
MCR1Haha - I think I've found the best (and proper) solution for bug 966099 \o/17:10
ubot5bug 966099 in unity (Ubuntu) "Keyboard shortcut overlay says Ctrl+Super+Down "minimises" the current window, but it doesn't" [High,Triaged] https://launchpad.net/bugs/96609917:10
MCR1andyrock, bregma, bschaefer, Trevinho: ^^ http://pastebin.com/K0ZzRLyZ17:13
MCR1I am proud I found a easy solution17:13
andyrockMCR1, do you have a branch?17:14
MCR1not yet17:14
andyrockMCR1, ping me when it's ready17:14
bschaeferMCR1, can you do a bzr diff > x and paste bin that? Or a branch would be nice17:14
MCR1but in a moment (gotta eat somethin' before)17:14
TrevinhoMCR1: nice! ;)17:14
MCR1bschaefer: Sure, I'll make a branch17:14
bschaeferMCR1, and nice, go eat17:14
bschaefer:)17:14
MCR1andyrock, bschaefer, Trevinho: Thx (& will do ;))17:15
andyrockMCR1, btw my overlay does not say "ctrl+super+down"17:15
MCR1what does it say ?17:15
MCR1if it is just the shortcut itself, which is different - it does not matter17:16
seb128so, does anyone know why libdbusmenu is not daily landing in raring?17:38
seb128mterry, ^?17:38
seb128or is that a didrocks secret?17:38
mterryseb128, I don't know how to tell yet if the "auto push" button is on or not.  But it seems OK according to the autopilot tests17:39
mterryseb128, unity stack isn't OK.  But libdbusmenu isn't part of that17:39
seb128ok17:39
mterryseb128, so didrocks secret I guess  :-/17:39
seb128I wonder if we should just manually backport the 2 leak fixes from chrisccoulson17:39
seb128and wait for didrocks then17:40
mterryseb128, we can manually push....17:40
seb128mterry, do you know how to do that?17:40
mterryseb128, oh wait.  No.  We thought we could, but there's something missing that we have to investigate to fix.  We need a credentials file that we have to create.  Let me see17:40
seb128mterry, I just want those leaks fixed in raring so people don't get confused by then in their new valgrinds logs17:41
jibelseb128, didrocks disabled autolanding of libdbusmenu on Jan. 15th but the commit message doesn't say the reason: "  disable dbusmenu and better wording for publishing not happening"17:42
seb128jibel, at the time the tests were still broken, cyphermox fixed them since17:42
mterryjibel, seb128: it was already re-enabled17:44
mterryjibel, maybe that bzr commit wasn't propogated to the jenkins?17:44
mterryjibel, does jenkins have commit 197?17:46
jibelmterry, right, didier reenabled it a week ago but last run was 18 days ago. He apparently didn't republish the stack.17:47
jibelmterry, do you know how to do it or should I do it?17:47
mterryjibel, I don't know how to publish from web UI.  And I need credentials file for script (working on it).  So if you could do it, it would be swell17:49
ryebschaefer: hi, re bug #1101310 - it looks like indicators are no longer clickable when dash is open, not sure whether they were intended to be clickable though17:49
ubot5bug 1101310 in unity (Ubuntu) "[staging][raring] Dash Maximise/Restore button depends on the launcher icon size." [High,Fix committed] https://launchpad.net/bugs/110131017:49
bschaeferrye, hmm well, that could be why the dash was designed to close when clicking on the panel17:50
jibelmterry, ok, I have to run now but I'll do that this evening before next run of the indicators.17:50
bschaeferor how it was...shoot17:50
* bschaefer has broken indicators atm17:50
seb128jibel, thanks17:51
bschaeferrye, let me fix that17:52
seb128cyphermox, https://launchpadlibrarian.net/129872662/buildlog_ubuntu-raring-amd64.libdbusmenu_12.10.3-0ubuntu1%2Bbzr440%2Bpkg0~raring1_FAILEDTOBUILD.txt.gz18:03
seb128cyphermox, one test failed there ... not sure if that's concerning the daily landing though18:04
cyphermoxhuh, right18:04
cyphermoxyeah, it would18:04
cyphermoxwhere was that?18:04
cyphermoxcharles_: ^^ did you end up figuring out what's up with that glib-events test?18:04
charles_cyphermox, I think so. Also I've got the libappindicator fix for you, just getting the bug tickets all up-to-date before I MP18:06
cyphermoxok18:07
ryebschaefer: hud still works with indicators (closes when they are clicked). But in unity2d indicator menus could pop-up over opened overlay so I don't know what's the proper way18:07
bschaeferrye, yeah, ill talk with design18:07
bschaeferrye, I removed that, because I thought that was causing the problem at first, and left it in18:08
bschaeferrye, its an easy fix, if its needed18:08
bschaeferrye, and actually the dash just closed when clicking on the panel, the Hud can handle things much nicer :)18:08
bschaefer(as it doesn't eat up the panel height)18:08
ryebschaefer: erm.. why does dash eat the panel height? for min/max/close buttons?18:11
bschaeferrye, for the dash dash preview18:11
bschaeferrye, as the ghost image of the preview is drawn over the panel, and when the launcher is resized the dash is over top the window buttons18:12
ryebschaefer: oooh18:14
bschaeferyeeah, otherwise it would have been an easy fix by add the panel height to the dash view (which is how the Hud can handle things easily)18:16
ian__605#0  0x00007fbb283f74b3 in ccsGSettingsValueChanged () from /usr/lib/libcompizconfig_gsettings_backend.so18:29
ian__605Can anyone give me some advice for debugging an app? A user gets inconsistent crashes with a SIGSEGV in ccsGSettingsValueChanged(). I can't reproduce it but it seems to happen every 7-8 program starts for him. Starting with gdb never reproduces it, but an apport bug report gives some info. seems that SegvReason is reading unknown VMA18:29
ian__605Sorry first post was the beginning of the stacktrace. Problem seems to come from compiz-gnome package18:29
fginthermterry, have you seen the ps-unity-autopilot-release-testing results? There is a noticable uptick in failed tests18:31
mterryfginther, hadn't looked yet.  let me see18:32
mterryfginther, they seem to be all a similar problem.  sil2100 , you around?  A lot of autopilot errors like:18:35
mterryTraceback (most recent call last):18:35
mterry  File "/usr/lib/python2.7/dist-packages/unity/tests/test_switcher.py", line 125, in test_switcher_move_next18:35
mterry    start = self.switcher.selection_index18:35
mterry  File "/usr/lib/python2.7/dist-packages/unity/emulators/switcher.py", line 72, in selection_index18:35
mterry    return self.controller.model.selection_index18:35
mterry  File "/usr/lib/python2.7/dist-packages/unity/emulators/switcher.py", line 240, in model18:35
mterry    i = self.get_children_by_type(SwitcherControllerImpl)[0]18:35
mterryIndexError: list index out of range18:35
mterry}}}18:35
fginthermterry, agreed. There are 28 failures due to that IndexError trace18:36
fgintherin the ati test18:36
sil2100Ok, so18:38
sil2100It looks like another problem with switcher refactoring18:38
sil2100mterry: what packages are those tests based on? What versions?18:38
fginthersil2100, unity is revision 308318:39
sil2100That's strange18:39
sil2100Need to check if they changed anything in the source again18:41
fginthersil2100, https://code.launchpad.net/~bregma/unity/refactor-switcher-controller-2/+merge/145025 removed switchercontrollerImpl from introspection18:42
bregmacripes18:43
sil2100Ah, eh18:43
sil2100fginther: just found the same thing18:43
* sil2100 sighs18:43
sil2100Again need to modify the tests to fit the new source changes18:43
sil2100Eh eh18:45
bregmasorry about that18:45
bobweaverHello there should I fix bugs that are in unity 2d ?  or is it a waste of time ? No I do not want to program NUX and C++ , I just like qml18:45
sil2100fginther: can I deal with it tomorrow?18:45
bregmathe code under test got modified to fix the tests and the tests were modified to fit the hidden internals of the code again and there was a mid-air collision18:46
sil2100mterry, fginther: will you guys have anything against the fix waiting for tomorrow morning my time?18:46
bregmabobweaver, you can propose fixes to unity-2d, they'll require SRUs to get them picked up in 12.04 or 12.10 though18:47
mterrysil2100, no that's fine18:47
sil2100mterry, fginther: thanks! See you tomorrow then ;)18:47
fginthersil2100, later18:47
bobweaverthanks bregma18:51
rperierthis patch looks good http://bazaar.launchpad.net/~rperier/unity/command-lens-switching/revision/3083 ?18:59
rperier(I am not sure at all about the autopilot test :\ )19:00
rperierI will propose a merge a link the corresponding bug to it19:04
MCR1andyrock: Ping :) https://code.launchpad.net/~mc-return/compiz/compiz.merge-fix966099-add-unmaximize-or-minimize-window-key/+merge/14546419:20
MCR1andyrock: The Unity overlay shortcut also needs adjustment... but functionality is provided by Compiz19:20
MCR1*Unity Help Overlay->change unmaximize_window_key to unmaximize_or_minimize_window_key19:21
rperierwhen you have time, someone could confirm bug 1108956 ?19:24
ubot5bug 1108956 in Unity "The dash closes when trying to switch from Command lens to home lens" [Undecided,New] https://launchpad.net/bugs/110895619:24
bschaeferrperier, hmm im not sure if thats a bug, for example if you do a Super + a, then press super the dash closes as wel19:26
bschaeferrperier, though, just to be safe, ill assign the design team to take a look at it19:27
MCR1the problem is (once again) -> too many functionality on one key :)19:27
MCR1*functionalities19:27
MCR1bschaefer: Do you have time to check the branch above ?19:29
MCR1It just adds a shortcut, which will act intelligently and restore a maximized window, while also being able to minimize a non-fullscreen window...19:31
MCR1*non-maximized19:31
MCR1maximized != fullscreen19:31
bschaeferMCR1, hmm sure, ill have to compile a bunch of things so it'll take a bit19:34
MCR1no stress...19:34
andyrockMCR1, first of all you need to fix the indentation19:34
MCR1ok19:34
MCR1andyrock: Should be fixed - well, the CompScreenImpl::* all have wrong indentation, but I've not fixed the others, just the newly added stuff...19:42
rperierbschaefer: so that's a bug when switching from the dash to command lens but not from the command lens to the dash ?19:42
rperiermhh19:42
rperierlet check the design team yeah19:43
bschaeferrperier, i assigned the design team, as they have the real say in it19:43
rperieryeah I saw19:43
bschaeferbut you would think going from Super + a -> Super should go to home dash, if your bug is a bug19:43
bschaeferunless they are both bugs19:43
rperierwhat Super+a does exactly ? (I am not on unity right now)19:44
bschaeferit goes to the application lens19:45
bschaefersuper+m, the music, super+f the file lens19:45
rperieruseful shortcuts :)19:48
MCR1andyrock: Here the corresponding Unity MP: https://code.launchpad.net/~mc-return/unity/unity.merge-fix966099-shortcut-fails-to-minimize-just-restores/+merge/14547419:59
andyrockMCR1, adding the review to my TODO list20:00
andyrockthanks ;)20:00
MCR1thx20:00
MCR1&& ofc I hope your TODO list is small, but I guess it's not20:01
MCR1;)20:01
MCR1:-X20:02
andyrockMCR1, well reviews have the priority ;)20:03
MCR1\o/20:03
* bschaefer is almost done compiling everything...20:03
* MCR1 also needs some double xeon machines ;)20:04
bschaeferMCR1, hmm sometimes when I do that shortcut the window goes to a different workspace...20:21
MCR1bschaefer: strange - can you define sometimes ? is it generally working ?20:24
bschaeferMCR1, well it seems to only do it after a compiz --replace20:24
MCR1hmm20:24
MCR1if you look at the compiz code -20:25
bschaeferwhich makes me think the windows don't update their workspace when forced onto the first one20:25
MCR1I've just merged the functions minimize and restore to one - so it should not be able to do that...20:25
bschaeferMCR1, also, I don't think its always guaranteed that the window on top is the focused window20:26
bschaeferthough it does seem to ignore the correct ontop windows20:27
bschaeferMCR1, also why did you move those functions together? Usually when a function name as an Or in it, it should be split up20:30
bschaeferSRP20:30
=== andyrock is now known as andyrock|window
=== andyrock|window is now known as andyrock|dinner
* bschaefer goes to eat lunch20:36
MCR1I moved them together to make it a core functionality as CCSM and thus Compiz rebell if two different actions are mapped to one shortcut20:38
MCR1I made it act intelligent, depending on the state of the window and mapped that combined-smart-function on one key to avoid CCSM from rebelling20:40
MCR1To not change any users' shortcuts it is a new shortcut and name now20:41
MCR1I think it is the most intelligent way to fix the bug, otherwise CCSM would probably have to be rewritten, now we simply got a smart shortcut... :)20:43
bschaeferMCR1, hmm well what I mean is there is a shortcut for Ctrl+Super+Up and a different one for Ctrl+Super+Down21:07
bschaeferMCR1, which each one should have a different callback function21:07
MCR1bschaefer: Have you seen the unity MP and the bug report ?21:08
MCR1The problem is that Ctrl+Super+Down should restore a maximized window, but also minimize a non-maximized window21:09
bschaeferMCR1, hmm I just saw the compiz branch, and i haven't looked at the bug report :(21:09
MCR1ah okay - then everything is clear now :)21:10
bschaeferMCR1, o hmm, well Ctrl+Super+Down doesn't minimize a window for me :(21:12
MCR1please open CCSM->General21:12
MCR1there (in Key bindings tab) you should find the new shortcut21:13
MCR1try to change it and see if it works 100% then21:13
MCR1"Unmaximize or Minimize Window"21:14
* bschaefer looks21:14
MCR1if you change it CCSM will warn if it collides with something else21:14
MCR1so we can exclude that21:15
bschaeferMCR1, oo i see, its Alt+F5 for me...21:15
jibelmterry, FYI, i reconfigured the indicator stack and verified the libdbusmenu is enabled.21:15
bschaeferMCR1, hmm well i just don't like having an OR in a name of a function :(21:15
MCR1probably it is some ubuntu patch that is missing then (which is changing shortcut keys) -> gotta find that -> I kept the default the same (did not even look what it was)21:17
MCR1the Alt+F5 problem21:18
MCR1I will have to change that to "Ctrl+Super+Down" ? then and remove "Ctrl+Super+Down" from the ubuntu distro patch also...21:19
bschaeferMCR1, well thats if its what is wanted, why don't we just have a toggle bool for having both?21:20
MCR1The Or in the function is somehow needed to describe what it does as it will Restore OR Minimize, depending on the state of the window21:20
bschaeferMCR1, 67privateScreen.optionSetUnmaximizeWindowKeyInitiate (CompScreenImpl::unmaximizeWin);21:20
bschaeferdoes that function do the same thing that your function does?21:21
bschaeferminus the21:21
MCR1ups, typo21:21
MCR1sry21:21
bschaeferminimize?21:21
MCR1I'll fix it immediately21:21
bschaeferMCR1, wait, that function is already there...21:21
bschaeferMCR1, im just trying not repeat our selfs21:21
MCR1no it is not the same21:21
bschaeferMCR1, i mean does it do what half your function does...21:22
MCR1this will JUST restore21:22
bschaeferyes21:22
MCR1yes21:22
bschaeferwhich means that functions does half of what your does, which means we are repeating our selfs, which means we should step back and re think this21:22
MCR1I have re-thought this, but give me your idea21:22
bschaeferMCR1, so, instead of a function callback for what you want, can we have a toggle button to say minimize21:23
mterryjibel, thanks21:23
bschaeferMCR1, so we remove that other function that is there, and use your function21:23
bschaeferbut check that the toggle is true before minimizing?21:24
* bschaefer isn't sure how clear this is sounding21:24
bschaeferthis way, the ctrl+super+down wont have to change anywhere, and we just have to put in the ubuntu patch to enable this feature by default21:25
MCR1bschaefer: I have found room for optimization, I could check for (w) only once, but I do not understand your toggle thing as this function is not toggling any state21:27
bschaeferMCR1, the toggling would be in CCSM (a bool)21:28
MCR1bschaefer: Ah, -> toggle behaviour21:28
bschaeferMCR1, and you would remove the other unmaximizewin function, and only use your own21:28
bschaeferyes, this way we don't repeat our selfs here21:28
MCR1the shortcut would have the same problem then...21:28
bschaeferand this way there will only be 1 shortcut you can do for this, but enable the minimizing feature21:29
MCR1the help overlay hint21:29
bschaeferMCR1, but we are going to enable this by default in ubuntu21:29
MCR1but if you toggle it off -> the shortcut hint will be false again21:30
bschaeferI just don't like implementing a whole new shortcut just to have a nicer wording in a shortcut overlay21:30
bschaeferand then we repeat our selfs in the code21:30
MCR1we are also adding functionality21:30
bschaeferurg, dam you shortcut overlay21:30
bschaeferDRY (Dont repeat yourself)21:30
bschaeferhmm21:30
MCR1it is a new, more powerful shortcut, than the old one -> used as default now in Ubuntu -> while for old-time Compiz users and their shortcuts nothing changes21:32
MCR1another advantage21:32
bschaeferbut its doing 50% of the same thing...21:32
MCR1look at the others there -> all do 50% of some other function ;)21:33
MCR1but they are still not the same21:33
bschaeferthis decision shouldn't be made based on the the overlay shortcut21:33
bschaeferhmm21:33
bschaeferyeah, but they should reuse that function :)21:33
MCR1no, I mean actions.cpp21:33
MCR1this would bring another problem21:34
bschaeferyes?21:34
MCR1CCSM does not allow 2 functions on the same key21:34
MCR1It would start rebelling LOUD21:34
bschaeferMCR1, we would be dropping 1 function key with how im imaging it21:34
bschaeferimagining21:34
bschaeferMCR1, also the OR should be an AND or THEN21:35
bschaeferreally I like Then21:35
bschaeferbut hmm21:35
MCR1bschaefer, if you worry about that bit of code duplication please take another look at actions.cpp and the other cases21:36
bschaeferyeah, Im worried to now :)21:36
bschaeferMCR1, it isn't much of a duplication anyway21:36
MCR1but you are right -> we can discuss the name and also the logic could be simplified a bit21:36
bschaeferYes, im just trying to get the best readability,21:37
bschaeferso the problem with AND or THEN is it implies those happen at the same time so hmm21:37
bschaeferI personally don't like shortcuts that can do more then 1 thing21:38
MCR1yep, it is not easy to find a *short* name for it21:38
bschaefer(ie. the ShowDesktop has a problem with this as well)21:38
bschaeferMCR1, because a problem I see, is a semi max window, then your function will minimize it :(21:38
MCR1design wanted it that way (and IMHO makes sense in this case)21:39
bschaeferwhich some people will think hey, it should restore this!21:39
bschaefers/will/may21:39
MCR1yep that is true21:39
MCR1and adjustable21:39
bschaeferi just don't like overloading to much on short cuts *cought* *cough* super21:39
bschaeferbut can we check if its semi maxed?21:40
MCR1yep, hehe21:40
MCR1yep also21:40
MCR1the check is easy21:40
bschaefercool, well if we add that check in, that should reduce confusion, from there it doesn't look to bad21:40
bschaeferits a small readable function21:40
MCR1but this was not part of the original bug and I did not want to change too much21:40
bschaeferwell to me maxed and semi max share the same state of being maxed21:41
bschaeferMCR1, we can open a new bug about that21:41
MCR1yeah, I also think we should restore them then...21:41
bschaeferand get design to have input on it21:41
MCR1yeah probably we should21:41
MCR1:)21:41
bschaeferyup :), you can do it in a bit, so lets look at the function21:42
bschaeferone thing, is you check if (w) twice21:42
MCR1yep, noticed that21:42
bschaeferlets just check it once, then check which state it is21:42
bschaeferif (w) { if (max) else if (min) }21:42
MCR1yep - it is done that way in the toggle functions as well21:42
MCR1good point21:42
* MCR1 is fixing21:43
MCR1I will have to find the Ubuntu patch that modifies the shortcuts also21:43
MCR1and remove the standard there21:44
bschaeferMCR1, and once you find the ubuntu patch, disable the old shortcut, and enable your new one with Ctrl+Super+Down21:44
bschaeferyeah21:44
MCR1yep21:44
bschaeferim not 100% sure where that is21:44
MCR1hehe21:44
MCR1I'll find it21:44
MCR1once I fixed all of that I'll ping you21:44
bschaeferalright :), i've seen it somewhere...it might be in a different branch21:44
bschaeferalright :)21:44
MCR1urgh - could you remove the shortcut then ?21:45
* MCR1 hates those patches21:45
bschaeferhmm ill talk with duflu when he gets on21:45
bschaeferas he'll have to review it anyway21:46
MCR1I think I know where those are...21:46
bschaeferif its in a different branch21:46
MCR1yeah21:46
* MCR1 is repeating himself: ***MCR1 hates those patches21:47
bschaeferyeeah, they are like dark magic ...21:48
MCR1*dark magic that tends to break and introduce bugs no coder ever finds the reason for... :P21:49
bschaeferyeah, well its hard to find a bug in the patches when you don't know the patches exists :)21:51
* bschaefer spent a day doing just that21:51
MCR1I still have not succeeded in convincing the merger to accept my branch, because of some magic patch failures :P21:52
bregmamore.  dark.  magic.21:55
bschaeferbregma, its everywhere21:56
bregmathere wasn't a witch or wizard who went bad that hadn't patched compiz21:56
MCR1haha21:57
MCR1In reality we need a lot more wizards hacking on Compiz ;)21:58
MCR1bschaefer: Thinking about it - we probably should implement the same functionality for semi-maximized windows as well -> first restore -> then minimize22:03
MCR1bschaefer: Otherwise it would be inconsistent22:03
bschaeferMCR1, yeah, throw it in22:03
MCR1ok22:03
bschaeferit makes more sense to me22:04
MCR1yes22:04
MCR1good point (again)22:04
MCR1bschaefer: Found a new bug - haha22:11
bschaeferMCR1, sweet, what is it?22:11
MCR1bschaefer: Currently restoring semi-maximized windows does not work via shortcut22:12
MCR1it will first maximize22:12
MCR1and then restore22:12
MCR1as it is not recognized as maximized22:12
bschaeferMCR1, hmm it should though, because when you do a semi maxed window, then drag it, it restores22:12
bschaeferMCR1, hmm possibly look at how the draging from a semi maxed window works22:12
MCR1yeah - and fortunately for us the correct coordinates are saved22:13
bschaeferMCR1, I noticed, when doing a max, then doing a unmax, it doesn't restore it to its real state22:14
=== salem_ is now known as _salem
bschaeferMCR1, eek, and try this with your branch22:15
bschaeferMCR1, drag the window to a maxed state, then drag it away22:15
bschaeferit goes to the correct window state, then goes to the larger state22:15
MCR1toggle win maximized works here22:16
MCR1maximize->unmaximize as well22:17
bschaeferwell im talking about draging it22:17
bschaeferit might not be your branch, but im noticing something odd with restoring windows22:18
MCR1grid needs more fixes22:18
bschaeferyeah, wasn't the grid fixed not to long ago?22:18
MCR1it is still in a bad state22:18
* bschaefer just implemented a feature in there and this wasn't the case22:18
MCR1grid also works22:19
MCR1x and y size restore correctly with grid22:20
mhall119http://www.iloveubuntu.net/ubuntu-1304s-unity-demoed-and-available-installation-archlinux22:20
bschaefermhall119, awesome!22:20
bschaeferMCR1, hmm well ill make a video22:21
bschaefer /me ubuntu ones isn't working22:34
bschaefermeh22:34
bschaeferMCR1, umm so yeah, something for me is messed up with draging a maxed window to get it to restore22:35
bschaeferit flashes from its correct state it was before being maxed, to the size of the maxed window wiht out being maxed22:35
bschaeferMCR1, i sent you a video...22:36
MCR1bschaefer: I saw, but I do not know what to tell you... looks ugly22:40
MCR1looks like grid weirdness22:41
bschaeferMCR1, o well, i just didn't think you could repro it haha22:41
bschaeferyeah22:41
* bschaefer isn't blaming your branch22:41
MCR1:)22:41
MCR1bschaefer: Everything ready (except the patching the ubuntu-config.patch in ubuntu/compiz/debian/patches)23:28
bschaeferMCR1, cool, Ill review the branch and talk to duflu when he gets around23:29
MCR1bschaefer: TBH, I do not know how to hack that patch without a bad hack ;)23:29
bschaefer:(23:30
bschaeferwell Ill talk with duflu about it, hopefully he'll have some experience with it23:30
MCR1yes - he has for sure ;)23:30
MCR1unmaximize_window_key is in the file ubuntu-config.patch -> simply removing that manipulation of this key from the patch should be enough to make everything work smoothly *hopefully*23:33
bschaeferMCR1, well you'll have to make sure your new shortcut is disabled by default for compiz atm23:33
bschaefer(trunk that is)23:33
bschaeferand for the patch you'll have to make sure its pointed to Ctrl+Super+Down, because you can't change the default behaviour of compiz trunk23:34
MCR1I did not change default, because it was not there before ;)23:34
MCR1but I get your point23:34
MCR1so I'll clear it for now, and it will have to be added via this patch black magic later23:35
bschaeferwell if the default behaviour is it isn't there, then it shouldn't be enabled :)23:35
MCR1yep, probably - it might collide with someone's other shortcut23:36
bschaeferdang, and you broke the ABI, now ill have to rebuild unity haha23:36
MCR1I wanted to ask if this will break the ABI...23:36
MCR1:-[23:36
bschaeferumm yes it will, but and you'll have to bump the ABI number23:36
MCR1ok, ack23:37
bschaeferi was mostly joking because i had built your branch, but some fixing you did changed it, so ill have to rebuild :)23:37
bschaeferhmm well actually I need to look at it some more, if you are just adding new functions then it might not23:37
MCR1ok - could you please test functionality again (with grid disabled) ?23:42
MCR1for the grid stuff I did some investigation also23:42
MCR1problem seems to be that core does not know about grid resized windows, just grid seems to do...23:43
MCR1but it needs further investigation23:43
bschaeferwell if thats the problem, then for now we should just leave of the semi, and file a bug about it23:43
MCR1no, no problem with this branch as it is completely unrelated23:43
MCR1I just looked at grid if those windows also get a flag set23:44
MCR1and it seems they get one, but not globally (it seems)23:44
MCR1but this is completely unrelated to this branch23:45
MCR1but I would like to be able to restore grid windows via this shortcut also, which is currently not possible for those that are in the corners for example23:45
MCR1but - once again - this is unrelated to the shortcut, it is a problem that is reproducable in current trunk as well23:46
bschaeferalright23:48
MCR1for example: place a window via grid shortcut or mouse in a corner - now try to restore it with Ctrl+Super+Down -> does not work23:48
MCR1but will work with semi-maximized23:48
MCR1should work with all of them23:49
MCR1but that is another grid story...23:49
MCR1Does it work as intended ?23:50
bschaefersorry, im looking at something else23:51
bschaeferMCR1, yeah, i've noticed that with semi max23:52
bschaeferMCR1, like you were saying, it doesn't think its maxed, so it mins it23:52
MCR1yep23:52
MCR1semimax *should* be fixed now23:53
bschaefero nice, let me try that23:53
bschaeferMCR1, hmm you set your shortcut to ctrl+super+down?23:54
bschaeferas I have 2 shortcuts that are set to the same thing...23:54
MCR1in the latest version I've cleared the default, should be empty23:55
bschaeferits not for me, and we can't change the shortcut of another shortcut23:55
MCR1I have removed it23:56
bschaeferMCR1, we can't change the default behaviour...we have to change it in the patch23:57
bschaeferfor ubuntu only23:57
MCR1sure23:57
MCR1I know23:57
MCR1It is empty in my MP :)23:57
bschaeferMCR1, hmm why is it set to Ctrl+Super+Down in the branch I compiled then...odd23:58
MCR1look at the diff -> no shortcut changed, none added -> https://code.launchpad.net/~mc-return/compiz/compiz.merge-fix966099-add-unmaximize-or-minimize-window-key/+merge/14546423:58

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