[00:26] racarr, what's the input library called and where can I find it? [00:38] bregma: It lives in the mir tree. [00:38] bregma: 3rd_party/android-input [00:38] by itself it doesn't do a whole lot so we have the classes in mir::input (i.e. src/input, src/input/android) [00:38] to tie it together [03:09] RAOF: got a second? [03:10] Getting a second nexus 4 in the mail tomorrow so I will finally have a full time Ubuntu phone :D [03:10] also it will be white [03:10] I'm sorry I just come back in the evening and distract the southern hemisphere [03:12] working on getting platform-api branch proposable so ricmm can review it in his morning [03:12] racarr: you're doing things on the platform api project? [03:13] racarr: platform-api stuff probably relates to my cry-for-help email..? :) [03:13] when I get back from holiday I'm going to try and get you guys to take on the python bindings for said project... at the moment every time you guys change the API the entire testing stack on the phone breaks :-/ [03:13] racarr: Cool, white, easy to distinuish [03:13] +g [03:14] thomi: We've been very aware of the need to stabilize it for a long time. Sorry to say I changed the client API again last night. But such changes are becoming rarer [03:15] hey, that's OK... it just sucks that the python bindings are in a totally separate project. It makes them rather too invisible [03:15] I'd like to have them in the same project, building at the same time as trunk [03:17] thomi: duflu: Yes. There is a new version of the API, so I did new mir backends [03:17] in the process of removing TODOs and testing the various qtubuntu configurations (mirserver/mirclient desktop/phone) [03:18] Hopefully it should land with qtubuntu changes tomorrow or monday unless something horrible happens [03:18] and things will just work tm* [03:19] duflu: So the short answer on the email is if the old instructions don't work (you could still try building ~robertcarr/platform-api/mirserver and ~robertcarr/qtubuntu/mirserver then unity/phablet/integrate-mir) [03:19] probably just wait until monday [03:19] I guess there is no reason why that wouldn't work [03:20] racarr: So long as you're not *aware* of anything those branches are missing... [03:20] duflu: They should work, at least on desktop [03:20] platform API, you need to build with [03:21] -DPLATFORM_API_IMPLEMENTATION=mirserver [03:21] qtubuntu, the whole source tree doesn't build (just the qpa plugin) so you go to src/platforms/ [03:21] then qmake -o Makefile [03:21] make [03:22] my system seems to have problems with finding the platform api so at this point becaue qmake isn't looking in arch specific directories so I have symlinks XD [03:22] um but then that builds [03:22] there was a fix that landed a few weeks ago in mesa that this will require so make sure your mesa is recent [03:22] then to build unity you can [03:22] I think it's [03:23] ./build -s [03:23] racarr: Recent as in PPA? [03:23] then just go to a VT and ./run -m [03:23] yeah PPA is perfect [03:24] racarr: Copy, paste. I'll let you know how I went on Monday [03:24] Cool :) [03:25] im really hoping by the end of monday to have things in a state where it's as simple as add repository, run unity/build unity [03:25] and then guess what it's time for!! [03:25] FEATURES! [03:35] thomi: Yo. [03:36] RAOF: hey - I sent an email to the mir-devel list, which is what I was gonna ask you about anyway. tl;dr version: mir client library leaks FDs again [03:36] Woot! [03:36] yeah... not so much [03:39] So you'd like me to un Bork Bork Bork! the client library? :) [03:40] RAOF: Yes pleese-a feex zee cleeent epee [03:40] can't really do swedish chef in text... [03:40] bork bork! [03:41] Hmm, sometimes I thing Swedish Chef would make more sense than GCC, or Mir errors [03:42] I'm sure we could patch clang [03:42] warnings would print "bork", and errors would print "bork bork!" [03:42] segfaults... [03:45] GCC had drawn a fish on the screen... ? [03:45] *has [03:46] Free the fish! [04:28] oh yay trying to run unity on new platform-api leaves my system unresponsive [04:29] do I A. Try and extract a stack trace failing and rebooting 10 times before suceeding [04:29] or B. Speculate at possible causes failing and rebooting 10 times before suceeding [04:41] ok the segfault was kind of funny, qtubuntu is now using a static keycode list, we can remove this because we use xkbcommon and its actually unused on phablet because the OSK goes through a different code path [04:41] but I hadnt removed it yet in the new qtubuntu because I figured it was harmless [04:41] but we send some pretty big keycodes apparently and it can overflow this array [04:41] unity still doesn't show up but hard to tell if its these QML errors or something broken [04:45] qt seems to be working but Shell.qml is broken [04:45] in the branch I have [04:46] going to get a clean build while I watch the daily show! [05:07] Whee it runs [05:07] apparently I only have 59 minutes of talk time left on my imac though [05:07] XD [09:06] alf_: could you check https://code.launchpad.net/~robertcarr/mir/dev-package-depends-on-boost/+merge/167807 [09:07] alan_g: looking [10:14] * duflu --> weekend === alan_g is now known as alan_g|lunch === alan_g|lunch is now known as alan_g === alan_g is now known as alan_g|lunch === olli_ is now known as olli === alan_g|lunch is now known as alan_g === alan_g is now known as alan_g|tea === alan_g|tea is now known as alan_g === mmrazik is now known as mmrazik|afk [15:04] good morning everyone [15:05] good afternoon kdub [15:12] kdub: hi! [15:14] kdub: alan_g: How about a hangout about swapper-swapper (and later with Robert about client-focus-notifications) to discuss how to proceed/speed things up? [15:16] alf_, sure [15:17] alf_: give me 5 minutes to review the current state of swapper-swapper [15:20] i have an update to take your simplification for the RWLockWriterBias class [15:20] i'll push after the meeting though [15:22] kdub: alan_g: hm, I can't find how to create a hangout :/ [15:22] alf_: that's the new improved interface. ;) [15:22] keep clicking [15:25] kdub: i don't think it compiles anyway. ;) [15:25] "voud" indeed [15:26] alan_g: kdub: https://plus.google.com/hangouts/_/ad562109a71a07604737b3453b414f1b29de4454?hl=en , let's give it a try [15:28] Morning [15:28] greyback: ricmm: Can we delay catchup 10 minutes? I am just finishing breakfast [15:30] oh we cancelled it today? [15:37] racarr: there? [15:45] racarr, want to talking about client-focus-notifications on the hangout? :) [15:45] Yes! sorry [16:06] all: Have a great weekend [16:10] You too alf [16:10] *excited for the mailman to come* [16:10] second nexus 4 [16:44] just in time for the nex7 to start working :P [16:46] really! [16:46] well that's ok nexus 4 is more useul anyway [16:56] I noticed last night how much time I spent looking for windows and how disorienting this was [16:56] since I switched to two monitors [16:56] hotcorner for scale windows -> really nice [17:01] Something I just noticed about filenames [17:01] session_manager_test.cpp is significantly more useful than [17:02] test_session_manager.cpp [17:03] * alan_g expects renames by next week [17:03] I named the mirplatform API impl files like, session_mir.cpp, application_options_mir.cpp etc, because I was copying the naming convention in there, but then I was tempted to rename the files to like, mir_session_options.cpp [17:03] and realized that was actually really annoying [17:03] because you have can't tab complete [17:03] Bye all. Have a good weekend! [17:03] Bye! [17:03] You too === alan_g is now known as alan_g|EOW [18:02] do you guys know how far RAOF got with packaging xmir? [18:03] improving the packages that is [18:03] Not I, sorry. [19:19] Whee flashing new nexus 4 [19:19] the white ones look funny [19:27] adb can not be run as root even after [19:27] fastboot unlock? [19:33] oh -b