=== MacSlow is now known as MacSlow|lunch === ara_ is now known as ara === MacSlow|lunch is now known as MacSlow [14:17] Jonii, dandrader is working on a bunch of new features for Unity for Ubuntu 12.10 [14:18] you can follow the raw development if you want [14:18] and ask questions on the multi-touch-dev@lists.launchpad.net mailing list [14:18] dandrader, bregma: howdy [14:19] yous is back [14:19] yeah [14:19] how went the sprint? [14:19] it was alright [14:20] half of us were in lex half in london [14:20] that was kinda a pain [14:20] excellent [14:20] but oh well [14:20] this is the future of collaboration, I suspect [14:20] I'm quite sore from helping my sister move in and put ikea furniture together though [14:20] allen key strain? [14:21] that and many others :) [14:21] assembling ikea furniture is fun [14:21] anyway, how was things last week? [14:21] anything notable? [14:21] * cnd hasn't gotten through email yet [14:21] it wwas quiet [14:22] cnd, big merge proposal waiting for your opinion :) -> https://code.launchpad.net/~dandrader/nux/geis/+merge/108239 [14:23] dandrader, cool! [14:23] I'll take a look today [14:23] last Monday was a holiday everywhere in the universe except Canada and the UK, nobody seemed able to get motivated after that [14:23] it also wasn't a holiday in lexington :( [14:23] swap days! [14:23] heh [14:24] I guess it's that New England work ethic at play [14:24] so about the big issue: I don't have any news yet, still waiting on answers from legal [14:25] * cnd is trying to be discreet about it on this channel [14:25] some good news is I SRUd a libgrip crasher [14:25] yay [14:25] it took me two days of intense debugging to track it down, it was a one-line fix [14:26] I find GTK challenging in a bad way [14:26] heh [14:27] it sounds like there are a lot of hidden issues in gtk that only a gtk developer would know about [14:27] only some subset of gtk developer, no one can know all of the hidden issues [14:28] I'm hoping the legal issue can be resolved in a way that avoids us having to change code [14:30] me too [14:31] I've made it clear that some changes will balloon the effort from a little inconvenience to a couple weeks of one FTE depending on how much change is needed [14:31] yes [14:32] cnd, what's FTE? [14:33] full time engineer [14:33] it's an abstract way of saying how much work would be required [14:34] yeah, like "man hours" [15:11] dandrader: engineers are not men. we are supermen. ;) [15:12] specially a full-time one :) === dandrader is now known as dandrader|afk === dandrader|afk is now known as dandrader [15:53] If a geis client specifies just one gesture class per subscription it will never get a gesture with more than one gesture class, right? [15:53] cnd, ^ [16:07] dandrader, bregma may know the full answer [16:07] I'm not sure [16:07] but I think you're correct [16:07] because grail won't specify a gesture type that wasn't in the subscription [16:09] dandrader, you are correct [16:10] ok [16:18] t [16:18] h [16:18] j [16:18] v [16:18] t [16:18] sorry === dandrader is now known as dandrader|lunch [17:48] is it possible to hook the touch stuff up to a random TUIO stream to get multitouch working? [17:58] doug, does tuio report data as uinput devices through the kernel? [17:59] or does it provide an X driver? === dandrader|lunch is now known as dandrader [18:35] cnd: neither, currently. i've just got a UDP stream of TUIO events... [18:46] doug, then you would either have to build a utouch-frame backend for TUIO [18:47] or you would have to create a TUIO xserver input module [18:47] or send the events through the kernel using uinput [19:02] hm, which is the easier route there? [19:25] doug, probably uinput [19:25] doug, I thought some people were already looking into that [19:25] but I don't know who [19:26] doug, you can look at the source of utouch-evemu to get an idea of how uinput works [19:26] it's a pretty straight forward interface [20:09] cnd++ [20:09] thanks [20:10] doug, np :) === dandrader is now known as dandrader|afk [20:12] this would seem to be it: http://lii-enac.fr/cgi-bin/gitweb.cgi?p=linux-input/tuiototouch.git [20:30] doug, sounds right [20:30] bregma, please take a look at bug 986886 to comment on what changes are expected and to figure out if there really is a bug we need to fix there [20:30] Launchpad bug 986886 in utouch-geis "GEIS returns invalid arguments" [Undecided,Incomplete] https://launchpad.net/bugs/986886 === dandrader|afk is now known as dandrader [20:36] bregma, also, the libgrip dailies are FTBFS: https://code.launchpad.net/~utouch-team/+recipe/libgrip-daily [20:52] yes, it looks like there is a problem in the PPA causing libgrip to not find any dependencies -- weird, right? [21:07] bregma, it looked to me like a patch wasn't applying [21:07] https://launchpadlibrarian.net/106513591/buildlog.txt.gz [21:07] perhaps it's a stale patch in the packaging that needs to be removed? [21:51] yeah, that log looks suspicious - it usually fails much earlier with a dependncy "... not going to be installed" error [21:52] keep in mind it also builds cleanly in local pbuidlers and in Ubuntu buildds [21:53] I guess the daily is trying to apply the packaging patch but the fix is already upstreamed in source [21:53] I suppose the solution is to release a new version of grip [21:53] sounds like a task for tomorrow [22:49] yeah, I think that sounds right