[02:31] <duflu> RAOF: What's this new pointer type you've mentioned?
[02:31] <RAOF> Overly complicated and abandoned.
[02:31] <RAOF> :)
[02:31]  * RAOF -> lunch.
[02:32] <duflu> RAOF: Oh. I think Google made a good choice actually in abbreviating their template to "sp"
[02:33] <duflu> We could have a "ref" or "ptr" similarly. It would improve the readability of many lines of code
[02:33] <duflu> Although C++ should have done this for us :(
[03:24] <racarr_> I would probably vote for an SP alias...
[03:25] <racarr_> its not so bad by itself but when its like
[03:25] <racarr_> std::vector<std::shared_ptr<graphics::....
[03:25] <racarr_> I start to struggle sometimes ;)
[09:03] <duflu> Bleh. Why does even the simplest task end up not being simple?
[09:37] <duflu> alan_g: Hey I think I probably failed to document some enhancements by other people that I didn't understand. Feel free to add more entries to the changelog
[09:38] <alan_g> duflu: ack
[10:12] <duflu> Oh did vivid just get kernel 3.18 today?
[10:15] <duflu> Apparently it did
[14:26] <Saviq> hey guys, does this trace ring any bells https://errors.ubuntu.com/problem/fc69b4f16d4d4571c83038bf9dc47b84135d9ced ?
[14:27] <Saviq> the exception msg I managed to find is:
[14:27] <Saviq> terminate called after throwing an instance of 'boost::exception_detail::clone_impl<boost::exception_detail::error_info_injector<std::runtime_error> >'
[14:27] <Saviq>   what():  Nested Mir Display Error: Failed to update EGL surface.
[14:34] <alan_g> Saviq: no bells. All it really says is "eglMakeCurrent() failed" (It doesn't fix the failure but we should at least record the error code!)
[14:35] <Saviq> alan_g, right, could be useful indeed
[14:38] <alan_g> I guess you've no context about when this happens?
[14:41] <Saviq> alan_g, during an AP test run
[14:42]  * Saviq tries to reproduce locall
[15:58] <mlankhorst> is rotation implemented on the server or do clients have to do their own rotation?
[16:28] <racarr_> Good morning!
[17:05] <racarr_> Ive been trying to figure out
[17:06] <racarr_> how to make the doxygen more useful...
[17:06] <racarr_> in particular the comments in tree
[17:06] <racarr_> I dont care that much abotu the generated output (perhaps thats a mistake)
[17:06] <racarr_> one idea is moving all the trivial comments
[17:06] <racarr_> e.g.
[17:06] <racarr_> mir_foo_get_baz
[17:06] <racarr_> retreives bazz from foo
[17:07] <racarr_> precondition foo is valid
[17:07] <racarr_> etc
[17:07] <racarr_> in to like
[17:07] <racarr_> trivial-comments.h
[17:07] <racarr_> so then in the header file there is only stuff you actually wanna read
[17:34] <racarr_> Landing https://code.launchpad.net/~mir-team/mir/fix-1407883/+merge/245703 to avoid CI failures