iOS build 4.9.3.1486

2020-04-21 Thread Hartley Horwitz via subsurface
I've done some basic checks on the latest iOS build. * loading a local file * syncing with cloud * edits, undo, redo (1 level only) * save back to cloud * restart app, reload from cloud to confirm changes All looks good. I'm probably not the best tester for this release because I wasn't experienc

Re: new CI builds

2020-04-21 Thread Chirana Gheorghita Eugeniu Theodor via subsurface
1495 seems to solve the crash issue on fedora 31. I was able to add a dive, edit it, save it delete it. More testing to follow. Any specific tests in your mind ? On Tue, Apr 21, 2020, 19:28 Dirk Hohndel wrote: > On Tue, 2020-04-21 at 07:29 -0700, Dirk Hohndel via subsurface wrote: > > On Tue, 2

Re: new CI builds

2020-04-21 Thread Dirk Hohndel via subsurface
On Tue, 2020-04-21 at 07:29 -0700, Dirk Hohndel via subsurface wrote: > On Tue, 2020-04-21 at 12:12 +0300, Chirana Gheorghita Eugeniu Theodor > wrote: > > push to obs pls. > > > > He can't. OBS builds are semi-automatically triggered by merges into > master - for reasons I haven't figured out so

Re: new CI builds

2020-04-21 Thread Dirk Hohndel via subsurface
On Tue, 2020-04-21 at 09:25 +0200, Dietrich Meyer via subsurface wrote: > > For Linux Fedora add new dive from Log > New dive still crashes the > > app: > > > Ok. > > > First test failed on linux. > > > I click Log > Add dive and app crashes. Where is the log of the > > > app in > > > linux? I can

Re: new CI builds

2020-04-21 Thread Dirk Hohndel via subsurface
On Tue, 2020-04-21 at 09:20 +0200, Berthold Stoeger wrote: > Ok, this one is weird. > > > That appears to be g++, right? To me this looks like it trips on an > assertion > of a "vector[n]" access. The one thing I found in the offending > function was a > construct of the type > "&mean[0]" > t

Re: new CI builds

2020-04-21 Thread Dirk Hohndel via subsurface
On Tue, 2020-04-21 at 12:12 +0300, Chirana Gheorghita Eugeniu Theodor wrote: > push to obs pls. > He can't. OBS builds are semi-automatically triggered by merges into master - for reasons I haven't figured out so far, I often have to manually nudge them to get started. Berthold has created a Pul

Re: new CI builds

2020-04-21 Thread Dirk Hohndel via subsurface
On Tue, 2020-04-21 at 14:13 +0200, Willem Ferguson wrote: > Using 4.9.3-1480 on Ubuntu 18.04 > > I downloaded this version from the Ubuntu ppa provided by Dirk. > > (one question in total ignorance. Is this package a snap?) No, the PPA gives you "native" binaries for Ubuntu. > I downloaded usin

Re: new CI builds

2020-04-21 Thread Willem Ferguson via subsurface
Using 4.9.3-1480 on Ubuntu 18.04 I downloaded this version from the Ubuntu ppa provided by Dirk. (one question in total ignorance. Is this package a snap?) I downloaded using both Bluetooth and iRDA, no problems. Edited these dives. No problem. Tested all the options I could on the dive list

Re: new CI builds

2020-04-21 Thread Chirana Gheorghita Eugeniu Theodor via subsurface
push to obs pls. On Tue, Apr 21, 2020, 10:20 Berthold Stoeger wrote: > Ok, this one is weird. > > On Dienstag, 21. April 2020 08:44:09 CEST Chirana Gheorghita Eugeniu > Theodor > wrote: > > > /usr/include/c++/9/bits/stl_vector.h:1042: std::vector<_Tp, > > _Alloc>::reference std::vector<_Tp, _All

Re: new CI builds

2020-04-21 Thread Dietrich Meyer via subsurface
> For Linux Fedora add new dive from Log > New dive still crashes the app: > > Ok. > > First test failed on linux. > > I click Log > Add dive and app crashes. Where is the log of the app in > > linux? I can dig it out but I am too lazy now. Cannot reproduce this on Opensuse 15.1 Updated to 4.9.

Re: new CI builds

2020-04-21 Thread Berthold Stoeger via subsurface
Ok, this one is weird. On Dienstag, 21. April 2020 08:44:09 CEST Chirana Gheorghita Eugeniu Theodor wrote: > /usr/include/c++/9/bits/stl_vector.h:1042: std::vector<_Tp, > _Alloc>::reference std::vector<_Tp, _Alloc>::operator[](std::vector<_Tp, > _Alloc>::size_type) [with _Tp = int; _Alloc = std: