It's interesting how far you can get if you ignore all the outstanding issues, 
bugs, and PRs and everything else that is trying to get your attention and just 
focus on getting a release out the door.
Five hours later and here we are:

- source tagged and pushed
- binaries built / builds triggered
  - Windows, AppImage, Launchpad/Ubuntu, Mac 10.13 and later are up
  - OBS/openSUSE/Fedora is still building
  - Ubuntu/Snap should be done soon, Michal pushed the update
  - macOS 10.12 is currently not supported -- see below
- website is updated
- new supported divecomputer list published
- download page updated (except macOS 10.12 is currently pointing at 
non-existing file)
- announcement in English has been posted (translations welcome, of course)
- checked that the update server backend gives reasonable responses

I'm sure I forgot a few things. I always do... but these were the things I 
could come up with today.

MAJOR kudos for this release go to Berthold, with strong supporting cast vote 
for Willem.
Their work (the statistics) is why I decided to bump the version to 5.0 (and 
also... I dislike double digit version numbers... 4.9.10 bugged me, and I 
wasn't going to do a 4.10...)

With the exception of Berthold's work there were a few bug fixes and changes 
here and there, but the statistics paint the picture that I think we are all 
aware of:

git shortlog -s -n v4.9.10..
   246  Berthold Stoeger
   213  Dirk Hohndel
    27  Robert C. Helling
     6  Michał Sawicz
     6  Willem Ferguson
     2  Miika Turkia
     1  henrik242
     1  Damian Zaremba
     1  Doug Junkins
     1  Linus Torvalds
     1  Andreas Buhr

Berthold fixed many issues and of course wrote nearly all of the statistics 
code. I bumble around with stuff, infrastructure, code management, the 
occasional mobile UI stuff. Robert works mostly on deco / planner, but added 
some things for the statistics as well. Michał did the Snap work, Willem does 
the user manual (and did a TON of the work that lead to the statistics, but 
much of this didn't end up being commits). Miika fixes importer bugs and the 
rest... yeah, small random fixes by a couple of people.

My gratitude, as always, goes to the translators that did an amazing job making 
sure things stayed well translated. To all the testers who helped us figure out 
the issues in the statistics code. And to everyone who has contributed in any 
way over those past 10+ years. This is an amazing project, filled with amazing 
people. I hope that when diving becomes more common again for more of us, that 
maybe a few more of us will be motivated to work on things and make Subsurface 
even better. But for now I'm just thrilled that I finally managed to get 5.0 
out the door.

I'll try to see how far I can get with the mobile version...

/D


As for macOS 10.12...

Berthold, the compiler that I have available there stumbles over 

subsurface/stats/./statsview.h:187:9: error: cannot refer to class template 
'ChartItemPtr' without a template argument list
        return ChartItemPtr(new T(*this, std::forward<Args>(args)...));

I thought we don't have a ton of people using the 10.12 binary, but I can 
actually see almost a hundred downloads for 4.9.10 (that's shockingly large for 
such an old release - but they all look like actual browser downloads, not 
bots). So I wonder if this is indeed worth addressing. I believe I am using the 
latest Xcode that is supported for macOS 10.12 builds...

_______________________________________________
subsurface mailing list
subsurface@subsurface-divelog.org
http://lists.subsurface-divelog.org/cgi-bin/mailman/listinfo/subsurface

Reply via email to