Subsurface deco calculations

[Linus Torvalds]

On Wed, Feb 8, 2017 at 2:39 PM, Linus Torvalds
<torvalds at linux-foundation.org> wrote:
>
> Why? We also do equipment merges not just across dive computers, but
> when you merge two consecutive dives into one dive with a surface
> interval.
>
> So then we'd need to have some other logic fir that case.

Naah, let's just say that if you want those cylinders to be merged,
you'd better make sure that the first one lacks and end pressure, and
the second one lacks a starting pressure.

Most people won't have added those in-between pressures anyway if they
really want to consider it a single dive. And if you had an
air-integrated computer, this is a non-issue. If you were really just
bobbing up-and-down on the surface for a while (which is why you want
to merge the two parts into one dive), you wouldn't have recorded
those intermediate pressures anyway.

I've pushed out my branch and made a github pull request for people. I
still encourage people to take a look, but it seems to handle both
Miika's and Willem's cases just fine, and it passes my "looks sane"
test.

But maybe I did something stupid.

                     Linus