Subsurface deco calculations

[Linus Torvalds]

On Wed, Feb 8, 2017 at 1:01 PM, Robert Helling <helling at atdotde.de> wrote:
>
> here  is the example that Willem sent a few days ago. Here, the gases differ both by o2 percentage and by start/end pressures but I would argue they should not end up with the same cylinder even if the o2 percentage would be the same.

Ok, my merge logic does actually largely work for those - after you
edit the start times so that they can be merged. But it does show that
there's one cylinder that has the same O2 percentage (36%) and then
two that do not (27/29% and 56/60%), and so the end result is a fairly
strange "five cylinders" with those 36% cylinders having been shared
between the different cases.

And yeah, with explicit manually added pressure data, that one merged
cylinder probably shouldn't be merged either.

I think I'll tweak my patch a bit and make a github pull request, but
it's pretty good as-is.

                       Linus