dive plan test

[Dirk Hohndel]

So I just pushed a commit that adds a TestPlan test. This one creates the
plan we discussed here before and runs Bühlmann against it, both in
imperial and in metric mode.

The test then makes sure we have the right gas switch depths (so the
imperial test fails right now since we only change to oxygen at 10ft, not
at 20ft as we should).

The thing that bothers me is that the plan that I create with the test is
different from the plan that Subsurface creates when I set up the same
preferences, etc. I must be missing one (or more?) parameter that
influences the plan, but even after staring at this for quite a while and
trying to print out useful debugging information, I fail to see what
triggers the difference. Granted, the differences are small, but the thing
is, there shouldn't BE a difference.

Here's what I get from the test:

<dive date='1970-01-01' time='00:00:00' duration='105:00 min'>
  <cylinder o2='15.0%' he='45.0%' />
  <cylinder o2='36.0%' />
  <cylinder o2='100.0%' />
  <divecomputer>
  <depth mean='30.716 m' />
  <event time='35:09 min' type='11' value='36' name='gaschange' cylinder='1' o2='36.0%' />
  <event time='69:21 min' type='11' value='100' name='gaschange' cylinder='2' o2='100.0%' />
  <sample time='0:00 min' depth='0.0 m' />
  <sample time='3:26 min' depth='79.0 m' />
  <sample time='30:00 min' depth='79.0 m' />
  <sample time='35:08 min' depth='33.0 m' />
  <sample time='35:09 min' depth='33.0 m' />
  <sample time='36:08 min' depth='33.0 m' />
  <sample time='37:08 min' depth='24.0 m' />
  <sample time='39:00 min' depth='24.0 m' />
  <sample time='39:20 min' depth='21.0 m' />
  <sample time='41:00 min' depth='21.0 m' />
  <sample time='41:20 min' depth='18.0 m' />
  <sample time='45:00 min' depth='18.0 m' />
  <sample time='45:20 min' depth='15.0 m' />
  <sample time='50:00 min' depth='15.0 m' />
  <sample time='50:20 min' depth='12.0 m' />
  <sample time='58:00 min' depth='12.0 m' />
  <sample time='58:20 min' depth='9.0 m' />
  <sample time='69:00 min' depth='9.0 m' />
  <sample time='69:20 min' depth='6.0 m' />
  <sample time='69:21 min' depth='6.0 m' />
  <sample time='70:20 min' depth='6.0 m' />
  <sample time='70:22 min' depth='6.0 m' />
  <sample time='103:00 min' depth='6.0 m' />
  <sample time='105:00 min' depth='0.0 m' />
  </divecomputer>
</dive>

And this is from Subsurface. All preferences removed and then UI changed
to match the data in the test:


<dive number='1' date='2015-07-05' time='10:17:33' duration='108:00 min'>
  <cylinder o2='15.0%' he='45.0%' />
  <cylinder o2='36.0%' />
  <cylinder o2='100.0%' />
  <divecomputer model='planned dive'>
  <depth max='79.0 m' mean='30.196 m' />
  <event time='35:09 min' type='11' value='36' name='gaschange' cylinder='1' o2='36.0%' />
  <event time='71:21 min' type='11' value='100' name='gaschange' cylinder='2' o2='100.0%' />
  <sample time='0:00 min' depth='0.0 m' />
  <sample time='3:26 min' depth='79.0 m' />
  <sample time='30:00 min' depth='79.0 m' />
  <sample time='35:08 min' depth='33.0 m' />
  <sample time='35:09 min' depth='33.0 m' />
  <sample time='36:08 min' depth='33.0 m' />
  <sample time='37:08 min' depth='24.0 m' />
  <sample time='39:00 min' depth='24.0 m' />
  <sample time='39:20 min' depth='21.0 m' />
  <sample time='42:00 min' depth='21.0 m' />  <---- first difference, one minute longer than in the test
  <sample time='42:20 min' depth='18.0 m' />
  <sample time='46:00 min' depth='18.0 m' />
  <sample time='46:20 min' depth='15.0 m' />
  <sample time='51:00 min' depth='15.0 m' />
  <sample time='51:20 min' depth='12.0 m' />
  <sample time='59:00 min' depth='12.0 m' />
  <sample time='59:20 min' depth='9.0 m' />
  <sample time='71:00 min' depth='9.0 m' />  <---- second stop that's a minute longer
  <sample time='71:20 min' depth='6.0 m' />
  <sample time='71:21 min' depth='6.0 m' />
  <sample time='72:20 min' depth='6.0 m' />
  <sample time='72:22 min' depth='6.0 m' />
  <sample time='106:00 min' depth='6.0 m' /> <---- third one that's a minute longer
  <sample time='108:00 min' depth='0.0 m' />
  </divecomputer>
</dive>

I pushed this as is, hoping that someone else will spot what I'm doing
differently that causes that discrepancy.

/D