Double DC strange behavior [was: Prev/Next DC]

[Dirk Hohndel]

> On Dec 23, 2014, at 4:10 AM, Davide DB <dbdavide at gmail.com> wrote:
> 
> I repeated the sequence and I was able to reproduce the bug.
> 
> I exported the dive before/after the new DC import in subsurface format.
> Comparing the two xml files I see that after importing the new DC
> profile it added a wrong air gas and it modified just one sample with
> a phantom "sensor '3'" attribute.

As I said on trac - that’s one thing that puzzles me and it’s clearly broken and wrong.

> This is the inserted cylinder:
> 
> <cylinder description='Air' start='218.25 bar' end='83.75 bar' />

So it takes the pressure data from the first DC, first tank, and turns that into the 
pressure data for the new air tank. Creative, but also wrong.

> This is the modified sample:
> 
> <sample time='0:12 min' depth='4.98 m' temp='18.4 C' pressure='218.25
> bar' sensor='3' />
> 
> BTW the new computer added also this:
> 
> <notes>" "</notes>

That’s because the UDDF from the deep contains these insightful lines:

	<notes>
	<para>" "</para>
    </notes>

> Before adding the new dc profile, the start/end pressure were
> calculated  (and correctly displayed) by galileo stored pressure
> samples.

Interestingly enough, even your ‘before’ XML already had very curious data
for the trimix cylinder that I couldn’t quite understand. Even though there
were samples with cylinder pressure it showed an (incorrect) end pressure
for that cylinder in the XML:

<cylinder size='17.0 l' workpressure='232.0 bar' description='D8.5 232 bar' o2='15.0%' he='60.0%' start='220.0 bar' end='200.0 bar' />

And as usual - your top post makes it very hard for people to keep the chronology 
of data in sight…

/D

> 
> On Tue, Dec 23, 2014 at 12:28 PM, Davide DB <dbdavide at gmail.com> wrote:
>> Hummm,
>> 
>> I think I got a nasty bug using two computers. I ask here before I file a bug.

I wouldn’t call this a nasty bug - it’s maybe a small inconsistency… :-)

>> 
>> I have a dive logged with a Galileo sol and a xdeep black Bottom timers.
>> 
>> 1) I downloaded Galileo data
>> 2) I filled all the relevant info as gas used, tanks, pressures, gas
>> switches....
>> 3) everything is fine.
>> 4) I import xdeep data
>> 5) I lost gas switches and gas used hence sac... One gas switch on
>> Galielo went to the xdeep.

I don’t see any indication of this. Can you explain where you think you see that?

>> What should I see in this complicated situation?
>> 
>> I was expecting to find one of the following:
>> 
>> #1 the same gas switches and gas I add on a profile are copied on the other

As I said on trac. NOPE. That’s NOT what you should see.
The UDDF from the xdeep clearly claims that it uses air and has NO gas switch at all.
Since the first DC has no tank with air, Subsurface invents a third cylinder with air and
switches the xdeep DC explicitly to it at the beginning of the dive. And has no gas
switch at all in the xdeep DC afterwards.

>> #2 The second profile is virgin so I have to insert again gas switches

I don’t understand. That’s what you get.

>> 
>> Actually option #2 would bring me in a dead end: different SAC so only
>> option #1 is correct

I don’t quite see how Subsurface could magically do the right thing with the data
it gets. Garbage in, garbage out.

What you COULD do to fix it is this:

Manually remove the gas switch event in the second DC structure (you can only do this
in the XML since we never show a forced gas switch at the start (since those are confusing
to the user)). And then simply add the gas switch to O2


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.subsurface-divelog.org/pipermail/subsurface/attachments/20141223/d902ff11/attachment-0001.html>