What's missing for 4.2?

[Jef Driesen]

On 2014-06-10 14:15, Anton Lundin wrote:
> On 10 June, 2014 - Jef Driesen wrote:
>> I'm a bit surprised you get a gas change event in gauge mode. The 
>> vyper
>> parser will only generate gas change events when they are recorded by 
>> the
>> device. But why would it bother with gas change events in gauge mode? 
>> Isn't
>> gauge mode supposed to turn your dive computer in a dumb data 
>> recorder,
>> which no longer cares about doing decompression calculations?
> 
> A quick glance at the libdivecomputer code, it sends gas as air if %O2
> is 0, and its probably that in gauge mode.

I'm not sure about the gauge mode, but for air mode, the stored O2 
percentage is certainly zero. So I can't treat zero as gauge mode, 
because that would break air dives.

> I did some googling and found:
> http://vyper.bunny.hu/eng/Vyper_protocol/

The most recent version is available here:

http://www.sarnau.info/papers:suunto_vyper

I maintain links to all known protocol documentation on the 
libdivecomputer website:

http://libdivecomputer.org/links.html

> It points out where the bits for the Air / Nitrox / Gauge mode are
> stored.

Except that these bits are useless for parsing the dive data. It's only 
the current value of the setting. For parsing the dives, you need that 
value for each individual dive. If the current setting is gauge, that 
doesn't mean all your past dives where gauge too.

> It also shows some other interesting bits in there like OLF (
> some Suunto speak for CNS as i read it) and some other goodies.

I believe OLF (Oxygen Limiting Factor) is either CNS or OTU, depending 
on which of the two is the highest, and thus the limiting factor.

> Another side note: it would be great if we stored settings used during
> the dive in subsurface too. For the Suunto Vyper that would be Altitude
> mode, Personal safety limit, and probably something more. For my OSTC3
> it would be great to grab GFLow/GFHigh and so on.

Supporting this is more complex than you may think. Almost every dive 
computer model has different settings, and stores them in some different 
way. Thus supporting this through some common and standardized interface 
is nearly impossible, and providing a custom interface for every single 
feature is a nightmare too. So for libdivecomputer I don't even bother 
trying, and I try to stick to those features that are more universally 
available.

>> >If you would like to look closer at how gauge mode is handled i can
>> >create dumps for you.
>> 
>> Extra data is always welcome :-)
>> 
> 
> Ill create some dumps for you.

Great.

Jef