4.8.2: processing offloaded Garmin Descent files

[Adric Norris]

This isn't really my issue (I came across it in the Garmin forums), but
caught my interest since the behaviour seems to vary by OS... and I'm not
sure if that's expected. To summarize, the question was "Is it possible to
process FIT files which have already been downloaded to my computer,
instead of requiring them to be present on the watch?".

What I found was that on Linux, simply placing them into a subdirectory
named Garmin/Activity seems to be adequate. Just choose the parent
directory as "Device or mount point", and all of the files are processed
without any apparent issue. The relevant output from "subsurface -vv" in
this case was:

Starting download from /home/adric/Descent
INFO: dc_device_open error value of 0
Finishing download thread: 50 dives downloaded
Set the current dive site: 0
Set the current dive site: 0
Set the current dive site: 2185751990
Set the current dive site: 2185751990
Set the current dive site: 2185751990
Set the current dive site: 1239251936

Doing the same thing on Windows 10, on the other hand, results in no files
being identified. Its output was:

Starting download from C:/Descent
INFO: dc_device_open error value of 0
EVENT_DEVINFO gave us the correct detected product (model 0 instead of 2859)
Finishing download thread: 0 dives downloaded
Set the current dive site: 0
Set the current dive site: 0

So I guess the real question is, should this approach work on Windows as
well? That doesn't seem like an unreasonable expectation, especially given
that the Descent presents itself as a mass storage device. But it's also
not a terribly pressing item, since this probably won't be the typical use
case and the workaround is super obvious. In the end, I'm merely curious. :)

I don't have access to a Mac, so no idea which side of this line it falls
under.

[Dirk Hohndel]

This isn't really my issue (I came across it in the Garmin forums), but caught my interest since the behaviour seems to vary by OS... and I'm not sure if that's expected. To summarize, the question was "Is it possible to process FIT files which have already been downloaded to my computer, instead of requiring them to be present on the watch?".
Starting download from /home/adric/Descent
INFO: dc_device_open error value of 0
Finishing download thread: 50 dives downloaded
Set the current dive site: 0
Set the current dive site: 0
Set the current dive site: 2185751990
Set the current dive site: 2185751990
Set the current dive site: 2185751990
Set the current dive site: 1239251936
Starting download from C:/Descent
INFO: dc_device_open error value of 0
EVENT_DEVINFO gave us the correct detected product (model 0 instead of 2859)
Finishing download thread: 0 dives downloaded
Set the current dive site: 0
Set the current dive site: 0
So I guess the real question is, should this approach work on Windows as well? That doesn't seem like an unreasonable expectation, especially given that the Descent presents itself as a mass storage device. But it's also not a terribly pressing item, since this probably won't be the typical use case and the workaround is super obvious. In the end, I'm merely curious. :)
I don't have access to a Mac, so no idea which side of this line it falls under.

Most of the development and testing of the Garmin code was done exactly this way - by downloading from a local folder.
But of course the developers aren't using Windows :-)
The "EVENT_DEVINFO" in the Windows output makes me wonder what's going on - the FIT file seems to have not been from a Garmin Descent.
I will have to try this myself and try to see if I can gleam anything useful by adding some debugging prints... but fundamentally I'd assume that this should work.

/D

[Adric Norris]

Post by Dirk Hohndel
Most of the development and testing of the Garmin code was done exactly
this way - by downloading from a local folder.
But of course the developers aren't using Windows :-)
The "EVENT_DEVINFO" in the Windows output makes me wonder what's going on
- the FIT file seems to have not been from a Garmin Descent.
I will have to try this myself and try to see if I can gleam anything
useful by adding some debugging prints... but fundamentally I'd assume that
this should work.

I was using the same 51 files (only 1 being a non-dive activity) for both
tests, all of which were copied off my Descent, so the files (as you
probably suspect) are known to be good. I'd be happy to provide them if you
like, although it doesn't sound likely to matter in this case.