Moving from Shearwater Cloud to Subsurface

[Josiah Vivona]

I have a couple hundred dives I have imported to Shearwater cloud from my Peregrine. In the Shearwater app I have added location info, buddy names, start and end pressure for dives, gear configurations and more. 

I really want to start using Subsurface as my primary dive log, but I am stuck on trying to export from Shearwater cloud to and import to Subsurface. What am I missing? I know I can import from my computer, but then I would have to reenter all the other data manually.

I have had the most luck with exporting UDDF from Shearwater and importing that to Subsurface, but I am missing a lot info. Some items such as location and dive buddy are not included in the exported data file. Other items such as the gas mix (mostly Nitrox 32) are not showing in subsurface despite being in the UDDF file in one form or another.

Everything shearwater proprietary is broken one way or other (really from the Shearwater export side).

Any help is appreciated 

[Ryan Holden]

I exported mine as CSV and manipulated the data in excel to ensure all the fields lined up (I had some issues due to adding a second transmitter when I switched to sidemount, which threw all the CSV columns out after that point). Then imported into Subsurface. You can confirm which fields match with which columns at the point of import.

[Josiah Vivona]

I have dug into this a little bit more and had the best luck with UDDF export from Shearwater cloud app. The biggest problem I was having was that my gas definitions were not being imported into subsurface. I found that the Shearwater exported UDDF did not include a link reference to the gas mix for the tankdata elements. I made a patch to the uddf.xslt file and everything that I care about is working for me now. Patch is attached, but I am not 100% this is totally correct.

Josiah Vivona

BTW getting subsurface to build on a M1 mac was a challenge

[miika....@gmail.com]

With a brief glimpse, the patch looks good to me. For the imports, we do assume things based on the samples we have seen. Usually it is undocumented data formats, but even with UDDF that does actually have specification, things are somewhat different between different vendors.

Is there any data stored by Shearwater in the vendor specific block? Or is the data you are still missing just not included in the export? If you can send me a sample dive exported from Shearwater, I could add it to my collection of format samples.

[Josiah Vivona]

The missing data I was most concerned about was gas mix, which my patch addresses. I am attaching a UDDF file exported from Shearwater Cloud.

I haven't tested the patch with dives having multiple tanks and gas mixes. I am not super familiar with the XSLT and XPATH syntax so my patch is copy-pasta.

Josiah Vivona

On Wed, Jun 21, 2023 at 6:52 PM miika....@gmail.com <miika....@gmail.com> wrote:

With a brief glimpse, the patch looks good to me. For the imports, we do assume things based on the samples we have seen. Usually it is undocumented data formats, but even with UDDF that does actually have specification, things are somewhat different between different vendors.

Is there any data stored by Shearwater in the vendor specific block? Or is the data you are still missing just not included in the export? If you can send me a sample dive exported from Shearwater, I could add it to my collection of format samples.

--
You received this message because you are subscribed to a topic in the Google Groups "Subsurface Divelog" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/subsurface-divelog/qGkPOhPhiWc/unsubscribe.
To unsubscribe from this group and all its topics, send an email to subsurface-dive...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/subsurface-divelog/6fd2575f-3ee1-4a17-803e-fb7a266b0f77n%40googlegroups.com.

[Michael Keller]

Hi Josiah.

Thanks for looking into this.

I think the reason that on Shearwater computers there is no link between entries in the list of gas mixes and entries in the list of tank sensors - Shearwater is correct in not adding links that do not exist. Also, since the list of gasmixes on Shearwater computers is dynamically ordered based on the O2 / He fractions in the gases the position of a given gas in the list will change when other gases are modified, requiring the user to manually edit the list of tank sensors every time they change the gas list if they want a certain tank sensor to stay associated with the same gas mix.

For this reason I believe that introducing a 1 : 1 mapping of gasmix list entries and tank sensor list entries that does not exist is not the correct way to go on about this. It will be better to just remove the limitation of non-linked `gasdefinition` entries not being imported if `tankdefinition` entries exist in https://github.com/subsurface/subsurface/blob/b5007bde674bd9b1100b24278250aab1790114d9/xslt/uddf.xslt#L302, and get the `gasdefnintion` entries imported as separate entities to `tankdefinition` entries, instead of incorrectly linking them into arbitrary pairs.

Ngā mihi

  Michael Keller

--
You received this message because you are subscribed to the Google Groups "Subsurface Divelog" group.
To unsubscribe from this group and stop receiving emails from it, send an email to subsurface-dive...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/subsurface-divelog/9a205d27-ba32-46b9-b096-f98a63d8aae2n%40googlegroups.com.