Update tapenade 3.16

[ewu63]

Purpose

This time I actually updated the Tapenade-generated source code to 3.16.

Type of change

• Bugfix (non-breaking change which fixes an issue)
• New feature (non-breaking change which adds functionality)
• Breaking change (non-backwards-compatible fix or feature)
• Code style update (formatting, renaming)
• Refactoring (no functional changes, no API changes)
• Documentation update
• Maintenance update
• Other (please describe)

Testing

Existing tests all pass. I kept a tiny diff in one AD'd file to make sure our CI checks are working as expected.

[codecov]

Codecov Report

Merging #63 (e36cd57) into master (4a8a6ed) will decrease coverage by 0.93%.
The diff coverage is n/a.

❗ Current head e36cd57 differs from pull request most recent head dd5d33a. Consider uploading reports for the commit dd5d33a to get more accurate results

@@            Coverage Diff             @@
##           master      #63      +/-   ##
==========================================
- Coverage   45.72%   44.78%   -0.94%     
==========================================
  Files           6        6              
  Lines         748      748              
==========================================
- Hits          342      335       -7     
- Misses        406      413       +7     

Impacted Files	Coverage Δ
idwarp/UnstructuredMesh_C.py	46.15% <0.00%> (-53.85%)	⬇️

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 4a8a6ed...dd5d33a. Read the comment docs.

[eirikurj]

Overall, the changes look fine. I am assuming you will update the yml file later to test with 3.16 later?

[eirikurj]

@nwu63 I guess you beat me to it. However, why did the tapende test not fail before?

[ewu63]

I think when I removed those lines in commit 5409a91, this prevented tapenade from generating any new files. Therefore, git did not catch anything as it was comparing nothing to the existing files. If we really want to do this properly, we will need to also delete all existing Fortran files in those directories before running Tapenade. The current check assumes that the same filenames are being touched by Tapenade each time.

[anilyil]

Changes look good. My only minor comment: looks like we had the actual build of tapenade printed in the headers after the version, i.e. Tapenade 3.10 (r5363) but the new version just has version and master. Is it possible to get a more informative thing printed there instead of master?

Might be a bit easier to match this run in the future, but I may be completely missing the point as well. Is the version number enough to fully define what state of tapenade we used?

[ewu63]

I don't think there's a way to change this because we're just using compiled binaries released by them. However, I think in addition to the version, the date there refer to the day that the binaries are built, so together with the version we should be able to uniquely identify which version of tapenade was used to generate these.

[anilyil]

Okay, then this is good to go for me.

[ewu63]

I'm not sure if this is waiting on anyone. @eirikurj can we merge this in?

[lamkina]

This looks good to me.

[sseraj]

Can you clarify what the autoEdit changes do?

[ewu63]

Can you clarify what the autoEdit changes do?

I added a few files so that they are not copied over to the actual output directories, meaning they will not be compiled or tracked by git. These files are generated as part of Tapenade (I'm sure there's a way to set it up such that they are not generated, but I didn't try this), but the AD'd code are not needed. The ignore list makes sure they are not copied over.

[sseraj]

That makes sense, thanks