Fix test for earlier SNOPT versions

[ewu63]

Purpose

Closes #369.

Expected time until merged

ASAP.

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

Checklist

• I have run flake8 and black to make sure the Python code adheres to PEP-8 and is consistently formatted
• I have formatted the Fortran code with fprettify or C/C++ code with clang-format as applicable
• I have run unit and regression tests which pass locally with my changes
• I have added new tests that prove my fix is effective or that my feature works
• I have added necessary documentation

[codecov]

Codecov Report

Attention: 1 lines in your changes are missing coverage. Please review.

Comparison is base (75e6479) 73.36% compared to head (5bedf92) 73.36%.

Files	Patch %	Lines
pyoptsparse/pySNOPT/pySNOPT.py	66.66%	1 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main     #370      +/-   ##
==========================================
- Coverage   73.36%   73.36%   -0.01%     
==========================================
  Files          22       22              
  Lines        3312     3315       +3     
==========================================
+ Hits         2430     2432       +2     
- Misses        882      883       +1     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[ewu63]

Can someone with access test whether the offending test passes with SNOPT 7.7.5? If so then I'd propose bumping the SNOPT version rather than trying to patch very outdated SNOPT versions within pySNOPT. CC @eirikurj @eytanadler

[eirikurj]

For completeness, I tried 7.7.5 and tests still fail. Is that however not expected since sol.objectives is now being overwritten with fStar, which is the buggy zero obj value returned from SNOPT?

[ewu63]

Hm I had thought that maybe 7.7.5 fixed the bug, but perhaps not. A few options:

1. add a version check in pySNOPT to not overwrite sol.objectives values for versions 7.7 below 7.7.7.
2. hack or skip the tests for buggy SNOPT versions

I am in favour of the first option since with the changes in #364, buggy SNOPT optimizations essentially lose the optimal objective value, while previously you can at least get the last queried value via sol.objectives which is often times close. I'm going to push a patch following 1) but happy to hear other opinions.

[ewu63]

Right, I forgot to un-scale the obj values. Thanks, LGTM

[lamkina]

LGTM

[ewu63]

@eirikurj or @sseraj let's merge this so we don't fail the nightly build?

[eirikurj]

Yeah, tests have been failing for a while. @sseraj do you have time to review this or anyone from @mdolab/pyoptsparse_maintainers? If not, I will merge at end of week.