Make xc_get() part of the API

[stigrj]

Questions

Is there any reason why this function is not part of the C++ API?

int xc_get(xc_functional fun, const char * name, double * value);

Motivation and Context

I would like to be able to retrieve the amount of exact exchange from XCFun. To me the only reason to keep the exx parameter in XCFun is to keep track of it for the host program, but currently I don't see any way to fetch this information (correct me if I'm wrong). I would like to avoid duplication of this parameter.

Alternative solution

Add specific function to fetch exx:

XCFun_API double xc_get_exx(xc_functional fun)

Description

Added XCFun_API to the xc_get() function in xcfun.h.

Types of changes

• [ ] Bug fix (non-breaking change which fixes an issue)
• [x] New feature (non-breaking change which adds functionality)
• [ ] Breaking change (fix or feature that would cause existing functionality to change)

Status

• [x] Ready to go

[stigrj]

Travis fails on an unrelated (?) Python issue.

[robertodr]

No particular reason to not have it exposed other than it was not used in the examples and tests. It would be good if:

1. You could add a test for it.
2. Document the API function with a Doxygen string. In particular, knowing which names are valid as second argument.

Travis failures are due to Pipenv vagaries. I think this is a minor change and can be pulled in and I can later look at greening Travis up. @bast what do you think?

[bast]

If this PR can wait a day or so then I would prefer if we fix the pipenv issue in a separate PR and rebase this on top of that. Does that sound OK?

[stigrj]

NP, I will work on @robertodr 's points in the mean time

[bast]

Good - I am looking that the pipenv side of things.

[bast]

Pipenv issue followed-up in #80.

[robertodr]

Yep, much better. Thanks all!

[bast]

I am working on the pipenv issue and will ping you here once this patch is ready to be rebased.

[bast]

81 is ready to go. Once merged, please rebase this patch.

[stigrj]

Rebased and ready to go :crossed_fingers:

[robertodr]

There are some genuine test failure now:

Numerical check failed: B3LYP contains  0% FOO
 got      1.1999998092651367e+00
 expected 0.0000000000000000e+00
 Absolute error is 1.2000e+00, relative error is inf

[stigrj]

Sorry, I thought even undefined functionals returned a 0.0 value.

[codecov-io]

Codecov Report

Merging #79 into master will increase coverage by 0.59%. The diff coverage is n/a.

@@            Coverage Diff             @@
##           master      #79      +/-   ##
==========================================
+ Coverage   44.37%   44.97%   +0.59%     
==========================================
  Files          78       78              
  Lines        1839     1839              
==========================================
+ Hits          816      827      +11     
+ Misses       1023     1012      -11

Impacted Files	Coverage Δ
src/xcfun.cpp	50.23% <ø> (+2.54%)	:arrow_up:

---

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 025d2e8...8372679. Read the comment docs.

[robertodr]

No problem. And coverage went up by a whopping 0.59%! Thanks.