Closed chjacob-tubs closed 6 years ago
Merging #67 into master will not change coverage. The diff coverage is
n/a
.
@@ Coverage Diff @@
## master #67 +/- ##
=======================================
Coverage 42.29% 42.29%
=======================================
Files 78 78
Lines 1785 1785
=======================================
Hits 755 755
Misses 1030 1030
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 16c90a1...18bc70e. Read the comment docs.
I am not familiar enough with SWIG to give comments off the top of my head. I'll check out this branch locally, test it out and give a more informed review.
I think short-term this is fine. Long-term either pybind11 or CFFI since they provide a thinner interface.
Thanks for the comments. Let me have another look before merging this, I will try to fix at least the copying/installation part ...
Apologies that it took quite a bit to review this one. Being in another timezone did not compose well with being unfamiliar with SWIG. Still, great learning opportunity for me!
I put some more work into this and now used UseSWIG instead of the manual setup that was used before. Please have a look, I think this is much nicer now. However, I didnt the RPATH stuff to work so I am still linking to the static library. To me, this is not a problem, but maybe you want to fix that.
I tested on MacOS and Linux with gcc. I am not sure whether switching to a different compiler will cause problems when building the Python shared library, because the Python libraries are usually built with gcc.
I also fixed the copying part, so the xcfun Python module can be used directly from the build directory.
Sorry for removing FinfPythonLibsNew - I thought that was something obscure that I added at some point. I stand corrected. The target_include_directories is indeed required, but I made the linking part a bit nicer
@bast I think once this is in we can:
release/v.2.0
v2.0.0
@robertodr @bast Let me fix the Python tests (#43) before doing the v2.0 release.
Oh yes, of course! Thanks for reining in my enthusiasm :) Merging this one and closing #67
This is my attempt at getting the SWIG Python bindings back to work. It does the job for me, but I am sure the setup is not perfect (maybe a cmake expert can have a look, see "possible improvements" below). (Mostly fixes #63)
Changes:
Possible improvements: