Feature: add ModuleIO::write_libxc_r() #5232
Conversation
…ce XC_Functional_Libxc
Conflicts: source/module_esolver/esolver_fp.cpp source/module_hamilt_general/module_xc/xc_functional_gradcorr.cpp source/module_hamilt_general/module_xc/xc_functional_libxc.h source/module_hamilt_general/module_xc/xc_functional_libxc_tools.cpp source/module_hamilt_general/module_xc/xc_functional_libxc_vxc.cpp source/module_hamilt_general/module_xc/xc_functional_libxc_wrapper_gcxc.cpp source/module_hamilt_general/module_xc/xc_functional_libxc_wrapper_tauxc.cpp source/module_hamilt_general/module_xc/xc_functional_vxc.cpp source/module_io/cube_io.h source/module_io/write_cube.cpp
mohanchen
added
the
Input&Output
| #include <stdexcept> | ||
| #include <unistd.h> | ||
|
|
||
| void ModuleIO::write_libxc_r( |
There was a problem hiding this comment.
could you add tests for this function? since there are many cases, I guess we need more than one test.
| } // end switch( func.info->family ) | ||
| } // end for( xc_func_type &func : funcs ) | ||
|
|
||
| auto write_data = [&esolver]( |
There was a problem hiding this comment.
I think it's better to let
write_databe a free function ( named likewrite_xc_r) instead of a lambda in such a big functionwrite_libxc_r. It will help someone like me already having data likefxc(nrxx, 3)calculated elsewhere to write. The dependences can be pw_big, pw_rhod, nspin, ld(number_spin), directory and precision, independent ofPARAMandesolver.
After discussion, I think the write_cube function needs more refactor (#5191). The parameters need to be reduced for users' convenience. Then I can directly call write_cube at my need.
|
The PR lasts too long, please resubmit the PR again. |
3 participants
No description provided.