Inconsistencies between interface specification / CLI-interface / C-API

[schmoelder]

This is a collection of smaller issues that need to be addressed to ensure consistency in the simulation output.

• [CLI] SOLUTION_VOLUME has shape (NTIME, 1), should have (NTIME)
• CSTR returns axial_coordinates and particle_coordinates. Should not return coordinates
  • [CLI]
  • [C-API]
• LRMP returns axial_coordinates and particle_coordinates. Should only return axial_coordinates
  • [CLI]
  • [C-API]
• [C-API] GRM squashes dimensions for solution_inlet/outlet if NCOMP=1. Should not do that. (Fixed in cadet/CADET-Python@2211aab)

[schmoelder]

Damn, I didn't properly cover some of the edge cases in the C-API, so there are still some issues regarding the coordinates.

For reference, these are the relevant lines:

CADET-Core/src/libcadet/api/CAPIv1.cpp

Line 907 in 8fc5a7b

cdtResult getPrimaryCoordinates(cdtDriver* drv, int unitOpId, double const** data, int* nCoords)

CADET-Core/src/libcadet/api/CAPIv1.cpp

Line 927 in 8fc5a7b

cdtResult getSecondaryCoordinates(cdtDriver* drv, int unitOpId, double const** data, int* nCoords)

CADET-Core/src/libcadet/api/CAPIv1.cpp

Line 947 in 8fc5a7b

cdtResult getParticleCoordinates(cdtDriver* drv, int unitOpId, int parType, double const** data, int* nCoords)

If coordinates are not stored or singleton dimensions are squeezed, it should return cdtDataNotStored.

To test this, see also cadet/CADET-Python#42