WIP: The eclipse gride preprocessing is not allowed to deactivate cells #393
Conversation
|
|
||
| { | ||
| const auto& ecl_actnum = ecl_grid.getACTNUM(); | ||
| for (std::size_t g = 0; g < ecl_grid.getCartesianSize(); g++) { |
There was a problem hiding this comment.
Using g for cell index is a little weird...? Other than that trivial observation, I'll try to run this through some cases to see if it is triggered.
There was a problem hiding this comment.
I like using g for global_index - happy to change it thougth.
| @@ -231,6 +231,14 @@ namespace cpgrid | |||
| // Make the grid. | |||
| processEclipseFormat(g, nnc_cells, z_tolerance, false, turn_normals); | |||
| } | |||
|
|
|||
There was a problem hiding this comment.
Should we maybe deactivate this test in production by #ifndef NDEDUG
There was a problem hiding this comment.
Better safe than sorry - I would prefer to leave that test in - even in release mode; can add a commented of #ifndef NDEBUG - which can be activated at some point in the future.
| { | ||
| const auto& ecl_actnum = ecl_grid.getACTNUM(); | ||
| for (std::size_t g = 0; g < ecl_grid.getCartesianSize(); g++) { | ||
| if (ecl_actnum[g] != g.actnum[g]) |
There was a problem hiding this comment.
Are you sure this compiles? Looks like you are calling a member function on a g which is of type size_t here.
There was a problem hiding this comment.
Hehe - no this probably does not compile ... :-(
joakim-hove
force-pushed
the
invalid-to-deactivate-cells
branch
from
630b3b3
to
3eb9dc3
Compare
joakim-hove
force-pushed
the
invalid-to-deactivate-cells
branch
from
3eb9dc3
to
9f861fd
Compare
This switch is added to be able to support the grid construction from opm-upscaling.
Related to: OPM/opm-common#1059
Observe that the uglyness in the second commit is to support opm-upscaling: OPM/opm-upscaling#279