The non-cartesian connections are required by the output facilities,
while the truly non-neighbour connections (meaning those not in the grid)
should be used by all other code.
* FlowMain now gets NNC by `geoprops_->nnc()`
* Refactored update()
* Moved big chunk of unused code to new private method pinchProcess_
* Reordered logic to unite the NNC logic
* Compute NNC by face2cell information from a grid
* Pass NNC information to EclipseWriter
* Made NNC respect ECLIPSE input data
* nncStructure() is now exportNncStructure()
the hydrocarbonstate is stored in the reservoir state and
used instead of the primalvariable_
The initial hydrocarbonstate is computed using
initHydroCarbonState(...);
Enables flow to accept a basename for a case by appending a .DATA suffix
should it not be provided. It already supported reading the basename
from a .DATA extension file, but not opening said file by handing it to
the parser.
opm-parser#677 changes the return types for the Deck family of classes.
This patch fixes all broken code from that patch set.
https://github.com/OPM/opm-parser/pull/677
Since the support for threshold pressures running Norne with
flow_mpi aborted error messages like
```
Program threw an exception: [/home/mblatt/DUNE-test/opm-autodiff/opm/autodiff/BlackoilModelBase_impl.hpp:383] Illegal size of threshold_pressures input ( 153924 ), must be equal to number of faces + nncs ( 78316 + 0 ).
```
This commit now distributes the threshold pressures (if present) just like the rest
of the model properties and Norne does not abort here any more.
Please note:
1. If there are NNCs flow_mpi will abort with an error.
2. We might want to resort to reading and calculating the threshold pressure
(and maybe other properties) on distributed grids instead of using communication.
Several files stopped compiling due to relying on opm-parser headers doing
includes. From opm-parser PR-656 https://github.com/OPM/opm-parser/pull/656
this assumption is no longer valid.
This was left out when rebasing (it had been introduced in the flow etc. mains).
This also fixes a bug: the nnc thresholds should be inserted at the end of the
vector to be consistent with the operators, not at the beginning.
