Closed georgmichel closed 7 years ago
Well this still contains the changes you should not have in the commit. But I took the liberty of just correcting it myself. If have pushed the corrected commit... Please have a look and chat that it is alright. The only disadvantage I see is that this change now credits me as the author which is of course not true. But since your patches all did not have a proper user name and email in the first place, I hope that you don't mind... I'm closing all PR's
OK, I think the commits are now as I intended. Obviously I still have to master git. But I prefer fixing bugs in openEMS. At least I have now updated my profile so the name and Email address should show up correctly next time. In my latest pull request it looked like there was only the "return true;" and nothing else for openems.cpp. Anyway, what matters is the improvement of the software and I don't care whether my name is mentioned for the patches.
Well this are the cases where I like to use git gui to create the commits. You can just easily skip parts of files or select only parts for committing. And usually I also use gitk to look at the tree and review my changes before pushing them. I'm looking forward to more bug fixes and improvements ...
You mean like this?
BTW the spurious chrash of the highest rank MPI instance occurs only after the final return in main (which is somehow mysterious) and only if there is a NF2FF box in the model. So probably this is due to a destructor relating to nf2ff (and MPI).