Open moritzdannhauer opened 9 years ago
It depends. There are various workarounds. Not sure we can modify tetgen code ourselves though.
Yes, changing tetgen code would not be an option. 1) However, we could make it clear that tetgen could take literally forever to create a mesh very explicitly in the documentation. 2) Can we put tetgen or any other module that likes to crash into something(!?) like a try ... catch structure. In case a crash was detected (catch), the tetgen instance gets removed safely / all memory is given back to the system (progress bar gets to 100%).
Unfortunately, tetgen tends to segfault. I'm familiar with how to detect and report the error on Unix type systems (see http://www.cplusplus.com/forum/unices/16430/ for example), but haven't looked into Windows.
I found a case that crashes pretty reliably. IF there are point clouds and surfaces as input, the supress surface splitting must be checked, otherwise tetgen and SCIRun crashes with a Segmentation fault.
Stale issue message
This issue is stale because it has been open 240 days with no activity. Remove the stale label or comment, or this will be closed in 60 days.
Is there any way to make sure if an external library crashes SCIRun5 will not be affected by it (=terminates it)?