Opened 16 years ago
Closed 16 years ago
#314 closed defect (fixed)
set_quantity crashes
| Reported by: | ole | Owned by: | ole |
|---|---|---|---|
| Priority: | normal | Milestone: | |
| Component: | Functionality and features | Version: | |
| Severity: | normal | Keywords: | |
| Cc: |
Description
After reusing the mesh object with fitting, I observed examples where set_quantity crashed with a message about 'Could not find triangle for point'
However, if mesh is being recreated from nodes and triangles no exception is thrown.
I am investigating.
Change History (8)
comment:1 Changed 16 years ago by ole
comment:2 Changed 16 years ago by ole
A test revealing this problem was derived and committed in changeset:6244 The test will pass now, but if the flag in quantity.py selecting reuse of computational mesh for fitting is set to True it will fail to locate triangle for offending point.
comment:3 Changed 16 years ago by ole
- Owner changed from ole to duncan
comment:4 Changed 16 years ago by ole
In changeset:6488 I added another example boiled down from a real life crash which causes fit to crash due a problem with the building of matrix AtA. The point of crash is slightly different for two different datasets included in the test.
comment:5 Changed 16 years ago by ole
- Resolution set to fixed
- Status changed from new to closed
Changes fixing this ticket were implemented in changeset:6534 to changeset:6538 and changeset:6541
See also ticket:168 which reported a related problem.
comment:6 Changed 16 years ago by ole
- Resolution fixed deleted
- Status changed from closed to reopened
comment:7 Changed 16 years ago by ole
- Owner changed from duncan to ole
- Status changed from reopened to new
comment:8 Changed 16 years ago by ole
- Resolution set to fixed
- Status changed from new to closed
Optimised a bit more and cleaned up in search_functions in changeset:6544 and changeset:6545
When this is done, see to ticket:315