Opened 7 years ago
Closed 7 years ago
#608 closed defect (fixed)
Numerical issue in Preflow with floating-point data
| Reported by: | Peter Kovacs | Owned by: | Peter Kovacs |
|---|---|---|---|
| Priority: | major | Milestone: | LEMON 1.4 release |
| Component: | core | Version: | hg main |
| Keywords: | Cc: | ||
| Revision id: |
Description
The attached patch contains the fix and test cases to reproduce the issue.
Other changes are also included to improve the test file.
Attachments (2)
Change History (7)
Changed 7 years ago by
| Attachment: | 608-all-538d666d8cb5--55febc0faa6d.patch added |
|---|
comment:1 Changed 7 years ago by
comment:2 follow-up: 3 Changed 7 years ago by
I cannot apply 608-all-538d666d8cb5--55febc0faa6d.patch with --exact, mercurial says this patch is corrupt.
Changed 7 years ago by
| Attachment: | 608-all-e0ccc1f0268f--8db773f19586.patch added |
|---|
comment:3 Changed 7 years ago by
Replying to alpar:
I cannot apply 608-all-538d666d8cb5--55febc0faa6d.patch with --exact, mercurial says this patch is corrupt.
I'm sorry, that's because of a strange issue with Mercurial (see also #602). I recreated the patches, the new versions can be imported with --exact.
- [e0ccc1f0268f] Remove unused typedefs in max_flow_test.cc (#608)
- [e2732b9da429] Refactoring and code formatting in max_flow_test.cc (#608)
- [e018899c2926] Use tolerance in max_flow_test.cc (#608)
- [259e3a90ad97] Improve max flow test method: set expected flow value (#608)
- [8db773f19586] Fix tolerance usage in Preflow algorithm (#608)
The actual bug fix is the last one. The previous ones are refactoring and extensions for the test class to allow the addition of the new test cases that reveal the bug.
comment:5 Changed 7 years ago by
| Resolution: | → fixed |
|---|---|
| Status: | new → closed |
The bugfix has been merged to the main branch, as well as into 1.3.
Attached changesets: