#16 closed defect (wontfix)
WHIZARD 1 bug in MD5 checksums
| Reported by: | Juergen Reuter | Owned by: | Juergen Reuter |
|---|---|---|---|
| Priority: | P4 | Milestone: | |
| Component: | core | Version: | 1.93 |
| Severity: | minor | Keywords: | MD5 checksums |
| Cc: |
Description
due to an error in the implementation of the MD5 checksums in WHIZARD 1 there is need to always use the "force" flag for the read statements. Furthermore, gfortran is forced to compile with --fno-range-check.
Change History (6)
comment:1 Changed 16 years ago by
| Milestone: | → v1.94 |
|---|---|
| Status: | new → assigned |
comment:2 Changed 16 years ago by
comment:3 Changed 16 years ago by
I tried to implement the MD5 checksums as they are in WHIZARD 2. Unfortunately, now the self-tests fail, probably because the character string for the MD5 calculation are read in wrongly. For the time being, I disabled all MD5 checking in order to make the 1.xx tree usable again.
comment:4 Changed 16 years ago by
| Milestone: | v1.94 → v1.9x |
|---|
In r1224 I resetted the state back to the one in 1.93. Furthermore, read_grids_force and read_events_force are now set to .true. as default values. Fixing MD5 sums has been moved from 1.94 to 1.9x milestone.
comment:5 Changed 15 years ago by
| Resolution: | → wontfix |
|---|---|
| Status: | assigned → closed |
This is a known bug of WHIZARD series 1.x that will not be fixed. The force flags are set to true as a default such that the MD5 checksums will not be checked at all. Closing the ticket.
Strangely, this error occurs when running the WHIZARD binary directly or doing 'make run' in the results directory where the corresponding grid or event file resides, but _not_ if doing 'make run' in the main directory.