Ticket #1577 (closed: fixed)
CrossCorrelate input checking
Reported by: | Peter Peterson | Owned by: | Vickie Lynch |
---|---|---|---|
Priority: | minor | Milestone: | Iteration 25 |
Component: | Mantid | Keywords: | |
Cc: | Blocked By: | ||
Blocking: | Tester: | Michael Whitty |
Description
This algorithm has more required parameters than it lets on. Upgrade the algorithm so it rejects invalid inputs before trying to execute.
Change History
comment:1 Changed 10 years ago by Peter Peterson
- Owner set to Peter Peterson
- Status changed from new to accepted
comment:3 Changed 10 years ago by Peter Peterson
- Status changed from accepted to verify
- Resolution set to fixed
comment:4 Changed 10 years ago by Russell Taylor
Pete, now that you understand this algorithm 100%, would you like to write the missing unit test? :)
comment:5 Changed 10 years ago by Vickie Lynch
- Status changed from verify to reopened
- Resolution fixed deleted
New input error checking requires reference spectra to be cross-correlated with itself which means the wiki example will fail.
comment:6 Changed 10 years ago by Vickie Lynch
- Status changed from reopened to accepted
- Owner changed from Peter Peterson to Vickie Lynch
comment:8 Changed 10 years ago by Vickie Lynch
- Status changed from accepted to verify
- Resolution set to fixed
comment:9 Changed 10 years ago by Michael Whitty
- Status changed from verify to verifying
- Component set to Mantid
- Tester set to Michael Whitty
comment:10 Changed 10 years ago by Michael Whitty
- Status changed from verifying to closed
Some points based on observations in r[7139]
The documentation page says that ReferenceSpectra is the workspace index of the spectra to cross-correlate with. It would seem that this is actually the spectra number.
The same would seem to be true for the WorkspaceIndexMin and WorkspaceIndexMax properties. I know these are minor points, and not at all urgent, but I think it's important we're consistent with users when we use phrases such as this.
Also, getting error message like: "No Workspaces in range between0 and 7" - shouldn't this say "histograms" or "spectra data" rather than workspaces?
That the algorithm rejected my input for various reasons until I got it "right" says the ticket is complete though. Better than in [5988] when it would take a nonsense range, etc. I do think an update to the documentation is required though.
comment:11 Changed 5 years ago by Stuart Campbell
This ticket has been transferred to github issue 2424