Crash on Windows after baseline setting to None
| Project: | Multipeak Fit version 2 for Beta Test |
| Version: | IGOR.6.10.x-2.0.0-B07 |
| Component: | Baseline functions |
| Category: | bug report |
| Priority: | normal |
| Assigned: | Unassigned |
| Status: | patch (ready to commit) |
First I would like to say that the multi peak fitting package 2 works great and does the job very well!
I stumbled upon a bug when working under Windows (XP). I selected a part of my wave using the graph cursors, and select 'Use Graph Cursors' in the Multi peak fit window. Then I want to fit the selected area without a baseline (set baseline to 'None'). Then I did 'Autolocate Peaks Now'. Normally when you do this you already see the residuals coming. This time however, no residuals information was shown in the area that was selected by the cursors. When I opened the 'Res_smoothmua' wave, there are a lot of zeros (0) and a part of the wave there was nothing, probably the part that was selected by the curcors. There was nothing means: no number, no sign, no NaN, just empty fields. When I then hit the 'Do Fit' button, Igor crashes and Windows asks if I want to send Microsoft a bug report (which, of course, I didn't). I tried this several times, and every time it went like this. I hope you have enough information to re-create the problem so that you can find what goes wrong.
Good luck!
Jos
Updates
Ah, yes. I can reproduce this. As far as I can tell, it happens only if you select None as the baseline before clicking Autolocate Peaks Now. So that would be a work-around: click Auto-locate Peaks Now before you select None as the baseline.
Since I can reproduce the problem, I should be able to fix it. Thank you for the report.
| Status: | active | » closed |
OK- I have fixed the cause of the crash. It will be in the next release of Igor Pro 6.10 pre-beta. Monitor the Igor Mailing list for announcements.
I have also fixed the problem with the None baseline and Auto-locate Peaks. I have uploaded it as a patch to B07.
| Status: | closed | » patch (ready to commit) |
| Attachment: | MultipeakFit200b07P1.zip (38.1 KB) |
I should not have marked this closed. I have re-opened it, and marked it "patch (ready to commit)". I have attached a fixed version of the Multi-peak Fit 2.ipf procedure file to this bug report.