Small bugs in the SANS reduction software

[Steve Williams]

I some times recieve bugs that Richard Heenan and Stephen King sent to Mantid help.

[Steve Williams]

Problem with the labeling of monitors in the SANS GUI

Subject: Re: [Mantid-help] SANS GUI: fatal crash bug
Ahhh… I think I understand why now!
 
Under the analysis tab in bank selection there are two boxes to specify spectrum numbers.  By chance I hit reduce without changing the spectrum for transmission monitor (as I had been doing yesterday) and everything worked fine and looked right.  When I realized what I’d done I changed the spectrum number for the transmission monitor from 2 to 3 as I’d been doing and Mantid bombed.
 
The way I’d interpreted these boxes – and I think anyone without a priori knowledge would given the wording alongside - was that these were the spectrum numbers containing the incident beam monitor spectrum and the transmission monitor spectrum, respectively.
 
Whereas, what I now suspect is that these boxes are actually BOTH supposed to specify incident beam monitor spectra but one to be used normalizing the SANS data and one to be used normalizing the TRANS.  Can you confirm if this is the case, please?!
 
If you concur, can I suggest we change the wording such that:
 
Monitor spectrum -> Incident monitor (SANS)
Transmission monitor -> Incident monitor (TRANS)
 
Thanks,
Steve
 
From: Draper, Nick (-,RAL,ISIS) 
Sent: 02 July 2010 15:24
To: Mantid Help; Williams, Steve (STFC,RAL,ISIS)
Cc: King, Stephen (STFC,RAL,ISIS)
Subject: RE: [Mantid-help] SANS GUI: fatal crash bug
 
Steve, Can you have a look at  this please.
 
Regards,
 
Nick Draper
Mantid Project Manager
R78, 1.02
 
Phone +44 1296 445228
Skype: nicholas_draper
 
________________________________________
From: mantid-help-bounces@mantidproject.org on behalf of stephen.king@stfc.ac.uk
Sent: Thu 7/1/2010 5:33 PM
To: Mantid Help
Cc: King, Stephen (STFC,RAL,ISIS)
Subject: [Mantid-help] SANS GUI: fatal crash bug
I’ve discovered a bug that crashes Mantid.
SANS2D /cycle_09_2 data.  Mask attached.
Sample SANS: 1017, Sample TRANS: 990, Sample DIRECT: 1002
Trans monitor is Sp 3
Mantid falls over during the transmission calculation. I think it is because the transmission runs have slightly different chopper phasing (our fault) so it hits a divide by zero.  But the condition is not trapped.
 
Steve

[Steve Williams]

(In [5079]) When CalculateTransmission is called from a SANS reduction the same monitor can't be specified as both the incident and transmitted re #1364

[Steve Williams]

(In [6059]) The SANS interface now checks if the Python MaskFile() function completed successfully, if not an error is shown to the user. This improves the situation when the LoadEmptyInstrument() takes a long time to load blocking Python closes #1364

[Mathieu Doucet]

(In [6064]) Copied Steve's edits to the new ISISReducer, and corrected minor bugs. Re #1364.

[Mathieu Doucet]

(In [6070]) Added TransmissionSample and TransmissionCan. Re #1364.

[Mathieu Doucet]

Punched in the wrong ticket ID for [6070]. Meant #1447

[Mathieu Doucet]

(In [6091]) Added AssignSample. Re #1364.

[Mathieu Doucet]

(In [6092]) Added AssignSample tests. Re #1364.

[Martyn Gigg]

Confirmed fixed.

[Stuart Campbell]

This ticket has been transferred to github issue ​2211