Handle the case where loading same inst. data with diff number hist

[Anders Markvardsen]

Steve have come across a case that we may not have tested/considered before that is loading: same instrument data in succession where the number of histogram varies.

Make muon interface handle this according to Steve's email below:

There seems to be a new problem with the grouping in the Muon Analysis interface. When loading a run with x histograms and then a run with y histograms, the group table is not updated correctly when the second run is loaded if x<>y and the interface becomes unusable. Try this with the attached files … 30604 has 192 histograms and 34997 has 96 histograms. Load 30604 and create four groups, 1: 1-48, 2: 49-96, 3: 97-144, 4: 145-192, and a two pairs (1-2) and (3-4). All works fine. Now try and load the second file and there’s an error ‘Unable to group loaded workspace’ … if the number of histograms has changed the previous grouping should be abandoned. In fact, you still end up with all the groups showing, so I’m not sure what’s going on inside at this stage (which probably explains why it seems to stop working). I got into trouble with this bug earlier today going the other way, i.e. I loaded a run with 32 detectors and then a run with 96 detectors … it carried over the grouping from the first run to the second which made the data look very strange.

[Nick Draper]

Bulk move of tickets out of triage (new) to assigned at the introduction of the triage state

[Arturs Bekasovs]

The interface currently loads the new data to the used workspace directly. It should be much easier when I load it to a tmp workspace. I will then just compare the number of detectors of the loaded one and the current one.

[Arturs Bekasovs]

Refs #8860. Re-load grouping if no. spectra changes.

Changeset: 92aaa279288a9a5d3c9501a4284b77085f431875

[Arturs Bekasovs]

Tester:

Check that what Steve talks about in the description is not happening any more:

1. Save all attached files to the disk.

2. Launch MuonAnalysis interface.

3. Load 30604 (with 192 histograms).

4. Go to Grouping Options and load emu_192_grouping.xml grouping. Try to plot the group4 to make sure everything was grouped correctly.

5. Go back to Home and try to load 34997 (with 96 histograms). It should be loaded fine and the grouping should be replaced with the one with 96 detectors. Try to plot something to make sure it was grouped correctly.

6. Load emu0006473 from AutoTestData. It should be loaded fine, 96 detector grouping should be abandoned, 32 detector one should be loaded.

7. Load downloaded 34997 again. 32 detector grouping should be abandoned for 96 detector one.

8. Load MUSR00015189 from AutoTestData. MUSR grouping should loaded (check the description on the Grouping Options).

[Arturs Bekasovs]

Grouping for EMU data with 192 histograms

[Arturs Bekasovs]

Purely interface stuff, so system tests couldn't possible be broken.

[Karl Palmen]

Grouping appears to behave sensibly and there are no 'Unable to group loaded workspace' errors.

[Arturs Bekasovs]

Refs #8860. Re-load grouping if no. spectra changes.

Changeset: 6542373d6cdf6e1856363f0eb2a06c1fcdf40680

[Karl Palmen]

Merge remote-tracking branch 'origin/bugfix/8860_muon_same_inst_diff_no_spectra'

Full changeset: ​c6ff3b3e9efb45df34d1c657a635a069b5ae6514

[Stuart Campbell]

This ticket has been transferred to github issue ​9704