[Muon] have psi GPD data can be read by muon interface

[Nick Draper]

Moved to the Backlog after R2.6

[Anders Markvardsen]

To be able to read in systemtests/Data/GPD900.nxs

[Arturs Bekasovs]

Can wait.

[Arturs Bekasovs]

This should be done by allowing opening the interface with facility set to SmuS (don't really now why not PSI) which the instrument belongs to. Then the instrument box should be populated correctly and allow to load GPD data.

Apart from that, GPD900.nxs looks like a common Muon Nexus v2 and should have all the information we need.

[Arturs Bekasovs]

SmuS stands for Swiss Muon Source, which is a part of PSI (Paul Scherrer Institute), so now it makes sense. :-)

[Arturs Bekasovs]

The file is loaded correctly if facility is set to SmuS.

Warning message when opening the interface should be removed though.

The problem with a given file is that it doesn't have any log values in it, which means we couldn't get our results table. This should be solved in #8163, as then we will at least have run_number to plot against.

[Arturs Bekasovs]

Trac playing tricks again.

[Arturs Bekasovs]

Refs #7367. Added support for SmuS facility to MuonAnalysis.

Plus made warning for unsupported facilities more informative.

Changeset: add752e01e8dc3f970de1d61acaf4625052cdc5c

[Arturs Bekasovs]

Tester:

First of all, make sure that MuonAnalysis shows a warning when your facility is set to anything which is not SmuS or ISIS. Check that warning is informative enough.

Check that "Load current run" button is disabled for any facility apart from ISIS.

Now, set your facility to SmuS, open MuonAnalysis and load systemtests/Data/GPD900.nxs. Verify that it loads successfully. The plot should look a bit like a sin function.

Now go through all the general analysis steps:

1. On the Home tab, update Time Zero to something like 0.5. Data file doesn't contain that information, so plot has some garbage at the front. This should be solved.
2. Go to Grouping Options tab. Add new pair called e.g. long_inv with Forward Group = bwd and Backward Group = fwd. Plot it, and notice that graph is mirrored on axis Y = 0.
3. Now go to Data Analysis tab. Just to make it more realistic, try to fit the data to function loaded from the string "name=UserFunction,Formula=sin(x * A) * B,A=4,B=0". This should produce a sin curve which fits the data quite nicely.
4. Now go to Results Table. GPD900.nxs doesn't have any log data in it, so the only "log" value we can use is Run Number. Make sure it is possible to create a table with your fitting results.

[Arturs Bekasovs]

Just realized that SmuS does not support Live Data loading, so need to disable "Load current run" for it.

[Arturs Bekasovs]

Refs #7367. Don't allow loading current run for SmuS.

Changeset: f4bea697798ed9fd2d6a4fd593afa5bbf2ddf1a7

[Arturs Bekasovs]

Not that important.

[Karl Palmen]

Test passes, but

(1) A grouping name with '_' causes plotting to fail on Windows. This could be fixed in a separate ticket. It worked with longInv instead of long_inv.

(2) When fitting, only the original graph and the sinusoidal fit are displayed. The error is not displayed. This might be because the error is optional and has not been chosen or it may be a problem that could be dealt with in a separate ticket.

[Karl Palmen]

Merge remote-tracking branch 'origin/feature/7367_muon_gpd_loading'

Full changeset: ​b9735c4ee1517fd41c81c158441b30ae42857904

[Stuart Campbell]

This ticket has been transferred to github issue ​8213