Several SetSampleMaterialproperties should provided as output properties

[Nick Draper]

The scattering values that are either provided or calculated should be passed out as output properties for use in scripts and visibility in the history window.

[Nick Draper]

re #7358 add output properties

Changeset: b0973d7c8cebf264b31c8402b9f5889a9e4ea447

[Nick Draper]

To test

1. run SetSampleMaterial using directly provided values (look at the wiki page for usage).
2. use the history of the resulting workspace to check the output property values
3. repeat using a chemical formula

[Peter Peterson]

Generally I think the ticket itself is invalid. The information calculated by the algorithm is stored in the Workspace->Sample->Material in the provided workspace and accessible through that as fully decorated python objects (see #6827).

Some comments. You need to implement what you deem is the "correct" one.

1. Rather than having duplicate output properties, set the calculated values in the existing ones with the same name (favorite option).
2. Make the output properties as a group, using setPropertyGroup and a heading like "calculated values"
3. Disable the output properties so people cannot enter anything in them.

[Nick Draper]

The ticket is not invalid, the request came from spencer who had users that wanted to be able to see the resulting values in something other than the log window, and were not willing to use python. Setting the calculated values as output properties allows them to be visible in the history view of a workspace. This also follows the precedent of GetEI.

The part I had missed and many thanks for Pete for picking this up, was that I had not correctly marked the direction of the properties as outputs. this way they should not appear on dialogs or the python signature.

[Nick Draper]

re #7358 correctly mark parameter direction

Changeset: ab0ebabde171a661bfe96da2dc24d81aed68f4d1

[Owen Arnold]

I've been testing this and had to reopen because the CoherentXSectionResult being returned doesn't look right. I had to create my own testing script for this ticket.

ws = Load(Filename='IRS26173')
SetSampleMaterial(InputWorkspace=ws,AttenuationXSection=2.56,ScatteringXSection=11.62,SampleNumberDensity=0.0849106, ChemicalFormula="H4-N2-C3")
sam = ws.sample()
mat = sam.getMaterial()

history = ws.getHistory()
setSampleMaterialAlgorithm=history.lastAlgorithm()
print mat.absorbXSection(), setSampleMaterialAlgorithm.getPropertyValue("AbsorptionXSectionResult")
print mat.cohScatterXSection(), setSampleMaterialAlgorithm.getPropertyValue("CoherentXSectionResult")
print mat.incohScatterXSection(),  setSampleMaterialAlgorithm.getPropertyValue("IncoherentXSectionResult")
print mat.totalScatterXSection(), setSampleMaterialAlgorithm.getPropertyValue("TotalXSectionResult")

In addition to the CoherentXSection not coming out right. There are the following issues that need to be addressed.

• There's also no unit test updates to cover the new functionality introduced here, so these new features have zero test coverage.
• The SetSampleMaterials wikipage example is wrong, because it is not possible to run the SetSampleMaterials algorithm (as per the example) without specifying a chemical formula.

[Nick Draper]

refs #7358 Fix the test failures

Specifically, one of the ouput properies was not being set. Unit tests checks of the output properties have been added. The failure of the example was due to a change made by Pete in a previous ticket, but I've update the help text.

Changeset: 1065a8b50a5beb1619384b7f5e91c28e79afd4c9

[Owen Arnold]

Code changes look OK. Unit tests in place, and passing, and my previous failing python script (see above) is now outputting what I expect.

[Stuart Campbell]

This ticket has been transferred to github issue ​8204