Add a third additional spectra to a tosca workspace loaded through the C2E interface

[Peter Parker]

Timmy would like a third spectra added to any tosca workspace loaded through the Convert to Energy interface.

This third spectra would just be the average (mean) of the two that are currently loaded.

[Peter Parker]

Refs #4531. Third histogram (spectra) created for Tosca Runs via CTE.

"Summary" property with value "Average" added to TOSCA parameter file. Values of Average and Sum add respective histograms to the resultant workspace of any run loaded through the CTE interface.

Changeset: 9d477c492a41445c45ae553417e9a603166403f4

[Peter Parker]

Refs #4531 - Fix for failed unit test. :/

Changeset: 1af04d86f3cd5ec4502317c59f4c69af86d7dff2

[Peter Parker]

To test, load a Tosca run via the CTE interface, and make sure it has 3 histograms (one more than the usual 2).

[Martyn Gigg]

Refs #4531. Update system test result.

Changeset: b95f45cd270ce3dccdf12feefa484df8c5bd7a0b

[Peter Parker]

Refs #4531 - Fix so that we actually compute the average when asked.

Changeset: e0231641ce6945f86287b9dba93138538558569f

[Stuart Campbell]

Do you have a dataset that I can use ?

[Stuart Campbell]

I used TSC10076.raw from the old test data subversion repo.

I did get an error with permission denied writing the fixedGrp,xml file. This happens when the default save path doesn't have the trailing slash on it.

[Stuart Campbell]

Construct the path correctly. refs #4531

Use os.path.join() instead of simply adding the strings.

Changeset: 127c7bd785574a60de59fe58b7410e311d1e86bf

[Stuart Campbell]

I know someone else should really test this now - but the change was so small!

[Stuart Campbell]

I just noticed after closing this that the spectrum numbers in the output workspace don't make sense, they are 0,1,1 shouldn't they be 1,2,3 ?

[Peter Parker]

I noticed this earlier, and now have a fix for it, which is basically ripping out every single one of my changes in this ticket (!), and replacing it all with a very small change to one of the properties in the instrument params file. (Obviously leaving your changes intact.)

I'd need to mess on with system tests, etc and make sure everything works 100% so I'll implement the change tomorrow as it's quite late here.

[Stuart Campbell]

There was a ticket at some point to be able to add a 'target' spectrum ID to the group in the XML file. I'm not sure that the work was completely finished - but it might be an easier way to go and wouldn't need a big re-work of code.

If you want to close this and create a ticket on the next iteration then I am happy with that.

[Peter Parker]

Refs #4531 - Change method used to get third spectra.

An undo of changes made previously on this ticket, and replacing them with a simple change to the Workflow.FixedGrouping property value in TOSCA_parameters.xml.

This gives us the same functionality, but removes the problem of having more than one spectra with the same ID.

Change to System Tests to follow...

Changeset: 2880712915fc3967b06c2e7af412cae54952cfdb

[Peter Parker]

Refs #4531 - Update system test result.

Changeset: 6b28b3b14b2a583d51fcc2858111617e847c4f45

[Stuart Campbell]

The spectra now have different numbers, although (and this is a problem in GroupDetectors) the output spectra start at 0 rather than 1. This might cause some confusion (at ISIS) as spectra #0 has a special meaning there.

Otherwise, all looks good to me.

[Stuart Campbell]

This ticket has been transferred to github issue ​5378