C2E: Swap fixed grouping to use a file.

[Samuel Jackson]

We have a workflow parameter in our parameters files called Workflow.FixedGrouping which allows comma separated ranges to be defined for detector groupings. This routine then creates the mapping file on the fly for every run.

There are only two instruments that use this parameter (TOSCA and VISION). We should remove this method (possibly moving the code for generating a grouping file with overlapping groups to an algorithm) and change it to using the Workflow.GroupingFile parameter and create a grouping file (like the one currently used BASIS).

[Samuel Jackson]

Not only is this code not efficient, it appears that the mapping between workspace indices to detector IDs is just plain wrong.

[Samuel Jackson]

Refs #9621 Remove fixed grouping. Use file definition instead.

Changeset: 99fe1f0bde14a684ba502bb2f135ca59f3f2ff51

[Samuel Jackson]

Refs #9621 Update Parameter files and add grouping files.

Changeset: 6f3a915f85b98aaafd97a7a28a6a77588c5d1f21

[Samuel Jackson]

Merge remote-tracking branch 'origin/master' into feature/9621_fixed_grouping_file

Refs #9621

Conflicts:

Code/Mantid/scripts/Inelastic/inelastic_indirect_reduction_steps.py

Changeset: 19961ca2ab76240fcddcc0b10f251e279b0fe6d1

[Samuel Jackson]

Refs #9621 Update reference result.

Changeset: 73fdcd759fdee4ffd6aaad985457fec065dff52e

[Samuel Jackson]

Merge remote-tracking branch 'origin/master' into feature/9621_fixed_grouping_file

Refs #9621

Conflicts:

SystemTests/AnalysisTests/ReferenceResults/II.TOSCAReductionFromFile.nxs

Changeset: d917c384cef147e7770e6c1ef1865cd9810dd34b

[Samuel Jackson]

Refs #9621 Swap to using spectrum numbers.

This follows a more natural mapping with the interface and instrument scientists perception of what range they are using.

Changeset: c2f6b398837bb62b4daf0594b2b87e861ee222e5

[Samuel Jackson]

TOSCA energy reduction was writing out a grouping file based off of some workspace index lists defined in the parameter file for every single sample run, but with some spectra removed depending on the masking used.

This has now been removed and we're using a fixed grouping file and the MaskDetectors algorithm to mask noisy detectors.

To Test

The output from TOSCA energy reduction should be the same regardless of the changes. I would do the following:

• On the 3.1 version of Mantid run an energy conversion with a TOSCA data file (there is one in the AutoTestData with prefix: TSC). Keep all options as default. Save this file somewhere sensible

• Now open a version with the changes applied. Do the exact same thing as step one, then compare the results and check they are equal.

• Check system tests are passing and do a code review

Remember to merge both the main repo and the system tests

[Peter Parker]

Merge remote-tracking branch 'origin/feature/9621_fixed_grouping_file'

Full changeset: ​22043d45386c1f03a40b28fc01f398ebf913d153

[Peter Parker]

Merge remote-tracking branch 'origin/feature/9621_fixed_grouping_file'

Full changeset: ​45ba9320449d628177564cdc2116164a61993a52

[Peter Parker]

Nice changes. We actually had this on our wish list when I was here last, so it's good you spotted it and finally got it cleared up.

[Stuart Campbell]

This ticket has been transferred to github issue ​10464