SANS support reduction in event mode

[Gesner Passos]

Larmor has the plan to collect data only in event mode. Besides, SANS2D has been collecting data in event mode, but the SANS interface uses only the histogram data inside it to perform the reduction.

The goal of this ticket is to use the data in event mode ensuring that the result is compatible with the current one that uses histogram data.

[Gesner Passos]

There was a considerable work in this ticket. So, I decided to split it, in order to help the tester to follow what was done and test this ticket.

[Gesner Passos]

re #8505: Correct loadMonitorsFromFile function SANSUtility

Changeset: 2780d987a2c5d1a4fe6408a8ee481e709c480831

[Gesner Passos]

re #8505: Small improvements on fromEvent2Histogram

Changeset: 0bce8889c4c16c8e399c765d6dfd7013525467be

[Gesner Passos]

re #8505: Avoid destructing the monitor workspace

SANSUtility: fromEvent2Histogram

Changeset: e7ef92989c15084598ba510248e4a77d7b23e9f1

[Gesner Passos]

re #8505: Preserve information of movements from event workspace

The conjoinworkspace does not copy the parameters, this means, that the centralization of the workspace would be lost. To avoid this, the method now copies the instrument parameters

Changeset: 17370e9dd21c8657070affffc679a50e263c9c21

[Gesner Passos]

re #8505: Avoid the 'special' behaviour of CropDetBank

This harmonise the behaviour of the ReductionSteps that, when need the sample, uses the input name of the worksapce given through execute method.

Changeset: b3d25d1ebc579e8b4221c60314be57d886992449

[Gesner Passos]

re #8505: Creation of the SliceEvent step.

Changeset: 6f4e9ada450fda6a07cabdeaec0177fe4be92355

[Gesner Passos]

re #8505: Introduces SliceEvent to Reduction steps.

Changeset: b8b864ab9085929c686bcd528c0c51ea61903486

[Gesner Passos]

re #8505: Allow testing Reduction of Event mode data.

by doing: ws22 = LoadEventNexus('22048') ici.AssignSample(ws22,reload=False)

Changeset: 1e80c01a6b24fdd1a1537e29ae04356fc402e906

[Gesner Passos]

re #8505: Fix issue of NormalizeToMonitor in EventMode

Changeset: 91064fea6f6f094756c1721a1fb609ea9ec242f1

[Gesner Passos]

re #8505: Improve the load to load in event mode

This has another improvement. It now uses Load to load the worksapce, allowing mantid to chose the best load to use. As a consequence, this may facilitate introducing others instruments in the future (for example McStats simulation is being considered).

Changeset: 74cadec0c757b4920c165a10dee3398dddb68cfa

[Gesner Passos]

re #8505: Relax the tolerance for the system test

I do not want to change the systemtest, but, because it now uses the EventMode data, instead of using the histogram, small changes can occur. I have changed the tolerance to require that the difference between the current data and the result of the reduction using the event mode data can not be greater than 1%.

Changeset: d63ec6d1c6b4b1a5f45e68fd9b9a6e12bab8f2cc

[Gesner Passos]

Tester:

This ticket changes the way we do reduction for SANS2D and prepare the way for LARMOR. We now propose using the event mode data from the data file. And in the next ticket, we will extend to allow them to slice the data in time.

I've prepared a script (compare_hist_event.py) to show you that if you provide an workspace directly you will maintain the same result we used to have. But, if you let the reduction to chose the loader, it will load the event mode data and the final reduced workspace will be just slightly different (Being honesty you will not be able to spot the difference by looking at the ploted lines).

Run this script and check that what I have said is true.

Them, you can also exercise the GUI (although it was not touched). You may use the check_gui_batchmode.csv with the MaskFile MaskSANS2DReductionGUI.txt, to execute three reduction. Check that the results are compatible and that the follow statement is 'almost' valid

trans_test_read == sample_test_rear - can_test_rear

[Gesner Passos]

Do not forget that you must update mantid and systemtests.

[Peter Parker]

Merge remote-tracking branch 'origin/feature/8505_slicing'

Full changeset: ​5935a2234e523dbda1d7e37f6c6a8af1de832ac5

[Peter Parker]

Merge remote-tracking branch 'origin/feature/8505_event_mode'

Full changeset: ​6fec68ed09d68c8f5e41703befbf1bc2a5db6c9c

[Gesner Passos]

Merge branch 'feature/8505_slicing' into feature/8528_slice_event

Full changeset: ​12a1b07ff54dbde427d706234760314f14c1df7a

[Stuart Campbell]

This ticket has been transferred to github issue ​9349