RebinByTimeAtSample first implementation

[Owen Arnold]

Rebin a TOF event workspace according to TatSample = PulseTime + TOF * L1/(L1 +L2)

Make the documentation clear that the first implementation of this is only to work for the elastic case. The inelastic cases will have to be added later.

[Owen Arnold]

refs #10102. Sort by time at sample.

Changeset: 2f7078388a6235a5fec084501e595e25bcb06864

[Owen Arnold]

refs #10102. Consolidate calcuations.

The splitters vectors, and the new sorting require the same time at sample calculation. I've made the calculation into a free funciton, which is used by both. This should help at a very fundamental level, because the rebinning (new working on) functionality should match the filtering functionality.

Changeset: e841d5a4c79bf29883eba3773f8277c54d7f119b

[Owen Arnold]

refs #10102. Filtering implemented and tested.

Changeset: 674828d318a7499ac247a0108941aad6fee13edc

[Owen Arnold]

refs #10102. Implement and test histogramming.

We need to be able to histogram according to time at sample.

Changeset: 67493b534803aeb9d560bbab1c5a75f738077105

[Owen Arnold]

refs #10102. Find min and max timeatsample in workspace.

Implemented and tested. This will allow us to find the X-range for our rebinning if the user does not supply all arguments to params.

Changeset: b38d28f2dd487f451153613d3c19debbff6b4ed9

[Owen Arnold]

refs #10102. RebinByTimeAtSample - works like RebinByPulseTime.

Added new algorithm. Shamelessly copy pasted a large portion of this (including the tests). Later refactoring step will be required to really reduce the duplication here, both for the tests and for the algorithms themselves.

Changeset: 1e2c6559c954b7ddd482eedb43781b78e1540347

[Owen Arnold]

refs #10102. Base class for algorithms common

Changeset: 080f99b9254988154bd5bf40ea4a01749d083ea5

[Owen Arnold]

refs #10102. Move more to the base class.

Changeset: a980d64fa6fd89a51c9e051bcf7de1c61f780ac1

[Owen Arnold]

I still need to refactor and extend tests, and update the rst for the new algorithm.

[Owen Arnold]

refs #10102. More reuse achieved.

Made the method to find the min and max values across all spectrum virtual and therefore polymorphic amongst each subtype of the base algorithm.

Changeset: 7a6434f42399d4bbc10aec4f244cf64e0c141fda

[Owen Arnold]

refs #10102. Refactor to base test class.

Templated base class taking algorithm as type argument. Shared test code is much neater now. Duplication is at an absolute minimal.

Changeset: 915fe44cd8f8a018f6bede60f473739d6a17252c

[Owen Arnold]

refs #10102. Add sample time filtering test.

Discount the PulseTime entirely in a unit test and use the L1, L2 distances as the control factors for the filtering.

Changeset: bbaa6f175e819c0d15c39d09e74261fea9960571

[Owen Arnold]

refs #10102. Explicitly provide template argument.

Changeset: ab399210d9b0079e1f490ff1e81bcc373564f139

[Owen Arnold]

refs #10102. Documentation.

Changeset: 0234c3cb271817184ad71600ed5a3576dccb4e6c

[Owen Arnold]

refs #10102. Kill type conversion warning.

Changeset: 4ec5f8f94d2ab7191df99ca361c6da682e65f24e

[Owen Arnold]

refs #10102. Amend documentation.

Changeset: 764192025edf750afc6027df16376e6bb3f693cd

[Owen Arnold]

refs #10102. Fix doxygen and cpp warnings.

Changeset: 702938e0f3ce4e467a95e00f36f32408078f2df4

[Owen Arnold]

Tester. This algorithm is ready, although usage examples with real data are hard to supply because both the pulse time and the event tof are taken into account, as well as the instrument geometry. I suggest the following:

• Code review the changes
• Check the functional unit tests, which do a good job of establishing whether the rebinning is working properly by using virtual instruments around fake event data.

I will be working with users here at ISIS who want to plot in absolute time. so there will be further opportunities to verify this prototype.

[Martyn Gigg]

There is a Sphinx warning about in RebinByTimeAtSample-v1.rst: ​http://builds.mantidproject.org/job/develop_clean/label=rhel6-build/287/warnings3Result/. It looks there's an extra 's' on :refs: when it should be :ref: on line 47

[Martyn Gigg]

There are also a couple of compiler warnings in EventList.cpp: ​http://builds.mantidproject.org/job/develop_clean/label=ubuntu-12.04-build/287/warnings19Result/

[Owen Arnold]

refs #10102. Fix sphinx warnings

Changeset: 9d6f74fcd97625597fba12b8ac4d8321d6830317

[Wenduo Zhou]

It works well. So the ticket is closed.

[Owen Arnold]

Merge remote-tracking branch 'origin/feature/10102_rebin_by_t_at_sample' into feature/10102_rebin_by_t_at_sample

Full changeset: ​08c6337aef5f698799f3adff71a14f2ef32681fd

[Wenduo Zhou]

Merge remote-tracking branch 'origin/feature/10102_rebin_by_t_at_sample'

Full changeset: ​9013d2f543857185ec179cdff42320f56ad3f8ea

[Stuart Campbell]

This ticket has been transferred to github issue ​10944