Create Algorithm to integrate up the detector hierarchy

[Stuart Campbell]

The initial example of this is for dividing data by a normalisation measurement where we want to actually divide each data pixel by the norm run integrated along the tube.

This would basically take the integral over the Nth level up the detector hierarchy (over the appropriate X-axis values) and then set the value for each pixel in this to be the integral.

This should take a parameter to specify how far up the detector hierarchy to go.

[Nick Draper]

"New" tickets moved at the code freeze of iteration 29

[Nick Draper]

Bulk move of tickets to iteration 31 at the iteration 30 code freeze

[Nick Draper]

Moved to iteration 33 at iteration 32 code freeze

[Nick Draper]

Moved at end of release 2.1

[Nick Draper]

Moved at the end of release 2.2

[Nick Draper]

Moved to milestone 2.4

[Nick Draper]

Stuart,

Is this still a requirement?

[Andrei Savici]

Algorithm to average over bank, tube, ... Refs #2710

Changeset: 6c601983e75c6eb90e18046bfd86a960cc2867ce

[Andrei Savici]

Merge branch 'feature/2710_average_tube' into develop

[Andrei Savici]

Added an extra include statement. Refs #2710

Changeset: a5893b90a54619fa38a1a01c273e610fd8c88d20

[Andrei Savici]

Merge branch 'feature/2710_average_tube' into develop

[Andrei Savici]

To test - load a file with an instrument that has a detector hierarchy. CNCS file is OK. Then try to run the algorithm, with different "LevelUp" parameters. 0 means that it just integrates, 1 it will average along the tube, 2 will average along an eightpack, 5 will average over the whole instrument. Go back to the original workspace and mask some detectors (in instrument view for example). Repeat the test. The masked detectors should be kept as masked

[Andrei Savici]

Merge branch 'feature/2710_average_tube' into develop into 6856_ConvertToDiffractionMDWS_v2

[Andrei Savici]

Merge branch 'feature/2710_average_tube' into develop into 6856_ConvertToDiffractionMDWS_v2

[Andrei Savici]

Merge branch 'feature/2710_average_tube' into develop into 6856_ConvertToDiffractionMDWS_v2

[Andrei Savici]

Merge branch 'feature/2710_average_tube' into develop into 6856_ConvertToDiffractionMDWS_v2

[Roman Tolchenov]

It seems to work but I have a couple of suggestions.

The x vectors aren't set in the output workspace. The arguments of the private methods aren't documented. The progress bar goes only up to 20%.

[Andrei Savici]

AverageByComponent small improvements. Refs #2710

Changeset: 55da71ba3413e570b61dd0f9800fd4fbc99ab7a7

[Andrei Savici]

Merge branch 'feature/2710_average_tube' into develop

[Andrei Savici]

Please test issues raised when the ticket was reopened

[Nick Draper]

works for CNCS, but fails for GEM and MAPS. There are files in the systemtest area to test this with.

Also this algorithm is doing two things in the one algorithm, one of which you would not assume by the name.

1. grouping spectra based in the instrument hierarchy (this I would expect).
2. integrating the resulting spectra (this I wouldn't expect AverageByComponent to do).

I suggest the following:

1. Make this algorithm just group spectra based in the instrument hierarchy (this I would expect). I would like to see it use GroupDetectors for that.
2. Create a second algorithm to use this and integration to acheive the result of both together.

The reason I'm suggesting this is that I think the 1st algorithm (still called AverageByComponent) would be usefull in other situations and quite widely used.

[Andrei Savici]

Renamed AverageByComponent to IntegrateByComponent. Refs #2710

Changeset: cf87d48d6c6555a7bc3147564723e9be5fcb35d3

[Andrei Savici]

Add to wiki description. Refs #2710

Changeset: abfd65d01ab6fbfaec038fac4c3e72a9c84ce697

[Andrei Savici]

Changed name, and error message. The rest of the comments will be dealt in ticket #7236

[Nick Draper]

re #2710 allow grouped workspaces

Changeset: 2daea1fcd57e17adb7bcb3a0604b1654cd4084b2

[Nick Draper]

re #2710 allow grouped workspaces

Changeset: 16277ce4b55e659bb8e90a1846f3100cac4100ba

[Nick Draper]

to tester:

From Andre:

to test - load a file with an instrument that has a detector hierarchy. CNCS file is OK. Then try to run the algorithm, with different "LevelUp" parameters. 0 means that it just integrates, 1 it will average along the tube, 2 will average along an eightpack, 5 will average over the whole instrument. Go back to the original workspace and mask some detectors (in instrument view for example). Repeat the test. The masked detectors should be kept as masked

Fron Nick:

Also test ISIS files such as GEM, HRPD and at least one other.

[Andrei Savici]

Algorithm to average over bank, tube, ... Refs #2710

Changeset: 6dca2b64bbf45b00fe17a87b2e17b393d1132a04

[Andrei Savici]

Added an extra include statement. Refs #2710

Changeset: 4c057cda26834d210d76dfa1944efc8f43a313f7

[Andrei Savici]

AverageByComponent small improvements. Refs #2710

Changeset: f8d22fae20b8b0e00072009b599ccf45d1983d3d

[Andrei Savici]

Renamed AverageByComponent to IntegrateByComponent. Refs #2710

Changeset: f498373300c5aa45d0719d82e02988a57f3c1783

[Andrei Savici]

Add to wiki description. Refs #2710

Changeset: 46e1f88c2d7e81cea15f78f9c7901af4f2942f1a

[Nick Draper]

re #2710 correct wiki doc

Changeset: 1d24ecd83cad691f9571cd6cbc04fe8711c011f8

[Nick Draper]

Corrected the wiki docs slightly, but passed testing with SNS and ISIS data.

[Stuart Campbell]

This ticket has been transferred to github issue ​3557