Ticket #8614 (closed: fixed)

Opened 7 years ago

Last modified 5 years ago

fix abs units reduction for LET when mono vanadium is measured in event mode

Reported by: Alex Buts Owned by: Alex Buts
Priority: major Milestone: Release 3.1.1
Component: Direct Inelastic Keywords: PatchCandidate
Cc: tatiana.guidi@… Blocked By: #8640
Blocking: #9061 Tester:

Description


Change History

comment:1 Changed 7 years ago by Alex Buts

  • Status changed from new to inprogress

refs #8614 Should fix the issue.

Though careful checks are needed. (and system tests as well)

Changeset: a41f3e45c2a0abbe6a1c9f93f9a26d6140265ad8

comment:2 Changed 7 years ago by Alex Buts

  • Cc tatiana.guidi@… added
  • Milestone changed from Backlog to Release 3.1

comment:3 Changed 7 years ago by Alex Buts

refs #8614 temporary reverted changes

as they cause strange behaviour

Revert "refs #8614 Should fix the issue."

This reverts commit a41f3e45c2a0abbe6a1c9f93f9a26d6140265ad8.

Changeset: ca613e94b45365e6d927368fbdca3c4ca73a567e

comment:4 Changed 7 years ago by Alex Buts

refs #8614 trying to nail down reason for MARI tests failing

Changeset: ee3380c534b3125f07472a0e80d472aa63861a7d

comment:5 Changed 7 years ago by Alex Buts

refs #8614 Now it works as expected for LET

and changed LET_Parameters.xml to have default nexus files to be nexus, not event nexus files.

Changeset: 66b5ea4aa97f04f804483c9fbc173bc6871f68d5

comment:6 Changed 7 years ago by Alex Buts

refs #8614 Minor comments

Changeset: 1f6b9b2e470fae1e6cbe92b51a36d4096e0e2627

comment:7 Changed 7 years ago by Alex Buts

refs #8614 Removed unnecessary parameter in instrument definition file

Generic loader now distinguish between nexus and event nexus file.

Changeset: 30d851b62df4c23388026c94451c0a38a329f567

comment:8 Changed 7 years ago by Alex Buts

  • Blocked By 8640 added

comment:9 Changed 7 years ago by Alex Buts

refs #8614 Dealing with consequences of ill defined merge

Some code pieces came not valid

Changeset: b969201684376b5cab33fb2bd975a86151c69713

comment:10 Changed 7 years ago by Nick Draper

  • Component changed from Framework to Direct Inelastic

comment:11 Changed 7 years ago by Nick Draper

  • Milestone changed from Release 3.1 to Backlog

Moved to backlog at the end of Release 3.1

comment:12 Changed 7 years ago by Alex Buts

  • Milestone changed from Backlog to Release 3.2

comment:13 Changed 7 years ago by Alex Buts

refs #8614 conflicts between ticket and recent changes to reduction

Merge branch 'master' into bugfix/8614_letReduction

Conflicts:

Code/Mantid/scripts/Inelastic/CommonFunctions.py Code/Mantid/scripts/Inelastic/DirectEnergyConversion.py

Changeset: 21d76d4896d0e2b00b29b62e3f22f8bc845506bf

comment:14 Changed 7 years ago by Alex Buts

refs #8614 finished with merge errors

moved copy spectra function to DirectEnergyConversion and prepared to unit test these

Changeset: 29c44a72e62daced26c4aaa41621b6e205d97cca

comment:15 Changed 7 years ago by Alex Buts

refs #8614 spectra_list property for reduction

and unit test for this property.

Added new property which contains list of the monitors to copy to detectors in event mode. Testes assignment for this property

Changeset: ee898da84d1bd39e63a792a911a208f98a2b0e7f

comment:16 Changed 7 years ago by Alex Buts

refs #8614 renamed copy_specta_to_monitors property

to spectra_to_monitors_list -- hope it is more vocal and does not interfere with copy_spectra2monitors function

Changeset: 1fb265745485fe163db9c116eda84c2c996df081

comment:17 Changed 7 years ago by Alex Buts

refs #8614 looks like all working fine

need to run system tests and add system test for this kind of workflow.

Changeset: 933cea9d8a5a9b6b72e9ae1d36f9a6a1e57cce76

comment:18 Changed 7 years ago by Alex Buts

refs #8614 minor changes (comments, log formats)

Changeset: 56ed7d56e19d781800bedf457b5a4d77ffb15ac7

comment:19 Changed 7 years ago by Alex Buts

refs #8614 should fix broken system test

Changeset: 93e052a46516c21a3d3e0e03b7f63f904cc7b3e0

comment:20 Changed 7 years ago by Alex Buts

refs #8614 fixed invalid get method for spectra_to_monitors_list

and unit test checking it is correct

Changeset: d8dbd59b339b10d8ab6f20e7e5319870ae4e97a8

comment:21 Changed 7 years ago by Alex Buts

refs #8614 Minor: comments and output formatting.

Changeset: 45d013e8b732eca83dade36e0d79a2033c070a9a

comment:22 Changed 7 years ago by Alex Buts

refs #8614 Attempt to remove part of generic code out of LET test

Changeset: 57667a26d6fa21e245fc3f539995562476e363cb

comment:23 Changed 7 years ago by Alex Buts

refs #8614 new system test for recent LET

Changeset: 1c54adb1201655e3651843791b54a90eca2e6955

comment:24 Changed 7 years ago by Alex Buts

refs #8614 Minimal set of changes to LET system test

Here are the changes to system test caused by GIT refusing to accept large raw file. I use existing raw wb file (though the data would be physically incorrect) and need to modify the reference results. As the results are physically incorrect, I am trying to make minor data set relating to the existing one and still testing the reduction

Changeset: db71fdd3c0232295a110500f0741b1072c38357c

comment:25 Changed 7 years ago by Alex Buts

refs #8614 Changes to stresstesting allowing checkin mutlirep

and ISISDirectInelastic working with old raw/map/mask files

Changeset: b7e4e5e9e098e4f1c39a2110167cfb557d430919

comment:26 Changed 7 years ago by Alex Buts

refs #8614 data files for event multirep mode

Changeset: 75565bca3f7def63d908c496b6e6f184978f3dca

comment:27 Changed 7 years ago by Alex Buts

refs #8614 modified reference result files

using old mask map and det.dat file

Changeset: 8fdf1b160c45afa55d245e85c3076c13fc4a3c40

comment:28 Changed 7 years ago by Alex Buts

refs #8614 fixed unit test

Changeset: e976dc0cd428f5044d3cf25d755c4c3162582c12

comment:29 Changed 7 years ago by Alex Buts

refs #8614 Trying to fix system test by rounding integration limits

Suspected that different results on different platforms come from different round-off procedure and different integration limits as the result.

Changeset: f2968b086581e3aab7e5e9f9f505a8f38828e67b

comment:30 Changed 7 years ago by Alex Buts

refs #8614 fixing case of single ws provided.

Changeset: 8b6a2313a343db438dbd8ad95a333e11197eaf1a

comment:31 Changed 7 years ago by Alex Buts

refs #8614 Trying to fix system tests on Unix.

It is certainly binning issue but needs further thinking on how to deal with it on different machines.

Changeset: ad3f137f2d95b7f7c3600de1fefad89a8ca30d0a

comment:32 Changed 7 years ago by Alex Buts

refs #8614 it looks like it is wrong comma.

Changeset: 7cf81efe0bf264cefad6e37b0d626afaedacaa21

comment:33 Changed 7 years ago by Martyn Gigg

Can I suggest a few bits of tidying up for the code in stresstesting.py:validateWorkspaceToNeXus(self):

  • nunRezToCheck=len(valNames) --> numResToCheck = len(valNames)
  • WorkspaceName=None; --> mismatchName = None
  • validatrionRez=True; --> validationResult = True

They basically just tidy up some spelling.

comment:34 Changed 7 years ago by Alex Buts

refs #8614 Spelling corrections in stresstesting

Changeset: b54addf1c13e286b2daf8d8b8592ccfac1ef3c38

comment:35 Changed 7 years ago by Alex Buts

refs #8614 typing error

Changeset: cf90937aec497bffdd85ed64793219ecd7fc01da

comment:36 Changed 7 years ago by Alex Buts

  • Status changed from inprogress to verify
  • Resolution set to fixed

This ticket is about bringing changes I made on LET for absolute reduction to work and converting the reduction currently used on LET into a system test. There are small changes to the reduction and stress testing to fix merge conflicts and allow multirep mode testing.

Changes are both in Mantid and in SystemTest branch.

As this reduction is currently used on LET, it is not much to test. IS believe that code is generally right but a code review would be helpful.

I tried to transform the reduction script to highlight further changes necessary to reduction but ultimately, all these changes are subject to discussions with instrument scientists who introduced the initial changes.

comment:37 Changed 7 years ago by Alex Buts

refs #8614 Different OS interpret return by reference in Python

differently.

Changeset: fd900c5748042f3d087c8f1af3dbffd435e58d85

comment:38 Changed 7 years ago by Alex Buts

refs #8614 Should fix discrepancies in running ISISDirectInelastic

on dirrefent OS-es (though sample may need to be changed too)

Changeset: 489ba0f66ba0fe4e8260b516c42fadb8bfa962b2

comment:39 Changed 7 years ago by Alex Buts

refs #8614 Changed sample LETmultirep reduction results

to be compatible with det.dat file written in nxs format.

Changeset: a392dd6f248f594fbbda774a7d1319493861effa

comment:40 Changed 7 years ago by Alex Buts

refs #8614 fixing LoadLotsOfFiles system test

Changeset: 23c64cd9578de228ffbb823ff1b51fd25ab4f515

comment:41 Changed 7 years ago by Alex Buts

  • Keywords PatchCandidate added

I am adding this ticket as PatchCandidate because its initial form is currently used on LET and Merlin together with Mantid 3.0

Update to 3.1 would kill current reduction and these changes have to be brought into to installed Mantid manually.

comment:42 Changed 7 years ago by Alex Buts

  • Blocking 9061 added

comment:43 Changed 7 years ago by Alex Buts

refs #8614 missing comma in LoadLotsOfFiles

Changeset: 9d8540f2c521bf612f7dfe968dec6053fbda21b3

comment:44 Changed 7 years ago by Nick Draper

  • Milestone changed from Release 3.2 to Release 3.1.1

Moved to patch release 3.1.1

comment:45 Changed 7 years ago by Nick Draper

  • Status changed from verify to closed

Merge remote-tracking branch 'origin/bugfix/8614_letReduction'

Full changeset: e374647b3780a922cfba34fbf248b5ed5cae5497

comment:46 Changed 7 years ago by Nick Draper

Tested by code review and the fact that the unit and system tests still pass.

comment:47 Changed 7 years ago by Nick Draper

Merge remote-tracking branch 'origin/bugfix/8614_letReduction'

Full changeset: dd317fcfdc9f49ac2852079d2dbe0422018f4e28

comment:48 Changed 7 years ago by Russell Taylor

Re #8614 & Re #9061. Bring all changes over from these tickets.

Since none of the affected files have been changed by any other tickets since the release, it is OK to bring over their current state (in master) in one shot. This avoids merge conflicts, which otherwise arise from bringing over individual commits.

Changeset: f0e117217c85895c58ce1a90343049d81be3aee0

comment:49 Changed 7 years ago by Russell Taylor

Re #8614 & Re #9061. Bring all changes over from these tickets.

Since none of the affected files have been changed by any other tickets since the release, it is OK to bring over their current state (in master) in one shot.

Changeset: e0bfd9fbcc74226836b4189a2426eecd33e907ea

comment:50 Changed 5 years ago by Stuart Campbell

This ticket has been transferred to github issue 9458

Note: See TracTickets for help on using tickets.