Make the MuonAnalysis less sensitive to filenames

[Arturs Bekasovs]

Instead of getting the instrument name from the filename, we should get it from the loaded workspace, letting Load algorithm deal with how it gets it.

In that case, we can use files like 00015189.nxs instead of MUSR00015189.nxs.

[Arturs Bekasovs]

That's really annoying scientists.

[Arturs Bekasovs]

There is a slight difficulty with that one: if we are not using the filename to determine the instrument, we can't actually say that the file user is trying to load is not a Muon file before actually loading it. This makes it much easier to reproduce the problem described in #8020. So it makes sense to do that other ticket first. Afterwards, I will be able to just throw an error if I determine that the instrument of the loaded workspace is not a Muon instrument and nothing will get affected.

[Arturs Bekasovs]

Mixed up "Blocking" and "Blocked By".

[Arturs Bekasovs]

Trac playing tricks.

[Arturs Bekasovs]

Trac playing tricks.

[Nick Draper]

Bulk move of tickets out of triage (new) to assigned at the introduction of the triage state

[Arturs Bekasovs]

There is slight disadvantage with that approach: files with run_number set to 0 will have names like EMU00000000. Asked Steve Cottrell whether this can be a problem. Waiting for the answer.

[Arturs Bekasovs]

Steve haven't answered, so I am going ahead with this. If zero run numbers will become a problem, I'll fix it separately.

[Arturs Bekasovs]

Refs #8547. Methods to get run[s] label from ws and tests for them.

Changeset: 93c1a972c6acc8b78ea5e783dac9e63d84de1dd1

[Arturs Bekasovs]

Refs #8547. Move sumWorkspaces to helper, fix it and add tests.

sumWorkspaces method was modifying the first workspace in the list passed in, which caused problems.

Changeset: ef9cb6d5d694c15c2c39758995f30ed8ef7f6136

[Arturs Bekasovs]

Refs #8547. Make load() return the label.

Changeset: 1028041705953ceed6302e73bca5ce5ec570ee34

[Arturs Bekasovs]

Refs #8547. Use the returned label, create and store a group for it.

Instead of getting group name and retrieving it on every plot, we can just store a pointer to the group.

Changeset: a7a05f1927c6592ee0997c1c26deea9cf688228d

[Arturs Bekasovs]

Refs #8547. Remove unused functions.

Changeset: 77d712a1b10042105081cf55caf94b346110804e

[Arturs Bekasovs]

Refs #8547. Different zero-padding width for ARGUS.

Changeset: afd9a93945675ae382a134060d61aac84422923f

[Arturs Bekasovs]

Refs #8547. Re-use logic of getLabel.

Changeset: 61c926d24cd0142487daea3a00e0d7a4e77236aa

[Arturs Bekasovs]

Refs #8547. Remove numerous Plus algorithm log messages.

Changeset: 7822bee681945bebc3ce62e760004d389ac826f4

[Arturs Bekasovs]

Refs #8547. Better error-reporting.

Changeset: 1e0b2c3a7cd238f63f67a2a2f1695d6779183562

[Arturs Bekasovs]

Refs #8547. Check group membership by name, not pointer.

Otherwise we end up with multiple workspaces with the same name in the group.

Changeset: 10ab90f7c1867df24bae51030ce3737c4c5aa734

[Arturs Bekasovs]

Refs #8547. Remove another unused function.

Changeset: c51fc72da94dc26cc6872396ad907b5ffd128cb9

[Arturs Bekasovs]

Refs #8547. Use the proper way to get zero padding for the instrument

Changeset: 35d2164ac6ae58345026868fbab69d1c7bc5ba28

[Arturs Bekasovs]

Refs #8547. Merge 'origin/master' into 8547

Conflicts:

Code/Mantid/MantidQt/CustomInterfaces/inc/MantidQtCustomInterfaces/MuonAnalysis.h Code/Mantid/MantidQt/CustomInterfaces/src/MuonAnalysis.cpp Code/Mantid/MantidQt/CustomInterfaces/src/MuonAnalysisHelper.cpp

Changeset: 09a0125611e691e53d1300b45f3c711b06a3ce27

[Arturs Bekasovs]

Refs #8547. Use the proper DLL export macro

Should fix windows builds.

Changeset: 7325e87e810656ce40bdbedf7900c3439eb6ac44

[Arturs Bekasovs]

Refs #8547. Move local comparison structure to comparison function

Should fix RHEL builds

Changeset: c93603ad1402147115c7c4d9d30c9c2842b95670

[Arturs Bekasovs]

Tester:

Copy MUSR00015189.nxs, MUSR00015190.nxs and MUSR00015191.nxs files from AutoTestData directory to a different one and rename them to something like my_musr_file1.nxs, my_musr_files2.nxs and my_musr_file3.nxs.

In MuonAnalysis, try to load my_musr_file1.nxs. Check that the file is loaded fine and the workspaces group is named correctly (MUSR00015189). Make a few plots. Make sure they end up in the right group.

Now try to load all 3 files at once. Again, check that thet are loaded fine and the group name is sensible (MUSR00015189-91). Plot Counts of some group and verify that numbers seem to be correct (should be approx. 3 times higher that the ones from a single file).

Try to load a non-Muon instrument file. Try to load files of different Muon instruments at once. Do more checks to make sure the right files are loaded and the wrong ones are discarded, no matter what the filename is.

[Arturs Bekasovs]

(In #2369) The problems appears to be solved #8547. When it is closed, the tester will need to re-check it, though I was able to load the file, fit it and produce the results table.

[Roman Tolchenov]

Merge remote-tracking branch 'origin/feature/8547_muonanalysis_filenames'

Full changeset: ​726a4f8d6d76290b2bf82357538c952a458a6238

[Stuart Campbell]

This ticket has been transferred to github issue ​9391