Ticket #11690 (closed: fixed)

Opened 5 years ago

Last modified 5 years ago

Beta Test issue with VSI

Reported by: Owen Arnold Owned by: Anton Piccardo-Selg
Priority: blocker Milestone: Release 3.4
Component: GUI Keywords: vates
Cc: Blocked By:
Blocking: Tester: Owen Arnold

Description

  • Open a 3D MDEW dataset in the VSI
  • Go to the standard view and choose Rebin
  • Select BinMD, Axis Aligned, and set all but one of the binning n-bins to 1
  • Run the algorithm and you'll get a fatal error. See attachment.

There are several possible fixes for this, but I think we should allow the rebin to proceed if possible and ensure that the cascade mechanism for the factories is set up as we expect and worry about the display in the following possible ways.

1) Refuse to show it (put a warning message out to this effect), but then what do we show, and what state is the VSI in? 2) Alternately, hook in the vtkDataSetLineFactory, but then should we switch to a 2D view, or 1D plot view accordingly?

Either one of these is where the issue is currently being generated. https://github.com/mantidproject/mantid/blob/master/Code/Mantid/Vates/VatesAPI/inc/MantidVatesAPI/vtkDataSetFactory.h#L178:L179 https://github.com/mantidproject/mantid/blob/master/Code/Mantid/Vates/VatesAPI/inc/MantidVatesAPI/vtkDataSetFactory.h#L204:L205

Attachments

Picture1.png (9.4 KB) - added by Owen Arnold 5 years ago.
MDEvent_Osiris.nxs (2.3 MB) - added by Anton Piccardo-Selg 5 years ago.

Change History

Changed 5 years ago by Owen Arnold

comment:1 Changed 5 years ago by Anton Piccardo-Selg

Regarding the chain of responsibility

The source plugin for MDEvent data seems to have a full chain:

https://github.com/mantidproject/mantid/blob/master/Code/Mantid/Vates/ParaviewPlugins/ParaViewSources/MDEWSource/vtkMDEWSource.cxx#L186::L195

For MDHisto data this is not the case: https://github.com/mantidproject/mantid/blob/master/Code/Mantid/Vates/ParaviewPlugins/ParaViewSources/MDHWSource/vtkMDHWSource.cxx#L176:L180

We might get away with completing the chain at this point (at least regarding the crash).

Extending chain

Extending the chain in the MDHisto source plugin fixes the crash.

Next steps

We need to hedge for 0D rebinning which still throws the unhandled exception

Last edited 5 years ago by Anton Piccardo-Selg (previous) (diff)

comment:2 Changed 5 years ago by Anton Piccardo-Selg

  • Status changed from new to inprogress

Refs #11690 Extend chain for MDHisto to 1D

Changeset: 5084aa4dbbdae8fd7868134a9e9675ce09cf66a7

comment:3 Changed 5 years ago by Anton Piccardo-Selg

Refs #11690 Add 0D factory for chain of resp.

Changeset: 1f72ffa654f770e941c8c17cbdf061047932c254

comment:4 Changed 5 years ago by Anton Piccardo-Selg

Refs #11690 Add unit test

Changeset: 0b0df2c0d3bed2119d9f1450ebe21739d3ab4d1b

comment:5 Changed 5 years ago by Anton Piccardo-Selg

Refs #11690 Add warnings

Changeset: 00858129d457b63acf1a7da1c3ee528bf786c5c9

comment:6 Changed 5 years ago by Anton Piccardo-Selg

Refs #11690 Refactor warnings

Changeset: f19c21430e6e6dfbb9036cf99c72388110aaae61

comment:7 Changed 5 years ago by Anton Piccardo-Selg

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

This is being verified as pull request #716.

Changed 5 years ago by Anton Piccardo-Selg

comment:8 Changed 5 years ago by Anton Piccardo-Selg

Refs #11690 Fix gnu compiler warnings

Changeset: ba6d01e9856022cce7d1a5f0acdf744f787634c4

comment:9 Changed 5 years ago by Owen Arnold

  • Status changed from verify to closed
  • Tester set to Owen Arnold

Merge pull request #716 from mantidproject/feature/11690_fix_chain_of_responsibility_for_MDHisto_in_vsi

Fix broken chain of responsibility for MDHisto in the VSI

Full changeset: b02a1a06628f0126824273b4d6f9509a98fb70fd

comment:10 Changed 5 years ago by Stuart Campbell

This ticket has been transferred to github issue 12528

Note: See TracTickets for help on using tickets.