-
Notifications
You must be signed in to change notification settings - Fork 128
Open
Labels
ISIS: LSSIssue and pull requests relating to SANS and Reflectometry (Large Scale Structures) at ISISIssue and pull requests relating to SANS and Reflectometry (Large Scale Structures) at ISISReflectometryIssues and pull requests related to reflectometryIssues and pull requests related to reflectometry
Milestone
Description
@36573 Adds the capability to the sliceviewer plot to retain ROI selections when new runs are loaded on the preview tab.
To do this we check for matching settings in the look-up table on the experiment settings tab and, if retain ROIs if as these are still relevant. See the issue for more info.
Adding this to the instrument view plot has further complications. As per Rachel:
Getting this to work for the instrument view plot has the following complications:
We would need to identify from the experiment settings ROI detector string whether to add multiple region selectors or just one large one. We would need to be able to do this in a performant way that ideally didn't rely on a specific setup in the IDF (i.e. requiring use of a RectangularDetector) or a particular detector ID numbering sequence. We could save the string using a different pattern to allow us to distinguish separate regions, but this would require various changes, from how we retrieve the list of selected detectors from the instrument view (i.e. being able to get detector IDs grouped per mask shape rather than a single combined list from all the separate shapes) to how we parse it in the reduction.
We would need to be able to convert a detector ID to its corresponding pixel on the instrument view plot. There are some methods provided for this (see AlFInstrumentWidget example, which uses method ProjectionSurface::drawShape2D) but for various technical reasons we can't make use of these in the preview tab implementation. I think that investigating and implementing a resolution would likely be a fair amount of work.
If there is a switch to a per bank reduction in future then this will require a review of the preview tab to see how scientists will work with it when there is not longer a bank summing step. As a result changes that are relatively complex in nature are probably not advisable at this stage.
Some of the required changes on the instrument view side might be better covered either as part of the instrument view epic or added after the new instrument view is available. James has confirmed that selecting a pixel from a detector ID (i.e. the requirement under point 2 above) is easy in the new instrument view.
I have confirmed with Max that if we can only add the selectors on the slice viewer plot in the short-term then that is still beneficial. A solution for the instrument view plot could come later.
Metadata
Metadata
Assignees
Labels
ISIS: LSSIssue and pull requests relating to SANS and Reflectometry (Large Scale Structures) at ISISIssue and pull requests relating to SANS and Reflectometry (Large Scale Structures) at ISISReflectometryIssues and pull requests related to reflectometryIssues and pull requests related to reflectometry
Type
Projects
Status
Backlog