carli/1758_multiprofile_intensity_unit_conversion

[crocka]

Description

The pr resolves #1758 and resolves #1907. It tackles three problems. First, the spectral profiles for different frames when matched should convert to the same intensity unit. Second, spectral matching should be avoided if two images do not have intersecting intensity unit. Third, the intensity unit option should only show the ones that are usable for each of the spectrally matched images.

Originally in the SpectralProfileWidgetStore, the intensityConversion observable is singular, mapping the effectiveFrame to the selected intensity unit. This is undoubtedly incorrect if we have more than one frame. To solve this issue, the intensityConfig object was move to FrameStore, and the intensityConversion is uniquely produced in plotData for each of the frames' profile. To access the frame that correspond to each profile, the frame store of the frame is added to the profile object in the profiles array in SpectralProfileSelectionStore.

Another problem that the pr solves is that the program should avoid spectral matching when the spectrally matched images do not have common intensity unit. This condition is added to the if-statement in setSpectralReference function in FrameStore using the getCommonIntensityOptions function. The getCommonIntensityOptions function searches for the common or intersecting intensity unit with 1) a second frame's intensityConfig, if the arguments is a single IntensityConfig, 2) an array of frame's intensityConfig, if the argument is an array rather than a single instance of IntensityConfig.

The third problem is to allow only the intensity units available for all of the spectrally matched images to show in the intensity option. Knowing that the getCommonIntensityOptions function finds the intersecting intensity unit, the intensityOptions in SpectralProfileWidgetStore simple made use of this function.

Checklist

For linked issues (if there are):

• assignee and label added
• ZenHub issue connection, board status, and estimate updated

For the pull request:

• reviewers and assignee added
• ZenHub estimate, milestone, and release (if needed) added
• e2e test passing / corresponding fix added
• changelog updated / no changelog update needed
• protobuf updated to the latest dev commit / no protobuf update needed
• BackendService unchanged / BackendService changed and corresponding ICD test fix added

[YuHsuan-Hwang]

Overall the code looks good, and I have some initial comments. Please also update the description in the PR with the new changes.

[kswang1029]

@crocka If we just load one image and apply intensity unit conversion, it appears that the function is broken. I can see the ylabel of the spectral plot changes accordingly but not the values. Could you have a look?

[kswang1029]

looks good now. no regression from e2e tests. 👍