Make instrument optional in LoadDiffCal

[peterfpeterson]

VULCAN added more detector banks and we are unable to process any of their data, including the previously existing banks, until a new IDF is created. This get past the issue in terms of loading the calibration file.

There is no associated issue, but this is related to EWM15052.

To test:

Pick your favorite diff-cal file and load it. The "data" can be viewed and the right-click menu item for showing the instrument is disabled.

---

Reviewer

Your comments will be used as part of the gatekeeper process. Comment clearly on what you have checked and tested during your review. Provide an audit trail for any changes requested.

As per the review guidelines:

• Is the code of an acceptable quality? (Code standards/GUI standards)
• Has a thorough functional test been performed? Do the changes handle unexpected input/situations?
• Are appropriately scoped unit and/or system tests provided?
• Do the release notes conform to the guidelines and describe the changes appropriately?
• Has the relevant (user and developer) documentation been added/updated?
• If the PR author isn’t in the mantid-developers or mantid-contributors teams, add a review comment rerun ci to authorize/rerun the CI

Gatekeeper

As per the gatekeeping guidelines:

• Has a thorough first line review been conducted, including functional testing?
• At a high-level, is the code quality sufficient?
• Are the base, milestone and labels correct?

[rboston628]

I think inlining the new constructors is a good idea.
Are we sure there's no risk of nullptr dereference?

[peterfpeterson]

The detids is passed by reference so I don't think you can have a nullptr

[rboston628]

Was able to build and run. I ran the following script

file = "/SNS/SNAP/shared/Calibration/Powder/04bd2c53f6bf6754/lite/diffraction/v_0001/diffract_consts_046680_v0001.h5"
LoadDiffCal(Filename = file, WorkspaceName = "lite")
file = "/SNS/SNAP/shared/Calibration/Powder/04bd2c53f6bf6754/native/diffraction/v_0/diffract_consts_default_v0.h5"
LoadDiffCal(Filename = file, WorkspaceName = "difc")
LoadDiffCal(file)

The first uses a SNAPLite diffcal file. The second uses a SNAP diffcal file. The third uses a SNAP diffcal file without a workspace name.

In the SNAPLite case, the following error output:

LoadDiffCal started
Extension ['.nxs.h5'] not found.
Searching for other facility extensions.
The instrument definition filename does not contain _Definition. Your instrument name will be set to: SNAP_46680.lite.nxs.h5
Error in execution of algorithm LoadInstrument:
FileDescriptor() - File 'SNAP_46680.lite.nxs.h5' does not exist
LoadDiffCal successful, Duration 0.12 seconds

In the SNAP cases, the following:

LoadDiffCal started
Either the InstrumentName or Filename property of LoadInstrument most be specified
Error in execution of algorithm LoadInstrument:
Either the InstrumentName or Filename property of LoadInstrument must be specified to load an instrument in ""
LoadDiffCal successful, Duration 2.33 seconds

At the end, the expected workspaces are created. For the case where WorkspaceName is not given, the workspaces are _cal, _group, and _mask.

In all three cases, the matrix workspaces have "Show Instrument" disabled. Selecting to "Show Data" shows the table data. For the calibration tables, double clicking will show the table data as well.

However, double-clicking on any of the matrix workspaces causes a segmentation fault:

Received signal: SIGSEGV
Segmentation fault

[rboston628]

Re-ran the above script. Everything still worked. Double-clicking the workspaces brought up a plot without segfaulting.