[IDA] ConvFit - Fix error thrown while using this tab after applying corrections.

[Peter Parker]

From Franz:

The error occurs when I am running the abs corrected file, created from abs correction tab (attached with the resolution file) in conv fit with 1 Lorz + el line.

Also it writes out the data again into an “osi95111_graphite_conv” file without the “correct”

The abs corr is running fortran code and seems to produce a not conform workspace....

The error:

Runtime error on line 12: “Matrixworkspace::getdetector(): no detectors for this workspace index. 
Search object” in file IndirectCommon.py at line 32

Fit started
Fit successful, Duration 0.12 seconds
PlotPeakByLogValue started
PlotPeakByLogValue successful, Duration 3.65 seconds
Unexpected exception: Workspace2D::getSpectrum, histogram number out of range

[Peter Parker]

Spencer has suggested that this probably has to do with the instrument file not being loaded before carrying out the step that produces the attached files.

Awaiting feedback from user to see if opening the CTE interface and then running the corrections again produces workspaces with valid detector info.

[Nick Draper]

Moved at end of release 2.1

[Peter Parker]

A more recent results log output is as follows:

RuntimeError: MatrixWorkspace::getDetector(): No detectors for this workspace index. search object 
  at line 12 in '<Interface>'
  caused by line 110 in 'C:/MantidInstall/scripts/Inelastic\IndirectDataAnalysis.py'
  caused by line 48 in 'C:/MantidInstall/scripts/Inelastic\IndirectDataAnalysis.py'
  caused by line 84 in 'C:/MantidInstall/scripts/Inelastic\IndirectCommon.py'

It seems the corrected workspaces coming out of CalcCorr do not contain detector info, and this is causing problems.

Two possible workarounds:

1. Produce conforming workspaces in CalcCorr.
2. Get detector information straight from the instrument.

The latter is probably the easiest, but wont solve the core issue, which is we are producing non-conform workspaces. These workspaces wont work elsewhere (like in SofQW) either.

[Nick Draper]

Moved at the end of release 2.2

[Nick Draper]

moved to Release 2.4

[Nick Draper]

Moved at the code freeze for release 2.4

[Nick Draper]

Moved to r2.6 at the end of r2.5

[Nick Draper]

Moved to the Backlog after R2.6

[Peter Parker]

This works fine for me now, by using the attached files as well as generating a new one from the calculate corrections tab.

It has obviously been fixed elsewhere. Closing.

[Samuel Jackson]

I tried this with both the files provided and some more recent files walked through CalCor. It does indeed appear that whatever was causing the issue has now been fixed.

[Stuart Campbell]

This ticket has been transferred to github issue ​5507