Get the SANS analysis to run under Mathieu's new frame work

[Steve Williams]

There should be system tests that import ISISCommandInterface

[Steve Williams]

(In [6396]) Extended the Python Mask ReductionStep to handle different shapes re #1686

[Steve Williams]

(In [6401]) More xml support for the Python Mask ReductionStep, tests will be added for this re #1686

[Steve Williams]

(In [6402]) xml support in the Python Mask ReductionStep re #1686

[Steve Williams]

(In [6504]) Get the SANS Python to work re #1686

[Steve Williams]

(In [6572]) SANS Python ReductionStep modules implementation -modules getting their data from each other re #1686

[Steve Williams]

(In [6617]) SANS Python ReductionStep modules implementation -mask module complete re #1686

[Steve Williams]

(In [6648]) Fix broken system tests re #1686

[Steve Williams]

(In [6669]) The new SANS framework can do a simple reduction and get the correct answers re #1686

[Mathieu Doucet]

(In [6673]) Revert back to old Normalize code that doesn't break HFIR unit tests. Re #1686

[Steve Williams]

(In [6683]) Fix broken SANS tests used at HFIR re #1686

[Steve Williams]

(In [6692]) New ISIS SANS interface supports more commands and extracting a single period re #1686

[Mathieu Doucet]

(In [6699]) Fixed minor copy-paste error. Re #1686

[Steve Williams]

(In [6700]) The new SANS runs existing ISIS command scripts (still no can, transmission and center finding support) re #1686

[Mathieu Doucet]

(In [6702]) Minor change to ReductionSingleton.run() to ensure that the method leaves the Reducer singleton instance in a known state. Re #1686

[Steve Williams]

(In [6754]) SANS reduction bug fixes re #1686

[Steve Williams]

(In [6775]) Analysis on the front detector now works re #1686

[Steve Williams]

(In [6778]) Fix broken SANS test re #1686

[Steve Williams]

(In [6779]) Fix broken SANS test re #1686

[Steve Williams]

(In [6780]) Fix broken SANS test re #1686

[Steve Williams]

(In [6812]) The new SANS framework gets the correct answers for reductions with can and transmission corrections re #1686

[Steve Williams]

(In [6879]) Fix test problem re #1686

[Steve Williams]

(In [6880]) Fix test problem re #1686

[Steve Williams]

(In [6888]) Fix test problem re #1686

[Steve Williams]

(In [6890]) Fix broken system test re #1686

[Steve Williams]

(In [6944]) Get ViewCurrentMask() working re #1686

[Martyn Gigg]

(In [6994]) Fix system tests. Re #1686

[Steve Williams]

(In [7153]) The SANS GUI now works with the new framework, except centre finding which still uses the old fixes #1686

[Steve Williams]

This stuff needds some thorough testing again the original code!

[Steve Williams]

(In [7163]) SANS system tests, replace changes I didn't want to commit and remove commented out code re #1686

[Steve Williams]

(In [7167]) Fix SANS system test re #1686

[Steve Williams]

(In [7179]) Fix SANS system test re #1686

[Steve Williams]

(In [7182]) Fix SANS system tests re #1686

[Steve Williams]

(In [7183]) Get phi masking working again re #1686

[Steve Williams]

(In [7196]) SANS new framework fixing some beam centre finder and batch mode problems fixes #1686

[Steve Williams]

(In [7204]) Things missing from ISISCommands functions fixes #1686

[Michael Whitty]

Notes:

• 'save result' button brings up a dialog saying that the workspace failed to save, when this is not the case.

[Steve Williams]

(In [7214]) Remove the error when saving the results workspace from the SANS GUI re #1686

[Michael Whitty]

More Notes (don't know if these fall under this ticket but putting them here for a notice to work on later, perhaps):

• Geometry tab, after loading raw data - Click on 'Run', get dialog message on fatal python error and SyntaxError: "EOL while scanning single-quoted string" in scripting console. Sorry, no line number.

• Running 2D Reduce with "Plot Result" checked, Python error - "NameError: "global name 'qti' is not defined" in file 'ISISCommands.py' at line 415"

• Running 2d reduce with LOQ 54431 sample - error

Component with name some-sample-holder was not found.
Error in execution of algorithm MoveInstrumentComponent:
Component with name some-sample-holder was not found.

Reproduce by: Run 2d with plot result checked, uncheck plot result and run again

[Michael Whitty]

Additional to point 1 (geometry tab) above: it seems to clear the information from the Masking tab. Don't know if this is helpful.

[Steve Williams]

(In [7215]) More errors on the SANS interface GUI re #1686

[Michael Whitty]

this line:

if reducer.to_Q.output_type == '1D':
}}

on line 412 of ISISCommands.py seems to return True for 2D as well as 1D. This shouldn't affect the analysis, but will stop the plotting of 2D ever working (you need to cut the "qti.app.MantidUI." off the start of line 415 for this to happen).

Otherwise, the issues I've noticed seem to be resolved. The geometry tab works, saving nexus file works, and (since plot will always think it's 1d and won't go to the name error), that doesn't cause a problem running the reduction.

[Michael Whitty]

the problems I've noted have been resolved.

[Michael Whitty]

the problems I've noted have been resolved. hopefully there will be some documentation for this interface in the next release?

[Stuart Campbell]

This ticket has been transferred to github issue ​2533