Use sanCAS v1.1 in SANS isis reduction

[Gesner Passos]

The way I would envisage the new SaveCansas1D algorithm working is that when called to write a reduced 1D data workspace, Mantid checks to see if either or both of the transmissions (sample & can) used to calculate that data have been generated as part of the reduction process. If they have then they should be automatically inserted into the XML file. If they are not present (eg, the user has chosen to reduce data without transmisisons or has deleted both of the transmission workspaces associated with the reduction before choosing to write the file) then the file is written without any SAStransmission_spectrum blocks.

[Gesner Passos]

(In #6839) This ticket will only implement the second version of load and save canSAS. It will let the introduction of this new feature inside the Reduction to the ticked #6929

[Nick Draper]

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

[Gesner Passos]

Add the transmission information to the reduced workspace

Due to some limitations imposed by the way the Singleton is used, there is no way to get the transmission workspaces after the reduction is finished from the user interface, except using the attempt to guess the name of the transmission workspaces (what we should not do). So, I propose here keeping the information about the workspaces of the transmission at the SampleLog of the Reduced workspace. This information will be them available whenever is necessary. It would even allow us to use this information inside the SaveCan algorithm itself. But, I think it would seem a litle bit of magic. So, i prefer to use it outside.

re #6929

Changeset: e1e05575f99bc5659c2a61e1b809d65f25b056f5

[Gesner Passos]

Extend SANSBatchMode to use the transmission workspace for CanSAS

It will pass on the parameters to SaveCanSAS1D if they are available.

re #6929

Changeset: 424db0d8700aedf2ef8f8bf2fec2d48aeae4aca9

[Gesner Passos]

Extend SingleMode to use the transmission workspace for CanSAS

It will pass on the parameters to SaveCanSAS1D if they are available. Changes on SANSRunWindow.cpp

re #6929

Changeset: 747c82c73d31e3287bf47bd2c30bd5f9e43b9f94

[Gesner Passos]

Merge branch 'feature/6929_use_sancas_v1_1' into develop

[Gesner Passos]

Current limitations of using SaveCanSAS in ISIS SANS:

• Reduction in single mode, selecting both detector, will allow you to save only the Front reduction using the SaveResult on the interface. (Because you save just the last reduced workspace)
• Reduction in single mode or batch mode, selecting merged detector, will not allow you to save transmission workspaces

I'm discussing with the interested people to decide if these limitations are acceptable.

[Gesner Passos]

For the first issue (Both detectors selected), the user said:

I do not like this current behaviour because it could easily catch a User out.

I think what I would suggest is this: only populate the Filename box with a 
workspace name if a single detector has been reduced (front/rear/main/hab) 
**or** if it is the MERGED workspace. In the case of BOTH, leave it empty and 
force the User to select a Save algorithm.

For the second item, the merged option, the problem was that we have 2 sample and 2 transmisison, so, it was not clear which to choose. But, the user explained that they are both the same, and they have the same value. So, we can use any one we want to add as transmission values.

[Gesner Passos]

Add the transmission information to the reduced workspace

Due to some limitations imposed by the way the Singleton is used, there is no way to get the transmission workspaces after the reduction is finished from the user interface, except using the attempt to guess the name of the transmission workspaces (what we should not do). So, I propose here keeping the information about the workspaces of the transmission at the SampleLog of the Reduced workspace. This information will be them available whenever is necessary. It would even allow us to use this information inside the SaveCan algorithm itself. But, I think it would seem a litle bit of magic. So, i prefer to use it outside.

re #6929

Changeset: e1e05575f99bc5659c2a61e1b809d65f25b056f5

[Gesner Passos]

Extend SANSBatchMode to use the transmission workspace for CanSAS

It will pass on the parameters to SaveCanSAS1D if they are available.

re #6929

Changeset: 424db0d8700aedf2ef8f8bf2fec2d48aeae4aca9

[Gesner Passos]

Extend SingleMode to use the transmission workspace for CanSAS

It will pass on the parameters to SaveCanSAS1D if they are available. Changes on SANSRunWindow.cpp

re #6929

Changeset: 747c82c73d31e3287bf47bd2c30bd5f9e43b9f94

[Gesner Passos]

correct the number of bins at transmission

Correct a wrong implementation of saving cansas that does not take the correct number of values to save them.

re #6929 re #6839

Changeset: a542e0b51658f4f324fbf0ac649aed3380469fd7

[Gesner Passos]

Implement the saving transmission and transmissioncan to merged option

add the properties transmission and transmissioncan to the merged workspace this allows the savecansas to save those values.

re #6929

Changeset: c34499bb41bbf1995a0652b4720934d88de4da8f

[Gesner Passos]

Correct the Transmission and TransmissionCan to take the unfitted data

saves the unfitted data and not the fitted one.

re #6929

Changeset: 0b091ba7ef3d14cfce9d09a8172a1f7541e7d991

[Gesner Passos]

Correct the unit test to adjust the units

the commit a542e0b51658f4f324fbf0ac649aed3380469fd7 changed the units, and this commit, updates the unit test.

re #6929 re #6839

Changeset: 94126a9398bef8b9c69d796fee5ca48ab0c7a508

[Gesner Passos]

input for testing

[Gesner Passos]

Tester if you do not have access to the isis data, them, you can only test this:

• Select LOQ instrument
• Select user file: systemtests/Data/LOQ/MASK.094AA
• change to batch mode and select the file for_can_sas_test.csv (do not forget to add Data/LOQ to the directory manager)
• select the output CanSAS at the bottom right.
• change again to the Single mode (the batch mode does not allow you to reduce .raw data).
• press Reduce (it will save the file 54431main_1D_2.2_10.0.xml at your output directory)
• You can load these files using the LoadCanSAS1D algorithm to see if it is correct as well.
• check that there is a SAStransmission_spectrum entry inside the xml file.
• There is not much more you can do, because it will produce some exceptions that are not directly related to this ticket.

If you have access to the ISIS data, then, you may:

• Open ISIS Sans interface
• Select SANS2D instrument
• Set the searching directories to cycle_13_1
• Open the Mask at: ndxsans2d/user/Masks/MASKSANS2D_131L_4m_12mm_Lakey_B.txt
• Load for_can_sas_sans2d.csv in batch mode
• Select the CanSAS as output format.
• Press Reduce it will save: 4m_12mm_lakey_rear.xml
• check that there is a SAStransmission_spectrum inside
• change to single mode and press reduce
• It will save 18617rear_1D_1.75_16.5.xml file
• check that there is a SAStransmission_spectrum inside
• Change at the Reduction Settings the detector to front-detector and to merged
• make sure that the SAStransmission_spectrum is still there.
• Finally, check that in both mode, the batch mode will save two files, while in single mode, you will not have the opportunity to save no file. (the reason is defined in comment:10)

[Gesner Passos]

Disable saving output on both & single mode

user request:

only populate the Filename box with a workspace name if a single detector has been reduced (front/rear/main/hab) or if it is the MERGED workspace. In the case of BOTH, leave it empty and force the User to select a Save algorithm.

re #6929

Changeset: e44a880a94ec5138105bdc552a7806bf23e2ec19

[Mathieu Doucet]

Reopening ticket to get rid of warning, as agreed.

[Mathieu Doucet]

Re #6929 Remove warning as agreed

Changeset: 922789aa5b42d62bbd02b440e7a3cf722fcbeeec

[Mathieu Doucet]

Fixed the warning as agreed with the developer and closing the ticket after successful testing.

[Peter Peterson]

Mathieu told me to "close" this.

[Stuart Campbell]

This ticket has been transferred to github issue ​7775