Extend Mask Syntax to support this one

[Gesner Passos]

MASK/REAR/LEFT V0>V2
MASK/REAR /RIGHT V0>V2
MASK/REAR /LEFT V509>V511   or more likely V254>V256 if there are 256 spectra per tube etc
MASK/REAR /RIGHT V509>V511

[Gesner Passos]

Peter: The first thing is to try to convince Richard to give up this idea. If you fail in this attempt...

Strings like http://www.mantidproject.org/SANS_User_File_Commands#MASK:

MASK[/REAR/FRONT/HAB] Hn[>Hm]  or  MASK Vn[>Vm]  - to mask single wires or 'strips'
MASK[/REAR/FRONT/HAB] Hn>Hm+Vn>Vm                - to mask a rectangular 'box'

Are being processed by add_mask_string after passing through parse_instruction (Mask_ISIS). That now, only append all the masks to its attribute: spec_mask_r or spec_mask_f.

In the execution of the Mask (exec), it takes this speck_mask_r and convert them to spectra list, using _ConvertToSpecList.

Hence, one possibility would be to correct parse_instruction to accept MASK/REAR/LEFT XXX instruction. Use them to attach to the spec_mask_r adding a new key information. For example, processing the line:

MASK/REAR/RIGHT V0>V2

Would attach to spec_mask_r the string:

, V0>V2 #RIGHT#

Then, _ConvertToSpecList could be extended to understand this instruction.

It is a sort of ugly solution because it does not improve the 'non-easy-to-understand' process of Mask does to parse the instruction, but it follows the current pattern.

[Nick Draper]

Ownership handed over to Peter

[Peter Parker]

Tester:

The idea of this ticket was to allow users of the ISIS SANS interface to mask vertical columns of detectors using a new command in the User Files that get read in during the reduction process.

Due to complexities introduced by the gas tubes of SANS2D being staggered along the x-axis, this is actually quite a difficult way of masking.

We've managed to convince the scientists that using the Instrument View is a much easier way (for them as well as us!) to go about achieving the same effect. See the child tickets of #8847 for the work done to facilitate this.

I'm therefore closing this ticket as invalid.

[Jay Rainey]

As indicated above, it has been recommended to scientists to use instrument view instead. Closing as invalid.

[Stuart Campbell]

This ticket has been transferred to github issue ​9815