Eliminate the use of HRP39182 in unit tests

[Russell Taylor]

This is part of the drive to reduce the size of our subversion repository and elimate the reading of data files in unit tests unless absolutely necessary.

Its use features in:

• PointByPointVCorrectionTest
• IkedaCarpenterPVTest
• LinearBackgroundTest
• QuadraticTest

Tests that have to load a file (i.e. those testing file loading algorithms) should switch to using a much smaller file. Others should create fictional data by using routines like WorkspaceCreationHelper.hh. If an algorithm under test requires the instrument geometry in any way then you may have to wait for completion of ticket #1899.

[Anders Markvardsen]

(In [8062]) substituted all occ of HRP39182 with a nexus version which is more than a factor 10 smaller. re #1917

[Anders Markvardsen]

(In [8063]) substituted all occ of HRP39182 with a nexus version which is more than a factor 10 smaller. re #1917

[Russell Taylor]

Anders - the changes you've made here have addressed one the aims of this exercise, which is reducing the size of the test data directory, but have missed the other point, which is that loading real data is much slower than just creating a workspace in memory.

In the case of the curve fitting tests here all you need is numbers in a single-spectrum workspace (no need for an instrument etc.). So you can create a workspace using WorkspaceCreationHelper.hh and fill the spectrum with appropriate numbers - you could even just create an array using the exact same numbers from the HRPD file.

What you did in the other ticket (#1912) - in the case of things like LoadRawInfo you obviously do have to look at a real file.

[Anders Markvardsen]

(In [8090]) Remove HRP test from IkedaCarpenter. Not required since dublicate test on focussed GEM data.

Replaced HRP workspaces in Linear and quadratic with made up ones.

Should the HRP test in PointByPointVCorrectionTest be a system test?

re #1917

[Anders Markvardsen]

(In [8091]) Test failure perhaps caused by two workspaces named the same in two different test. re #1917

[Nick Draper]

Bulk move of tickets to iteration 27, if your ticket is essential for Iteration 26 then move it back.

[Nick Draper]

Bulk move of tickets at the end of iteration 27

[Anders Markvardsen]

When I resurrected the BackToBackExp and IkedaCarpender tests I must also deleted any reference to HRP39182 since no reference to this dataset shows up from a search of the entire Mantid code

[Michael Whitty]

(In [11089]) refs #1917 - remove unused dataset from repository at Anders' request

[Michael Whitty]

this has been done.

[Stuart Campbell]

This ticket has been transferred to github issue ​2764