Separate off run information, logs etc, in to a Run class

[Martyn Gigg]

Currently the Sample class holds data about the sample but also holds the run logs. This means that a user has to get hold of the Sample object to query the run data which is counter intuitive.

Create a separate Run class and move all non-Sample related information to this.

[Martyn Gigg]

(In [5309]) Created a new Run class to hold run information, logs etc. Added a test but not propagated its use out yet. Fixed a bug in property manager, that was found here, where removeProperty would cause a crash when the manager's destructor was run since it wasn't actually removing the entry from the property map. Updated the property manager test as a result. Re #1416

[Martyn Gigg]

(In [5310]) Forgot to check in changes to Sample test. Re #1416

[Martyn Gigg]

(In [5348]) Converted all uses of the logs on the Sample class to use the new Run class. Also cleared some warnings. Re #1416

[Martyn Gigg]

(In [5349]) Checked in local modifications by accident. Re #1416

[Martyn Gigg]

(In [5350]) Windows is unhappy with a compiler generated copy constructor. Added that and an assignment operator plus a test. Re #1416

[Michael Whitty]

(In [5353]) re #1416 - adjusting solution/project files for Windows build in response to revision 5349

[Martyn Gigg]

(In [5357]) Parsing ISIS run information from Raw and Nexus fields into new run object. Also fixed an with old getSampleDetails from Python. New method getSampleInfo now does this job. Re #1416

[Martyn Gigg]

(In [5364]) Implemented run parameter reading for ISIS Nexus and Raw loading routines. Re #1416

[Martyn Gigg]

(In [5370]) Modified Nexus processed routines to deal with single-valued log entries, previously these were only coping with time series logs. The sample geometry has been moved to single-valued entries instead of a time series but LoadNexusProcessed still checks for an entry as a time series. Re #1416

[Martyn Gigg]

(In [5372]) Fix build. My version of gcc let this through but gcc44 and other don't. Re #1416

[Martyn Gigg]

(In [5373]) Fix a test. Re #1416

[Martyn Gigg]

(In [5374]) Fix windows build. Re #1416

[Martyn Gigg]

(In [5388]) Fix for setting the proton charge. It's stored as a float in the Nexus file and trying to retrieve it as a double gives you zero. The unit test wasn't testing this as I thought it was. Updated it to do so. Re #1416

[Nick Draper]

Moved to verify state

[Peter Peterson]

This was done correctly.

[Stuart Campbell]

This ticket has been transferred to github issue ​2263