Some tweaks to the message output for the script window and interpeter to better handle new lines etc. Re #946

Various changes to IFunction, Function, CompositeFunction and ParameterTie to improve function editing. re #1033

Avoid potential crashes when moving curves in and out of graphs using the CurvesDialog window. Re #950.

visible. Also separated out the scale type options into a separate file so that their 
definitions are not copied. Re #1030

Changed layer selection on 1D plots so that only shift+left-click actually selects the whole layer. Fixes #1028

First real IConstraint test where a constraint is put on the centre point of a Lorentzian away from the optimum value. Refs #961.

Windows installer now puts PyQt4 and sip in the Mantid/bin directory. This prevents conflicts between MantidPlot and existing versions of PyQt4 or sip. Re #1027

  “This application has failed to start because MSVCR80.dll was not found. Re-installing the application may fix this problem.”

This is likely to be a Mixing of retail and debug components problem.

In main.cpp a #pragma line has been put to fix this problem if this occur for you, which by default
is commented out. Refs #963.

Add ability to plot (or copy) non-contiguous selection of rows or columns from a Mantid Matrix. Re #950.

Add ability to plot spectra by dragging a selection from the workspace browser to an existing graph. Re #950.

Went back to copying x arrays to gsl in order to make sure that masked bins are ignored. 

Commented out one line in saveSPE.cpp to make MARI files saved as SPE readable by MATLAB mslice, works for MARI files but fails elsewhere? ?? refs #841

Improve the slickness of plotting simultaneously from multiple workspaces. There's now a plotSpectraList function that takes a list of workspace/index pairs, eliminating the need for a mergePlots. Re #950.

Corrected a small error in DetectorEfficiencyCor which now also respects detector masking. We now have a script that write SPE files that should be compatible with the Excitations Group Matlab application Homer refs #841

New algorithm DetectorEfficiencyCor corrects numbers of counts in a workspace for the efficiency of detectors, which is a function of neutron angle and energy as well as the detectors size and gas pressure. Only basic tests exist so far. Bugs in other algorithms were correct during development refs #841

A few changes for stability of SANS script. The transmission files now only have the first 8 spectra loaded which improves memory usage a substantial amount. Re #1013