1.0 [December 2006] Initial release

1.1 [July 2007] Fixes a bug in pycplex.py for 64-bit platforms.
Thanks to Christoph Lampert for the fix.

1.2 [October 2007] 
Added checks to CPX.c to make sure the right size integer arrays are passed
to CPLEX on 64-bit platforms.

Added qpopt(), addrows(), and copyquad() functions. Thanks to Fabio De Bona for the 
additions.

1.3 [February 2008]
Added addsos(), copysos(), delsetsos() functions
These functions are untested, please let me know if they work correctly.

Added chgbds() function

Thanks to Christina Fabritius Eskebaek for adding test3 case in pycplex.py 
for quadratic and quadratic integer programs.

1.4 [February 2008]
Added MPProb.cplexsparse() function to convert a given matrix to CPLEX sparse 
representation. Only non-zero entries are passed to CPLEX. 
Thanks to Stephen Hartke for this contribution.

1.5 [March 2008]
Fixed a memory leak in getx()
Added setdpblparam function
Added getslack, getpi, getdj, getbase, and copystart functions (for warmstarting)
All these changes were made by Stephen Hartke

1.6 [April 2008]
***Important***: fixed a bug where CPX.getdj was calling CPX.getx instead
Created distutils package, removed Makefile
Install should now work on Windows, too.
Thanks to Vivian De Smedt for contributions leading to the distutils packaging
Added CPX.chgrhs()


2.0a [September 2008]
- renamed cplex_const.py to cplexcodes.py
- added mpprob.py (replacement for pycplex.py)
- moved tests to tests.py
- disabled (for now) addsos and copysos because of CPLEX 9 and CPLEX 11 incompatibility