[Coin-discuss] OsiOsl resolve( )

[Matthew Galati]

I am curious why in OsiOslSolverInterface::resolve( ) - it seems to use 
primal simplex rather than dual simplex. For using OsiOsl with BCP (for 
example) to do branch and cut - it would seem better to call dual 
simplex as the default on resolve( ). I guess the interface should 
"figure out" which to use based on how the problem has been recently 
modified. Or perhaps we should be more generic and have both dualSolve( 
) and primalSolve( ) methods and have BCP do the work to decide which to 
call. Thoughts?

----
OsiOslSolverInterface()::resolve( )
#if 0
   rtcod=ekk_dualSimplex(model); // *FIXME* : why not 0 (eitherSimplex)
#else
   rtcod=ekk_simplex(model, 256 + 32); // no presolve and no scaling
#endif

Matt


-- 
Matthew Galati
ISE Lehigh University
IBM Service Parts Solutions
610.758.4042 (Office)
610.882.0779 (Home)
magh at lehigh.edu, magal11 at us.ibm.com
http://sagan.ie.lehigh.edu/mgalati/