I started having a hard look at the values handed to Python by the gnubg.match() interface and noticed that the no double and double-take equities seemed to be completely off. As far as I can see, these are raw copies of values from arOutput, while when they are displayed by gnubg, for example in the analysis window, those values are run through mwc2eq before being used. I'd like to fix this, as I want to be able to find all the mistakes in an analysed match within a Python routine.
The problem is that mwc2eq needs a cubeinfo structure, which in turn needs a matchinfo structure. When the data structures are being built by walking the list of move records, it's not immediately clear to me where to get the applicable matchstate and cubeinfo records. If anyone knows this better than I do, it would save a bit of work so I can fix the values in the gnubg.match() return structure. -- Jim Segrave [EMAIL PROTECTED] _______________________________________________ Bug-gnubg mailing list [email protected] http://lists.gnu.org/mailman/listinfo/bug-gnubg
