UCI could be simplified wrt en-passant

[LionelCons]

I think that the en-passant indication (E) is not needed.

If a White Pawn unmoves from d6 to e5 (i.e. uci=d6e5) we know that this was a capture (diagonal move). Since the previously captured piece is not given, we can assume an en-passant capture (i.e. e5xd6 e.p.) to undo.

In contrast, if a White Pawn unmoves from d6 to e5 resurrecting a Pawn (i.e. uci=Pd6e5) we know that this was a normal capture (i.e. e5xd6) to undo.

The same is true for https://github.com/kraktus/retroboard-chess but it seems that the Python code is inactive...

[kraktus]

Hey,thanks for the feedback! Indeed it could be stripped, but that would make the rUCI position sensitive (cannot be parsed without a matching board), which was inconvenient to use.

The python code is indeed inactive but pretty much done 😄

[LionelCons]

Thanks for the reply.

I'm afraid I don't understand why the parsing would be board sensitive.

FWIW, I'm trying some ideas in Python to merge Move and UnMove as well as Board and RetrogradeBoard.

This seems to work fine and I don't need the E or even the U in the extended UCI.

In my test code, the move parsing is board independent. The board is only needed to check if the move is legal or not.