Python FEN chess Validation with regular expression

Python FEN chess Validation with regular expression

def fenPass(fen):
    """
    """
    regexMatch=re.match('\s*^(((?:[rnbqkpRNBQKP1-8]+\/){7})[rnbqkpRNBQKP1-8]+)\s([b|w])\s([K|Q|k|q]{1,4})\s(-|[a-h][1-8])\s(\d+\s\d+)$', fen)
    if  regexMatch:
        regexList = regexMatch.groups()
        fen = regexList[0].split("/")
        if len(fen) != 8:
            raise ValueError("expected 8 rows in position part of fen: {0}".format(repr(fen)))

        for fenPart in fen:
            field_sum = 0
            previous_was_digit, previous_was_piece = False,False

            for c in fenPart:
                if c in ["1", "2", "3", "4", "5", "6", "7", "8"]:
                    if previous_was_digit:
                        raise ValueError("two subsequent digits in position part of fen: {0}".format(repr(fen)))
                    field_sum += int(c)
                    previous_was_digit = True
                    previous_was_piece = False
                elif c == "~":
                    if not previous_was_piece:
                        raise ValueError("~ not after piece in position part of fen: {0}".format(repr(fen)))
                    previous_was_digit, previous_was_piece = False,False
                elif c.lower() in ["p", "n", "b", "r", "q", "k"]:
                    field_sum += 1
                    previous_was_digit = False
                    previous_was_piece = True
                else:
                    raise ValueError("invalid character in position part of fen: {0}".format(repr(fen)))

            if field_sum != 8:
                raise ValueError("expected 8 columns per row in position part of fen: {0}".format(repr(fen)))  

    else: raise ValueError("fen doesn`t match follow this example: rnbqkbnr/pppppppp/8/8/8/8/PPPPPPPP/RNBQKBNR w KQkq - 0 1 ")  


def main():
    try:
        fenPass('rnbqkbnr/pppppppp/8/8/8/8/PPPPPPPP/RNBQKBNR w KQkq - 0 1')
    except ValueError as e:
        print e.message
        
if __name__ == '__main__':
    main()

This is really helpful. Thanks!

Great work! Thanks for sharing!

One missing piece that I noticed, is that on the Castle portion of the FEN regex, there's also the option of no Castle at all, which would be '-'.
To fix that, just add the -| like in the following two portions. Here is the fixed regex:

\s*^(((?:[rnbqkpRNBQKP1-8]+\/){7})[rnbqkpRNBQKP1-8]+)\s([b|w])\s(-|[K|Q|k|q]{1,4})\s(-|[a-h][1-8])\s(\d+\s\d+)$

Here is a FEN that can be used to test that:

r4rk1/pbqn1pp1/1pn1p2p/2ppP2Q/3P4/2PBP1B1/PP1N2PP/R4RK1 w - - 0 14

I'm really confused by the lines 22-25.
In no documentation I can find something about a valid character "~".
Can you explain, whats it meaning is?

@PanCave You can remove the condition since it is redundant. The regex expression matching will never allow ~ in the FEN notation

For regexList = regexMatch.groups(), it may be better to replace this with regexList = fen.split().

For this FEN: "rnbqkbnr/pppppppp/8/8/8/8/PPPPPPPP/RNBQKBNR w KQkq - 0 1", regexMatch.groups() incorrectly set regexList equal to:

["rnbqkbnr/pppppppp/8/8/8/8/PPPPPPPP/RNBQKBNR", "rnbqkbnr/pppppppp/8/8/8/8/PPPPPPPP/", "w", "KQkq", "-", "0 1"]

The second element shouldn't exist ("w" should be there), and the "0" and "1" should be the 5th and 6th elements instead of being together.