http://codingdojo.org/cgi-bin/wiki.pl?KataBankOCR

story1.py

import io


numbers = {
    (' _ ', '| |', '|_|'): 0,
    ('   ', '  |', '  |'): 1,
    (' _ ', ' _|', '|_ '): 2,
    (' _ ', ' _|', ' _|'): 3,
    ('   ', '|_|', '  |'): 4,
    (' _ ', '|_ ', ' _|'): 5,
    (' _ ', '|_ ', '|_|'): 6,
    (' _ ', '  |', '  |'): 7,
    (' _ ', '|_|', '|_|'): 8,
    (' _ ', '|_|', ' _|'): 9,
}

def parsenumber(tpl):
    return numbers[tpl]


def readtuples(msg):
    top, mid, btm = msg.readline(), msg.readline(), msg.readline()
    print top
    print mid
    print btm
    for i in range(0, len(top), 3):
        if i+3 > len(top): break
        if i+3 > len(mid): break
        if i+3 > len(btm): break
        yield (top[i:i+3], mid[i:i+3], btm[i:i+3])


sample = \
'    _  _     _  _  _  _  _ \n' + \
'  | _| _||_||_ |_   ||_||_|\n' + \
'  ||_  _|  | _||_|  ||_| _|\n'



def test():
    infile = io.BytesIO(sample)
    for tpl in readtuples(infile):
        print parsenumber(tpl)

if __name__ == "__main__":
    test()