trim extra data off end of zipfile

trim_zip.py

import struct

ECD_SIG = "\x50\x4B\x05\x06"
ECD_LEN = 22

def trim_zip(filename_in, filename_out):
    """
    trims the zipfile give by filename `filename_in`, removing any bytes
    after the end of central directory record and writing the result to
    a file with filename `filename_out`.
    
    returns the number of entries in the file according to the end of central
    directory record.
    """
    
    # read the file into memory for quick searching
    file_in = open(filename_in)
    data = file_in.read()
    
    # seek the end of central directory
    file_in.seek(data.index(ECD_SIG))
    ecd = file_in.read(ECD_LEN)
    ecd_fields = struct.unpack("<IHHHHIIH", ecd)
    
    # just some sanity checks
    assert ecd_fields[0] == 0x06054b50
    assert ecd_fields[1] == 0
    assert ecd_fields[2] == 0
    assert ecd_fields[3] == ecd_fields[4]
    
    # ignore everything after the end of central directory
    trimmed_data = data[:file_in.tell()]
    
    file_out = open(filename_out, "wb")
    file_out.write(trimmed_data)
    file_out.close()
    
    return ecd_fields[4]

if __name__ == "__main__":
    print trim_zip("ziptest.zip", "ziptest2.zip")
    
    import zipfile
    print zipfile.is_zipfile("ziptest.zip")
    print zipfile.is_zipfile("ziptest2.zip")