Python re.sub不替代

Question

I'm trying to replace some fragments in file with code:

#!/usr/bin/python
import re
import binascii

def replace(match):
    nw = hex(int(match.group(2),16)+int('43',16))[-2:]
    nl = match.group(2)+match.group(3)+match.group(4)+nw+match.group(6)
    print(match.group(1),nl,nw,len(match.group(1)),len(nl))
    return nl

def main():
    pattern = re.compile(r'(?=((\w{2})(\w{2}0C)(0202010302)(\w{2})(03)))')
    f = open('txt_dump')

    for line in f:
#        for match in re.finditer(pattern, line):
#            print(match.group(1),match.group(2),match.group(3),match.group(4),match.group(5),match.group(6), hex(int(match.group(2),16)+int('43',16)))
#            nw = hex(int(match.group(3),16)+int('43',16))[-2:]
        a = pattern.subn(repl=replace, string=line)        
    print(a[1])
    with open("out", "wb") as of:
        of.write(binascii.unhexlify(a[0]))

    with open("out_txt", "w") as of:
        of.write(a[0])



if __name__ == '__main__':
    main()

As a result I'm getting such output:

7D030C02020103020003 7D030C0202010302c003 c0 20 20
00030C02020103020003 00030C02020103024303 43 20 20
A7030C02020103020303 A7030C0202010302ea03 ea 20 20
03030C02020103020303 03030C02020103024603 46 20 20
22030C02020103020303 22030C02020103026503 65 20 20
03030C02020103020303 03030C02020103024603 46 20 20
5A030C02020103020003 5A030C02020103029d03 9d 20 20
00030C02020103020003 00030C02020103024303 43 20 20
00030C02020103020303 00030C02020103024303 43 20 20
03030C02020103020303 03030C02020103024603 46 20 20
4F030C02020103020203 4F030C02020103029203 92 20 20
D5030C02020103020003 D5030C02020103021803 18 20 20
00030C02020103020003 00030C02020103024303 43 20 20
00030C02020103020003 00030C02020103024303 43 20 20
00030C02020103020103 00030C02020103024303 43 20 20
01030C02020103020503 01030C02020103024403 44 20 20
0D030C02020103020003 0D030C02020103025003 50 20 20
00030C02020103020003 00030C02020103024303 43 20 20
00030C02020103020003 00030C02020103024303 43 20 20
00030C02020103020003 00030C02020103024303 43 20 20
00030C02020103020003 00030C02020103024303 43 20 20
00030C02020103020103 00030C02020103024303 43 20 20
01030C02020103020203 01030C02020103024403 44 20 20
02030C02020103020203 02030C02020103024503 45 20 20
CB030C02020103020003 CB030C02020103020e03 0e 20 20
00030C02020103020003 00030C02020103024303 43 20 20
00030C02020103020603 00030C02020103024303 43 20 20
06030C02020103020203 06030C02020103024903 49 20 20
02030C02020103020103 02030C02020103024503 45 20 20
89030C02020103020003 89030C0202010302cc03 cc 20 20
00030C02020103020003 00030C02020103024303 43 20 20
62030C02020103020003 62030C0202010302a503 a5 20 20
32

So I'm sure it found all fragments properly. But in resulting file it does not replace matches, it placing new value near the old. For example 7D030C0202010302c0037D030C02020103020003 instead of 7D030C0202010302c003 .

What am I doing wrong?

Answer 1

In your pattern, one of the group is the whole pattern (?=())

For example for

7D030C0202010302c003 c0 20 20

As a result you get this as a match:

('7D030C0202010302c003', '7D', '030C', '0202010302', 'c0', '03')

If you remove the outer group your pattern is something like this :

pattern = re.compile(r'(\w{2})(\w{2}0C)(0202010302)(\w{2})(03)')

You get this :

('7D', '030C', '0202010302', 'c0', '03')

And the end result :

030C0202010302c04f03 c0 20 20