linux 重复文件 图片,【图片】查找重复文件Py版本【linux吧】_百度贴吧

该楼层疑似违规已被系统折叠 隐藏此楼查看此楼

#!/bin/env python3

import hashlib

import os

import sys

g_buffer = bytearray(8192)

g_buffer_view = memoryview(g_buffer)

def get_digest(path, size):

global g_buffer

global g_buffer_view

md5 = hashlib.md5()

with open(path, 'rb', buffering=0) as fp:

while size > 0:

n = fp.readinto(g_buffer)

md5.update(g_buffer_view[:n])

size -= n

return md5.digest()

def main():

if len(sys.argv) > 1:

START_DIR = os.path.abspath(sys.argv[1])

else:

START_DIR = os.getcwd()

if not os.path.isdir(START_DIR):

print('"{}" is not a directory'.format(START_DIR), file=sys.stderr)

sys.exit(1)

all_ = {}

for root, dirs, files in os.walk(START_DIR, followlinks=False):

for f in files:

p = os.path.join('/', root, f)

if os.path.islink(p): continue

sz = os.path.getsize(p)

if sz == 0: continue

all_.setdefault(sz, []).append(p)

for sz, paths in all_.items():

if len(paths) < 2: continue

dup = {}

for p in paths:

d = get_digest(p, sz)

dup.setdefault(d, []).append(p)

for ps in dup.values():

if len(ps) < 2: continue

print(len(ps))

for p in ps: print(p)

if __name__ == '__main__':

main()