kindle delete useless sdr files

kindle_del_sdr.py

#!/usr/bin/env python
# -*- coding: utf-8 -*-
# Create on: 2015年10月25日 下午4:48:30
# 说明：删除重复文件 dfs递归搜索目录判断有垃圾sdr文件
# Author：lixingcong

import os
import shutil
import time

# Generator of un-duplicated files
def gen_a_dup_list(sdr_, normal_file_):
	for sdr in sdr_:
		sdr_cut = sdr[:-4]
		if sdr_cut not in normal_file_:
			yield sdr
			
# Super remove not-empty files
def rm_dir(path, count):
	try:
		shutil.rmtree(path)
		print "#%d Deleted: %s" % (count, path)
	except:
		print "#%d ERROR deleting dir: %s" % (count, path)

# Delete all useless sdr files
def main_delete(path):
	print '== ==' * 10
	print "Now Scan for trash in this folder:", path
	sdr = []
	normal_files = []
	sub_dirs = []
	
	list_dir = os.listdir(path)
	for t in list_dir:
		# Collect sdr files
		if t.endswith('.sdr'):
			sdr.append(t)
			# print t[:-4]

		# Collect book files
		elif os.path.isfile(path + t):
			len_of_filename = len(t)
			for rev_x in xrange(2, len_of_filename):
				# extand name will be removed
				if t[-rev_x] == '.':
					t1 = t[0:-rev_x]
					normal_files.append(t1)
					break
					
		# Collect sub-folders
		elif os.path.isdir(path + t):
			sub_dirs.append(t)

	sdr_to_del = gen_a_dup_list(sdr, normal_files)
	# To check if the generator is empty, you should change it to list
	# If you have not changed, you should use method next() then throw a exception 
	del_list = list(sdr_to_del)
	if del_list == []:
		print "Yes, Sir. There is no useless sdr!"
	else:
		count = 1
		for sdr_ in del_list:
			rm_dir(path + sdr_, count)
			count = count + 1
			
	# Recusive delete
	if sub_dirs != []:
		for sub_dirs_ in sub_dirs:
			# You should not forget to add '/' as its tail
			# to prevent some serious problem in windows
			main_delete(path + sub_dirs_ + '/')
	
if __name__ == "__main__":
	start_time=time.time()
	# You should not forget to add '/' as its tail
	main_delete('./documents/')
	end_time=time.time()
	print '== =='*10
	print 'Time elapsed: %.2fs'%(end_time-start_time)