Python处理超大文件排序：数据切割与归并排序实战

def get_file_lines(file_path):
 # 目标文件的路径
 file_path = str(file_path)
 with open(file_path, 'rb') as file:
  # 定义行数
  i = 0
  while True:
   # 一次读取一行数据
   line = file.readline()
   if not line :
    break
   else:
    # 每读一行，行数加一
    i += 1
   #设置进度条，每当i读取1000000行时打印一次i
   # 每当读取1000000的整数倍行时，打印行数（进度条）
   if i % 1000000 == 0:
    print(i)
  # 打印总行数
  print(i)
  return i

# 定义均等切割函数，num是待切割的文件的行数的值，n为切割份数,file_path是待切割的文件，file_dir是切割好的文件写入的目录
def evg_split(num, n, file_path, file_dir):
 last_list = []
 # 如果样本刚好可以整除为n份
 if num % n == 0:
  for i in range(n):
   # 则直接将样本分为n分没份对应num/n个
   last_list.append(num / n)
 # 如果不能整除
 if num % n != 0:
  # 如果不能整除，则先将num整除n-1并取n-1份,余下的单独做一份
  evg = (num - num % n) // (n - 1)
  for i in range(n):
   last_list.append(evg)
  last_list.append(num % (n - 1))
 print(last_list)
 # return last_list
 # 对应于切割后的每一份数据
 with open(file_path, 'rb') as path:
  for i in range(n):
   # 创建临时文件
   tmp_file = file_dir + str(i) + '.txt'
   # 打开临时文件，将内容一条一条的写入
   file = open(tmp_file, 'wb')
   for j in range(int(last_list[i])):
    line = path.readline()
    file.write(line)
    print(line)
   print('------------')
   file.close()

def merge( mylist1, mylist2, file1):
 while len(mylist1) > 0 and len(mylist2) > 0:
  if mylist1[0]<mylist2[0]:
   with open(file1,'a') as file:
    file.write(str(mylist1[0]))
    del mylist1[0]
  elif mylist1[0] > mylist2[0]:
   with open(file1,'a') as file:
    file.write(str(mylist2[0]))
    del mylist2[0]
  else:
   with open(file1,'a') as file:
    file.write(str(mylist1[0]))
    file.write(str(mylist2[0]))
    del mylist1[0]
    del mylist2[0]
 with open(file1, 'a') as file:
  for i in mylist1:
   file.write(str(i))
  for i in mylist2:
   file.write(str(i))