1. 最基本的读文件方法:
# File: readline-example-1.py
file = open("sample.txt")
while 1:
line = file.readline()
if not line:
break
pass # do something
一行一行得从文件读数据,显然比较慢;不过很省内存。
在我的机器上读10M的sample.txt文件,每秒大约读32000行
2. 用fileinput模块
# File: readline-example-2.py
import fileinput
for line in fileinput.input("sample.txt"):
pass
写法简单一些,不过测试以后发现每秒只能读13000行数据,效率比上一种方法慢了两倍多……
3. 带缓存的文件读取
# File: readline-example-3.py
file = open("sample.txt")
while 1:
lines = file.readlines(100000)
if not lines:
break
for line in lines:
pass # do something
这个方法真的更好吗?事实证明,用同样的数据测试,它每秒可以读96900行数据!效率是第一种方法的3倍,第二种方法的7倍!
————————————————————————————————————————————————————————————
在Python 2.2以后,我们可以直接对一个file对象使用for循环读每行数据:
# File: readline-example-5.py
file = open("sample.txt")
for line in file:
pass # do something
而在Python 2.1里,你只能用xreadlines迭代器来实现:
# File: readline-example-4.py
file = open("sample.txt")
for line in file.xreadlines():
pass # do something