处理任意格式的文本文件

Posted 2020-12-03

# 内存--硬盘内容  序列话
# 手动挡
# f = open(‘文件名或类似文件的东西‘， ‘文件打卡模式‘)
# f是文件对象或指针，用来进行读写操作
# f.close()

# 三种模式：
# w. write 写
# r read 读
# a append  追加内容

import os 
%pwd

‘C:\study\jupyter‘

f = open(‘coop.txt‘, ‘w‘)  # 用w的方式打开文件，不存在则创建
f.write(‘coop‘ * 7)  # 向文件写入字符串
f.close()

with open(‘coop-1.txt‘, ‘w‘) as f:
    f.write(‘coop‘ * 7)

with open(‘coop.txt‘) as f: # 文件名后面的r是默认模式
    data = f.read()# 读出所有内容，保存到一个变量
    print(data)

coopcoopcoopcoopcoopcoopcoop

# 在打开文件时要考虑此文件是否存在，使用try except

with open(‘coop.txt‘, ‘a‘) as f:
    f.write(‘coop1
‘)
    f.write(‘coop2
‘)
    f.write(‘
111‘)
    f.write(‘
222‘)

with open(‘coop.txt‘) as f:  
    print(f.readline())  # 每次读一行
    print(f.readline())
    print(f.readline())
    print(f.readline())

coopcoopcoopcoopcoopcoopcoopcoop

coop

coop1

coop2

with open(‘coop.txt‘) as f:  # 当文件不是很大时用readlines
    print(f.readlines())   # 如何去掉

[‘coopcoopcoopcoopcoopcoopcoopcoop
‘, ‘coop
‘, ‘coop1
‘, ‘coop2
‘, ‘
‘, ‘111
‘, ‘222‘]

with open(‘coop.txt‘) as f:  
    print(f.tell())  # tell()告诉我们光标现在的位置(列的位置)
    print(f.readline())  # 每次读一行
    print(f.tell())
    print(f.readline())
    print(f.tell())
    print(f.seek(0))  # seek（0）让光标返回到初始0的位置
    print(f.readline())
    print(f.readline())
    f.seek(5)

    print(f.readline())
    print(f.tell())

0
coopcoopcoopcoopcoopcoopcoopcoop

34
coop

40
0
coopcoopcoopcoopcoopcoopcoopcoop

coop

oopcoopcoopcoopcoopcoopcoop

34

f = open(‘coop.txt‘, ‘a‘)
f.write(‘append
‘)
# print(f.readlines())

7

with open(‘coop.txt‘,) as f:
    data = f.read()
    print(data)

coopcoopcoopcoopcoopcoopcoopcoop
coop
coop1
coop2

111
222appendappendappendappendappendappendappend
append
append
append
append
append

##############

# 匹配相应后缀名的文件
import fnmatch
for f in os.listdir(‘.‘):
    if fnmatch.fnmatch(f, ‘*.txt‘):
        print(f)
    elif fnmatch.fnmatch(f, ‘*.pdf)‘):
        print(‘find pdf‘, f)

coop-1.txt
coop.txt

# 匹配相应后缀名的文件
import fnmatch
for f in os.listdir(‘.‘):
    if fnmatch.fnmatch(f, ‘?+.txt‘):  # 正则？，一个字符
        print(f)
    elif fnmatch.fnmatch(f, ‘?.pdf)‘):
        print(‘find pdf‘, f)

#################

import fnmatch
for f in os.listdir(‘.‘):
    if fnmatch.fnmatch(f, ‘w+.txt‘):  # 正则？，一个字符
        print(f)
    elif fnmatch.fnmatch(f, ‘?.pdf)‘):
        print(‘find pdf‘, f)

# 单纯匹配某种命名规则的文件
import glob
for f in glob.glob(‘[0-9].txt‘):
    print(f)

0.txt
1.txt

import glob
for f in glob.glob(‘[0-9]+.txt‘):  # 不可以加+号，已匹配更多字符
    print(f)

############################

# 序列化 picle ，持久化， 存盘
# 后缀名随意，推荐使用pkl
# 存储python的数据结构
name_list = [‘coop‘, ‘fang‘, ‘beijing‘]
data = {‘name‘:name_list, ‘age‘:(2,3,4)}

import pickle
with open(‘data.pkl‘, ‘wb‘) as f: # 使用wb，通用二进制存储
    pickle.dump(data, f)

with open(‘data.pkl‘, ‘rb‘) as f:
    data = pickle.load(f)
    print(data)

{‘name‘: [‘coop‘, ‘fang‘, ‘beijing‘], ‘age‘: (2, 3, 4)}

############################

# 虚拟文件，临时文件，不需要真的存到磁盘
import io
output = io.StringIO()
output.write(‘the first code
‘)
print(‘ddd‘, file=output)

# 去除内容
contents = output.getvalue()
print(contents)

#关闭文件，清理缓存
output.close()   # 打印顺序为什么是那个样子

the first code
ddd

# 用类似字典的方式存储任意的python对象  pickle存储的是数据结构
import shelve
with shelve.open(‘coop.she‘) as so:
    so[‘coop‘] = ‘fang‘  # 生成三个文件

with shelve.open(‘coop.she‘) as so:
    print(so[‘coop‘])

fang