6_文件IO

1. 基本文件读取

        readline()，readlines()，write()，writelines()

        f.read(size)，指定读取文件的字节数，须要注意的是，对于同一个文件，一直调用这个函数，文件指针会一直移动直到文件尾。

        因此若是要中途修改文件指针的位置，那么须要调用seek()函数。经常使用的是seek(0)：移动到文件头；seek(n)：移动到文件位置n处；seek(0,n)：移动到文件尾。

        f.tell()函数存储文件指针当前的位置

---

---

2. 文件操做

       2.1 文件复制

              除了手动编写，也能够使用现成模块

import shutil

shutil.copyfile('1.txt','2.txt')
##复制文件2.txt到1.txt中

       2.2 文件删除

import os,os.path

filename = 'test.txt'
if os.path.exists(filename):                ##判断文件是否存在
    os.remove(filename)
else:
    print('not exist!')

       2.3 文件重命名

import os

filename = 'test.txt'
rename = 'write.txt'

os.rename(filename,rename)

---

---

3. 目录操做

       3.1 目录建立

import os

os.listdir('f:/')

os.mkdir('f:/newdir')                       ##建立一个新目录

os.mkdirs('f:/newdir1/newdir2')             ##建立一个二级目录

       3.2 目录删除

import os

os.rmdir('f:/newdir')                       ##只能删除空目录

os.removedirs('f:/newdir1/newdir2')         ##删除两级目录

shutil.rmtree('f:/newdir')                  ##能够删除非空目录

       3.3 目录遍历

###递归法

import os

def visitdir1(path):
    for lists in os.listdir(path):
        sub_path = os.path.join(path,lists)
        print(sub_path)
        if os.path.isdir(sub_path):
            visitdir(sub_path)

​`````````````````````````````````````````````````````````````````````````````````````
###os.walk法

import os
def visitdir2(path):
    list_dirs = os.walk(path)               ##返回三元组：路径名，目录列表，文件列表
    for root,dirs,files in list_dirs:
        for d in dirs:
            print(os.path.join(root,d))
        
        for f in files:
            print(os.path.join(root,f))

---

---

4. 综合应用

       题目

              音频文件的根目录在：/dnn4_added/0_ASR/01_resource/original_speech_data/AISHELL-ASR009-zh-cn-mandarin178/data_aishell/wav

              说话信息在/dnn4_added/0_ASR/01_resource/original_speech_data/AISHELL-ASR009-zh-cn-mandarin178/data_aishell/transcript/aishell_transcript_v0.8.txt

              请获得一个列表文件，文件里面显示音频全路径和说话信息，并把说话信息里的空格都去掉。而且把音频文件的后缀名改为.mfcc

       代码

import os
import os.path
import json
import sys


filter = [".wav"]           ##设置文件过滤后缀
postfix = '.mfcc'           ##题目要求的后缀

dirrname = '/dnn4_added/0_ASR/01_resource/original_speech_data/AISHELL-ASR009-zh-cn-mandarin178/data_aishell/wav'
filename = '/dnn4_added/0_ASR/01_resource/original_speech_data/AISHELL-ASR009-zh-cn-mandarin178/data_aishell/transcript/aishell_transcript_v0.8.txt'

##目录遍历，把wav文件都找出来
def all_path(dirname):
    PATHS = []
    for maindir, subdir, file_name_list in os.walk(dirname):
        for filename in file_name_list:
            apath = os.path.join(maindir,filename)
            ext = os.path.splitext(apath)[1]
            
            if ext in filter:
                PATHS.append(apath)
    
    return PATHS

PATHS = all_path(dirrname)

##read txt as dictionary
dict = {}
with open(filename,'r') as file_object:
    for line in file_object:
        split_list = line.split(' ',1)
        split_list[1] = ''.join(split_list[1].split())
        dict[split_list[0]] = split_list[1] 

##find key in dictionary and update it 
for path in PATHS:
    split_dirname = os.path.splitext(path)[0]
    split_basename = os.path.basename(split_dirname)
    if dict.has_key(split_basename):
        split_dirname += postfix
        dict[split_dirname] = dict.pop(split_basename)

##write the new dictionary as txt
keys = list(dict.keys())
values = list(dict.values())
z = list(zip(keys,values))

outputfile = open('dic.txt','w')
for row in z:
    rowtext = '{} {}'.format(row[0],row[1])
    outputfile.write(rowtext)
    outputfile.write('\n')
outputfile.close