为什么grep在子进程stdin中返回所有输出？

Question

ubuntu@ubuntu:/home/ubuntuUser$ cat test.txt 
This is a test file
used to validate file handling programs
#pyName: test.txt
this is the last line
ubuntu@ubuntu:/home/ubuntuUser$ cat test.txt | grep "#pyName"
#pyName: test.txt
ubuntu@ubuntu:/home/ubuntuUser$ "

#1  >>> a = subprocess.Popen(['cat test.txt'], 
                             stdin=subprocess.PIPE,
                             stdout=subprocess.PIPE, 
                             stderr=subprocess.PIPE, 
                             shell=True)
#2  >>> o, e = a.communicate(input='grep #pyName')
#3  >>> print o
#3  This is a test file
#3  used to validate file handling programs
#3  #pyName: test.txt
#3  this is the last line
#3  
#3  >>> 

问题：

问题1： 文件上的shell grep命令只打印匹配的行，而grep via子进程打印整个文件。怎么了？在

问题2： 如何将通过communicate（）发送的输入添加到 初始命令（'cat测试.txt')? 在#2之后，在“|”和shell命令变得像cat test.txt | grep #pyName之后，初始命令是否会被来自communicate的输入字符串追加？在

Answer 1

@prasath如果您正在寻找使用communicate（）的示例

[root@ichristo_dev]# cat process.py    A program that reads stdin for input
#! /usr/bin/python

inp = 0  
while(int(inp) != 10):  
    print "Enter a value: "  
    inp = raw_input()  
    print "Got", inp  


[root@ichristo_dev]# cat communicate.py
#! /usr/bin/python

from subprocess import Popen, PIPE  

p = Popen("./process.py", stdin=PIPE, stdout=PIPE, stderr=PIPE, shell=True)  
o, e = p.communicate("10")  
print o  


[root@ichristo_dev]#./communicate.py  
Enter a value:   
Got 10