write python script to run the shell script?

Question

I run the svn status got the modified files :

 svn status
 ? .settings
 ? .buildpath
 ? .directory
 A A.php
 M B.php
 D html/C.html
 M html/D.fr
 M api/E.api
 M F.php
 ..

After I want to keep all of these files

zcvf MY.tar.gz all files that svn stat display (not include ? just M,A,D)

My idea is to create the python script can run the shell,because right now to do this I just copy the file name one by one.

zcvf MY.tar.gz all the files that we run svn stat

Anybody could guide how to do this or some related tutorial? But if you think it difficult than copy && paste I will ignore my trying?

thanks

• user2665694: Redirect the the output of "svn status" then parse the output file using Python. Where is the problem?
• kurumi: if you can, do it with the shell..Why the need to write a Python script?
• Steven Rumbalski: Your question is answered here: stackoverflow.com/questions/4760215/…

Answer 1

I don't see why you would use python for this if you can do it in a single line of code in the shell.

svn status | grep "^[AMD]" | sed 's/^.\{8\}//' | xargs zcvf My.tar.gz

I used grep to only select lines that are modified, if you want all files that svn status lists (also those that are added / deleted) you can leave that part out. I've used sed here to delete the first part of every line, I'm sure there is an easier way to do that but I can't think of one right now.

• tree em: you said "I've used sed here to delete the first part of every line" why?
• Yexo: Because "svn status" gives lines like M file.txt. You are only interested in the file.txt part.
• Yexo: Instead of sed you could also use cut -b9

Answer 2

Once you figure out your command as a string you can just call it with subprocess

subprocess

This module spawns called processes and allows you to control them. From there its up to you.

• Jakob Bowyer: I don't need examples :L you should edit and add the examples to someones post

Answer 3

You could use check_output() and check_call() functions:

#!/usr/bin/env python
from subprocess import check_call, check_output as qx
filenames = [line[8:] # extract filename
 for line in qx(['svn', 'status']).splitlines()
 if not line.startswith('?')] # exclude files that are not under VC
check_call(['tar', 'cvfz', 'MY.tar.gz'] + filenames)

On Python < 2.7 see check_output() recipe.

Answer 4

subprocess is the Pythonic way, but using a small bash one-liner could be a shorter idea.

Bash one-liner

svn status | egrep "^ M" | awk '{s=s 2ドル " "} END {print "tar cvfz MY.tar "s}'

Subprocess

import subprocess as sp
p=sp.Popen('svn status', shell=True, stdout=sp.PIPE, 
 stderr=sp.PIPE).communicate()[0]
files=[]
for line in p:
 if line.strip().find('M')==0:
 files.append(line.split(' ')[1].strip())
files=' '.join(files)
sp.Popen('tar cvfz MY.tar.gz '+files, shell=True).communicate()

• Yexo: "grep M" is not correct, it'll also select lines like "A MyNewFile.txt"
• Adam Matan: Fixed to an egrep regex pattern.
• Yexo: In your new python code you make exactly the same mistake. You shouldn't compare to the result of line.find('M') to !=-1 but to ==0 instead. Your python code will also fail for files with spaces in the name.