current community

your communities

more stack exchange communities

Stack Exchange
tour help
careers 2.0
Take the 2-minute tour ×
Stack Overflow is a question and answer site for professional and enthusiast programmers. It's 100% free, no registration required.
up vote 2 down vote favorite

I have two simple scripts - I am trying to pass some information (date as input into the python script) to the bash script. Here's the python one:

#!/usr/local/bin/python

import os
import sys
import subprocess

year = "2012"
month = "5"
month_name = "may"

file = open('date.tmp','w')
file.write(year + "\n")
file.write(month + "\n")
file.write(month_name + "\n")
file.close

subprocess.call("/home/lukasz/bashdate.sh")

And here's the bash one:

#!/bin/bash

cat /home/lukasz/date.tmp | \
while read CMD; do
    echo -e $CMD
done
rm /home/lukasz/date.tmp

Python script works fine without issues. It calls the bash script but it looks like the while loop just does not run. I know the bash script does run overall because the rm command gets executed and the date.tmp file is removed. However if I comment out the subprocess call in python then run the bash script manually it works fine displaying each line.

Brief explanation of what I am trying to accomplish. I have a python script that exports a very large DB to CSV (almost 300 tables and a few gigs of data) which then calls the bash script to zip the CSVs into one file and move it to another location. I need to pass the month and year supplied to the python script to the bash script.

share|improve this question
    
As an aside, are those three variables all you need to pass in your actual script? If so, I'd suggest passing them as command-line arguments rather than mucking about with temporary files. Something like subprocess.call(["/home/lukasz/bashdate.sh", year, month, month_name]) in Python, and then year=$1; month=$2; month_name=$3 in Bash. To check the number of arguments passed to the Bash script, use if [ $# -lt 3 ]; then echo "More arguments please."; exit 1; fi (or change -lt to -eq if you want exactly three arguments). –  Blair Jan 2 '13 at 21:46
    
Why do you need a bash script to zip the CSVs into one file and move it to another location? Do you not know about shutil and/or zipfile and os in the standard library? –  abarnert Jan 2 '13 at 22:53
    
@abarnert This is directly from shutil page on python.org: Warning Even the higher-level file copying functions (shutil.copy(), shutil.copy2()) can’t copy all file metadata. Besides this (which is an issue) the other problem is the CSVs are in /tmp (it's the only location to which mysqld can write files easily due to permissions). In the end the solution to just call a bash script was far simpler and it works. Considering this is a thing that is done once a month manually (it requires some other user input for documentation) its good enough for my needs. –  Lukasz Jan 3 '13 at 17:59
    
@Lukasz: Since you didn't show us how you do the copying in your bash script, it's impossible to know whether it's possible to do the same thing in Python. But I don't understand why you think Python can't read files in /tmp. I'm also not sure why you think a solution you can't figure out yourself is "far simpler" than something you probably could figure out with no problems. –  abarnert Jan 3 '13 at 19:43

3 Answers 3

up vote 2 down vote accepted

Instead of writing a temp file, send the values of year, month, and month-name to the bash script as parameters. Ie, in the Python code remove all the lines with file in them, and replace
subprocess.call("/home/lukasz/bashdate.sh")
with
subprocess.call(['/home/lukasz/bashdate.sh', year, month, month_name])

and in the bash script, replace the cat ... rm lines with (eg)
y=$1; m=$2; mn=$3
which puts the year, month, and month-name into shell variables y, m, and mn.

share|improve this answer
    
You could even re-work the bash script to use getopt if you really wanted to. It wouldn't be too much work. –  mgilson Jan 2 '13 at 21:59
    
Oh I had no idea that can be done. Probably should have looked at subprocess.call more. Thanks! –  Lukasz Jan 2 '13 at 21:59
up vote 5 down vote

I believe that you need file.close() instead of file.close. With the latter, you're not actually closing the file since you don't call the method. Since you haven't actually closed the file yet, it might not be flushed and so the entire contents of the file might be buffered rather than written to disk.

As a side note, these things are taken care of automatically if you use a context manager:

with open('foofile','w') as fout:
    fout.write("this data")
    fout.write("that data")

#Sleep well tonight knowing that python guarantees your file is closed properly
do_more_stuff(blah,foo,bar,baz,qux)
share|improve this answer
    
+1 - beaten to the punch. Tested it on my system and this made it work. –  Blair Jan 2 '13 at 21:34
    
That worked - I will mark as answered once StackOverflow will let me. –  Lukasz Jan 2 '13 at 21:40
up vote -1 down vote

Maybe try adding shell=True to the call: 

subprocess.call("/home/lukasz/bashdate.sh", shell=True)
share|improve this answer
2  
I really don't think that will make any difference in this case (other than spawning a shell unnecessarily) –  mgilson Jan 2 '13 at 21:38

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.