Question

Python 2.4
For my example I have a 2 column csv file

Eg:

HOST, FILE
server1, /path/to/file1
server2, /path/to/file2
server3, /path/to/file3

I would like to get the file size of the object at PATH for each row in the csv FILE, then add that value to the csv FILE on a new column.
Making it:

 HOST, PATH, FILESIZE
 server1, /path/to/file1, 6546542
 server2, /path/to/file2, 46546343
 server3, /path/to/file3, 87523

Ive tried a couple methods but havnt had much success.

The code below executes fileSizeCmd (du -b) on the PATH and outputs the filezie correctly, but I havnt figured out how to use the data to add to the csv FILE

 import datetime
 import csv
 import os, time
 from subprocess import Popen, PIPE, STDOUT

 now = datetime.datetime.now()
 fileSizeCmd = "du -b"
 SP = " "

 # Try to get disk size and append to another row after entry above
 #st = os.stat(row[3])
 #except IOError:
 #print "failed to get information about", file
 #else:
 #print "file size:", st[ST_SIZE]
 #print "file modified:", time.asctime(time.localtime(st[ST_MTIME]))

 incsv = open('my_list.csv', 'rb')
 try:
     reader = csv.reader(incsv)
     outcsv = open('results/results_' + now.strftime("%m-%d-%Y") + '.csv', 'wb')
     try:
         writer = csv.writer(outcsv)

         for row in reader:
         p = Popen(fileSizeCmd + SP + row[1], shell=True, stdin=PIPE, stdout=PIPE, stderr=PIPE)
         stdout, empty = p.communicate()


         print 'Command: %s\nOutput: %s\n' % (fileSizeCmd + SP + row[1], stdout)

         #  Results in bytes example
         #
         #  Output:
         #  8589935104      /path/to/file
         #

     #  Write 8589935104 to new column of csv FILE

   finally:
      outcsv.close()

 finally:
incsv.close()

Answer 1

Sketch w/o error handling:

#!/usr/bin/env python

import csv
import os

filename = "sample.csv"
# localhost, 01.html.bak
# localhost, 01.htmlbak
# ...

def filesize(filename):
    # no need to shell out for filesize
    return os.stat(filename).st_size

with open(filename, 'rb') as handle:
    reader = csv.reader(handle)
    # result is written to sample.csv.updated.csv
    writer = csv.writer(open('%s.updated.csv' % filename, 'w'))
    for row in reader:
        # need to strip filename, just in case
        writer.writerow(row + [ filesize(row[1].strip()) ])

# result
# localhost, 01.html.bak,10021
# localhost, 01.htmlbak,218982
# ...