Command line 'cut' utility that can handle csv quoting. This allows proper handling of fields that contain delimiters, both field and record delimiters like commas and newlines. Thanks to github.com/JoeGermuska for the initial version of the code.

csvcut

#!/usr/bin/env python
"""
from https://gist.github.com/JoeGermuska/561347
Like cut, but for CSVs. To be used from a shell command line.
 
Note that fields are 1-based, similar to the UNIX 'cut' command.
 
Should use something better than getopt, but this works...
 
Usage: 
    csvcut foobar.csv
    (prints the first column of each row of foobar.csv)
    
    head -10 foobar.csv | csvcut -f 1,3
    (prints the first and third columns of the first ten lines of foobar.csv)
 
    csvcut -f 1,3 -d "|" foobar.csv
    (prints the first and third columns of the pipe-delimited foobar.csv)
 
    csvcut -f 1,3 -t foobar.csv
    (prints the first and third columns of the tab-delimited foobar.csv
     if present, the -d option will be ignored.)
 
    csvcut -f 1,3 -T foobar.csv
    (prints the first and third columns of the csv-delimited foobar.csv
     output will be tab-delimited, and if present, the -o option will be ignored.)
 
    csvcut -h foobar.csv
    (prints the values of the first line of foobar.csv, preceded by the field index which would 
     be used to display that column.  If present, the -f option will be ignored.)
 
    csvcut -f 1,2,3 -d "|" -o , foobar.csv
    (prints the first three columns of the pipe-delimited foobar.csv; output
    will be comma-delimited.)
 
    csvcut -f 1,2,3 -o "|" foobar.csv
    (prints the first three columns of the comma-delimited foobar.csv; output
    will be pipe-delimited.)
 
    csvcut -o "|" foobar.csv
    (prints all the columns of the comma-delimited foobar.csv; output will be
    pipe-delimited.)
 
    csvcut -f 1,2 -d "," -q "|" foobar.csv
    (prints the first two columns of the comma-delimited, pipe-quoted foorbar.csv.)
"""
import sys, csv, getopt
opts, args = getopt.getopt(sys.argv[1:], "f:d:o:q:htT", [])
if args:
    i = open(args[0],"U")
else:
    i = sys.stdin
 
delimiter = ','
output_delimiter = ','
cols = []
show_headers = False
 
if opts:
    opts = dict(opts)
    show_headers = '-h' in opts
    if '-f' in opts:
        # reduce field index values by 1 to work with python's 0-based indexing
        cols = [(int(c) - 1) for c in opts['-f'].split(",")]
    if '-t' in opts:
        delimiter = "\t"
    elif '-d' in opts:
        delimiter = opts['-d']
    if '-T' in opts:
        output_delimiter = "\t"
    elif '-o' in opts:
        output_delimiter = opts['-o']
    if '-q' in opts:
        quotechar = opts['-q']
    else:
        quotechar = None
 
writer = csv.writer(sys.stdout, delimiter=output_delimiter)
for row in csv.reader(i, delimiter=delimiter, quotechar=quotechar):
    if show_headers:
        for i,c in enumerate(row):
            print "%3i: %s" % (i+1,c) # add 1 to make 1-based
        break
    if len(cols) == 0:
        # print all fields
        writer.writerow(row)
    else:
        # If some request fields are missing, skip them and keep going
        writer.writerow([row[c] for c in cols if c < len(row)])

This version can tolerate incomplete rows. For example, if column 10 is requested, but some rows only have 5 columns, then the program will continue without error. No missing values will be filled in. In the original version of the code, a fatal error would stop the program.

Other changes:

• moved the writer definition and int() casting outside the loop, assuming it would speed things up.
• the behavior of the "-f" parameter is different; to print all fields, omit this parameter. It now uses 1-based indexing to be consistent with UNIX 'cut' usage
• the default output delimiter has been changed to a comma.
• enabled tab-delimited output with new "-T" flag