def main():
r"""
Reads a csv file, keeping only certain columns. Prints to stdout.
Examples
---------
Read a comma delimited csv file, data.csv, keep the 'name' column
$ python cut.py -l name,age test/commafile.csv
Use a tab delimited dataset
$ python cut.py -d'\t' -l name test/tabfile.csv
Note that -dt -dtab -d\t -d'\t' -d\\t also work
"""
usage = "usage: %prog [options] dataset"
usage += '\n'+main.__doc__
parser = OptionParser(usage=usage)
parser.add_option(
"-l", "--keep_list",
help="Only keep variables in this (comma delimited) list."
" [default: %default] ",
action="store", dest='keep_list', default=None)
parser.add_option(
"-d", "--delimiter",
help="Use DELIMITER as the column delimiter. [default: %default]",
action="store", dest='delimiter', default=',')
parser.add_option(
"-o", "--outfilename",
help="Write to this file rather than stdout. [default: %default]",
action="store", dest='outfilename', default=None)
(options, args) = parser.parse_args()
### Parse args
# Raise an exception if the length of args is greater than 1
assert len(args) <= 1
# If an argument is given, then it is the 'infilename'
# If no arguments are given, set infilename equal to None
infilename = args[0] if args else None
## Handle the options
# Change keep_list to a Python list
keep_list = options.keep_list.split(',') if options.keep_list else None
# Deal with tabs
if options.delimiter in ['t', '\\t', '\t', 'tab']:
options.delimiter = '\t'
## Get the infile/outfile
infile, outfile = common.get_inout_files(infilename, options.outfilename)
## Call the function that does the real work
cut_file(infile, outfile, delimiter=options.delimiter, keep_list=keep_list)
## Close the files iff not stdin, stdout
common.close_files(infile, outfile)
def main():
r"""
DESCRIPTION
-----------
Subsample files or stdin and write to stdout.
NOTES
-----
Assumes the first row is a header.
EXAMPLES
---------
Subsample a comma delimited dataset and redirect output to a new file
$ python subsample.py data.csv > subsampled_data.csv
Subsample, keeping only 10% of rows
$ python subsample.py -r 0.1 data.csv
"""
usage = "usage: %prog [options] dataset"
usage += '\n'+main.__doc__
parser = OptionParser(usage=usage)
parser.add_option(
"-r", "--subsample_rate",
help="Subsample subsample_rate, 0 <= r <= 1. E.g. r = 0.1 keeps 10% "
"of rows. [default: %default] ",
action="store", dest='subsample_rate', type=float, default=0.01)
parser.add_option(
"-n", "--subsample_number",
help="Subsample subsample_number, It's latent unless -r not specified"
"and -n specified. E.g. n = 100 keeps 100 of rows. [default: %default] ",
action="store", dest='subsample_number', type=int, default=1)
parser.add_option(
"-d", "--delimiter",
help="Use DELIMITER as the column delimiter. [default: %default]",
action="store", dest='delimiter', default=',')
parser.add_option(
"-s", "--seed",
help="Integer to seed the random number generator with. "
"[default: %default] ",
action="store", dest='seed', type=int, default=None)
parser.add_option(
"-o", "--outfilename",
help="Write to this file rather than stdout. [default: %default]",
action="store", dest='outfilename', default=None)
(options, args) = parser.parse_args()
### Parse args
# Raise an exception if the length of args is greater than 1
assert len(args) <= 1
# If an argument is given, then it is the 'infilename'
# If no arguments are given, set infilename equal to None
infilename = args[0] if args else None
## Handle the options
# Deal with tabs
if options.delimiter in ['t', '\\t', '\t', 'tab']:
options.delimiter = '\t'
## Get the infile/outfile
infile, outfile = common.get_inout_files(infilename, options.outfilename)
## Call the function that does the real work
subsample(
infile, outfile, options.subsample_rate, options.subsample_number, options.delimiter,
options.seed)
## Close the files iff not stdin, stdout
common.close_files(infile, outfile)