def main(argv=None):
"""script main.
parses command line options in sys.argv, unless *argv* is given.
"""
if argv is None:
argv = sys.argv
parser = E.OptionParser(
version=
"%prog version: $Id: csv_cut.py 2782 2009-09-10 11:40:29Z andreas $",
usage=globals()["__doc__"])
parser.add_option("-r",
"--remove",
dest="remove",
action="store_true",
help="remove specified columns, keep all others.")
parser.add_option("-u",
"--unique",
dest="unique",
action="store_true",
help="output rows are uniq.")
parser.add_option(
"-l",
"--large",
dest="large",
action="store_true",
help=
"large columns. Do not use native python CSV module [default=%default]."
)
parser.add_option("-f",
"--filename-fields",
dest="filename_fields",
type="string",
help="filename with field information.")
parser.set_defaults(
remove=False,
unique=False,
filename_fields=None,
)
(options, args) = E.Start(parser, add_csv_options=True, quiet=True)
input_fields = args
if options.filename_fields:
input_fields = map(
lambda x: x[:-1].split("\t")[0],
filter(lambda x: x[0] != "#",
open(options.filename_fields, "r").readlines()))
if options.unique:
outfile = UniqueBuffer(sys.stdout)
else:
outfile = options.stdout
while 1:
line = sys.stdin.readline()
if not line:
E.Stop()
sys.exit(0)
if line[0] == "#":
continue
first_line = line
break
old_fields = first_line[:-1].split("\t")
fields = []
for f in input_fields:
# do pattern search
if f[0] == "%" and f[-1] == "%":
pattern = re.compile(f[1:-1])
for o in old_fields:
if pattern.search(o) and o not in fields:
fields.append(o)
else:
if f in old_fields:
fields.append(f)
if options.remove:
fields = set(fields)
fields = [x for x in old_fields if x not in fields]
if options.large:
reader = CSV.DictReaderLarge(CommentStripper(sys.stdin),
fieldnames=old_fields,
dialect=options.csv_dialect)
else:
reader = csv.DictReader(CommentStripper(sys.stdin),
fieldnames=old_fields,
dialect=options.csv_dialect)
writer = csv.DictWriter(outfile,
fields,
dialect=options.csv_dialect,
lineterminator=options.csv_lineterminator,
extrasaction='ignore')
print "\t".join(fields)
first_row = True
ninput, noutput, nerrors = 0, 0, 0
while 1:
ninput += 1
try:
row = reader.next()
except _csv.Error, msg:
options.stderr.write("# error while parsing: %s\n" % (msg))
nerrors += 1
continue
except StopIteration:
break
def main(argv=None):
"""script main.
parses command line options in sys.argv, unless *argv* is given.
"""
if argv is None:
argv = sys.argv
parser = E.OptionParser(version="%prog version: $Id: csv_cut.py 2782 2009-09-10 11:40:29Z andreas $",
usage=globals()["__doc__"])
parser.add_option("-r", "--remove", dest="remove", action="store_true",
help="remove specified columns, keep all others.")
parser.add_option("-u", "--unique", dest="unique", action="store_true",
help="output rows are uniq.")
parser.add_option("-l", "--large", dest="large", action="store_true",
help="large columns. Do not use native python CSV module [default=%default].")
parser.add_option("-f", "--filename-fields", dest="filename_fields", type="string",
help="filename with field information.")
parser.set_defaults(
remove=False,
unique=False,
filename_fields=None,
)
(options, args) = E.Start(parser,
add_csv_options=True,
quiet=True)
statement = " ".join(args)
if options.large:
reader = CSV.DictReaderLarge(CommentStripper(sys.stdin),
dialect=options.csv_dialect)
else:
reader = csv.DictReader(CommentStripper(sys.stdin),
dialect=options.csv_dialect)
exec "f = lambda r: %s" % statement in locals()
counter = E.Counter()
writer = csv.DictWriter(options.stdout,
reader.fieldnames,
dialect=options.csv_dialect,
lineterminator=options.csv_lineterminator)
writer.writerow(dict((fn, fn) for fn in reader.fieldnames))
while 1:
counter.input += 1
try:
row = reader.next()
except _csv.Error, msg:
options.stderr.write("# error while parsing: %s\n" % (msg))
counter.errors += 1
continue
except StopIteration:
break
