def action(args):
# for debugging:
# pandas.set_option('display.max_columns', None)
# pd.set_option('display.max_rows', None)
dfs = []
columns = None # to preserve column order
for csv in args.csv:
df = utils.read_csv(csv,
dtype=str,
nrows=args.limit,
comment='#',
na_filter=False,
header=None if args.no_header else 0)
columns = df.columns
dfs.append(df)
df = pandas.concat(dfs, ignore_index=True)
if not args.stack:
if args.on:
on = args.on.split(',')
if args.no_header:
on = map(int, on)
df = df.groupby(by=on, sort=False)
df = df.tail(1) if args.take_last else df.head(1)
else:
df = df.drop_duplicates(take_last=args.take_last)
df.to_csv(args.out, columns=columns, index=False)
def action(args):
# for debugging:
# pandas.set_option('display.max_columns', None)
# pd.set_option('display.max_rows', None)
dfs = []
for csv in args.csv:
dfs.append(
utils.read_csv(csv,
dtype=str,
comment=args.comment,
na_filter=False,
header=None if args.no_header else 0))
if args.on:
on = args.on.split(',')
if args.no_header:
on = map(int, on)
else:
on = None
df = dfs.pop(0)
for d in dfs:
df = df.merge(d, how=args.how, on=on)
df.to_csv(args.out, header=not args.no_header, index=False)
def action(args):
# for debugging:
# pandas.set_option('display.max_columns', None)
# pd.set_option('display.max_rows', None)
dfs = []
for csv in args.csv:
df = utils.read_csv(csv,
dtype=str,
nrows=args.limit,
sep=args.from_delimiter,
comment=args.comment,
na_filter=False,
header=None if args.no_header else 0)
dfs.append(df)
df = pandas.concat(dfs, ignore_index=True)
if args.header:
df.columns = args.header.split(',')
df.to_csv(args.out,
header=args.header or not args.no_header,
sep=args.to_delimiter,
index=False)
def action(args):
# for debugging:
# pandas.set_option('display.max_columns', None)
# pd.set_option('display.max_rows', None)
dfs = []
for csv in args.csv:
df = utils.read_csv(
csv,
dtype=str,
nrows=args.limit,
sep=args.from_delimiter,
comment=args.comment,
na_filter=False,
header=None if args.no_header else 0)
dfs.append(df)
df = pandas.concat(dfs, ignore_index=True)
if args.header:
df.columns = args.header.split(',')
df.to_csv(
args.out,
header=args.header or not args.no_header,
sep=args.to_delimiter,
index=False)
def action(args):
# for debugging:
# pandas.set_option('display.max_columns', None)
# pd.set_option('display.max_rows', None)
dfs = []
for csv in args.csv:
dfs.append(utils.read_csv(
csv,
dtype=str,
comment=args.comment,
na_filter=False,
header=None if args.no_header else 0))
if args.on:
on = args.on.split(',')
if args.no_header:
on = map(int, on)
else:
on = None
df = dfs.pop(0)
for d in dfs:
df = df.merge(d, how=args.how, on=on)
df.to_csv(args.out, header=not args.no_header, index=False)
