def test_read_csv_filepath_dtype(self):
result = rcf.read_csv(self.testdata, dtype={0: str, 1: int, 2: float})
self.assertEqual(
[], result, "Non-empty list returned by correct call to read_csv!")
result = rcf.read_csv(self.testdata,
dtype={
"A": str,
"B": int,
"C": float
})
self.assertEqual(
[], result, "Non-empty list returned by correct call to read_csv!")
result = rcf.read_csv("data/data_noHeader.csv",
names=["Alpha", "Beta", "Gamma"],
dtype={
"Alpha": str,
"Beta": int,
"Gamma": float
})
self.assertEqual(
[], result, "Non-empty list returned by correct call to read_csv!")
result = rcf.read_csv(self.testdata, dtype={0: str, 4: int})
self.assertEqual(["read_csv.filepath-dtype"], result,
"Wrong result when dtype contains too high index!")
result = rcf.read_csv(self.testdata, dtype={"B": str, "Q": int})
self.assertEqual(["read_csv.filepath-dtype"], result,
"Wrong result when dtype contains non-existing key!")
def test_read_csv_filepath_skiprows_skipfooter(self):
result = rcf.read_csv(self.testdata,
skiprows=2,
skipfooter=2,
engine="python")
self.assertEqual(
[], result, "Non-empty list returned by correct call to read_csv!")
result = rcf.read_csv(self.testdata,
skiprows=3,
skipfooter=4,
engine="python")
self.assertEqual(
["read_csv.filepath-skiprows-skipfooter"], result,
"Wrong result returned when skiprows + skipfooter > number of rows!"
)
def test_read_csv_filepath_names(self):
result = rcf.read_csv(self.testdata, names=["D", "E", "F"])
self.assertEqual(
[], result, "Non-empty list returned by correct call to read_csv!")
result = rcf.read_csv("data/data_tabbed.csv",
sep="\t",
names=["D", "E", "F"])
self.assertEqual(
[], result, "Non-empty list returned by correct call to read_csv!")
result = rcf.read_csv("data/data_plus.csv",
skiprows=[0],
names=['Alpha', 'Beta', 'Gamma'])
self.assertEqual(
[], result, "Non-empty list returned by correct call to read_csv!")
result = rcf.read_csv(self.testdata, names=["D", "E"])
self.assertEqual(
[], result, "Non-empty list returned by correct call to read_csv!")
result = rcf.read_csv(self.testdata, names=["D"])
self.assertEqual(["read_csv.filepath-names"], result,
"Wrong result when names has too few elements!")
result = rcf.read_csv(self.testdata, names=["D", "E", "F", "G"])
self.assertEqual(["read_csv.filepath-names"], result,
"Wrong result when names has too many elements!")
result = rcf.read_csv(self.testdata, names="string")
self.assertEqual(
["read_csv.filepath-names"], result,
"Wrong result when names is a string longer than number of columns!"
)
def test_read_csv_filepath_parse_dates(self):
result = rcf.read_csv("data/date_data.csv",
index_col=0,
parse_dates=True)
self.assertEqual(
[], result, "Non-empty list returned by correct call to read_csv!")
result = rcf.read_csv("data/date_data.csv", parse_dates=[0])
self.assertEqual(
[], result, "Non-empty list returned by correct call to read_csv!")
result = rcf.read_csv("data/date_data.csv", parse_dates=["A"])
self.assertEqual(
[], result, "Non-empty list returned by correct call to read_csv!")
result = rcf.read_csv("data/date_data.csv", parse_dates=[[2, 3, 4]])
self.assertEqual(
[], result, "Non-empty list returned by correct call to read_csv!")
result = rcf.read_csv("data/date_data.csv",
parse_dates=[[2, 3, 4]],
index_col=1)
self.assertEqual(
[], result, "Non-empty list returned by correct call to read_csv!")
result = rcf.read_csv("data/date_data.csv",
parse_dates={"P": [2, 3, 4]})
self.assertEqual(
[], result, "Non-empty list returned by correct call to read_csv!")
result = rcf.read_csv("data/date_data.csv", parse_dates=True)
self.assertEqual(
["read_csv.filepath-parse_dates"], result,
"Wrong result returned with parse_dates=True for non-date index!")
result = rcf.read_csv("data/date_data.csv", parse_dates=[1])
self.assertEqual(
["read_csv.filepath-parse_dates"], result,
"Wrong result returned when parse_dates points at non-date column!"
)
result = rcf.read_csv("data/date_data.csv", parse_dates=[[0, 3, 4]])
self.assertEqual(
["read_csv.filepath-parse_dates"], result,
"Wrong result returned when parse_dates points at columns that cannot be combined as a date!"
)
result = rcf.read_csv("data/date_data.csv",
parse_dates={"P": [0, 3, 4]})
self.assertEqual(
["read_csv.filepath-parse_dates"], result,
"Wrong result returned when parse_dates points at columns that cannot be combined as a date!"
)
def test_read_csv_na(self):
result = rcf.read_csv(self.testdata,
na_values=["NaN", "Nan"],
keep_default_na=True)
self.assertEqual(
[], result, "Non-empty list returned by correct call to read_csv!")
result = rcf.read_csv(self.testdata,
na_filter=False,
na_values=["NaN", "Nan"])
self.assertEqual(
["read_csv.na_filter-na_values"], result,
"Wrong result when calling read_csv with na_value specified, but na_filter=False!"
)
result = rcf.read_csv(self.testdata,
na_filter=False,
keep_default_na=True)
self.assertEqual(
["read_csv.na_filter-keep_default_na"], result,
"Wrong result when calling read_csv with keep_default_na specified, but na_filter=False!"
)
def test_read_csv_quotechar_doublequote(self):
result = rcf.read_csv(self.testdata, doublequote=True)
self.assertEqual(
[], result, "Non-empty list returned by correct call to read_csv!")
result = rcf.read_csv(self.testdata, doublequote=False)
self.assertEqual(
[], result, "Non-empty list returned by correct call to read_csv!")
result = rcf.read_csv(self.testdata,
quoting=csv.QUOTE_NONE,
doublequote=True)
self.assertEqual(
["read_csv.quoting-doublequote"], result,
"Wrong result when doublequote specified despite quoting being QOUTE_NONE!"
)
result = rcf.read_csv(self.testdata,
quoting=csv.QUOTE_NONE,
doublequote=False)
self.assertEqual(
["read_csv.quoting-doublequote"], result,
"Wrong result when doublequote specified despite quoting being QOUTE_NONE!"
)
def test_read_csv_bad_lines(self):
result = rcf.read_csv(self.testdata)
self.assertEqual(
[], result, "Non-empty list returned by correct call to read_csv!")
result = rcf.read_csv(self.testdata,
warn_bad_lines=False,
error_bad_lines=False)
self.assertEqual(
[], result, "Non-empty list returned by correct call to read_csv!")
result = rcf.read_csv(self.testdata, error_bad_lines=True)
self.assertEqual(
[], result, "Non-empty list returned by correct call to read_csv!")
result = rcf.read_csv(self.testdata,
warn_bad_lines=False,
error_bad_lines=True)
self.assertEqual(
[], result, "Non-empty list returned by correct call to read_csv!")
result = rcf.read_csv(self.testdata, warn_bad_lines=True)
self.assertEqual(
["read_csv.error_bad_lines-warn_bad_lines"], result,
"Wrong result when read_csv is called with warn_bad_lines=True without setting error_bad_lines to False!"
)
result = rcf.read_csv(self.testdata,
warn_bad_lines=True,
error_bad_lines=True)
self.assertEqual(
["read_csv.error_bad_lines-warn_bad_lines"], result,
"Wrong result when read_csv is called with both error_bad_lines and warn_bad_lines explicitly set to True!"
)
def test_read_csv_header(self):
result = rcf.read_csv(self.testdata, header=1)
self.assertEqual(
[], result, "Non-empty list returned by correct call to read_csv!")
result = rcf.read_csv(self.testdata, prefix="X", header=None)
self.assertEqual(
[], result, "Non-empty list returned by correct call to read_csv!")
result = rcf.read_csv(self.testdata, names=["A", "B", "C"])
self.assertEqual(
[], result, "Non-empty list returned by correct call to read_csv!")
result = rcf.read_csv(self.testdata,
names=["A", "B", "C"],
header=None,
prefix="X")
self.assertEqual(
["read_csv.names-prefix", "read_csv.names-header"], result,
"Wrong result when calling read_csv with both names and prefix, and None-header!"
)
result = rcf.read_csv(self.testdata, names=["A", "B", "C"], header=1)
self.assertEqual(
["read_csv.names-header"], result,
"Wrong result when calling read_csv with both names and header!")
result = rcf.read_csv(self.testdata, prefix="X")
self.assertEqual(
["read_csv.header-prefix"], result,
"Wrong result when calling read_csv with both prefix without setting header=None!"
)
result = rcf.read_csv(self.testdata, header=0, prefix="X")
self.assertEqual(
["read_csv.header-prefix"], result,
"Wrong result when calling read_csv with both non-None header and prefix!"
)
result = rcf.read_csv(self.testdata,
names=["A", "B", "C"],
header=0,
prefix="X")
expectedResult = [
"read_csv.names-prefix", "read_csv.names-header",
"read_csv.header-prefix"
]
self.assertEqual(
expectedResult, result,
"Wrong result when read_csv is called with both names, prefix and non-None header!"
)
def test_read_csv_parse_dates(self):
def f(d):
""" Create date from string on format YYYY-mm-dd """
date_parts = d.split("-")
return datetime.date(int(date_parts[0]), int(date_parts[1]),
int(date_parts[2]))
result = rcf.read_csv("data/date_data.csv",
parse_dates=[0],
infer_datetime_format=True,
keep_date_col=False,
date_parser=f,
dayfirst=True,
cache_dates=True)
self.assertEqual(
[], result, "Non-empty list returned by correct call to read_csv!")
result = rcf.read_csv(self.testdata, infer_datetime_format=True)
self.assertEqual(
["read_csv.parse_dates-infer_datetime_format"], result,
"Wrong result returned when infer_datetime_format is specified but not parse_dates!"
)
result = rcf.read_csv(self.testdata,
parse_dates=False,
infer_datetime_format=True)
self.assertEqual(
["read_csv.parse_dates-infer_datetime_format"], result,
"Wrong result returned when infer_datetime_format is specified but parse_dates is False!"
)
result = rcf.read_csv(self.testdata, keep_date_col=True)
self.assertEqual(
["read_csv.parse_dates-keep_date_col"], result,
"Wrong result returned when keep_date_col is specified but not parse_dates!"
)
result = rcf.read_csv(self.testdata, date_parser=f)
self.assertEqual(
["read_csv.parse_dates-date_parser"], result,
"Wrong result returned when date_parser is specified but not parse_dates!"
)
result = rcf.read_csv(self.testdata, dayfirst=True)
self.assertEqual(
["read_csv.parse_dates-dayfirst"], result,
"Wrong result returned when dayfirst is specified but not parse_dates!"
)
result = rcf.read_csv(self.testdata, cache_dates=True)
self.assertEqual(
["read_csv.parse_dates-cache_dates"], result,
"Wrong result returned when cache_dates is specified but not parse_dates!"
)
def test_read_csv_duplicated_value_ignored(self):
result = rcf.read_csv(self.testdata)
self.assertEqual(
[], result, "Non-empty list returned by correct call to read_csv!")
result = rcf.read_csv(self.testdata,
lineterminator='\n',
escapechar='\\',
delimiter=',',
comment='#',
thousands=' ',
decimal='.',
quotechar='\'',
na_values=["NaN", "nan"],
true_values=["True", "1"],
false_values=["False", "0"])
self.assertEqual(
[], result, "Non-empty list returned by correct call to read_csv!")
result = rcf.read_csv(self.testdata, lineterminator=',', sep=',')
self.assertEqual(
["read_csv.lineterminator-sep"], result,
"Wrong result when calling read_csv with lineterminator=sep!")
result = rcf.read_csv(self.testdata, delimiter=',', thousands=',')
self.assertEqual(
["read_csv.delimiter-thousands"], result,
"Wrong result when calling read_csv with delimiter=thousands!")
result = rcf.read_csv(self.testdata,
true_values=["1", "true"],
false_values=["1", "false"])
self.assertEqual(
["read_csv.true_values-false_values"], result,
"Wrong result when calling read_csv with same value in true_values and false_values!"
)
result = rcf.read_csv(self.testdata,
quotechar="#",
na_values=["NA", "#", "NaN"])
self.assertEqual(
["read_csv.quotechar-na_values"], result,
"Wrong result when calling read_csv with quotechar value in na_values!"
)
result = rcf.read_csv(
self.testdata, sep=',',
delimiter=',') # Should only contain sep-delimiter once
self.assertEqual(
["read_csv.sep-delimiter"], result,
"Wrong result when calling read_csv with sep and delimiter set to same value!"
)
def test_read_csv_delims(self):
result = rcf.read_csv(self.testdata, sep=",")
self.assertEqual(
[], result, "Non-empty list returned by correct call to read_csv!")
result = rcf.read_csv(self.testdata, delimiter=",")
self.assertEqual(
[], result, "Non-empty list returned by correct call to read_csv!")
result = rcf.read_csv(self.testdata, delim_whitespace=True)
self.assertEqual(
[], result, "Non-empty list returned by correct call to read_csv!")
result = rcf.read_csv(self.testdata, delim_whitespace=False, sep='.')
self.assertEqual(
[], result, "Non-empty list returned by correct call to read_csv!")
result = rcf.read_csv(self.testdata, delim_whitespace=True, sep=',')
self.assertEqual(
["read_csv.delim_whitespace-sep"], result,
"Wrong result when calling read_csv with delim_whitespace=True and sep defined!"
)
result = rcf.read_csv(self.testdata,
delim_whitespace=True,
delimiter=',')
self.assertEqual(
["read_csv.delim_whitespace-delimiter"], result,
"Wrong result when calling read_csv with delim_whitespace=True and delimiter defined!"
)
result = rcf.read_csv(self.testdata, sep=" ", delimiter=':')
self.assertEqual(
["read_csv.sep-delimiter"], result,
"Wrong result when calling read_csv with both sep and delimiter defined!"
)
result = rcf.read_csv(self.testdata,
sep=" ",
delimiter=',',
delim_whitespace=True)
self.assertEqual(
[
"read_csv.delim_whitespace-sep",
"read_csv.delim_whitespace-delimiter", "read_csv.sep-delimiter"
], result,
"Wrong result when calling read_csv with both sep and delimiter defined!"
)
def test_read_csv_quoting_doublequote(self):
result = rcf.read_csv(self.testdata, doublequote=True)
self.assertEqual(
[], result, "Non-empty list returned by correct call to read_csv!")
result = rcf.read_csv(self.testdata,
quoting=csv.QUOTE_ALL,
doublequote=True)
self.assertEqual(
[], result, "Non-empty list returned by correct call to read_csv!")
result = rcf.read_csv(self.testdata,
quoting=csv.QUOTE_NONE,
doublequote=True)
self.assertEqual(
["read_csv.quoting-doublequote"], result,
"Wrong result when quoting is QUOTE_NONE and doublequote is set to True!"
)
result = rcf.read_csv(self.testdata,
quoting=csv.QUOTE_NONE,
quotechar="",
doublequote=True)
self.assertEqual(
["read_csv.quoting-doublequote"], result,
"Wrong result when quoting is QUOTE_NONE, quotechar is empty and doublequote is set to True!"
)
result = rcf.read_csv(self.testdata,
quoting=csv.QUOTE_NONE,
quotechar="",
doublequote=False)
self.assertEqual(
["read_csv.quoting-doublequote"], result,
"Wrong result when quoting is QUOTE_NONE, quotechar is empty and doublequote is set to False!"
)
result = rcf.read_csv(self.testdata,
quoting=csv.QUOTE_NONE,
doublequote=False)
self.assertEqual(
["read_csv.quoting-doublequote"], result,
"Wrong result when quoting is QUOTE_NONE and doublequote is set to False!"
)
