def test_from_rows(self):
c = ColumnGroup.from_rows((int, str), [("colA", "colB"), (1, "a"),
(2, "b"), (3, "c")])
self.assertEqual(
ColumnGroup([
Column("colA", int, [1, 2, 3]),
Column("colB", str, ["a", "b", "c"])
]), c)
def _append_other_columns_and_put_rows(self, other_handle: "DataHandle",
new_rows):
columns = [
Column(col.name, col.col_type, []) for col in itertools.chain(
self._data_column_group, other_handle._data_column_group)
]
column_group = ColumnGroup(columns)
column_group.replace_rows(new_rows)
self._data_column_group.replace_columns(column_group.columns)
def read_csv(csv_path: Union[str, Path],
mcsv_path: Optional[Union[str, Path]] = None,
nrows=100) -> Optional[Data]:
if isinstance(csv_path, str):
csv_path = Path(csv_path)
if mcsv_path is None:
mcsv_path = csv_path.with_suffix(".mcsv")
if not mcsv_path.is_file():
missing_mcsv(csv_path) # util command to open a window
return None
with mcsv.open_csv(csv_path, "r", mcsv_path) as mcsv_reader:
if nrows >= 0:
reader = islice(mcsv_reader, nrows)
else:
reader = mcsv_reader
header = [to_standard(n) for n in next(reader)]
column_group = ColumnGroup([
Column(name, description,
values) for name, description, *values in zip(
header, mcsv_reader.descriptions, *reader)
])
return Data(
column_group,
DataSource.create(to_standard(csv_path.stem), csv_path,
mcsv_reader.meta_csv_data))
def data_from_rows(col_types, rows):
columns = list(zip(*rows))
return Data(
ColumnGroup([
Column(col[0], col_type, col[1:])
for col_type, col in zip(col_types, columns)
]), None)
def __getitem__(self, item):
indices = to_indices(len(self._column_group), item)
column_group = ColumnGroup([self._column_group[i] for i in indices])
return DataHandle(self._column_group, indices)
description = TextFieldDescription.INSTANCE
b.description_by_col_index(i, description)
meta_csv_data = b.build()
else:
b = MetaCSVDataBuilder()
cur_data = self._data_source.meta_csv_data
b.encoding(cur_data.encoding)
b.bom(cur_data.bom)
b._dialect = cur_data.dialect
for i, col in enumerate(self._column_group):
description = self._data_source.get_description(col.col_info)
b.description_by_col_index(i, description)
meta_csv_data = b.build()
return meta_csv_data
if __name__ == "__main__":
import doctest
doctest.testmod(
extraglobs={
'original_test_data':
Data(
ColumnGroup([
Column("A", int, [1, 5, 3]),
Column("B", int, [3, 2, 4]),
Column("C", int, [2, 2, 7]),
Column("D", int, [4, 7, 8])
]), None)
})