def count(self, col: str):
""" Count the number of values in the specified column on the table and return the result in a table with one row
and one column.
Args:
col (str): the name of the column whose values to be counted
Returns:
a Table object
Raises:
DHError
"""
table_op = DedicatedAggOp(AggType.COUNT, count_column=col)
return self.table_op_handler(table_op)
def max_by(self, by: List[str] = []):
""" Perform max-by aggregation on the table and return the result table. Columns not used in the grouping
must be of numeric types.
Args:
by (List[str], optional): the group-by column names, default is empty
Returns:
a Table object
Raises:
DHError
"""
table_op = DedicatedAggOp(AggType.MAX, column_names=by)
return self.table_op_handler(table_op)
def last_by(self, by: List[str] = []):
""" Perform last-by aggregation on the table and return the result table which contains the last row of each
distinct group.
Args:
by (List[str], optional): the group-by column names, default is empty
Returns:
a Table object
Raises:
DHError
"""
table_op = DedicatedAggOp(AggType.LAST, column_names=by)
return self.table_op_handler(table_op)
def count_by(self, col: str, by: List[str] = []):
""" Perform count-by aggregation on the table and return the result table. The count of each group is stored in
a new column named after the 'col' parameter.
Args:
col (str): the name of the column to store the counts
by (List[str], optional): the group-by column names, default is empty
Returns:
a Table object
Raises:
DHError
"""
table_op = DedicatedAggOp(AggType.COUNT, column_names=by, count_column=col)
return self.table_op_handler(table_op)
def std_by(self, column_names: List[str] = []):
""" Perform std-by aggregation on the table and return the result table. Columns not used in the grouping
must be of numeric types.
Args:
column_names (List[str]): the group-by column names
Returns:
a Table object
Raises:
DHError
"""
table_op = DedicatedAggOp(AggType.STD, column_names=column_names)
return self.table_op_handler(table_op)
def group_by(self, by: List[str] = []):
""" Perform a group-by aggregation on the table and return the result table. After the operation,
the columns not in the group-by columns become array-type.
If no group-by column is given,the content of each column is grouped into its own array.
Args:
by (List[str], optional): the group-by column names; default is empty
Returns:
a Table object
Raises:
DHError
"""
table_op = DedicatedAggOp(AggType.GROUP, column_names=by)
return self.table_op_handler(table_op)