def execute_binary_op_series_group_by(op, left, right, **kwargs):
if left.index != right.index:
raise ValueError('Cannot perform {} operation on two series with '
'different groupings'.format(type(op).__name__))
result = execute_binary_op(op, make_selected_obj(left),
make_selected_obj(right), **kwargs)
return result.groupby(left.index)
def execute_series_regex_search_gb(op, data, pattern, **kwargs):
return execute_series_regex_search(
op,
make_selected_obj(data),
getattr(pattern, 'obj', pattern),
**kwargs,
).groupby(data.index)
def execute_udf_node_groupby(op, *args, **kwargs):
func = op.func
# all grouping keys must be identical
assert_identical_grouping_keys(*args)
# we're performing a scalar operation on grouped column, so
# perform the operation directly on the underlying Series
# and regroup after it's finished
args_objs = [make_selected_obj(arg) for arg in args]
groupings = args[0].index
return dd.map_partitions(func, *args_objs).groupby(groupings)
def execute_log_series_gb_series_gb(op, left, right, **kwargs):
result = execute_node(op, make_selected_obj(left),
make_selected_obj(right), **kwargs)
return result.groupby(left.index)
def execute_unary_op_series_gb(op, operand, **kwargs):
result = execute_node(op, make_selected_obj(operand), **kwargs)
return result.groupby(operand.index)
def execute_binary_op_simple_series_gb(op, left, right, **kwargs):
result = execute_binary_op(op, left, make_selected_obj(right), **kwargs)
return result.groupby(right.index)
def execute_cast_series_group_by(op, data, type, **kwargs):
result = execute_cast_series_generic(op, make_selected_obj(data), type,
**kwargs)
return result.groupby(data.index)
def execute_series_group_by_negate(op, data, **kwargs):
return execute_series_negate(
op, make_selected_obj(data), **kwargs
).groupby(data.index)
def execute_day_of_week_name_series_group_by(op, data, **kwargs):
return day_name(make_selected_obj(data).dt).groupby(data.index)
def execute_series_group_by_find_in_set(op, needle, haystack, **kwargs):
pieces = [getattr(piece, 'obj', piece) for piece in haystack]
return execute_series_find_in_set(op, make_selected_obj(needle), pieces,
**kwargs).groupby(needle.index)
def execute_series_right_gb(op, data, nchars, **kwargs):
return execute_series_right(op, make_selected_obj(data),
nchars).groupby(data.index)
def execute_series_regex_replace_gb(op, data, pattern, replacement, **kwargs):
return execute_series_regex_replace(make_selected_obj(data), pattern,
replacement,
**kwargs).groupby(data.index)
def execute_series_regex_extract_gb(op, data, pattern, index, **kwargs):
return execute_series_regex_extract(op, make_selected_obj(data), pattern,
index, **kwargs).groupby(data.index)
def execute_string_ascii_group_by(op, data, **kwargs):
return execute_string_ascii(op, make_selected_obj(data),
**kwargs).groupby(data.index)
def execute_string_like_series_groupby_string(op, data, pattern, escape,
**kwargs):
return execute_string_like_series_string(
op, make_selected_obj(data), pattern, escape,
**kwargs).groupby(data.grouper.groupings)
def execute_node_struct_field_series_group_by(op, data, **kwargs):
field = op.field
selected_obj = make_selected_obj(data)
return (selected_obj.map(operator.itemgetter(field),
meta=selected_obj._meta).rename(field).groupby(
data.index))
def _filtered_reduction(data, mask):
return make_selected_obj(data)[mask.obj].groupby(data.index)
def execute_day_of_week_index_series_group_by(op, data, **kwargs):
return (
make_selected_obj(data)
.dt.dayofweek.astype(np.int16)
.groupby(data.index)
)