def transform_dict_like(
obj: FrameOrSeries,
func: AggFuncTypeDict,
*args,
**kwargs,
):
"""
Compute transform in the case of a dict-like func
"""
from pandas.core.reshape.concat import concat
if len(func) == 0:
raise ValueError("No transform functions were provided")
if obj.ndim != 1:
# Check for missing columns on a frame
cols = sorted(set(func.keys()) - set(obj.columns))
if len(cols) > 0:
raise SpecificationError(f"Column(s) {cols} do not exist")
# Can't use func.values(); wouldn't work for a Series
if any(is_dict_like(v) for _, v in func.items()):
# GH 15931 - deprecation of renaming keys
raise SpecificationError("nested renamer is not supported")
is_aggregator = lambda x: isinstance(x, (list, tuple, dict))
# if we have a dict of any non-scalars
# eg. {'A' : ['mean']}, normalize all to
# be list-likes
# Cannot use func.values() because arg may be a Series
if any(is_aggregator(x) for _, x in func.items()):
new_func: AggFuncTypeDict = {}
for k, v in func.items():
if not is_aggregator(v):
# mypy can't realize v is not a list here
new_func[k] = [v] # type:ignore[list-item]
else:
new_func[k] = v
func = new_func
results: Dict[Label, FrameOrSeriesUnion] = {}
for name, how in func.items():
colg = obj._gotitem(name, ndim=1)
try:
results[name] = transform(colg, how, 0, *args, **kwargs)
except Exception as err:
if (
str(err) == "Function did not transform"
or str(err) == "No transform functions were provided"
):
raise err
# combine results
if len(results) == 0:
raise ValueError("Transform function failed")
return concat(results, axis=1)
def normalize_dictlike_arg(
self, how: str, obj: FrameOrSeriesUnion, func: AggFuncTypeDict
) -> AggFuncTypeDict:
"""
Handler for dict-like argument.
Ensures that necessary columns exist if obj is a DataFrame, and
that a nested renamer is not passed. Also normalizes to all lists
when values consists of a mix of list and non-lists.
"""
assert how in ("apply", "agg", "transform")
# Can't use func.values(); wouldn't work for a Series
if (
how == "agg"
and isinstance(obj, ABCSeries)
and any(is_list_like(v) for _, v in func.items())
) or (any(is_dict_like(v) for _, v in func.items())):
# GH 15931 - deprecation of renaming keys
raise SpecificationError("nested renamer is not supported")
if obj.ndim != 1:
# Check for missing columns on a frame
cols = set(func.keys()) - set(obj.columns)
if len(cols) > 0:
cols_sorted = list(safe_sort(list(cols)))
raise KeyError(f"Column(s) {cols_sorted} do not exist")
is_aggregator = lambda x: isinstance(x, (list, tuple, dict))
# if we have a dict of any non-scalars
# eg. {'A' : ['mean']}, normalize all to
# be list-likes
# Cannot use func.values() because arg may be a Series
if any(is_aggregator(x) for _, x in func.items()):
new_func: AggFuncTypeDict = {}
for k, v in func.items():
if not is_aggregator(v):
# mypy can't realize v is not a list here
new_func[k] = [v] # type:ignore[list-item]
else:
new_func[k] = v
func = new_func
return func
def transform_dict_like(
obj: FrameOrSeries,
func: AggFuncTypeDict,
*args,
**kwargs,
):
"""
Compute transform in the case of a dict-like func
"""
from pandas.core.reshape.concat import concat
if len(func) == 0:
raise ValueError("No transform functions were provided")
if obj.ndim != 1:
# Check for missing columns on a frame
cols = set(func.keys()) - set(obj.columns)
if len(cols) > 0:
cols_sorted = list(safe_sort(list(cols)))
raise SpecificationError(f"Column(s) {cols_sorted} do not exist")
# Can't use func.values(); wouldn't work for a Series
if any(is_dict_like(v) for _, v in func.items()):
# GH 15931 - deprecation of renaming keys
raise SpecificationError("nested renamer is not supported")
results: Dict[Hashable, FrameOrSeriesUnion] = {}
for name, how in func.items():
colg = obj._gotitem(name, ndim=1)
try:
results[name] = transform(colg, how, 0, *args, **kwargs)
except Exception as err:
if (
str(err) == "Function did not transform"
or str(err) == "No transform functions were provided"
):
raise err
# combine results
if len(results) == 0:
raise ValueError("Transform function failed")
return concat(results, axis=1)
def validate_dictlike_arg(self, how: str, obj: FrameOrSeriesUnion,
func: AggFuncTypeDict) -> None:
"""
Raise if dict-like argument is invalid.
Ensures that necessary columns exist if obj is a DataFrame, and
that a nested renamer is not passed.
"""
assert how in ("apply", "agg", "transform")
# Can't use func.values(); wouldn't work for a Series
if (how == "agg" and isinstance(obj, ABCSeries)
and any(is_list_like(v) for _, v in func.items())) or (any(
is_dict_like(v) for _, v in func.items())):
# GH 15931 - deprecation of renaming keys
raise SpecificationError("nested renamer is not supported")
if obj.ndim != 1:
# Check for missing columns on a frame
cols = set(func.keys()) - set(obj.columns)
if len(cols) > 0:
cols_sorted = list(safe_sort(list(cols)))
raise KeyError(f"Column(s) {cols_sorted} do not exist")
def agg_dict_like(
obj: AggObjType,
arg: AggFuncTypeDict,
_axis: int,
) -> FrameOrSeriesUnion:
"""
Compute aggregation in the case of a dict-like argument.
Parameters
----------
obj : Pandas object to compute aggregation on.
arg : dict
label-aggregation pairs to compute.
_axis : int, 0 or 1
Axis to compute aggregation on.
Returns
-------
Result of aggregation.
"""
is_aggregator = lambda x: isinstance(x, (list, tuple, dict))
if _axis != 0: # pragma: no cover
raise ValueError("Can only pass dict with axis=0")
selected_obj = obj._selected_obj
# if we have a dict of any non-scalars
# eg. {'A' : ['mean']}, normalize all to
# be list-likes
# Cannot use arg.values() because arg may be a Series
if any(is_aggregator(x) for _, x in arg.items()):
new_arg: AggFuncTypeDict = {}
for k, v in arg.items():
if not isinstance(v, (tuple, list, dict)):
new_arg[k] = [v]
else:
new_arg[k] = v
# the keys must be in the columns
# for ndim=2, or renamers for ndim=1
# ok for now, but deprecated
# {'A': { 'ra': 'mean' }}
# {'A': { 'ra': ['mean'] }}
# {'ra': ['mean']}
# not ok
# {'ra' : { 'A' : 'mean' }}
if isinstance(v, dict):
raise SpecificationError("nested renamer is not supported")
elif isinstance(selected_obj, ABCSeries):
raise SpecificationError("nested renamer is not supported")
elif (
isinstance(selected_obj, ABCDataFrame) and k not in selected_obj.columns
):
raise KeyError(f"Column '{k}' does not exist!")
arg = new_arg
else:
# deprecation of renaming keys
# GH 15931
keys = list(arg.keys())
if isinstance(selected_obj, ABCDataFrame) and len(
selected_obj.columns.intersection(keys)
) != len(keys):
cols = list(
safe_sort(
list(set(keys) - set(selected_obj.columns.intersection(keys))),
)
)
raise SpecificationError(f"Column(s) {cols} do not exist")
from pandas.core.reshape.concat import concat
if selected_obj.ndim == 1:
# key only used for output
colg = obj._gotitem(obj._selection, ndim=1)
results = {key: colg.agg(how) for key, how in arg.items()}
else:
# key used for column selection and output
results = {key: obj._gotitem(key, ndim=1).agg(how) for key, how in arg.items()}
# set the final keys
keys = list(arg.keys())
# Avoid making two isinstance calls in all and any below
is_ndframe = [isinstance(r, ABCNDFrame) for r in results.values()]
# combine results
if all(is_ndframe):
keys_to_use = [k for k in keys if not results[k].empty]
# Have to check, if at least one DataFrame is not empty.
keys_to_use = keys_to_use if keys_to_use != [] else keys
axis = 0 if isinstance(obj, ABCSeries) else 1
result = concat({k: results[k] for k in keys_to_use}, axis=axis)
elif any(is_ndframe):
# There is a mix of NDFrames and scalars
raise ValueError(
"cannot perform both aggregation "
"and transformation operations "
"simultaneously"
)
else:
from pandas import Series
# we have a dict of scalars
# GH 36212 use name only if obj is a series
if obj.ndim == 1:
obj = cast("Series", obj)
name = obj.name
else:
name = None
result = Series(results, name=name)
return result
