def replace_totals(data_frame):
index_names = data_frame.index.names
raw = data_frame.reset_index()
for name in index_names:
marker = get_totals_marker_for_dtype(raw[name].dtype)
raw[name].replace(marker, np.nan, inplace=True)
return raw
def _replace_nans_for_totals_values(data_frame, dtypes):
# some things are just easier to do without an index. Reset it temporarily to replace NaN values with the rollup
# marker values
index_names = data_frame.index.names
data_frame.reset_index(inplace=True)
for dimension_key, dtype in dtypes.items():
data_frame[dimension_key] = data_frame[dimension_key].fillna(
get_totals_marker_for_dtype(dtype))
return data_frame.set_index(index_names)
def _replace_rollup_constants_for_totals_markers(data_frame, dtypes):
# some things are just easier to do without an index. Reset it temporarily to replace Rollup constants with the
# rollup marker values
index_names = data_frame.index.names
data_frame.reset_index(inplace=True)
for dimension_key, dtype in dtypes.items():
data_frame[dimension_key] = data_frame[dimension_key].replace(
RollupValue.CONSTANT, get_totals_marker_for_dtype(dtype))
return data_frame.set_index(index_names)
def _totals(df):
if isinstance(df, pd.Series):
return df.sum()
totals_index_value = get_totals_marker_for_dtype(df.index.levels[-1].dtype)
return pd.DataFrame(
[df.sum()],
columns=columns,
index=pd.Index([totals_index_value], name=df.index.names[-1]),
)
def apply(self, data_frame, reference):
metric, over = self.args
f_metric_alias = alias_selector(reference_alias(metric, reference))
if over is None:
df = data_frame[f_metric_alias]
return 100 * df / df
if not isinstance(data_frame.index, pd.MultiIndex):
marker = get_totals_marker_for_dtype(data_frame.index.dtype)
totals = data_frame.loc[marker, f_metric_alias]
if totals == 0:
return np.nan
return 100 * data_frame[f_metric_alias] / totals
f_over_alias = alias_selector(over.alias)
idx = data_frame.index.names.index(f_over_alias)
group_levels = data_frame.index.names[idx:]
over_dim_value = get_totals_marker_for_dtype(
data_frame.index.levels[idx].dtype)
totals_alias = (slice(None), ) * idx + (slice(over_dim_value,
over_dim_value), )
totals = reduce_data_frame_levels(
data_frame.loc[totals_alias, f_metric_alias], group_levels)
def apply_totals(group_df):
if not isinstance(totals, pd.Series):
return 100 * group_df / totals
n_index_levels = len(totals.index.names)
extra_level_names = group_df.index.names[n_index_levels:]
group_df = group_df.reset_index(extra_level_names, drop=True)
share = 100 * group_df / totals[group_df.index]
return pd.Series(share.values, index=group_df.index)
return (data_frame[f_metric_alias].groupby(
level=group_levels).apply(apply_totals).reorder_levels(
order=data_frame.index.names).sort_index())
def totals(data_frame, dimensions, columns):
"""
Computes the totals across a dimension and adds the total as an extra row.
"""
if not isinstance(data_frame.index, pd.MultiIndex):
totals_marker = get_totals_marker_for_dtype(data_frame.index.dtype)
totals_df = pd.DataFrame(
[data_frame.sum()],
index=pd.Index([totals_marker], name=data_frame.index.name),
)
return data_frame.append(totals_df)
def _totals(df):
if isinstance(df, pd.Series):
return df.sum()
totals_index_value = get_totals_marker_for_dtype(
df.index.levels[-1].dtype)
return pd.DataFrame(
[df.sum()],
columns=columns,
index=pd.Index([totals_index_value], name=df.index.names[-1]),
)
totals_df = None
for i in range(-1, -1 - len(dimensions), -1):
groupby_levels = data_frame.index.names[:i]
if groupby_levels:
level_totals_df = (data_frame[columns].groupby(
level=groupby_levels).apply(_totals))
missing_dims = set(data_frame.index.names) - set(
level_totals_df.index.names)
if missing_dims:
for dim in missing_dims:
dtype = data_frame.index.levels[
data_frame.index.names.index(dim)].dtype
level_totals_df[dim] = get_totals_marker_for_dtype(dtype)
level_totals_df.set_index(dim, append=True, inplace=True)
level_totals_df = level_totals_df.reorder_levels(
data_frame.index.names)
else:
totals_index_values = [
get_totals_marker_for_dtype(level.dtype)
for level in data_frame.index.levels
]
level_totals_df = pd.DataFrame(
[data_frame[columns].apply(_totals)],
columns=columns,
index=pd.MultiIndex.from_tuples([totals_index_values],
names=data_frame.index.names),
)
totals_df = (totals_df.append(level_totals_df)
if totals_df is not None else level_totals_df)
return data_frame.append(totals_df).sort_index()
