Groupby limitations (#782)

Author: 1e-to

sdc/datatypes/hpat_pandas_groupby_functions.py

@@ -52,6 +52,11 @@
 from sdc.str_ext import string_type
 
 
+performance_limitation = "This function may reveal slower performance than Pandas* on user system.\
+    Users should exercise a tradeoff between staying in JIT-region with that function or\
+    going back to interpreter mode."
+
+
 @sdc_register_jitable
 def merge_groupby_dicts_inplace(left, right):
     """ Merges one internal groupby dictionary (right) into another one (left) and returns
@@ -625,7 +630,12 @@ def sdc_pandas_series_groupby_var(self, ddof=1, *args):
 sdc_pandas_dataframe_groupby_count.__doc__ = sdc_pandas_dataframe_groupby_docstring_tmpl.format(**{
     'method_name': 'count',
     'example_caption': 'Compute count of group, excluding missing values.',
-    'limitations_block': '',
+    'limitations_block':
+        f"""
+        Limitations
+        -----------
+        - {performance_limitation}
+        """,
     'see_also':
     """
     .. seealso::
@@ -641,7 +651,12 @@ def sdc_pandas_series_groupby_var(self, ddof=1, *args):
 sdc_pandas_dataframe_groupby_max.__doc__ = sdc_pandas_dataframe_groupby_docstring_tmpl.format(**{
     'method_name': 'max',
     'example_caption': 'Compute max of group values.',
-    'limitations_block': '',
+    'limitations_block':
+        f"""
+        Limitations
+        -----------
+        - {performance_limitation}
+        """,
     'see_also': '',
     'extra_params': ''
 })
@@ -650,7 +665,12 @@ def sdc_pandas_series_groupby_var(self, ddof=1, *args):
 sdc_pandas_dataframe_groupby_mean.__doc__ = sdc_pandas_dataframe_groupby_docstring_tmpl.format(**{
     'method_name': 'mean',
     'example_caption': 'Compute mean of groups, excluding missing values.',
-    'limitations_block': '',
+    'limitations_block':
+        f"""
+        Limitations
+        -----------
+        - {performance_limitation}
+        """,
     'see_also':
     """
     .. seealso::
@@ -666,7 +686,12 @@ def sdc_pandas_series_groupby_var(self, ddof=1, *args):
 sdc_pandas_dataframe_groupby_median.__doc__ = sdc_pandas_dataframe_groupby_docstring_tmpl.format(**{
     'method_name': 'median',
     'example_caption': 'Compute median of groups, excluding missing values.',
-    'limitations_block': '',
+    'limitations_block':
+        f"""
+        Limitations
+        -----------
+        - {performance_limitation}
+        """,
     'see_also':
     """
     .. seealso::
@@ -682,7 +707,12 @@ def sdc_pandas_series_groupby_var(self, ddof=1, *args):
 sdc_pandas_dataframe_groupby_min.__doc__ = sdc_pandas_dataframe_groupby_docstring_tmpl.format(**{
     'method_name': 'min',
     'example_caption': 'Compute min of group values.',
-    'limitations_block': '',
+    'limitations_block':
+        f"""
+        Limitations
+        -----------
+        - {performance_limitation}
+        """,
     'see_also': '',
     'extra_params': ''
 })
@@ -691,7 +721,12 @@ def sdc_pandas_series_groupby_var(self, ddof=1, *args):
 sdc_pandas_dataframe_groupby_prod.__doc__ = sdc_pandas_dataframe_groupby_docstring_tmpl.format(**{
     'method_name': 'prod',
     'example_caption': 'Compute prod of group values.',
-    'limitations_block': '',
+    'limitations_block':
+        f"""
+        Limitations
+        -----------
+        - {performance_limitation}
+        """,
     'see_also': '',
     'extra_params': ''
 })
@@ -700,7 +735,12 @@ def sdc_pandas_series_groupby_var(self, ddof=1, *args):
 sdc_pandas_dataframe_groupby_std.__doc__ = sdc_pandas_dataframe_groupby_docstring_tmpl.format(**{
     'method_name': 'std',
     'example_caption': 'Compute standard deviation of groups, excluding missing values.',
-    'limitations_block': '',
+    'limitations_block':
+        f"""
+        Limitations
+        -----------
+        - {performance_limitation}
+        """,
     'see_also':
     """
     .. seealso::
@@ -716,16 +756,26 @@ def sdc_pandas_series_groupby_var(self, ddof=1, *args):
 sdc_pandas_dataframe_groupby_sum.__doc__ = sdc_pandas_dataframe_groupby_docstring_tmpl.format(**{
     'method_name': 'sum',
     'example_caption': 'Compute sum of groups, excluding missing values.',
-    'limitations_block': '',
+    'limitations_block':
+        f"""
+        Limitations
+        -----------
+        - {performance_limitation}
+        """,
     'see_also': '',
     'extra_params': ''
 })
 
 
 sdc_pandas_dataframe_groupby_var.__doc__ = sdc_pandas_dataframe_groupby_docstring_tmpl.format(**{
-    'method_name': 'sum',
+    'method_name': 'var',
     'example_caption': 'Compute variance of groups, excluding missing values.',
-    'limitations_block': '',
+    'limitations_block':
+        f"""
+        Limitations
+        -----------
+        - {performance_limitation}
+        """,
     'see_also':
     """
     .. seealso::