Improve user-guide doc in UDF

docs/src/python/user-guide/expressions/user-defined-functions.py

@@ -16,7 +16,7 @@
 
 # --8<-- [start:shift_map_batches]
 out = df.group_by("keys", maintain_order=True).agg(
-    pl.col("values").map_batches(lambda s: s.shift()).alias("shift_map_batches"),
+    pl.col("values").map_batches(lambda s: s.shift(), is_elementwise=True).alias("shift_map_batches"),
     pl.col("values").shift().alias("shift_expression"),
 )
 print(out)
@@ -25,7 +25,7 @@
 
 # --8<-- [start:map_elements]
 out = df.group_by("keys", maintain_order=True).agg(
-    pl.col("values").map_elements(lambda s: s.shift()).alias("shift_map_elements"),
+    pl.col("values").map_elements(lambda s: s.shift(), return_dtype=pl.List(int)).alias("shift_map_elements"),
     pl.col("values").shift().alias("shift_expression"),
 )
 print(out)

docs/user-guide/expressions/user-defined-functions.md

@@ -74,7 +74,9 @@ Let's try that out and see what we get:
 
 Ouch.. we clearly get the wrong results here. Group `"b"` even got a value from group `"a"` 😵.
 
-This went horribly wrong, because the `map_batches` applies the function before we aggregate! So that means the whole column `[10, 7, 1`\] got shifted to `[null, 10, 7]` and was then aggregated.
+This went horribly wrong because `map_batches` applied the function before aggregation, due to the `is_elementwise=True` parameter being provided. So that means the whole column `[10, 7, 1]` got shifted to `[null, 10, 7]` and was then aggregated.
+
+However, the good news is that the default value for `is_elementwise` is `False`. The trade-off is that it won't run in the streaming engine anymore.
 
 So my advice is to never use `map_batches` in the `group_by` context unless you know you need it and know what you are doing.