Add support for min_success_ratio for local map_task execution

flytekit/core/array_node_map_task.py

@@ -2,6 +2,7 @@
 import functools
 import hashlib
 import logging
+import math
 import os  # TODO: use flytekit logger
 from contextlib import contextmanager
 from typing import Any, Dict, List, Optional, Set, Union, cast
@@ -268,25 +269,38 @@
             outputs_expected = False
         outputs = []
 
-        mapped_input_value_len = 0
+        mapped_tasks_count = 0
         if self._run_task.interface.inputs.items():
             for k in self._run_task.interface.inputs.keys():
                 v = kwargs[k]
                 if isinstance(v, list) and k not in self.bound_inputs:
-                    mapped_input_value_len = len(v)
+                    mapped_tasks_count = len(v)
                     break
 
-        for i in range(mapped_input_value_len):
+        failed_count = 0
+        min_successes = mapped_tasks_count
+        if self._min_successes:
+            min_successes = self._min_successes
+        elif self._min_success_ratio:
+            min_successes = math.ceil(min_successes * self._min_success_ratio)
+
+        for i in range(mapped_tasks_count):
             single_instance_inputs = {}
             for k in self.interface.inputs.keys():
                 v = kwargs[k]
                 if isinstance(v, list) and k not in self._bound_inputs:
                     single_instance_inputs[k] = kwargs[k][i]
                 else:
                     single_instance_inputs[k] = kwargs[k]
-            o = exception_scopes.user_entry_point(self.python_function_task.execute)(**single_instance_inputs)
-            if outputs_expected:
-                outputs.append(o)
+            try:
+                o = exception_scopes.user_entry_point(self._run_task.execute)(**single_instance_inputs)
+                if outputs_expected:
+                    outputs.append(o)
+            except Exception as exc:
+                outputs.append(None)
+                failed_count += 1
+                if mapped_tasks_count - failed_count < min_successes:
+                    raise exc
 
         return outputs
 

flytekit/core/map_task.py

@@ -5,6 +5,7 @@
 import functools
 import hashlib
 import logging
+import math
 import os
 import typing
 from contextlib import contextmanager
@@ -263,25 +264,36 @@
             outputs_expected = False
         outputs = []
 
-        mapped_input_value_len = 0
+        mapped_tasks_count = 0
         if self._run_task.interface.inputs.items():
             for k in self._run_task.interface.inputs.keys():
                 v = kwargs[k]
                 if isinstance(v, list) and k not in self.bound_inputs:
-                    mapped_input_value_len = len(v)
+                    mapped_tasks_count = len(v)
                     break
 
-        for i in range(mapped_input_value_len):
+        failed_count = 0
+        min_successes = mapped_tasks_count
+        if self._min_success_ratio:
+            min_successes = math.ceil(min_successes * self._min_success_ratio)
+
+        for i in range(mapped_tasks_count):
             single_instance_inputs = {}
             for k in self.interface.inputs.keys():
                 v = kwargs[k]
                 if isinstance(v, list) and k not in self.bound_inputs:
                     single_instance_inputs[k] = kwargs[k][i]
                 else:
                     single_instance_inputs[k] = kwargs[k]
-            o = exception_scopes.user_entry_point(self._run_task.execute)(**single_instance_inputs)
-            if outputs_expected:
-                outputs.append(o)
+            try:
+                o = exception_scopes.user_entry_point(self._run_task.execute)(**single_instance_inputs)
+                if outputs_expected:
+                    outputs.append(o)
+            except Exception as exc:
+                outputs.append(None)
+                failed_count += 1
+                if mapped_tasks_count - failed_count < min_successes:
+                    raise exc
 
         return outputs
 

tests/flytekit/unit/core/test_array_node_map_task.py

@@ -1,4 +1,5 @@
 import functools
+import typing
 from collections import OrderedDict
 from typing import List
 
@@ -254,3 +255,30 @@ def task3(c: str, a: int, b: float) -> str:
     m3 = array_node_map_task(functools.partial(task3, c=param_c))(a=param_a, b=param_b)
 
     assert m1 == m2 == m3 == ["1 - 0.1 - c", "2 - 0.2 - c", "3 - 0.3 - c"]
+
+
+@pytest.mark.parametrize(
+    "min_success_ratio, should_raise_error",
+    [
+        (None, True),
+        (1, True),
+        (0.75, False),
+        (0.5, False),
+    ],
+)
+def test_raw_execute_with_min_success_ratio(min_success_ratio, should_raise_error):
+    @task
+    def some_task1(inputs: int) -> int:
+        if inputs == 2:
+            raise ValueError("Unexpected inputs: 2")
+        return inputs
+
+    @workflow
+    def my_wf1() -> typing.List[typing.Optional[int]]:
+        return array_node_map_task(some_task1, min_success_ratio=min_success_ratio)(inputs=[1, 2, 3, 4])
+
+    if should_raise_error:
+        with (pytest.raises(ValueError)):
+            my_wf1()
+    else:
+        assert my_wf1() == [1, None, 3, 4]

tests/flytekit/unit/core/test_map_task.py

@@ -306,3 +306,30 @@ def task3(c: str, a: int, b: float) -> str:
     m3 = map_task(functools.partial(task3, c=param_c))(a=param_a, b=param_b)
 
     assert m1 == m2 == m3 == ["1 - 0.1 - c", "2 - 0.2 - c", "3 - 0.3 - c"]
+
+
+@pytest.mark.parametrize(
+    "min_success_ratio, should_raise_error",
+    [
+        (None, True),
+        (1, True),
+        (0.75, False),
+        (0.5, False),
+    ],
+)
+def test_raw_execute_with_min_success_ratio(min_success_ratio, should_raise_error):
+    @task
+    def some_task1(inputs: int) -> int:
+        if inputs == 2:
+            raise ValueError("Unexpected inputs: 2")
+        return inputs
+
+    @workflow
+    def my_wf1() -> typing.List[typing.Optional[int]]:
+        return map_task(some_task1, min_success_ratio=min_success_ratio)(inputs=[1, 2, 3, 4])
+
+    if should_raise_error:
+        with (pytest.raises(ValueError)):
+            my_wf1()
+    else:
+        assert my_wf1() == [1, None, 3, 4]