[pyright] analyzeUnannotatedFunctions (#26392)

Sibling to dagster-io/internal#12806, enable
`analyzeUnannotatedFunctions` by using a script to insert comments on
newly created type errors.

`examples` is exempted from this as the type suppresion comments
negatively impact using terse code snippets in docs etc.


## How I Tested These Changes

make pyright

Author: alangenfeld

.buildkite/dagster-buildkite/dagster_buildkite/package_spec.py

@@ -232,7 +232,7 @@ def build_steps(self) -> List[BuildkiteTopLevelStep]:
     @property
     def requirements(self):
         # First try to infer requirements from the python package
-        package = PythonPackages.get(self.name)
+        package = PythonPackages.get(self.name)  # pyright: ignore[reportArgumentType]
         if package:
             return set.union(package.install_requires, *package.extras_require.values())
 

.buildkite/dagster-buildkite/dagster_buildkite/utils.py

@@ -363,7 +363,7 @@ def skip_if_no_docs_changes():
     if message_contains("NO_SKIP"):
         return None
 
-    if not is_feature_branch(os.getenv("BUILDKITE_BRANCH")):
+    if not is_feature_branch(os.getenv("BUILDKITE_BRANCH")):  # pyright: ignore[reportArgumentType]
         return None
 
     # If anything changes in the docs directory

docs/content/integrations/pandas.mdx

@@ -43,10 +43,10 @@ TripDataFrame = create_dagster_pandas_dataframe_type(
         PandasColumn.integer_column("bike_id", min_value=0),
         PandasColumn.categorical_column("color", categories={"red", "green", "blue"}),
         PandasColumn.datetime_column(
-            "start_time", min_datetime=datetime(year=2020, month=2, day=10)
+            "start_time", min_datetime=Timestamp(year=2020, month=2, day=10)
         ),
         PandasColumn.datetime_column(
-            "end_time", min_datetime=datetime(year=2020, month=2, day=10)
+            "end_time", min_datetime=Timestamp(year=2020, month=2, day=10)
         ),
         PandasColumn.string_column("station"),
         PandasColumn.exists("amount_paid"),

examples/docs_snippets/docs_snippets/legacy/dagster_pandas_guide/core_trip.py

@@ -1,7 +1,7 @@
 from datetime import datetime
 
 from dagster_pandas import PandasColumn, create_dagster_pandas_dataframe_type
-from pandas import DataFrame, read_csv
+from pandas import DataFrame, Timestamp, read_csv
 
 from dagster import Out, file_relative_path, job, op
 
@@ -12,10 +12,10 @@
         PandasColumn.integer_column("bike_id", min_value=0),
         PandasColumn.categorical_column("color", categories={"red", "green", "blue"}),
         PandasColumn.datetime_column(
-            "start_time", min_datetime=datetime(year=2020, month=2, day=10)
+            "start_time", min_datetime=Timestamp(year=2020, month=2, day=10)
         ),
         PandasColumn.datetime_column(
-            "end_time", min_datetime=datetime(year=2020, month=2, day=10)
+            "end_time", min_datetime=Timestamp(year=2020, month=2, day=10)
         ),
         PandasColumn.string_column("station"),
         PandasColumn.exists("amount_paid"),

examples/experimental/assets_yaml_dsl/assets_yaml_dsl_tests/test_stocks_dsl.py

@@ -19,7 +19,7 @@
 from dagster import AssetKey
 from dagster._core.definitions import materialize
 from dagster._core.pipes.subprocess import PipesSubprocessClient
-from examples.experimental.assets_yaml_dsl.assets_yaml_dsl.domain_specific_dsl.stocks_dsl import (
+from examples.experimental.assets_yaml_dsl.assets_yaml_dsl.domain_specific_dsl.stocks_dsl import (  # type: ignore
     build_stock_assets_object,
 )
 

integration_tests/python_modules/dagster-k8s-test-infra/dagster_k8s_test_infra/cluster.py

@@ -71,7 +71,7 @@ def _cluster_provider(request):
                             f"Found existing image tagged {docker_image}, skipping image build. To rebuild,"
                             f" first run: docker rmi {docker_image}"
                         )
-                    except docker.errors.ImageNotFound:
+                    except docker.errors.ImageNotFound:  # pyright: ignore[reportAttributeAccessIssue]
                         build_and_tag_test_image(docker_image)
                     kind_load_images(
                         cluster_name=cluster_config.name,
@@ -231,7 +231,7 @@ def check_export_runs(instance):
 
     # example PYTEST_CURRENT_TEST: test_user_code_deployments.py::test_execute_on_celery_k8s (teardown)
     current_test = (
-        os.environ.get("PYTEST_CURRENT_TEST").split()[0].replace("::", "-").replace(".", "-")
+        os.environ.get("PYTEST_CURRENT_TEST").split()[0].replace("::", "-").replace(".", "-")  # pyright: ignore[reportOptionalMemberAccess]
     )
 
     for run in instance.get_runs():

integration_tests/test_suites/backcompat-test-suite/webserver_service/defs_for_earliest_tested_release.py

@@ -1,5 +1,3 @@
-# type: ignore
-
 # Backcompat test definitions intended for use with our oldest testest release of Dagster. Does not
 # use `Definitions` because it is not available in our oldest supported releases.
 

integration_tests/test_suites/celery-k8s-test-suite/conftest.py

@@ -30,7 +30,7 @@ def dagster_docker_image():
                 f"Found existing image tagged {docker_image}, skipping image build. To rebuild, first run: "
                 f"docker rmi {docker_image}"
             )
-        except docker.errors.ImageNotFound:
+        except docker.errors.ImageNotFound:  # pyright: ignore[reportAttributeAccessIssue]
             build_and_tag_test_image(docker_image)
 
     return docker_image

integration_tests/test_suites/daemon-test-suite/monitoring_daemon_tests/test_monitoring.py

@@ -96,7 +96,7 @@ def test_docker_monitoring(aws_env):
         find_local_test_image(docker_image)
 
     run_config = merge_dicts(
-        load_yaml_from_path(os.path.join(get_test_project_environments_path(), "env_s3.yaml")),
+        load_yaml_from_path(os.path.join(get_test_project_environments_path(), "env_s3.yaml")),  # pyright: ignore[reportArgumentType]
         {
             "ops": {
                 "multiply_the_word_slow": {
@@ -139,21 +139,21 @@ def test_docker_monitoring(aws_env):
 
                     start_time = time.time()
                     while time.time() - start_time < 60:
-                        run = instance.get_run_by_id(run.run_id)
-                        if run.status == DagsterRunStatus.STARTED:
+                        run = instance.get_run_by_id(run.run_id)  # pyright: ignore[reportOptionalMemberAccess]
+                        if run.status == DagsterRunStatus.STARTED:  # pyright: ignore[reportOptionalMemberAccess]
                             break
-                        assert run.status == DagsterRunStatus.STARTING
+                        assert run.status == DagsterRunStatus.STARTING  # pyright: ignore[reportOptionalMemberAccess]
                         time.sleep(1)
 
                     time.sleep(3)
 
-                    instance.run_launcher._get_container(  # noqa: SLF001
-                        instance.get_run_by_id(run.run_id)
+                    instance.run_launcher._get_container(  # noqa: SLF001  # pyright: ignore[reportAttributeAccessIssue]
+                        instance.get_run_by_id(run.run_id)  # pyright: ignore[reportOptionalMemberAccess]
                     ).stop()
 
                     # daemon resumes the run
-                    poll_for_finished_run(instance, run.run_id, timeout=300)
-                    assert instance.get_run_by_id(run.run_id).status == DagsterRunStatus.SUCCESS
+                    poll_for_finished_run(instance, run.run_id, timeout=300)  # pyright: ignore[reportOptionalMemberAccess]
+                    assert instance.get_run_by_id(run.run_id).status == DagsterRunStatus.SUCCESS  # pyright: ignore[reportOptionalMemberAccess]
 
 
 @pytest.fixture
@@ -183,7 +183,7 @@ def test_docker_monitoring_run_out_of_attempts(aws_env):
         find_local_test_image(docker_image)
 
     run_config = merge_dicts(
-        load_yaml_from_path(os.path.join(get_test_project_environments_path(), "env_s3.yaml")),
+        load_yaml_from_path(os.path.join(get_test_project_environments_path(), "env_s3.yaml")),  # pyright: ignore[reportArgumentType]
         {
             "ops": {
                 "multiply_the_word_slow": {
@@ -230,17 +230,17 @@ def test_docker_monitoring_run_out_of_attempts(aws_env):
 
                     start_time = time.time()
                     while time.time() - start_time < 60:
-                        run = instance.get_run_by_id(run.run_id)
-                        if run.status == DagsterRunStatus.STARTED:
+                        run = instance.get_run_by_id(run.run_id)  # pyright: ignore[reportOptionalMemberAccess]
+                        if run.status == DagsterRunStatus.STARTED:  # pyright: ignore[reportOptionalMemberAccess]
                             break
-                        assert run.status == DagsterRunStatus.STARTING
+                        assert run.status == DagsterRunStatus.STARTING  # pyright: ignore[reportOptionalMemberAccess]
                         time.sleep(1)
 
                     time.sleep(3)
 
-                    instance.run_launcher._get_container(  # noqa: SLF001
-                        instance.get_run_by_id(run.run_id)
+                    instance.run_launcher._get_container(  # noqa: SLF001  # pyright: ignore[reportAttributeAccessIssue]
+                        instance.get_run_by_id(run.run_id)  # pyright: ignore[reportOptionalMemberAccess]
                     ).stop(timeout=0)
 
-                    poll_for_finished_run(instance, run.run_id, timeout=60)
-                    assert instance.get_run_by_id(run.run_id).status == DagsterRunStatus.FAILURE
+                    poll_for_finished_run(instance, run.run_id, timeout=60)  # pyright: ignore[reportOptionalMemberAccess]
+                    assert instance.get_run_by_id(run.run_id).status == DagsterRunStatus.FAILURE  # pyright: ignore[reportOptionalMemberAccess]

integration_tests/test_suites/daemon-test-suite/test_dagster_daemon_health.py

@@ -242,14 +242,14 @@ def _get_error_number(error):
                         )[SensorDaemon.daemon_type()]
 
                         # Errors build up until there are > 5, then pull off the last
-                        if status.healthy is False and len(status.last_heartbeat.errors) >= 5:
-                            first_error_number = _get_error_number(status.last_heartbeat.errors[0])
+                        if status.healthy is False and len(status.last_heartbeat.errors) >= 5:  # pyright: ignore[reportArgumentType,reportOptionalMemberAccess]
+                            first_error_number = _get_error_number(status.last_heartbeat.errors[0])  # pyright: ignore[reportOptionalSubscript,reportOptionalMemberAccess]
 
                             if first_error_number > 5:
                                 # Verify error numbers decrease consecutively
                                 assert [
                                     _get_error_number(error)
-                                    for error in status.last_heartbeat.errors
+                                    for error in status.last_heartbeat.errors  # pyright: ignore[reportOptionalIterable,reportOptionalMemberAccess]
                                 ] == list(range(first_error_number, first_error_number - 5, -1))
 
                                 assert not get_daemon_statuses(
@@ -276,10 +276,10 @@ def _get_error_number(error):
                                 )[SensorDaemon.daemon_type()]
 
                                 # Error count does not rise above 5, continues to increase
-                                assert len(status.last_heartbeat.errors) == 5
+                                assert len(status.last_heartbeat.errors) == 5  # pyright: ignore[reportArgumentType,reportOptionalMemberAccess]
 
                                 new_first_error_number = _get_error_number(
-                                    status.last_heartbeat.errors[0]
+                                    status.last_heartbeat.errors[0]  # pyright: ignore[reportOptionalSubscript,reportOptionalMemberAccess]
                                 )
 
                                 assert new_first_error_number > first_error_number
@@ -307,7 +307,7 @@ def _get_error_number(error):
                     )[SensorDaemon.daemon_type()]
 
                     # Error count does not rise above 5
-                    if len(status.last_heartbeat.errors) == 0:
+                    if len(status.last_heartbeat.errors) == 0:  # pyright: ignore[reportArgumentType,reportOptionalMemberAccess]
                         break
 
                     if (now - init_time).total_seconds() > 15:
@@ -322,8 +322,8 @@ def test_multiple_error_daemon(monkeypatch):
 
         def run_loop_error(_, _ctx, _shutdown_event):
             # ?message stack cls_name cause"
-            yield SerializableErrorInfo("foobar", None, None, None)
-            yield SerializableErrorInfo("bizbuz", None, None, None)
+            yield SerializableErrorInfo("foobar", None, None, None)  # pyright: ignore[reportArgumentType]
+            yield SerializableErrorInfo("bizbuz", None, None, None)  # pyright: ignore[reportArgumentType]
 
             while True:
                 yield
@@ -354,9 +354,9 @@ def run_loop_error(_, _ctx, _shutdown_event):
                         instance, [SensorDaemon.daemon_type()], now.timestamp()
                     )[SensorDaemon.daemon_type()]
 
-                    if status.healthy is False and len(status.last_heartbeat.errors) == 2:
-                        assert status.last_heartbeat.errors[0].message.strip() == "bizbuz"
-                        assert status.last_heartbeat.errors[1].message.strip() == "foobar"
+                    if status.healthy is False and len(status.last_heartbeat.errors) == 2:  # pyright: ignore[reportArgumentType,reportOptionalMemberAccess]
+                        assert status.last_heartbeat.errors[0].message.strip() == "bizbuz"  # pyright: ignore[reportOptionalSubscript,reportOptionalMemberAccess]
+                        assert status.last_heartbeat.errors[1].message.strip() == "foobar"  # pyright: ignore[reportOptionalSubscript,reportOptionalMemberAccess]
                         break
 
                 if (now - init_time).total_seconds() > 10:
@@ -403,7 +403,7 @@ def test_warn_multiple_daemons(capsys):
             now.timestamp(),
             heartbeat_interval_seconds=heartbeat_interval_seconds,
         )[SensorDaemon.daemon_type()]
-        last_heartbeat_time = status.last_heartbeat.timestamp
+        last_heartbeat_time = status.last_heartbeat.timestamp  # pyright: ignore[reportOptionalMemberAccess]
 
         # No warning when a second controller starts up again
         with daemon_controller_from_instance(
@@ -437,7 +437,7 @@ def test_warn_multiple_daemons(capsys):
                 now.timestamp(),
                 heartbeat_interval_seconds=heartbeat_interval_seconds,
             )[SensorDaemon.daemon_type()]
-            last_heartbeat_time = status.last_heartbeat.timestamp
+            last_heartbeat_time = status.last_heartbeat.timestamp  # pyright: ignore[reportOptionalMemberAccess]
 
             # Starting up a controller while one is running produces the warning though
             with daemon_controller_from_instance(