feat(similarity): Do not block on project platform (#82952)

This will allow attempting to backfill all projects. Only the event's
platform will prevent attempting a backfill.

src/sentry/seer/similarity/utils.py

@@ -36,16 +36,6 @@
         "ruby",
     ]
 )
-# Existing projects with these platforms shouldn't be backfilled and new projects with these
-# platforms shouldn't have Seer enabled.
-SEER_INELIGIBLE_PROJECT_PLATFORMS = frozenset(
-    [
-        # We have no clue what's in these projects
-        "other",
-        "",
-        None,
-    ]
-)
 BASE64_ENCODED_PREFIXES = [
     "data:text/html;base64",
     "data:text/javascript;base64",
@@ -455,11 +445,9 @@ def _is_snipped_context_line(context_line: str) -> bool:
 
 def project_is_seer_eligible(project: Project) -> bool:
     """
-    Return True if the project hasn't already been backfilled, is a Seer-eligible platform, and
-    the feature is enabled in the region.
+    Return True if the project hasn't already been backfilled and the feature is enabled in the region.
     """
     is_backfill_completed = project.get_option("sentry:similarity_backfill_completed")
-    is_seer_eligible_platform = project.platform not in SEER_INELIGIBLE_PROJECT_PLATFORMS
     is_region_enabled = options.get("similarity.new_project_seer_grouping.enabled")
 
-    return not is_backfill_completed and is_seer_eligible_platform and is_region_enabled
+    return not is_backfill_completed and is_region_enabled

tests/sentry/tasks/test_backfill_seer_grouping_records.py

@@ -23,7 +23,6 @@
 from sentry.grouping.enhancer.exceptions import InvalidEnhancerConfig
 from sentry.models.group import Group, GroupStatus
 from sentry.models.grouphash import GroupHash
-from sentry.models.project import Project
 from sentry.seer.similarity.grouping_records import CreateGroupingRecordData
 from sentry.seer.similarity.types import RawSeerSimilarIssueData
 from sentry.seer.similarity.utils import MAX_FRAME_COUNT
@@ -114,10 +113,6 @@
 
 @django_db_all
 class TestBackfillSeerGroupingRecords(SnubaTestCase, TestCase):
-    def create_project(self, **kwargs) -> Project:
-        """We overwrite the default create_project method to make sure that all created projects are seer eligible."""
-        return super().create_project(**kwargs, platform=kwargs.get("platform", "python"))
-
     def create_exception_values(self, function_name: str, type: str, value: str):
         return {
             "values": [
@@ -220,9 +215,6 @@ def assert_group_metadata_not_updated(self, group: Group) -> None:
 
     def setUp(self):
         super().setUp()
-        # Make the project seer eligible
-        self.project.platform = "python"
-        self.project.save()
         bulk_data = self.create_group_event_rows(5)
         self.event = bulk_data["events"][0]
         self.bulk_rows, self.bulk_events = (bulk_data["rows"], bulk_data["events"])
@@ -1716,8 +1708,6 @@ def test_backfill_seer_grouping_records_cohort_creation(
         project_same_cohort = self.create_project(
             organization=self.organization, id=self.project.id + thread_number
         )
-        project_same_cohort.platform = "javascript"
-        project_same_cohort.save()
         event_same_cohort = self.store_event(
             data={
                 "exception": EXCEPTION,
@@ -1806,93 +1796,6 @@ def test_backfill_seer_grouping_records_cohort_creation(
 
         assert mock_logger.info.call_args_list == expected_call_args_list
 
-    @override_options({"similarity.new_project_seer_grouping.enabled": True})
-    @patch("sentry.tasks.embeddings_grouping.backfill_seer_grouping_records_for_project.logger")
-    @patch("sentry.tasks.embeddings_grouping.utils.post_bulk_grouping_records")
-    def test_backfill_seer_grouping_records_cohort_creation_not_seer_eligible(
-        self, mock_post_bulk_grouping_records, mock_logger
-    ):
-        """
-        Test that non Seer eligible projects are not processed when worker_number is provided.
-        """
-        # Create 1 seer eligible project that project_id % thread_number == worker_number
-        thread_number = options.get("similarity.backfill_total_worker_count")
-        worker_number = self.project.id % thread_number
-
-        # Create 1 non seer eligible project that project_id % thread_number != worker_number
-        project_same_cohort_not_eligible = self.create_project(
-            organization=self.organization, id=self.project.id + thread_number
-        )
-        project_same_cohort_not_eligible.platform = "other"  # Not currently eligible
-        project_same_cohort_not_eligible.save()
-        self.create_event(project_same_cohort_not_eligible.id, times_seen=5)
-
-        # Create one project where project_id % thread_number != worker_number
-        self.create_project(organization=self.organization, id=self.project.id + 1)
-
-        mock_post_bulk_grouping_records.return_value = {"success": True, "groups_with_neighbor": {}}
-        with TaskRunner():
-            backfill_seer_grouping_records_for_project(
-                current_project_id=None,
-                worker_number=worker_number,
-            )
-
-        project_last_group_id = sorted(
-            [group.id for group in Group.objects.filter(project_id=self.project.id)]
-        )[0]
-
-        expected_cohort = [self.project.id, project_same_cohort_not_eligible.id]
-        expected_call_args_list = [
-            call(
-                "backfill_seer_grouping_records",
-                extra={
-                    "current_project_id": self.project.id,
-                    "last_processed_group_id": None,
-                    "cohort": expected_cohort,
-                    "last_processed_project_index": None,
-                    "only_delete": False,
-                    "skip_processed_projects": True,
-                    "skip_project_ids": None,
-                    "worker_number": worker_number,
-                },
-            ),
-            call(
-                "backfill_seer_grouping_records",
-                extra={
-                    "current_project_id": self.project.id,
-                    "last_processed_group_id": project_last_group_id,
-                    "cohort": expected_cohort,
-                    "last_processed_project_index": 0,
-                    "only_delete": False,
-                    "skip_processed_projects": True,
-                    "skip_project_ids": None,
-                    "worker_number": worker_number,
-                },
-            ),
-            call(
-                "backfill_seer_grouping_records",
-                extra={
-                    "current_project_id": project_same_cohort_not_eligible.id,
-                    "last_processed_group_id": None,
-                    "cohort": expected_cohort,
-                    "last_processed_project_index": 1,
-                    "only_delete": False,
-                    "skip_processed_projects": True,
-                    "skip_project_ids": None,
-                    "worker_number": worker_number,
-                },
-            ),
-            call(
-                "backfill_seer_grouping_records.project_is_not_seer_eligible",
-                extra={"project_id": project_same_cohort_not_eligible.id},
-            ),
-            call(
-                "reached the end of the projects in cohort", extra={"worker_number": worker_number}
-            ),
-        ]
-
-        assert mock_logger.info.call_args_list == expected_call_args_list
-
     @override_options({"similarity.new_project_seer_grouping.enabled": True})
     @override_options({"similarity.backfill_project_cohort_size": 1})
     @patch("sentry.tasks.embeddings_grouping.backfill_seer_grouping_records_for_project.logger")
@@ -1910,8 +1813,6 @@ def test_backfill_seer_grouping_records_cohort_creation_multiple_batches(
         project_same_worker = self.create_project(
             organization=self.organization, id=self.project.id + thread_number
         )
-        project_same_worker.platform = "javascript"
-        project_same_worker.save()
         event_same_worker = self.store_event(
             data={
                 "exception": EXCEPTION,