Add Event.grouping field and fill it

An event always has a single (automatically calculated) Grouping associated with it.
We add this info to the Event model (we'll soon display it in the UI, and as per the
now-removed comment it's simply the consistent thing to do)

events/factories.py

@@ -25,9 +25,14 @@ def create_event(project=None, issue=None, timestamp=None, event_data=None):
         Max("digest_order"))["digest_order__max"]
     issue_digest_order = max_current + 1 if max_current is not None else 1
 
+    # we get this via issue because we don't have manual merging yet; once we do, the following is more appropriate:
+    # Grouping.objects.filter(project=project, grouping_key=grouping_key).get()
+    grouping = issue.grouping_set.first()
+
     return Event.objects.create(
         project=project,
         issue=issue,
+        grouping=grouping,
         ingested_at=timestamp,
         digested_at=timestamp,
         timestamp=timestamp,

events/migrations/0014_event_grouping.py

@@ -0,0 +1,24 @@
+# Generated by Django 4.2.18 on 2025-01-31 14:31
+
+from django.db import migrations, models
+import django.db.models.deletion
+
+
+class Migration(migrations.Migration):
+
+    dependencies = [
+        ("issues", "0007_alter_turningpoint_options"),
+        ("events", "0013_harmonize_foogested_at"),
+    ]
+
+    operations = [
+        migrations.AddField(
+            model_name="event",
+            name="grouping",
+            field=models.ForeignKey(
+                null=True,
+                on_delete=django.db.models.deletion.SET_NULL,
+                to="issues.grouping",
+            ),
+        ),
+    ]

events/migrations/0015_set_event_grouping.py

@@ -0,0 +1,35 @@
+# Generated by Django 4.2.18 on 2025-01-31 14:40
+
+from django.db import migrations
+
+
+def set_event_grouping(apps, schema_editor):
+    # We can just deduce the grouping from the issue, because manual merging of issues has no UI yet, i.e. as it stands
+    # an issue always has exactly one grouping.
+
+    # This is slightly optimized under the assumption that in cases where there are very many events, there a orders of
+    # magnitude fewer issues, so we iterate over issues and then filter events by issue, rather than iterating over
+    # events and then getting the issue for each event.
+
+    # Naive impl.:
+    # Event = apps.get_model("events", "Event")
+    # for event in Event.objects.all():
+    #     event.grouping = event.issue.grouping_set.first()
+    #     event.save()
+
+    Event = apps.get_model("events", "Event")
+    Issue = apps.get_model("issues", "Issue")
+    for issue in Issue.objects.all():
+        grouping = issue.grouping_set.first()
+        Event.objects.filter(issue=issue).update(grouping=grouping)
+
+class Migration(migrations.Migration):
+
+    dependencies = [
+        ("events", "0014_event_grouping"),
+        ("issues", "0007_alter_turningpoint_options"),
+    ]
+
+    operations = [
+        migrations.RunPython(set_event_grouping),
+    ]

events/models.py

@@ -61,6 +61,9 @@ class Event(models.Model):
     # not actually expected to be null, but we want to be able to delete issues without deleting events (cleanup later)
     issue = models.ForeignKey("issues.Issue", blank=False, null=True, on_delete=models.SET_NULL)
 
+    # not actually expected to be null
+    grouping = models.ForeignKey("issues.Grouping", blank=False, null=True, on_delete=models.SET_NULL)
+
     # The docs say:
     # > Required. Hexadecimal string representing a uuid4 value. The length is exactly 32 characters. Dashes are not
     # > allowed. Has to be lowercase.
@@ -183,7 +186,7 @@ class Event(models.Model):
         return get_title_for_exception_type_and_value(self.calculated_type, self.calculated_value)
 
     @classmethod
-    def from_ingested(cls, event_metadata, digested_at, digest_order, stored_event_count, issue, parsed_data,
+    def from_ingested(cls, event_metadata, digested_at, digest_order, stored_event_count, issue, grouping, parsed_data,
                       denormalized_fields):
 
         # 'from_ingested' may be a bit of a misnomer... the full 'from_ingested' is done in 'digest_event' in the views.
@@ -197,6 +200,7 @@ class Event(models.Model):
                 event_id=event_metadata["event_id"],  # the metadata is the envelope's event_id, which takes precedence
                 project_id=event_metadata["project_id"],
                 issue=issue,
+                grouping=grouping,
                 ingested_at=event_metadata["ingested_at"],
                 digested_at=digested_at,
                 data=json.dumps(parsed_data),

ingest/views.py

@@ -270,14 +270,13 @@ class BaseIngestAPIView(View):
             # "what is a limit anyway, if you can go either over it, or work is done before the limit is reached")
             evict_for_max_events(project, digested_at, project_stored_event_count)
 
-        # NOTE: an event always has a single (automatically calculated) Grouping associated with it. Since we have that
-        # information available here, we could add it to the Event model.
         event, event_created = Event.from_ingested(
             event_metadata,
             ingested_at,
             issue.digested_event_count,
             issue_stored_event_count,
             issue,
+            grouping,
             event_data,
             denormalized_fields,
         )

issues/tests.py

@@ -22,6 +22,7 @@ from bsmain.management.commands.send_json import Command as SendJsonCommand
 from compat.dsn import get_header_value
 from events.models import Event
 from ingest.views import BaseIngestAPIView
+from issues.factories import get_or_create_issue
 
 from .models import Issue, IssueStateManager
 from .regressions import is_regression, is_regression_2, issue_is_regression
@@ -398,7 +399,7 @@ class ViewTests(TransactionTestCase):
         self.user = User.objects.create_user(username='test', password='test')
         self.project = Project.objects.create()
         ProjectMembership.objects.create(project=self.project, user=self.user)
-        self.issue = Issue.objects.create(project=self.project, **denormalized_issue_fields())
+        self.issue, _ = get_or_create_issue(self.project)
         self.event = create_event(self.project, self.issue)
         self.client.force_login(self.user)