feat(preprod): Add backfill migration for PreprodArtifactMobileAppInfo (#105883)

Backfill migration for the new PreprodArtifactMobileAppInfo. Only about
9900 rows of PreprodArtifact to backfill, so per the [migration
docs](https://develop.sentry.dev/backend/application-domains/database-migrations/#post-deploy-migrations-is_post_deployment),
we won't use a post-deploy migration (<50k rows).

Author: rbro112

migrations_lockfile.txt

@@ -23,7 +23,7 @@ nodestore: 0001_squashed_0002_nodestore_no_dictfield
 
 notifications: 0002_notificationmessage_jsonfield
 
-preprod: 0021_add_preprod_artifact_mobile_app_info
+preprod: 0022_backfill_preprod_artifact_mobile_app_info
 
 prevent: 0002_alter_integration_id_not_null
 

src/sentry/preprod/migrations/0022_backfill_preprod_artifact_mobile_app_info.py

@@ -0,0 +1,103 @@
+# Generated by Django 5.2.8 on 2026-01-07 10:40
+
+import logging
+
+from django.db import IntegrityError, migrations, router, transaction
+from django.db.backends.base.schema import BaseDatabaseSchemaEditor
+from django.db.migrations.state import StateApps
+
+from sentry.new_migrations.migrations import CheckedMigration
+from sentry.utils.query import RangeQuerySetWrapperWithProgressBar
+
+logger = logging.getLogger(__name__)
+
+
+def backfill_preprod_artifact_mobile_app_info(
+    apps: StateApps, schema_editor: BaseDatabaseSchemaEditor
+) -> None:
+    PreprodArtifact = apps.get_model("preprod", "PreprodArtifact")
+    PreprodArtifactMobileAppInfo = apps.get_model("preprod", "PreprodArtifactMobileAppInfo")
+
+    all_artifacts = PreprodArtifact.objects.all()
+    mobile_app_info_to_create = []
+
+    # Batch up the query for existing PreprodArtifactMobileAppInfo entries to avoid N+1 queries.
+    # Fetch all artifact IDs that already have a PreprodArtifactMobileAppInfo
+    existing_mobile_app_info_artifact_ids = set(
+        PreprodArtifactMobileAppInfo.objects.values_list("preprod_artifact_id", flat=True)
+    )
+
+    # Loop through all artifacts, but skip if already exists
+    for artifact in RangeQuerySetWrapperWithProgressBar(all_artifacts):
+        if artifact.id in existing_mobile_app_info_artifact_ids:
+            logger.info(
+                "Skipping artifact %s because it already has a PreprodArtifactMobileAppInfo entry",
+                artifact.id,
+            )
+            continue
+
+        # Only create if there's at least one mobile app field with data
+        has_mobile_app_data = any(
+            [
+                artifact.build_version,
+                artifact.build_number is not None,
+                artifact.app_name,
+                artifact.app_icon_id,
+            ]
+        )
+
+        if has_mobile_app_data:
+            mobile_app_info_to_create.append(
+                PreprodArtifactMobileAppInfo(
+                    preprod_artifact_id=artifact.id,
+                    build_version=artifact.build_version,
+                    build_number=artifact.build_number,
+                    app_name=artifact.app_name,
+                    app_icon_id=artifact.app_icon_id,
+                )
+            )
+
+    with transaction.atomic(router.db_for_write(PreprodArtifactMobileAppInfo)):
+        try:
+            PreprodArtifactMobileAppInfo.objects.bulk_create(
+                mobile_app_info_to_create,
+                ignore_conflicts=True,
+            )
+        except IntegrityError as e:
+            logger.exception(
+                "Error bulk creating preprod artifact mobile app info",
+                extra={
+                    "mobile_app_info_to_create": len(mobile_app_info_to_create),
+                    "error": e,
+                },
+            )
+
+
+class Migration(CheckedMigration):
+    # This flag is used to mark that a migration shouldn't be automatically run in production.
+    # This should only be used for operations where it's safe to run the migration after your
+    # code has deployed. So this should not be used for most operations that alter the schema
+    # of a table.
+    # Here are some things that make sense to mark as post deployment:
+    # - Large data migrations. Typically we want these to be run manually so that they can be
+    #   monitored and not block the deploy for a long period of time while they run.
+    # - Adding indexes to large tables. Since this can take a long time, we'd generally prefer to
+    #   run this outside deployments so that we don't block them. Note that while adding an index
+    #   is a schema change, it's completely safe to run the operation after the code has deployed.
+    # Once deployed, run these manually via: https://develop.sentry.dev/database-migrations/#migration-deployment
+
+    # This migration is safe to run after the code has deployed as the total amount of PreprodArtifacts is small (~6k records)
+    is_post_deployment = False
+
+    dependencies = [
+        ("preprod", "0021_add_preprod_artifact_mobile_app_info"),
+    ]
+
+    operations = [
+        migrations.RunPython(
+            backfill_preprod_artifact_mobile_app_info,
+            reverse_code=migrations.RunPython.noop,
+            elidable=True,
+            hints={"tables": ["sentry_preprodartifact", "sentry_preprodartifactmobileappinfo"]},
+        ),
+    ]

tests/sentry/preprod/migrations/test_0022_backfill_preprod_artifact_mobile_app_info.py

@@ -0,0 +1,70 @@
+import pytest
+
+from sentry.models.organization import Organization
+from sentry.preprod.models import PreprodArtifactMobileAppInfo
+from sentry.testutils.cases import TestMigrations
+
+
+@pytest.mark.migrations
+class BackfillPreprodArtifactMobileAppInfoTest(TestMigrations):
+    migrate_from = "0021_add_preprod_artifact_mobile_app_info"
+    migrate_to = "0022_backfill_preprod_artifact_mobile_app_info"
+    app = "preprod"
+
+    def setup_before_migration(self, apps):
+        PreprodArtifact = apps.get_model("preprod", "PreprodArtifact")
+
+        self.organization: Organization = self.create_organization(name="test", slug="test")
+        self.project = self.create_project(organization=self.organization)
+        self.user = self.create_user()
+
+        # Create artifact with all mobile app fields
+        # state=3 is PROCESSED
+        self.artifact_all_fields = PreprodArtifact.objects.create(
+            project_id=self.project.id,
+            build_version="1.2.3",
+            build_number=100,
+            app_name="Test App",
+            app_icon_id="icon123",
+            state=3,
+        )
+
+        # Create artifact with some mobile app fields
+        # state=3 is PROCESSED
+        self.artifact_partial_fields = PreprodArtifact.objects.create(
+            project_id=self.project.id,
+            build_version="2.0.0",
+            build_number=200,
+            state=3,
+        )
+
+        # Create artifact with no mobile app fields
+        # state=0 is UPLOADING
+        self.artifact_no_fields = PreprodArtifact.objects.create(
+            project_id=self.project.id,
+            state=0,
+        )
+
+    def test_backfills_mobile_app_info(self):
+        # Check artifact with all fields
+        mobile_app_info_all = PreprodArtifactMobileAppInfo.objects.get(
+            preprod_artifact_id=self.artifact_all_fields.id
+        )
+        assert mobile_app_info_all.build_version == "1.2.3"
+        assert mobile_app_info_all.build_number == 100
+        assert mobile_app_info_all.app_name == "Test App"
+        assert mobile_app_info_all.app_icon_id == "icon123"
+
+        # Check artifact with partial fields
+        mobile_app_info_partial = PreprodArtifactMobileAppInfo.objects.get(
+            preprod_artifact_id=self.artifact_partial_fields.id
+        )
+        assert mobile_app_info_partial.build_version == "2.0.0"
+        assert mobile_app_info_partial.build_number == 200
+        assert mobile_app_info_partial.app_name is None
+        assert mobile_app_info_partial.app_icon_id is None
+
+        # Check artifact with no fields - should not have mobile app info created
+        assert not PreprodArtifactMobileAppInfo.objects.filter(
+            preprod_artifact_id=self.artifact_no_fields.id
+        ).exists()