refactor(worker): Inject repository service into ReportService and optimize commit query

apps/worker/services/report/__init__.py

@@ -48,6 +48,7 @@
 from shared.reports.resources import Report
 from shared.reports.types import TOTALS_MAP
 from shared.storage.exceptions import FileNotInStorageError
+from shared.torngit.base import TorngitBaseAdapter
 from shared.torngit.exceptions import TorngitError
 from shared.upload.constants import UploadErrorCode
 from shared.utils.sessions import Session, SessionType
@@ -192,11 +193,15 @@ def create_report_upload(
 
 class ReportService(BaseReportService):
     def __init__(
-        self, current_yaml: UserYaml | dict, gh_app_installation_name: str | None = None
+        self,
+        current_yaml: UserYaml | dict,
+        gh_app_installation_name: str | None = None,
+        repository_service: TorngitBaseAdapter | None = None,
     ):
         super().__init__(current_yaml)
         self.flag_dict: dict[str, RepositoryFlag] | None = None
         self.gh_app_installation_name = gh_app_installation_name
+        self.repository_service = repository_service
 
     def has_initialized_report(self, commit: Commit) -> bool:
         """
@@ -413,7 +418,7 @@ def _possibly_shift_carryforward_report(
         self, carryforward_report: Report, base_commit: Commit, head_commit: Commit
     ) -> Report:
         try:
-            provider_service = get_repo_provider_service(
+            provider_service = self.repository_service or get_repo_provider_service(
                 repository=head_commit.repository,
                 installation_name_to_use=self.gh_app_installation_name,
             )

apps/worker/tasks/preprocess_upload.py

@@ -1,8 +1,11 @@
 import logging
 
+from sqlalchemy.orm import joinedload
+
 from app import celery_app
 from database.enums import CommitErrorTypes
 from database.models import Commit
+from database.models.core import Owner, Repository
 from helpers.exceptions import RepositoryWithoutValidBotError
 from helpers.github_installation import get_installation_name_for_owner_for_task
 from helpers.save_commit_error import save_commit_error
@@ -99,6 +102,11 @@ def process_impl_within_lock(self, db_session, repoid, commitid):
         commit = (
             db_session.query(Commit)
             .filter(Commit.repoid == repoid, Commit.commitid == commitid)
+            .options(
+                joinedload(Commit.repository)
+                .joinedload(Repository.author)
+                .joinedload(Owner.github_app_installations)
+            )
             .first()
         )
         assert commit, "Commit not found in database."
@@ -123,7 +131,9 @@ def process_impl_within_lock(self, db_session, repoid, commitid):
         )
         commit_yaml = fetch_commit_yaml_and_possibly_store(commit, repository_service)
         report_service = ReportService(
-            commit_yaml, gh_app_installation_name=installation_name_to_use
+            commit_yaml,
+            gh_app_installation_name=installation_name_to_use,
+            repository_service=repository_service,
         )
         commit_report = report_service.initialize_and_save_report(commit)
         # Persist changes from within the lock