Mitigate nonexistent user entity cases

Author: arkid15r

backend/apps/github/common.py

@@ -137,7 +137,8 @@ def sync_repository(
                 # Assignees.
                 issue.assignees.clear()
                 for gh_issue_assignee in gh_issue.assignees:
-                    issue.assignees.add(User.update_data(gh_issue_assignee))
+                    if issue_assignee := User.update_data(gh_issue_assignee):
+                        issue.assignees.add(issue_assignee)
 
                 # Labels.
                 issue.labels.clear()
@@ -184,7 +185,8 @@ def sync_repository(
             # Assignees.
             pull_request.assignees.clear()
             for gh_pull_request_assignee in gh_pull_request.assignees:
-                pull_request.assignees.add(User.update_data(gh_pull_request_assignee))
+                if pull_request_assignee := User.update_data(gh_pull_request_assignee):
+                    pull_request.assignees.add(pull_request_assignee)
 
             # Labels.
             pull_request.labels.clear()

backend/apps/github/management/commands/github_update_owasp_organization.py

@@ -82,31 +82,36 @@ def handle(self, *_args, **options) -> None:
         for idx, gh_repository in enumerate(gh_repositories[offset:]):
             prefix = f"{idx + offset + 1} of {gh_repositories_count}"
             entity_key = gh_repository.name.lower()
-            print(f"{prefix:<12} https://github.com/OWASP/{entity_key}")
-
-            owasp_organization, repository = sync_repository(
-                gh_repository,
-                organization=owasp_organization,
-                user=owasp_user,
-            )
-
-            # OWASP chapters.
-            if entity_key.startswith("www-chapter-"):
-                chapters.append(Chapter.update_data(gh_repository, repository, save=False))
-
-            # OWASP projects.
-            elif entity_key.startswith("www-project-"):
-                projects.append(Project.update_data(gh_repository, repository, save=False))
-
-            # OWASP committees.
-            elif entity_key.startswith("www-committee-"):
-                committees.append(Committee.update_data(gh_repository, repository, save=False))
+            repository_url = f"https://github.com/OWASP/{entity_key}"
+            print(f"{prefix:<12} {repository_url}")
+
+            try:
+                owasp_organization, repository = sync_repository(
+                    gh_repository,
+                    organization=owasp_organization,
+                    user=owasp_user,
+                )
+
+                # OWASP chapters.
+                if entity_key.startswith("www-chapter-"):
+                    chapters.append(Chapter.update_data(gh_repository, repository, save=False))
+
+                # OWASP projects.
+                elif entity_key.startswith("www-project-"):
+                    projects.append(Project.update_data(gh_repository, repository, save=False))
+
+                # OWASP committees.
+                elif entity_key.startswith("www-committee-"):
+                    committees.append(Committee.update_data(gh_repository, repository, save=False))
+            except Exception:
+                logger.exception("Error syncing repository %s", repository_url)
+                continue
 
         Chapter.bulk_save(chapters)
         Committee.bulk_save(committees)
         Project.bulk_save(projects)
 
-        if repository is None:
+        if repository is None:  # The entire organization is being synced.
             # Check repository counts.
             local_owasp_repositories_count = Repository.objects.filter(
                 is_owasp_repository=True,

backend/apps/github/management/commands/github_update_project_related_repositories.py

@@ -63,11 +63,15 @@ def handle(self, *args, **options) -> None:
                         project.save(update_fields=("invalid_urls", "related_urls"))
                         continue
 
-                organization, repository = sync_repository(gh_repository)
-                if organization is not None:
-                    organization.save()
+                try:
+                    organization, repository = sync_repository(gh_repository)
+                    if organization is not None:
+                        organization.save()
 
-                project.repositories.add(repository)
+                    project.repositories.add(repository)
+                except Exception:
+                    logger.exception("Error syncing repository %s", repository_url)
+                    continue
 
             projects.append(project)
 

backend/apps/github/models/common.py

@@ -1,6 +1,7 @@
 """Github app common models."""
 
 from django.db import models
+from github.GithubException import UnknownObjectException
 
 
 class GenericUserModel(models.Model):
@@ -78,4 +79,7 @@ class Meta:
     @staticmethod
     def get_node_id(node):
         """Extract node_id."""
-        return node.raw_data["node_id"]
+        try:
+            return node.raw_data["node_id"]
+        except UnknownObjectException:
+            pass

backend/apps/github/models/user.py

@@ -102,7 +102,7 @@ def get_non_indexable_logins() -> set:
         }
 
     @staticmethod
-    def update_data(gh_user, *, save: bool = True) -> User:
+    def update_data(gh_user, *, save: bool = True) -> User | None:
         """Update GitHub user data.
 
         Args:
@@ -114,6 +114,9 @@ def update_data(gh_user, *, save: bool = True) -> User:
 
         """
         user_node_id = User.get_node_id(gh_user)
+        if not user_node_id:
+            return None
+
         try:
             user = User.objects.get(node_id=user_node_id)
         except User.DoesNotExist: