Add field to store github_branch for challenges

apps/challenges/challenge_config_utils.py

@@ -312,6 +312,7 @@ def get_value_from_field(data, base_location, field_name):
     "challenge_metadata_schema_errors": "ERROR: Unable to serialize the challenge because of the following errors: {}.",
     "evaluation_script_not_zip": "ERROR: Please pass in a zip file as evaluation script. If using the `evaluation_script` directory (recommended), it should be `evaluation_script.zip`.",
     "docker_based_challenge": "ERROR: New Docker based challenges are not supported starting March 15, 2025.",
+    "invalid_github_branch_format": "ERROR: GitHub branch name '{branch}' is invalid. It must match the pattern 'challenge-<year>-<version>' (e.g., challenge-2024-1, challenge-2060-v2).",
 }
 
 
@@ -364,6 +365,23 @@ def __init__(
         self.phase_ids = []
         self.leaderboard_ids = []
 
+    def validate_github_branch_format(self):
+        """
+        Ensure the github branch name matches challenge-<year>-<version>
+        """
+        branch = self.request.data.get(
+            "GITHUB_BRANCH_NAME"
+        ) or self.request.data.get("BRANCH_NAME")
+        if not branch:
+            branch = "challenge"
+        pattern = r"^challenge-\d{4}-[a-zA-Z0-9]+$"
+        if not re.match(pattern, branch):
+            self.error_messages.append(
+                self.error_messages_dict[
+                    "invalid_github_branch_format"
+                ].format(branch=branch)
+            )
+
     def read_and_validate_yaml(self):
         if not self.yaml_file_count:
             message = self.error_messages_dict.get("no_yaml_file")
@@ -587,7 +605,8 @@ def validate_serializer(self):
                         "GITHUB_REPOSITORY"
                     ],
                     "github_branch": self.request.data.get(
-                        "GITHUB_BRANCH_NAME", ""
+                        "GITHUB_BRANCH_NAME", "challenge"
+
                     ),
                 },
             )
@@ -1134,6 +1153,9 @@ def validate_challenge_config_util(
 
     val_config_util.validate_serializer()
 
+    # Add branch format validation
+    val_config_util.validate_github_branch_format()
+
     # Get existing config IDs for leaderboards and dataset splits
     if current_challenge:
         current_challenge_phases = ChallengePhase.objects.filter(

apps/challenges/migrations/0113_add_github_branch_field_and_unique_constraint.py

@@ -3,6 +3,32 @@
 from django.db import migrations, models
 
 
+def populate_github_branch_default(apps, schema_editor):
+    """
+    Populate existing challenges with empty github_branch fields to use "challenge" as default.
+    """
+    Challenge = apps.get_model("challenges", "Challenge")
+    # Update all challenges that have github_repository but empty github_branch
+    Challenge.objects.filter(github_repository__isnull=False).exclude(
+        github_repository=""
+    ).filter(github_branch__isnull=True).update(github_branch="challenge")
+
+    # Also update challenges with empty string github_branch
+    Challenge.objects.filter(github_repository__isnull=False).exclude(
+        github_repository=""
+    ).filter(github_branch="").update(github_branch="challenge")
+
+
+def reverse_populate_github_branch_default(apps, schema_editor):
+    """
+    Reverse migration - set github_branch back to empty string for challenges that were set to "challenge".
+    """
+    Challenge = apps.get_model("challenges", "Challenge")
+    # Only reverse if the field was set to "challenge" by this migration
+    Challenge.objects.filter(github_repository__isnull=False).exclude(
+        github_repository=""
+    ).filter(github_branch="challenge").update(github_branch="")
+
 def fix_duplicate_github_fields(apps, schema_editor):
     """
     No data migration needed since we're using a partial unique constraint.
@@ -28,9 +54,14 @@ class Migration(migrations.Migration):
             model_name="challenge",
             name="github_branch",
             field=models.CharField(
-                blank=True, default="", max_length=200, null=True
+                blank=True, default="challenge", max_length=200, null=True
             ),
         ),
+        # Data migration to populate existing records
+        migrations.RunPython(
+            populate_github_branch_default,
+            reverse_populate_github_branch_default,
+        ),
         migrations.RunPython(
             fix_duplicate_github_fields,
             reverse_fix_duplicate_github_fields,

apps/challenges/models.py

@@ -188,7 +188,7 @@ def __init__(self, *args, **kwargs):
     )
     # The github branch name used to create/update the challenge
     github_branch = models.CharField(
-        max_length=200, null=True, blank=True, default=""
+        max_length=200, null=True, blank=True, default=challenge
     )
     # The number of vCPU for a Fargate worker for the challenge. Default value
     # is 0.25 vCPU.

apps/challenges/views.py

@@ -3898,8 +3898,9 @@ def create_or_update_github_challenge(request, challenge_host_team_pk):
         return Response(response_data, status=status.HTTP_406_NOT_ACCEPTABLE)
 
     # Get branch name with default fallback
-    github_branch = request.data.get("GITHUB_BRANCH_NAME", "")
-
+    github_branch = request.data.get("GITHUB_BRANCH_NAME") or request.data.get(
+        "BRANCH_NAME", "challenge"
+    )
     challenge_queryset = Challenge.objects.filter(
         github_repository=request.data["GITHUB_REPOSITORY"],
         github_branch=github_branch,
@@ -4288,6 +4289,8 @@ def create_or_update_github_challenge(request, challenge_host_team_pk):
                             "challenge_evaluation_script_file"
                         ],
                         "worker_image_url": worker_image_url,
+                        "github_repository": request.data["GITHUB_REPOSITORY"],
+                        "github_branch": github_branch,
                     },
                 )
                 if serializer.is_valid():

scripts/migration/populate_github_branch.py

@@ -0,0 +1,79 @@
+#!/usr/bin/env python
+# Command to run: python manage.py shell < scripts/migration/populate_github_branch.py
+"""
+Populate existing challenges with github_branch="challenge" for backward compatibility.
+
+This script should be run after the migration to ensure all existing challenges
+have the github_branch field populated with the default value.
+"""
+
+import traceback
+
+from challenges.models import Challenge
+from django.db import models
+
+
+def populate_github_branch_fields():
+    """
+    Populate existing challenges with empty github_branch fields to use "challenge" as default.
+    """
+    print("Starting github_branch field population...")
+
+    challenges_to_update = (
+        Challenge.objects.filter(github_repository__isnull=False)
+        .exclude(github_repository="")
+        .filter(
+            models.Q(github_branch__isnull=True) | models.Q(github_branch="")
+        )
+    )
+
+    count = challenges_to_update.count()
+
+    if count == 0:
+        print("No challenges found that need github_branch population.")
+        return
+
+    print(f"Found {count} challenges that need github_branch population.")
+
+    updated_count = challenges_to_update.update(github_branch="challenge")
+
+    print(
+        f"Successfully updated {updated_count} challenges with github_branch='challenge'"
+    )
+
+    remaining_empty = (
+        Challenge.objects.filter(github_repository__isnull=False)
+        .exclude(github_repository="")
+        .filter(
+            models.Q(github_branch__isnull=True) | models.Q(github_branch="")
+        )
+        .count()
+    )
+
+    if remaining_empty == 0:
+        print("✅ All challenges now have github_branch populated!")
+    else:
+        print(
+            f"⚠️  Warning: {remaining_empty} challenges still have empty github_branch fields"
+        )
+
+    sample_challenges = (
+        Challenge.objects.filter(github_repository__isnull=False)
+        .exclude(github_repository="")
+        .values("id", "title", "github_repository", "github_branch")[:5]
+    )
+
+    print("\nSample updated challenges:")
+    for challenge in sample_challenges:
+        print(
+            f"  ID: {challenge['id']}, Title: {challenge['title']}, "
+            f"Repo: {challenge['github_repository']}, Branch: {challenge['github_branch']}"
+        )
+
+
+try:
+    populate_github_branch_fields()
+    print("\n✅ Script completed successfully!")
+except Exception as e:
+    print(f"\n❌ Error occurred: {e}")
+    print(traceback.print_exc())

scripts/seed.py

@@ -256,7 +256,7 @@ def create_challenge(
         featured=is_featured,
         image=image_file,
         github_repository=f"evalai-examples/{slug}",
-        github_branch="main",
+        github_branch="challenge",
     )
     challenge.save()
 

tests/unit/challenges/test_views.py

@@ -6020,14 +6020,13 @@ def test_create_challenge_using_github_without_branch_name(self):
             self.assertEqual(response.status_code, 201)
             self.assertEqual(response.json(), expected)
 
-        # Verify github_branch defaults to empty string when not provided
+        # Verify github_branch defaults to "challenge" when not provided
         challenge = Challenge.objects.first()
         self.assertEqual(
             challenge.github_repository,
             "https://github.com/yourusername/repository",
         )
-        self.assertEqual(challenge.github_branch, "")
-
+        self.assertEqual(challenge.github_branch, "challenge")
 
 class ValidateChallengeTest(APITestCase):
     def setUp(self):