Backend: Optimize create_or_update_github_challenge function to prefetch challenge_phase_splits, reducing N+1 query issues during updates; add unit test to verify performance improvements. (#5001)

RishabhJain2018 · web-flow · commit 68f4a6193b2f · 2026-02-08T13:25:34.000-08:00
diff --git a/apps/challenges/views.py b/apps/challenges/views.py
@@ -4561,20 +4561,44 @@ def create_or_update_github_challenge(request, challenge_host_team_pk):
                         error_messages, status=status.HTTP_400_BAD_REQUEST
                     )
 
+                # Prefetch all challenge_phase_splits for this challenge to avoid
+                # N+1 queries in the leaderboard, dataset_split, and
+                # challenge_phase_split update loops.
+                all_challenge_phase_splits = list(
+                    ChallengePhaseSplit.objects.filter(
+                        challenge_phase__challenge__pk=challenge.pk
+                    ).select_related(
+                        "challenge_phase", "dataset_split", "leaderboard"
+                    )
+                )
+                cps_by_leaderboard_config_id = {
+                    cps.leaderboard.config_id: cps
+                    for cps in all_challenge_phase_splits
+                    if cps.leaderboard.config_id is not None
+                }
+                cps_by_dataset_split_config_id = {
+                    cps.dataset_split.config_id: cps
+                    for cps in all_challenge_phase_splits
+                    if cps.dataset_split.config_id is not None
+                }
+                cps_by_phase_split_leaderboard = {
+                    (
+                        cps.challenge_phase_id,
+                        cps.dataset_split_id,
+                        cps.leaderboard_id,
+                    ): cps
+                    for cps in all_challenge_phase_splits
+                }
+
                 # Updating Leaderboard object
                 leaderboard_ids = {}
                 yaml_file_data_of_leaderboard = yaml_file_data["leaderboard"]
                 for data in yaml_file_data_of_leaderboard:
-                    challenge_phase_split_qs = (
-                        ChallengePhaseSplit.objects.filter(
-                            challenge_phase__challenge__pk=challenge.pk,
-                            leaderboard__config_id=data["config_id"],
-                        )
+                    leaderboard_config_id = data.get("config_id", data["id"])
+                    challenge_phase_split = cps_by_leaderboard_config_id.get(
+                        leaderboard_config_id
                     )
-                    if challenge_phase_split_qs:
-                        challenge_phase_split = (
-                            challenge_phase_split_qs.first()
-                        )
+                    if challenge_phase_split is not None:
                         leaderboard = challenge_phase_split.leaderboard
                         serializer = LeaderboardSerializer(
                             leaderboard,
@@ -4672,16 +4696,10 @@ def create_or_update_github_challenge(request, challenge_host_team_pk):
                 ]
                 dataset_split_ids = {}
                 for data in yaml_file_data_of_dataset_split:
-                    challenge_phase_split_qs = (
-                        ChallengePhaseSplit.objects.filter(
-                            challenge_phase__challenge__pk=challenge.pk,
-                            dataset_split__config_id=data["id"],
-                        )
+                    challenge_phase_split = cps_by_dataset_split_config_id.get(
+                        data["id"]
                     )
-                    if challenge_phase_split_qs:
-                        challenge_phase_split = (
-                            challenge_phase_split_qs.first()
-                        )
+                    if challenge_phase_split is not None:
                         dataset_split = challenge_phase_split.dataset_split
                         serializer = DatasetSplitSerializer(
                             dataset_split,
@@ -4770,17 +4788,10 @@ def create_or_update_github_challenge(request, challenge_host_team_pk):
                         "leaderboard_decimal_precision": leaderboard_decimal_precision,
                     }
 
-                    challenge_phase_split_qs = (
-                        ChallengePhaseSplit.objects.filter(
-                            challenge_phase__pk=challenge_phase,
-                            dataset_split__pk=dataset_split,
-                            leaderboard__pk=leaderboard,
-                        )
+                    challenge_phase_split = cps_by_phase_split_leaderboard.get(
+                        (challenge_phase, dataset_split, leaderboard)
                     )
-                    if challenge_phase_split_qs:
-                        challenge_phase_split = (
-                            challenge_phase_split_qs.first()
-                        )
+                    if challenge_phase_split is not None:
                         serializer = ZipChallengePhaseSplitSerializer(
                             challenge_phase_split, data=data
                         )
@@ -4791,7 +4802,11 @@ def create_or_update_github_challenge(request, challenge_host_team_pk):
                     if serializer.is_valid():
                         serializer.save()
                     else:
-                        error_messages = f"challenge phase split update (phase:{data['challenge_phase_id']}, leaderboard:{data['leaderboard_id']}, dataset split: {data['dataset_split_id']}):{str(serializer.errors)}"
+                        error_messages = (
+                            f"challenge phase split update "
+                            f"(phase:{challenge_phase}, leaderboard:{leaderboard}, "
+                            f"dataset split: {dataset_split}):{str(serializer.errors)}"
+                        )
                         raise RuntimeError()
 
                 response_data = {
diff --git a/examples/example4/test_zip_file.zip b/examples/example4/test_zip_file.zip
diff --git a/examples/example4/test_zip_file/challenge_config.yaml b/examples/example4/test_zip_file/challenge_config.yaml
@@ -0,0 +1,76 @@
+title: Challenge Title
+short_description: Short description of the challenge (preferably 140 characters)
+description: description.html
+evaluation_details: evaluation_details.html
+terms_and_conditions: terms_and_conditions.html
+image : logo.png
+submission_guidelines: submission_guidelines.html
+evaluation_script: evaluation_script.zip
+start_date: 2017-06-09 20:00:00
+end_date: 2017-06-19 20:00:00
+
+leaderboard:
+  - id: 1
+    schema: {"labels": ["yes/no", "number", "others", "overall"], "default_order_by": "overall"}
+  - id: 2
+    schema: {"labels": ["metric_a", "metric_b"], "default_order_by": "metric_a"}
+
+challenge_phases:
+  - id: 1
+    name: Challenge Phase 1
+    description: challenge_phase_description.html
+    leaderboard_public: True
+    is_public: True
+    start_date: 2017-06-09 20:00:00
+    end_date: 2017-06-19 20:00:00
+    test_annotation_file: test_annotation.txt
+    codename: phase1
+    max_submissions_per_day: 100
+    max_submissions: 1000
+    max_submissions_per_month: 1000
+  - id: 2
+    name: Challenge Phase 2
+    description: challenge_phase_description.html
+    leaderboard_public: True
+    is_public: True
+    start_date: 2017-06-09 20:00:00
+    end_date: 2017-06-19 20:00:00
+    test_annotation_file: test_annotation.txt
+    codename: phase2
+    max_submissions_per_day: 100
+    max_submissions: 1000
+    max_submissions_per_month: 1000
+
+dataset_splits:
+  - id: 1
+    name: Name of the dataset split
+    codename: codename of dataset split
+  - id: 2
+    name: Name of the dataset split 2
+    codename: codename of dataset split 2
+
+challenge_phase_splits:
+  - challenge_phase_id: 1
+    leaderboard_id: 1
+    dataset_split_id: 1
+    visibility: 3
+    leaderboard_decimal_precision: 2
+    is_leaderboard_order_descending: True
+  - challenge_phase_id: 1
+    leaderboard_id: 2
+    dataset_split_id: 1
+    visibility: 3
+    leaderboard_decimal_precision: 2
+    is_leaderboard_order_descending: True
+  - challenge_phase_id: 2
+    leaderboard_id: 1
+    dataset_split_id: 1
+    visibility: 3
+    leaderboard_decimal_precision: 2
+    is_leaderboard_order_descending: True
+  - challenge_phase_id: 2
+    leaderboard_id: 2
+    dataset_split_id: 2
+    visibility: 3
+    leaderboard_decimal_precision: 2
+    is_leaderboard_order_descending: True
diff --git a/examples/example4/test_zip_file/challenge_phase_description.html b/examples/example4/test_zip_file/challenge_phase_description.html
@@ -0,0 +1,3 @@
+<div class="row">
+        <p><span style="font-weight: 400;">Lorem Ipsum is simply dummy text of the printing and typesetting industry. Lorem Ipsum has been the industry's standard dummy text ever since the 1500s, when an unknown printer took a galley of type and scrambled it to make a type specimen book. It has survived not only five centuries, but also the leap into electronic typesetting, remaining essentially unchanged. It was popularised in the 1960s with the release of Letraset sheets containing Lorem Ipsum passages, and more recently with desktop publishing software like Aldus PageMaker including versions of Lorem Ipsum.</span></p>
+</div>
diff --git a/examples/example4/test_zip_file/description.html b/examples/example4/test_zip_file/description.html
@@ -0,0 +1,3 @@
+<div class="row">
+        <p><span style="font-weight: 400;">Lorem Ipsum is simply dummy text of the printing and typesetting industry. Lorem Ipsum has been the industry's standard dummy text ever since the 1500s, when an unknown printer took a galley of type and scrambled it to make a type specimen book. It has survived not only five centuries, but also the leap into electronic typesetting, remaining essentially unchanged. It was popularised in the 1960s with the release of Letraset sheets containing Lorem Ipsum passages, and more recently with desktop publishing software like Aldus PageMaker including versions of Lorem Ipsum.</span></p>
+</div>
diff --git a/examples/example4/test_zip_file/evaluation_details.html b/examples/example4/test_zip_file/evaluation_details.html
@@ -0,0 +1,3 @@
+<div class="row">
+        <p><span style="font-weight: 400;">Lorem Ipsum is simply dummy text of the printing and typesetting industry. Lorem Ipsum has been the industry's standard dummy text ever since the 1500s, when an unknown printer took a galley of type and scrambled it to make a type specimen book. It has survived not only five centuries, but also the leap into electronic typesetting, remaining essentially unchanged. It was popularised in the 1960s with the release of Letraset sheets containing Lorem Ipsum passages, and more recently with desktop publishing software like Aldus PageMaker including versions of Lorem Ipsum.</span></p>
+</div>
diff --git a/examples/example4/test_zip_file/evaluation_script.zip b/examples/example4/test_zip_file/evaluation_script.zip
diff --git a/examples/example4/test_zip_file/logo.png b/examples/example4/test_zip_file/logo.png
diff --git a/examples/example4/test_zip_file/submission_guidelines.html b/examples/example4/test_zip_file/submission_guidelines.html
@@ -0,0 +1,3 @@
+<div class="row">
+        <p><span style="font-weight: 400;">Lorem Ipsum is simply dummy text of the printing and typesetting industry. Lorem Ipsum has been the industry's standard dummy text ever since the 1500s, when an unknown printer took a galley of type and scrambled it to make a type specimen book. It has survived not only five centuries, but also the leap into electronic typesetting, remaining essentially unchanged. It was popularised in the 1960s with the release of Letraset sheets containing Lorem Ipsum passages, and more recently with desktop publishing software like Aldus PageMaker including versions of Lorem Ipsum.</span></p>
+</div>
diff --git a/examples/example4/test_zip_file/terms_and_conditions.html b/examples/example4/test_zip_file/terms_and_conditions.html
@@ -0,0 +1,3 @@
+<div class="row">
+        <p><span style="font-weight: 400;">Lorem Ipsum is simply dummy text of the printing and typesetting industry. Lorem Ipsum has been the industry's standard dummy text ever since the 1500s, when an unknown printer took a galley of type and scrambled it to make a type specimen book. It has survived not only five centuries, but also the leap into electronic typesetting, remaining essentially unchanged. It was popularised in the 1960s with the release of Letraset sheets containing Lorem Ipsum passages, and more recently with desktop publishing software like Aldus PageMaker including versions of Lorem Ipsum.</span></p>
+</div>
diff --git a/examples/example4/test_zip_file/test_annotation.txt b/examples/example4/test_zip_file/test_annotation.txt
@@ -0,0 +1,10 @@
+1
+2
+3
+4
+5
+6
+7
+8
+9
+10
diff --git a/examples/example4/test_zip_file/zip_challenge.yaml b/examples/example4/test_zip_file/zip_challenge.yaml
@@ -0,0 +1,76 @@
+title: Challenge Title
+short_description: Short description of the challenge (preferably 140 characters)
+description: description.html
+evaluation_details: evaluation_details.html
+terms_and_conditions: terms_and_conditions.html
+image : logo.png
+submission_guidelines: submission_guidelines.html
+evaluation_script: evaluation_script.zip
+start_date: 2017-06-09 20:00:00
+end_date: 2017-06-19 20:00:00
+
+leaderboard:
+  - id: 1
+    schema: {"labels": ["yes/no", "number", "others", "overall"], "default_order_by": "overall"}
+  - id: 2
+    schema: {"labels": ["metric_a", "metric_b"], "default_order_by": "metric_a"}
+
+challenge_phases:
+  - id: 1
+    name: Challenge Phase 1
+    description: challenge_phase_description.html
+    leaderboard_public: True
+    is_public: True
+    start_date: 2017-06-09 20:00:00
+    end_date: 2017-06-19 20:00:00
+    test_annotation_file: test_annotation.txt
+    codename: phase1
+    max_submissions_per_day: 100
+    max_submissions: 1000
+    max_submissions_per_month: 1000
+  - id: 2
+    name: Challenge Phase 2
+    description: challenge_phase_description.html
+    leaderboard_public: True
+    is_public: True
+    start_date: 2017-06-09 20:00:00
+    end_date: 2017-06-19 20:00:00
+    test_annotation_file: test_annotation.txt
+    codename: phase2
+    max_submissions_per_day: 100
+    max_submissions: 1000
+    max_submissions_per_month: 1000
+
+dataset_splits:
+  - id: 1
+    name: Name of the dataset split
+    codename: codename of dataset split
+  - id: 2
+    name: Name of the dataset split 2
+    codename: codename of dataset split 2
+
+challenge_phase_splits:
+  - challenge_phase_id: 1
+    leaderboard_id: 1
+    dataset_split_id: 1
+    visibility: 3
+    leaderboard_decimal_precision: 2
+    is_leaderboard_order_descending: True
+  - challenge_phase_id: 1
+    leaderboard_id: 2
+    dataset_split_id: 1
+    visibility: 3
+    leaderboard_decimal_precision: 2
+    is_leaderboard_order_descending: True
+  - challenge_phase_id: 2
+    leaderboard_id: 1
+    dataset_split_id: 1
+    visibility: 3
+    leaderboard_decimal_precision: 2
+    is_leaderboard_order_descending: True
+  - challenge_phase_id: 2
+    leaderboard_id: 2
+    dataset_split_id: 2
+    visibility: 3
+    leaderboard_decimal_precision: 2
+    is_leaderboard_order_descending: True
diff --git a/tests/unit/challenges/test_views.py b/tests/unit/challenges/test_views.py

Original file line number	Diff line number	Diff line change
`@@ -0,0 +1,3 @@`
	`1`	`+<div class="row">`
	`2`	+ <p><span style="font-weight: 400;">Lorem Ipsum is simply dummy text of the printing and typesetting industry. Lorem Ipsum has been the industry's standard dummy text ever since the 1500s, when an unknown printer took a galley of type and scrambled it to make a type specimen book. It has survived not only five centuries, but also the leap into electronic typesetting, remaining essentially unchanged. It was popularised in the 1960s with the release of Letraset sheets containing Lorem Ipsum passages, and more recently with desktop publishing software like Aldus PageMaker including versions of Lorem Ipsum.</span></p>
	`3`	`+</div>`