Merge pull request #9738 from beeankha/update_schedule_sort

Enable ?page_size=1 to Fetch Correct Objects on JT Endpoint

SUMMARY

Problem:
When multiple schedules are made from a single JT and a URL sort is done via:
api/v2/job_templates/[id]/schedules/?page_size=1
... the first half of the results come back with the same ID. 🤔
Solution:
Implementing a backup sort via ID on the schedules model fixes this problem, and now all schedules on the job template endpoint are being fetched properly.

ISSUE TYPE


Bugfix Pull Request

COMPONENT NAME


API

AWX VERSION

awx: 18.0.0

ADDITIONAL INFORMATION

This might fix #9632

Reviewed-by: Seth Foster <None>
Reviewed-by: Rebeccah Hunter <rhunter@redhat.com>

awx/main/migrations/0135_schedule_sort_fallback_to_id.py

@@ -0,0 +1,18 @@
+# Generated by Django 2.2.16 on 2021-03-29 15:30
+
+from django.db import migrations
+import django.db.models.expressions
+
+
+class Migration(migrations.Migration):
+
+    dependencies = [
+        ('main', '0134_unifiedjob_ansible_version'),
+    ]
+
+    operations = [
+        migrations.AlterModelOptions(
+            name='schedule',
+            options={'ordering': [django.db.models.expressions.OrderBy(django.db.models.expressions.F('next_run'), descending=True, nulls_last=True), 'id']},
+        ),
+    ]

awx/main/models/schedules.py

@@ -63,7 +63,7 @@ class ScheduleManager(ScheduleFilterMethods, models.Manager):
 class Schedule(PrimordialModel, LaunchTimeConfig):
     class Meta:
         app_label = 'main'
-        ordering = ['-next_run']
+        ordering = [models.F('next_run').desc(nulls_last=True), 'id']
         unique_together = ('unified_job_template', 'name')
 
     objects = ScheduleManager()