chore: Suggest a fix for ordering access policies so assignments are …

…prioritized.

enterprise_access/apps/subsidy_access_policy/constants.py

@@ -30,8 +30,9 @@ class SegmentEvents:
 
 # Configure the priority of each policy type here.  When given multiple redeemable policies to select for redemption,
 # the policy resolution engine will select policies with the lowest priority number.
-CREDIT_POLICY_TYPE_PRIORITY = 1
-SUBSCRIPTION_POLICY_TYPE_PRIORITY = 2
+# BB - Some padding between the larger policy type grouping...
+CREDIT_POLICY_TYPE_PRIORITY = 10
+SUBSCRIPTION_POLICY_TYPE_PRIORITY = 20
 
 
 class PolicyTypes:

enterprise_access/apps/subsidy_access_policy/models.py

@@ -931,14 +931,15 @@ def __str__(self):
 class CreditPolicyMixin:
     """
     Mixin class for credit type policies.
+    BB - Something in the docs about how to use this? Potentially remove?
     """
 
     @property
     def priority(self):
         return CREDIT_POLICY_TYPE_PRIORITY
 
 
-class PerLearnerEnrollmentCreditAccessPolicy(CreditPolicyMixin, SubsidyAccessPolicy):
+class PerLearnerEnrollmentCreditAccessPolicy(SubsidyAccessPolicy):
     """
     Policy that limits the number of enrollments transactions for a learner in a subsidy.
 
@@ -1007,8 +1008,16 @@ def remaining_balance_per_user(self, lms_user_id):
         existing_transaction_count = len(self.transactions_for_learner(lms_user_id)['transactions'])
         return self.per_learner_enrollment_limit - existing_transaction_count
 
+    @property
+    def priority(self):
+        """
+        Ensure that Per learner enrollment policies are evaluated before spend cap but after assignment
+        based policy types.
+        """
+        return CREDIT_POLICY_TYPE_PRIORITY + 1
 
-class PerLearnerSpendCreditAccessPolicy(CreditPolicyMixin, SubsidyAccessPolicy):
+
+class PerLearnerSpendCreditAccessPolicy(SubsidyAccessPolicy):
     """
     Policy that limits the amount of learner spend for enrollment transactions.
 
@@ -1089,8 +1098,15 @@ def remaining_balance_per_user(self, lms_user_id=None):
         positive_spent_amount = spent_amount * -1
         return self.per_learner_spend_limit - positive_spent_amount
 
+    @property
+    def priority(self):
+        """
+        Ensure that Per leaner spend policies are evaluated last, after other similar policy types.
+        """
+        return CREDIT_POLICY_TYPE_PRIORITY + 2
+
 
-class AssignedLearnerCreditAccessPolicy(CreditPolicyMixin, SubsidyAccessPolicy):
+class AssignedLearnerCreditAccessPolicy(SubsidyAccessPolicy):
     """
     Policy based on LearnerContentAssignments, backed by a learner credit type of subsidy.
 
@@ -1427,6 +1443,7 @@ def allocate(self, learner_emails, content_key, content_price_cents):
             content_price_cents,
         )
 
+<<<<<<< Updated upstream
 
 class PolicyGroupAssociation(TimeStampedModel):
     """
@@ -1455,3 +1472,11 @@ class Meta:
         null=False,
         help_text='The uuid that uniquely identifies the associated group.',
     )
+=======
+    @property
+    def priority(self):
+        """
+        Ensure that assignment policies are evaluated before other similar policy types
+        """
+        return CREDIT_POLICY_TYPE_PRIORITY + 0
+>>>>>>> Stashed changes