Skip to main content

Associate a policy with an action

Linking a policy to an approval action​

Next, let's link the newly created policy to the approval action so that when a policy is evaluated against an approval action, each condition is checked.

To do this, you'll need to send a POST request to associate the policy with the action.

In the request, you'll need to send through:

  • the approval action ID as a path parameter (e.g.,98765).
  • the policy ID in the request body (e.g., 12345).

Example request​

POST - Associate policy with action
curl -L 'https://api.nullplatform.com/approval/action/98765/policy' \
-H 'Content-Type: application/json' \
-H 'Accept: application/json' \
-d '{
  "policy_id": 12345
}'

A successful response (200 OK) confirms the association.

How policy evaluation works​

How policy evaluation works Once you've linked your policy to an approval action, nullplatform evaluates it against relevant entity data. Each condition in the policy is checked against real-time metadata to determine whether the approval should proceed.

Below is an example of how this evaluation process works in practice.

Policy evaluation example
{
    "id": 12345,
    "name": "Enforce coverage, security, and scaling for deployments",
    "conditions": {
        "build.metadata.coverage": { "$gte": 80 },
        "build.metadata.security.issues": { "$lte": 4 },
        "scope.capabilities.auto_scaling": true,
        "$or": [
            { "scope.dimensions.environment": "dev" },
            { "scope.dimensions.environment": "qa" }
        ]
    },
    "evaluations": [
        {
            "criteria": {
                "build.metadata.coverage": { "$gte": 80 }
            },
            "result": "met"
        },
        {
            "criteria": {
                "build.metadata.security.issues": { "$lte": 4 }
            },
            "result": "not_met"
        },
        {
            "criteria": {
                "scope.capabilities.auto_scaling": true
            },
            "result": "met"
        },
        {
            "criteria": {
                "$or": [
                    { "scope.dimensions.environment": "dev" },
                    { "scope.dimensions.environment": "qa" }
                ]
            },
            "result": "met"
        }
    ],
    "passed": false
}

What's evaluated here:

  • Coverage (≥ 80%) → "met" → The condition was satisfied.
  • Security issues (≤ 4) → "not_met" → The condition was not met (perhaps there were more than 4 security issues).
  • Auto-scaling enabled → "met" → The resource supports auto-scaling.
  • Environment (dev or qa) → "met" → The deployment was in "dev" or "qa".

Since at least one condition was not met, the overall policy evaluation failed ("passed": false).