Testing rules

Rules are easier to test than code: every rule is a pure function of its input. The Rule Engine ships three layers of test tooling — syntactic validation, single-rule dry runs, and full rule-set executions with monitoring.

Validate syntax

Run validation before saving any rule set. Three flavors:

Validate parsed rule objects. Validate raw YAML content. Validate raw JSON content.

curl -X POST https://appengine.appmint.io/rules/validate-yaml \
  -H "x-org-id: $ORG" -H "x-customer: $USER" \
  -H "Content-Type: application/json" \
  -d '{"yamlContent": "version: 2.0\nrules:\n  - id: TEST\n    when: order.total > 100\n    then:\n      add_message: hi"}'

The response lists structural errors (missing id, malformed condition, unknown operator), unknown helper references, and warnings (e.g. a rule with no then block).

Dry-run a single rule

Pass the rule and a testData payload. The engine evaluates it and returns the resolved branches:

curl -X POST https://appengine.appmint.io/rules/test-rule \
  -H "x-org-id: $ORG" -H "x-customer: $USER" \
  -H "Content-Type: application/json" \
  -d '{
    "rule": {
      "id": "BULK",
      "when": { ">=": ["order.lineCount", 5] },
      "then": { "apply_discount": { "type": "percent", "value": 10 } },
      "else": { "apply_discount": { "type": "percent", "value": 0 } }
    },
    "testData": { "order": { "lineCount": 7 } }
  }'

Response:

{
  "passed": true,
  "branchTaken": "then",
  "actions": { "apply_discount": { "type": "percent", "value": 10 } },
  "evaluationTimeMs": 1.4,
  "vars": {}
}

Run a full rule set with monitoring

Pass options.logExecution: true and options.trackPerformance: true to capture per-rule timings and write the run to execution history.

curl -X POST https://appengine.appmint.io/rule-engine/execute \
  -H "x-org-id: $ORG" -H "x-customer: $USER" \
  -d '{
    "rules": [ /* array of rules */ ],
    "data": { /* input */ },
    "options": { "logExecution": true, "trackPerformance": true, "enableDebug": true, "timeout": 5000 }
  }'

enableDebug: true returns intermediate variable bindings and which branch each rule took.

Sample data fixtures

The repo ships canonical examples to copy from. Validate your data against the shape these expect:

Example	Path
Bulk discount	src/rule-engine/examples/bulk-discount.yaml
Fraud gate	src/rule-engine/examples/fraud-gate.yaml
Order pipeline	src/rule-engine/examples/order-pipeline.yaml
Shipping path	src/rule-engine/examples/shipping-path.yaml
Order all-fields fixture	src/rule-engine/examples/order-all-fields.json
Invalid order fixture	src/rule-engine/examples/invalid-order.json

Read execution history

Query	Description
executionId	One specific run
ruleSetId	Filter by rule set
dateFrom, dateTo	Time window
success	Only failures (false) or successes (true)
event	Trigger event filter
limit, page	Paging

Each record carries the rule set, input data hash, the resolved actions, and per-rule timings.

Aggregate counts. Latency, throughput, slow rules.