depends_on

The depends_on meta-argument allows you to explicitly define execution dependencies between test scenarios, ensuring that prerequisites are executed in the correct sequence using a topological Directed Acyclic Graph (DAG).

---

Gherkin Declaration (@depends_on)

Dependencies are declared as tags directly above a Scenario:, supporting canonical array brackets (matching Terraform lists) as well as comma-separated targets:

@checkout @e2e
@depends_on(["auth/login.feature.@login_success", "inventory/cart.feature.@add_item"])
Scenario: User completes checkout successfully
  Given shopping cart contains selected products
  When user confirms shipping and billing details
  Then order confirmation receipt is issued

You can also write @depends_on("target1", "target2") or use multiple separate @depends_on(...) tags.

---

Topological DAG Ordering

During testform plan and testform apply:

1. Testform builds a Directed Acyclic Graph (DAG) of all planned test resources.
2. Applies Kahn’s algorithm to resolve the exact linear execution sequence.
3. Prerequisite scenarios (auth/login) are verified before downstream scenarios (checkout/order).

---

Circular Dependency Prevention

If a circular dependency cycle is declared ($A \rightarrow B \rightarrow A$), Testform aborts during static analysis before executing any tests:

[ERROR] Circular dependency detected in DAG:
  auth/login.feature.@login_success -> checkout/order.feature.@checkout -> auth/login.feature.@login_success

---

Visualizing the Graph: testform graph

Output the complete dependency DAG in standard Graphviz DOT format:

testform graph

# Render to image using Graphviz:
testform graph | dot -Tpng -o dependencies.png