Ecosystem¶

behave-pool is designed to integrate with other packages in the Behave ecosystem. This page covers the optional integrations available.

behave-priority¶

behave-priority provides priority-based scenario ordering. Scenarios tagged with @priority.high, @priority.medium, or @priority.low are executed in priority order.

Installation¶

pip install "behave-pool[ecosystem]"

Or separately:

pip install behave-priority

Usage with behave-pool¶

behave-priority works as a Behave runner that wraps ParallelRunner. Register both in behave.ini:

[behave]
jobs = 4

[behave.runners]
parallel = behave_pool:ParallelRunner
priority = behave_priority:PriorityRunner

Run with priority ordering and parallel execution:

behave --runner=priority --parallel 4 features/

Example feature with priorities¶

Feature: User management

  @priority.high
  Scenario: Admin can create users
    Given I am an admin
    When I create a new user
    Then the user should exist

  @priority.medium
  Scenario: User can update profile
    Given I am a registered user
    When I update my profile
    Then my profile should be updated

  @priority.low
  Scenario: User can delete account
    Given I am a registered user
    When I delete my account
    Then my account should be removed

behave-modern-json-report¶

behave-modern-json-report provides a modern JSON report format for Behave test results, with richer output than Behave's built-in JSON formatter.

Installation¶

pip install "behave-pool[ecosystem]"

Or separately:

pip install behave-modern-json-report

Usage with behave-pool¶

Behave formatters (--format/--outfile) are not wired into the parallel workers, so use the built-in unified report instead — it implements the same ExecutionReport schema that behave-modern-json-report produces:

behave-pool --parallel 4 --parallel-report report.json features/

[behave]
jobs = 4
runner = behave_pool:ParallelRunner

[behave.userdata]
pool.report = report.json

Any tool that consumes the modern JSON ExecutionReport format can read this file directly — features, scenarios, steps, errors, durations and tag statistics included.

Output format¶

The unified report includes:

Feature, scenario, and step-level results
Durations for each level
Error messages and tracebacks
Tags and metadata

{
  "features": [
    {
      "name": "Login functionality",
      "filename": "features/login.feature",
      "status": "passed",
      "duration": 1.23,
      "scenarios": [
        {
          "name": "User logs in with valid credentials",
          "status": "passed",
          "duration": 0.45,
          "steps": [...]
        }
      ]
    }
  ],
  "summary": {
    "total_features": 1,
    "total_scenarios": 1,
    "passed": 1,
    "failed": 0,
    "duration": 1.23
  }
}

Using all ecosystem packages together¶

pip install "behave-pool[ecosystem]"

[behave]
jobs = 4
runner = behave_pool:ParallelRunner

[behave.userdata]
pool.balance = lpt
pool.report = report.json

behave-pool features/

This gives you:

Parallel execution with 4 worker processes
LPT load balancing for optimal wall-clock time
Unified JSON report (modern-json compatible schema) for CI integration and reporting

Compatibility¶

Package	Status	Notes
`behave-priority`	Compatible	Works as a wrapping runner
`behave-modern-json-report`	Schema-compatible	Formatters don't run inside workers; `--parallel-report` emits the same `ExecutionReport` schema
`behave`	Required (`>=1.3.0`)	Base framework
`pytest`	Compatible	For running `behave-pool`'s own test suite

Custom integrations¶

behave-pool exposes a stable public API for custom integrations:

from behave_pool import ParallelRunner
from behave_pool.config import ConfigSnapshot, snapshot_config
from behave_pool.work_unit import WorkUnit
from behave_pool.result import WorkerResult
from behave_pool.timing import TimingStore

See the API reference for full documentation of these classes.