Example

The directory example/ is an working project that sends requests to public REST API endpoints, please do not abuse these.

Structure

The project must have the follwoing structure:

.
└── features
    ├── environment.py
    ├── test.feature
    ├── requests
    │   └── ...
    └── steps
        └── steps.py

In this example there are two requirements*.txt files. The reason is that requirements.txt will be copied and installed in the container image if grizzly-cli is used. The container image should not contain grizzly-cli and should be installed where scenarios are started from.

After installing grizzly-cli the easiest way to get a correct project structure is to use the builtin init subcommand:

grizzly-cli init my-grizzly-project
cd my-grizzly-project/

Environment

features/environment.py should contain:

from grizzly.environment import *

This file can contain overloading of behave hooks to trigger events that should happen during different stages of running a feature file.

from grizzly.environment import before_feature as grizzly_before_feature, after_feature as grizzly_after_feature, before_scenario, after_scenario, before_step

def before_feature(context: Context, *args: Tuple[Any, ...], **kwargs: Dict[str, Any]) -> None:
    # custom code that should run before feature file is started, e.g. notify something that a test
    # is started

    grizzly_before_feature(context, *args, **kwargs)


def after_feature(context: Context, feature: Feature, *args: Tuple[Any, ...], **kwargs: Dict[str, Any]) -> None:
    grizzly_after_feature(context, feature, *args, **kwargs)

    # custom code that should run before feature file is started, e.g. notify something that a test
    # is finished

Steps

features/steps/steps.py should contain:

from grizzly.steps import *

This is where custom step implementation can be added, then should look something like:

from behave.runner import Context
from behave import then  # pylint: disable=no-name-in-module

from grizzly.steps import *
from grizzly.context import GrizzlyContext


@then(u'this custom step should be executed')
def step_custom_the_custom_step(context: Context) -> None:
    grizzly = cast(GrizzlyContext, context.grizzly)

    # custom step implementation

Request templates

features/requests can contain jinja2 templates used in requests. E.g., if the feature file contains the following step:

Then send request "payload.j2.json"

Then features/requests/payload.j2.json needs to exist.

Get

First do a sparse checkout of the example/ directory in the repository.

If you have git older than 2.25.0, follow these instructions on stackoverflow.com.

BashPowerShell

mkdir grizzly-example
cd grizzly-example
git init
git remote add -f origin https://github.com/Biometria-se/grizzly.git
git sparse-checkout init
git sparse-checkout set example/
git pull origin main
rm -rf .git/
cd example/

mkdir grizzly-example
cd .\grizzly-example\
git init
git remote add -f origin https://github.com/Biometria-se/grizzly.git
git sparse-checkout init
git sparse-checkout set example/
git pull origin main
rm -Recurse -Force .\.git\
cd .\example\

Create an python virtual environment and install dependencies:

BashPowerShell

python3 -m venv .env
source .env/bin/activate
pip3 install -r requirements.txt
pip3 install grizzly-loadtester-cli

python3 -m venv .env
.\.env\Scripts\activate
pip3 install -r .\requirements.txt
pip3 install grizzly-loadtester-cli

If you do not already have an working "IBM MQ" client setup and run grizzly-cli in local mode you will not be able to use MessageQueueUser. See grizzly-cli/static/Containerfile on how to get these. When that is done you need to install the extra dependencies:

pip3 install grizzly-loadtester[mq]

Run

grizzly has some runtime features which is easiliest handled by using the grizzly-cli. It provides a simple command line interface wrapping the behave command, for providing initial variable values, configuration etc.

To run the example, in local mode:

BashPowerShell

grizzly-cli local run -e environments/example.yaml features/example.feature

grizzly-cli local run -e .\environments\example.yaml .\features\example.feature

And in distributed mode (requires docker and docker-compose in PATH):

BashPowerShell

grizzly-cli dist run -e environments/example.yaml features/example.feature

grizzly-cli dist run -e .\environments\example.yaml .\features\example.feature

Develop

If you have Visual Studio Code installed, you can also install the grizzly extension to make your life easier when developing scenarios!

pip3 install grizzly-loadtester-ls
code --install-extension biometria-se.grizzly-loadtester-vscode