Skip to content

noosenergy/noos-invoke

Repository files navigation

CircleCI

Noos Invoke

Software development kit for sharing workflows across CI/CD pipelines.

Such a project aims to enforce parity and reproducability between local development and CI/CD workflows in remote containers (e.g. executable versions, command line calls, environment variables...) - developped with inv[oke].

Installation

Install the package from the PyPi repository:

```bash
$ pip install noos-inv
```

To enable shell completion, execute the following command (e.g. zsh),

```bash
$ noosinv --print-completion-script=zsh
```

And copy/paste its stdout into your shell config.

```bash
# NOOSINV completion script

_complete_noosinv() {
    collection_arg=''
    if [[ "${words}" =~ "(-c|--collection) [^ ]+" ]]; then
        collection_arg=$MATCH
    fi
    reply=( $(noosinv ${=collection_arg} --complete -- ${words}) )
}

compctl -K _complete_noosinv + -f noosinv
```

Finally, still in your shell config, enable automatic sub shell loading:

```bash
# ENV variable sub shell loading with command "source .env"

set -a
```

Usage as a command line tool

The noos-inv package installs a CLI binary, for managing common CI/CD tasks.

From the terminal,

```bash
$ noosinv

Usage: noosinv [--core-opts] <subcommand> [--subcommand-opts] ...

Subcommands:

docker.build       Build Docker image locally.
docker.buildx      Build and push x-platform Docker image to a remote registry.
docker.configure   Create and configure buildx builder for multi-platform.
docker.login       Login to Docker remote registry (AWS ECR or Dockerhub).
docker.pull        Pull Docker image from a remote registry.
docker.push        Push Docker image to a remote registry.
git.config         Setup git credentials with a Github token.
helm.install       Provision local Helm client (Chart Museum Plugin).
helm.lint          Check compliance of Helm charts / values.
helm.login         Login to Helm remote registry (AWS ECR or Chart Museum).
helm.push          Push Helm chart to a remote registry (AWS ECR or Chart Museum).
helm.test          Test local deployment in Minikube.
local.dotenv       Create local dotenv file.
local.ports        Forward ports for defined Kubernetes pods.
python.clean       Clean project from temp files / dirs.
python.coverage    Run coverage test report.
python.format      Auto-format source code.
python.lint        Run python linters.
python.package     Build project wheel distribution.
python.release     Publish wheel distribution to PyPi.
python.test        Run pytest with optional grouped tests.
terraform.run      Run a plan in Terraform cloud.
terraform.update   Update variable in Terraform cloud.
```

Source your environnement variables first for a seamless experience. (use command local.dotenv to create it from the provided template)

```bash
$ source .env
```

Special note on K8S port-forwards

Add the NOOSINV_LOCAL_CONFIG OS variable to your shell config, as the path to a local configuration file:

```json
{
    "podForwards": {
        "pod_1": {
            "podNamespace": "default",
            "podPrefix": "service-1-",
            "podPort": 80,
            "localPort": 8000
        },
        "pod_2": {
            "podNamespace": "test",
            "podPrefix": "service-2-",
            "podPort": 8080,
            "localPort": 8000,
            "localAddress": "0.0.0.0"
        }
    }
}
```

To start port forwarding a specific K8S cluster pod:

```bash
$ noosinv local.ports -p pod_1
```

To kill all previous port forward processes:

```bash
$ noosinv local.ports -u
```

Or previously opened port forward:

```bash
$ noosinv local.ports -p pod_1 -u
```

Development

Make sure poetry has been installed and pre-configured,

This project is shipped with a Makefile, which is ready to do basic common tasks.

```bash
$ make

help                           Display this auto-generated help message
update                         Lock and install build dependencies
clean                          Clean project from temp files / dirs
format                         Run auto-formatting linters
install                        Install build dependencies from lock file
lint                           Run python linters
test                           Run pytest with all tests
package                        Build project wheel distribution
release                        Publish wheel distribution to PyPi
```