search

Hydrospheredata / mist-cli

Mist command line interface
Apache License 2.0
2 stars 4 forks source link

readme

=========== Mist CLI

.. image:: https://codecov.io/gh/Hydrospheredata/mist-cli/branch/master/graph/badge.svg :target: https://codecov.io/github/Hydrospheredata/mist-cli :alt: codecov.io .. image:: https://travis-ci.org/Hydrospheredata/mist-cli.svg?branch=master :target: https://travis-ci.org/Hydrospheredata/mist-cli :alt: travis-ci.org

CLI interface for mist server for creating and updating mist function and context configuration.

Installation

.. code-block:: bash

pip install mist-cli

Autocompletion

Instantly run the script every time when bash starts that generates content of mist-cli-complete.sh.

.. code-block:: bash

echo 'eval "$(_MIST_CLI_COMPLETE=source mist-cli)"' >> ~/.bash_profile

Or you can just add mist-cli-complete.sh somewhere in the system and execute it when it needed.

Also, you can generate that script by yourself:

.. code-block:: bash

_MIST_CLI_COMPLETE=source mist-cli > mist-cli-complete.sh

Usage mist-cli apply

Apply method accepts -f or --file parameter that should contain file or folder with *.conf files that represent your mist configuration (Artifact, Function or Context).

The content of the file is a simple model that describes deployment of a particular entry of your config.

All conf files should follow some format (e.g 00artifact.conf <example/my-awesome-job/00artifact.conf>_) where field version is only supported in Artifact model type. By default, you can name your config files as you want and all configs will be processed without any order. So you can define this order with 2 numbers followed by a name. So, for example, you name your stage test-stage and want it run with priority 10 you should name the file like 10test-stage.conf.

For easy development process, you can skip validation of your configuration, for example, by default, Function models will be validated against context and artifact existence. So if you want to update Function with context = foo and path = test-job_0_0_1.py make sure that artifact with that key and context with that name exists in Mist. You can easily skip this kind of validation with --validate flag.