Last updated:
0 purchases
packme 0.8.0
packme
overview
packme is a python3 application for generating and running packer templates.
It works by generating manifest.json files out of manifest.yml files which contain the packer settings for the image to build. The manifest.yml files will be searched in a directory which contains as subdirectories the images to be built using packer. That directory has to be defined from the CLI (-t/--templates-dir option) and has the following structure:
my_templates_directory/
image1/
manifest.yml
...
imageN/
manifest.yml
Each manifest.yml has the following structure:
---
parameters:
param1: value1
...
paramN: valueN
packer:
description: packer description string
variables: packer variables dict
builders: packer builders list
provisioners: packer provisioners list
post-processors: packer post processors list
The packer section is strictly the YAML representation of a packer json file with the standards description, variables, builders, provisioners and post-processors packer fields. Those fields can contain Jinja2 templates which can be of two flavors: parameters or environment. parameters will be defined in the parameters section of the YAML file while environment will be either environment variables or variables passed through the packme CLI.
In order to run packme, an input configuration file has to be provided. This file is a YAML file which declares the list of the images for which manifest.json file will be built and run after with packer. It has the following structure:
---
templates:
image1:
packages: [package1, ..., packageN]
...
imageN:
extends: image1
where packages defines non standard packages that will added as extra provisioners when generating the manifest.json file for the corresponding image.
An image can extend another one by starting its building from another image. This is the aim of extends keyword which allows to declare that dependency.
Documentation
packme documentation
Prerequesites
packer
QEMU
Installation
pip install packme
For personal and professional use. You cannot resell or redistribute these repositories in their original state.
There are no reviews.