Packaging OpenG2P Docker

This page contains steps to be performed for packaging different components and addons, of OpenG2P and similar, into a docker image.
Instructions
Clone the https://github.com/OpenG2P/openg2p-packaging and go to packaging directory
Create text file, example my-package.txt . This signifies a package. This file should include all openg2p modules (repositories) to be packaged into a new docker. Each line describes one repository to include, and the structure of each line looks like this.
repo_name = git://<github branch name>//<github url to pull from>
repo2_name = file://<path of the package in local system>
Any underscore in the repository name will be converted to hyphen during installation. For example:
repo_name = git://<github branch name>//<github url to pull from>
This is internally converted to repo-name.
The above configuration can be made via environment variables also.
Any variable with the prefix G2P_PACKAGE_my_package_ will be considered as repository to install i.e., G2P_PACKAGE_<package_name>_<repo_name>. For example;
G2P_PACKAGE_my_package_repo3_name=git://<github branch name>//<github url to pull from>
These env variables can be added in .env file (in the same folder). The .env file will automatically be considered.
If same package is available in my-package.txt, .env and environment variable, then this will be the preference order in which they are considered (highest to lowest).
.env file
Environment variable.
my-package.txt
Use the .env to overload packages from my-package.txt
Run the following to download all packages:
./package.sh my-package.txt
After downloading packages, run the following to build docker image:
docker build . -t <docker image name>
Then push the image.
docker push <docker image name>
Notes:
The above uses bitnami's odoo image as base.
This script also pulls in any OCA dependencies configured, in oca_dependencies.txt inside each package. Use this env variable to change the version of OCA dependencies to be pulled, OCA_DEPENDENCY_VERSION (defaults to 15.0).
This also installs any python requirements configured in requirements.txt inside each package.
Reference packages can be found in packages directory inside packaging directory.
Access to Deployed Setup
Last modified 2mo ago