Odoo Development Tips

Development Setup

Assumptions:

• You are a Python developer
• You have at least basic Odoo development knowledge
• You are using Ubuntu 18.04 for development (or similar)

Pre-requisites:

• Python setuptools: sudo apt-get install python3-setuptools
• Python wheel: pip3 install wheel

1. Get Odoo one of 2 ways (choose just one, unless you know what you are doing):

   • Option 1: Clone the Odoo source code to a directory of your choice. You can use the following script to do this. Be sure to read the source code if you have security concerns. Just run the this command in a directory where you want to download the source code (be sure to choose the correct version of Odoo for you when prompted):

   source <(curl -s https://gitlab.com/sylnsr/linux-stuff/-/raw/master/dev/scripts/odoo/thin-clone-odoo.sh)

   • Option 2: Install Odoo 13 using the .deb installer.

2. Setup your Python 3.6 virtual environment by running the following in the root directory of the project:
   python3 -m venv ./venv

3. Activate the Python 3 virtual environment:
   source ./venv/bin/activate

4. Install the Odoo requirements into the virtual environment:
   pip3 install -r <path-to-odoo-source-code>/requirements.txt

5. Configure a very basic odoo.conf file and copy it to /etc/odoo/odoo.conf. Setup the dir paths so that they match what you have in your Odoo config file. For example, assuming you:

   • downloaded the source to ~/git/odoo/13/
   • have your local module project cloned to ~/git/my-repo/my-project/
   • are developing you modules in your project in a sub-dir named addons
   • created the dir path ~/tmp/var_lib_odoo for Odoo data files

   …here is a sample with default credentials for the demo database:

   [options]
   data_dir = ~/tmp/var_lib_odoo
   addons_path = ~/git/my-repo/my-project/addons,~/git/odoo/13/addons
   admin_passwd = odoo
       
   workers = 1
   max_cron_threads = 0
   list_db = True
   db_name = demo
   db_host = localhost
   db_user = odoo
   db_password = odoo
   

   Note: setting workers to 1 (one) ensures that your debugger will be hooked on the only thread Odoo is running.

6. Ensure your local Postgres database is running. If you want to use a Docker container for Postgres, you could just put this code in a bash script and run it:

   #!/usr/bin/env bash
   cd "$(dirname "$0")"
   docker run --rm -it --name dev13_pg12 \
   -e POSTGRES_PASSWORD=odoo \
   -e POSTGRES_USER=odoo \
   -e POSTGRES_DB=postgres \
   -p 5432:5432 \
   postgres:12
   

7. Start debugging Odoo in your IDE. For example, if you have Odoo source installed at ~/git/odoo/13, your debugger would be setup to start debugging this Python script: ~/git/odoo/13/odoo-bin. Note that your debugger must also be setup to use the Python environment in the virtual environment.

Common dev setup issues

“greenlet.error”

If you get the following error when debugging: greenlet.error: cannot switch to a different thread and the long-polling port is not working, then look at this: https://github.com/miguelgrinberg/Flask-SocketIO/issues/65#issuecomment-294375994

Beware though, enabling the “Gevent compatible” option may cause debugging to no longer work. See https://youtrack.jetbrains.com/issue/PY-11910

“port already in use”

If you install Odoo from the installer AND THEN you try to run it from source, you will probably get an error indicating that port 8069 is already in use. This is because the installer starts Odoo running as a service so the port will already be occupied. Stop the already running Odoo process.

Recommended Tools

• IDE: IntelliJ IDEA or Pycharm
• Python Shell: IPython
• Odoo-RPC-Client library for Python:

The IPython shell is especially useful with odoo-rpc-client for testing your class methods that are accessible over Odoo’s built-in JSON-RPC API. In other words, using these tools together gives you a productive live sandbox environment for testing your modules over the RPC-API without having to constantly start and stop any experimental Python scripts.

To quickly get started with IPython and Odoo RPC Client, just follow these steps:

1. Activate the virtual environment:
   source ~/venv/bin/activate

2. Start an IPython shell:
   ipython

3. Import Odoo RPC client:
   from odoo_rpc_client import Client

4. Setup the connection for Odoo RPC client:
   odoo = Client(host='127.0.0.1', dbname='demo', user='admin', pwd='admin', protocol='json-rpc')

Now you’ll be in a live Python environment for interacting with your Odoo instance from IPython.

For example, retrieve the first company email from the res.partner model:

odoo['res.partner'].browse(1).email

… which results in:

Out[1]: 'info@yourcompany.example.com'

Odoo with Vue

If, for whatever reason, you dont want to develop your Odoo UI solution using Odoo’s native QWeb framework, but want to use the simpler and much more ubiquitous Vue framework, then you should check out the Odoo In Vue repo.

It provides a working prototype of a Vue app that connects to Odoo in the back-end. Additionally, because creating a whole lot of components from scratch is a whole lot of unnecessary work, the solution uses the Quasar framework to make dramatically increase your development time. Quasar is also beneficial because it makes porting your Vue app to Electron or Cordova etc, extremely simple.

• odoo