Development: Getting Started
Be sure to use the Nightly version of the docs to ensure you're up to date with the latest changes.
After reading through the Code Contributions Guide and forking the repo you can start working. This project is developed with docker and as such you will be greatly aided by using docker for development. It's not necessary but it is helpful.
With VSCode Dev Containers
- Visual Studio Code
First ensure that docker is running. Then when you clone the repo and open with VS Code you should see a popup asking you to reopen the project inside a development container. Click yes and it will build the development container and run the setup required to run both the backend API and the frontend webserver. This also pre-configures pre-commit hooks to ensure that the code is up to date before committing.
Checkout the makefile for all of the available commands.
For slow terminal checkout the solution in this GitHub Issue
git config oh-my-zsh.hide-info 1
Without Dev Containers
Once the prerequisites are installed you can cd into the project base directory and run
make setup to install the python and node dependencies.
# Naviate To The Root Directory cd /path/to/project # Utilize the Makefile to Install Dependencies make setup
# Install Python Dependencies Set-Directory -Path "C:\path\to\project" poetry install # Install Node Dependencies Set-Directory frontend yarn install
Setting ENV Variables
Before you start the server you MUST copy the
frontend/template.env files to their respective locations with the name
frontend/.env respectively. The application will-not run without these files.
Starting The Server
Once that is complete you're ready to start the servers. You'll need two shells open, One for the server and one for the frontend.
# Terminal #1 make backend # Terminal #2 make frontend
# Terminal # 1 poetry run python mealie/db/init_db.py # Initialize the database poetry run python mealie/app.py # start application # Terminal # 2 Set-Directory frontend yarn run dev
Make File Reference
make help for reference. If you're on a system that doesn't support makefiles in most cases you can use the commands directly in your terminal by copy/pasting them from the Makefile.
docs 📄 Start Mkdocs Development Server code-gen 🤖 Run Code-Gen Scripts setup 🏗 Setup Development Instance setup-model 🤖 Get the latest NLP CRF++ Model clean-data ⚠️ Removes All Developer Data for a fresh server start clean-pyc 🧹 Remove Python file artifacts clean-test 🧹 Remove test and coverage artifacts backend-clean 🧹 Remove all build, test, coverage and Python artifacts backend-test 🧪 Run tests quickly with the default Python backend-format 🧺 Format, Check and Flake8 backend-all 🧪 Runs all the backend checks and tests backend-coverage ☂️ Check code coverage quickly with the default Python backend 🎬 Start Mealie Backend Development Server frontend 🎬 Start Mealie Frontend Development Server frontend-build 🏗 Build Frontend in frontend/dist frontend-generate 🏗 Generate Code for Frontend frontend-lint 🧺 Run yarn lint docker-dev 🐳 Build and Start Docker Development Stack (currently not functional, see #756, #1072) docker-prod 🐳 Build and Start Docker Production Stack
We use vue-i18n package for internationalization. Translations are stored in json format located in frontend/lang/messages.
Translations are stored in json format located in mealie/lang/messages.
Quick frontend localization with VS Code
i18n Ally for VScode is helpful for generating new strings to translate using Code Actions. It also has a nice feature, which shows translations in-place when editing code.
A few settings must be tweaked to make the most of its features. Some settings are stored on project level, but most of them have to be set manually in your workspace or user settings.\ We've found that the following settings work best:
"i18n-ally.enabledFrameworks": ["vue"], "i18n-ally.extract.autoDetect": true, "i18n-ally.dirStructure": "auto", "i18n-ally.extract.targetPickingStrategy": "global-previous", "i18n-ally.displayLanguage": "en-US", "i18n-ally.keystyle": "nested", "i18n-ally.sourceLanguage": "en-US",