Basic setup: Quickstart with Angular & Vue Templates

Prerequisites

• Basic understanding of HTML, JavaScript, and Angular or Vue for frontend development.

• Some experience with Python for backend operations.

• Code Studio Template is already configured on your Dataiku instance.

This tutorial teaches how to rapidly initiate a web application using pre-configured templates. These templates streamline your workflow by auto-configuring a Node server to host your Code Studio browser path while also routing API requests to the exposed Flask backend.

These templates are also designed for smooth deployment as a standard Dataiku web application.

Quick Project Setup with Cookiecutter

Note

• Cookiecutter serves as a command-line tool for swift project generation from templates. The Cookiecutter Python code environment will be employed in this scenario, as defined in the Code Studio template.

• All web application projects should be initiated inside the project-lib-versioned/webapps directory. Doing so enables seamless synchronization of your source code and build files with the Dataiku project library, thereby simplifying the deployment process.

Activating the Cookiecutter Environment

To activate your Cookiecutter environment in the Visual Studio Code (VS Code) UI panel, run the command below:

source /opt/dataiku/python-code-envs/cookiecutter/bin/activate

Note

• The default path for code environments specified in the Code Studio template is /opt/dataiku/python-code-envs/. If you’ve customized this location in the template settings, update the path in the command accordingly.

Generating a New Web Application

In this section, we will guide you through the process of creating a new web application using Cookiecutter. You have the option to choose between two templates, VueJS or Angular, for your web application project.

• Navigate to the following directory:

  mkdir ~/workspace/project-lib-versioned/webapps && cd ~/workspace/project-lib-versioned/webapps
  

• Execute the following command to instantiate a new project template with Cookiecutter:

  cookiecutter gh:dataiku/solutions-contrib
  

  Note

  Cookiecutter is a command-line utility that expedites the creation of project structures using templates. In this context, we utilize it with the gh:dataiku/solutions-contrib repository, a public GitHub repository storing templates for Vue and Angular web applications. These templates are specifically designed to streamline the setup of web applications in Dataiku’s Code Studio and facilitate smooth deployment to standard Dataiku web apps, ensuring a straightforward and consistent development process for developers.

• Cookiecutter will then prompt you to select a template:

  [1/1] Select template
    1 - Vue (./bs-templates/vue)
    2 - Angular (./bs-templates/angular)
    Choose from [1/2] (1):
  

  • Select the desired template by entering the corresponding number (1 for Vue and 2 for Angular).

• After template selection, you’ll be prompted to enter specific project parameters:

  [1/6] Choose your (Angular / Vue) project name (Angular Project): tutorial
  [2/6] version (0.0.1):
  [3/6] Choose your client serve port (default 4200) (4200):
  [4/6] Choose your flask backend port (default 5000) (5000):
  [5/6] dss_instance (default):
  [6/6] dku_project ():
  

  • “Project Name”: your Angular or Vue project name. Though the example uses tutorial, you can choose any name.

  • “Version”: the project’s version.

  • “Client serve port”: port on which your frontend (e.g., Angular) will run. The default is 4200.

  • “Flask backend port”: port your Flask backend will operate on. The default is 5000.

  • “DSS Instance and DKU Project”: these fields are generally optional while operating within Dataiku Code Studio. You can safely leave them blank unless you’re configuring an external development environment that necessitates their input.

• After entering these details, a new project folder will be generated in the project-lib-versioned/webapps directory.

Following these steps, you’ll successfully set up a new web application, ready for further development and deployment.

Launching the Frontend and Backend

Once you’ve generated your web application using Cookiecutter, you can proceed to launch the frontend and backend. The process is the same for both Angular and Vue templates.

Launching the Frontend

• Navigate to your project’s directory by replacing __PROJECT_NAME__ with your project folder’s name:

  cd ~/workspace/project-lib-versioned/webapps/__PROJECT_NAME__
  

• Install the required dependencies:

  yarn install
  

  The Code Studio template comes pre-configured with npm, yarn, and pnpm. You can use any of these package managers to install dependencies. For instance, you can replace yarn install with npm install or pnpm install based on your preference.

• Run the frontend server

  ng serve --host 127.0.0.1
  

• You can now preview the web application in the Dev panel of the Code Studio.

Launching the Backend

• Activate your backend code environment

  source /opt/dataiku/python-code-envs/infra37/bin/activate
  

  Note

  If you’ve imported the Code Studio template, an infra37 code environment with the essential requirements for launching the included Flask server will be available. This environment necessitates python >= 3.6 or higher and incorporates the following packages:

  Flask>=0.9
  git+https://github.com/dataiku/solutions-contrib.git@main#egg=webaiku&subdirectory=bs-infra
  python-dotenv>=0.19.0
  dataiku-api-client
  

• Navigate back to your project directory (again, replace __PROJECT_NAME__ with your specific folder name):

  cd ~/workspace/project-lib-versioned/webapps/__PROJECT_NAME__
  

• Execute the following command to launch the backend:

  python -m backend.wsgi
  

By following these steps, you’ll successfully initiate both the frontend and backend of your web application within Dataiku’s Code Studio environment.

Remember that you can choose between Angular and Vue templates for your project, and the instructions for launching the frontend and backend are the same regardless of your choice.