Get Started with GoodData for VS Code

This article will get you started with the VS Code extension and the CLI tool.

Prerequisites

Before getting started, ensure that you have the following:

VS Code 😉
Node.js version 16.17 or newer installed
GoodData workspace.

Use Trial

We recommend using the GoodData Trial. This makes it easy to follow this guide without risking tampering with your production environment.

Install VS Code Extension

Install the GoodData Visual Studio Code Extension (VS Code Extension). It provides tools for developer effectiveness, such as IntelliSense, syntax highlighting, and validation.

Steps:

Open the extensions tab in your VS Code.
In the search bar, look for “GoodData”.
Select the GoodData extension and install it.

Install CLI

To install the GoodData Command-line Interface (CLI), run:

npm i -g @gooddata/code-cli

Now that you have a new CLI command available, try running:

gd -v

This should return something along the lines of 0.5.0.

CLI shortcut

The CLI command is gd. Some terminal frameworks, e.g. OMZ, have preset the gd alias as a git diff.

If you experience problems with the alias, try to run unalias gd and consider adding this line to your .zshrc (or another) profile file.

How to Use

Initialize a local GoodData environment

When working with GoodData VS Code extension, you can work exclusively in VS Code, if you utilize the embedded terminal.

Steps:

In VS Code open a folder, that you will be using to build and store your analytics.
Run the following command to initialize your local environment:
```
gd init
```
Follow the instructions to create a gooddata.yaml configuration file.

If you are unsure about some of the steps, see the gooddata.yaml structure.
The result should look like this:
```
profiles:
dev:
    host: https://mamma_mia.trial.cloud.gooddata.com/
    token: $GOODDATA_API_TOKEN
    workspace_id: gdc_demo_730b50f8-8a93-4723-a872-64dca815a616
    data_source_id: gdc_demo_c2066bb5-bb14-419a-8172-239a380ffb49
source_dir: analytics
default_profile: dev
```
At the end of the procedure, you will be asked whether you want to clone. If you do so, you can skip then next step.

Run the following command to clone GoodData workspace to your local environment:

gd clone

You should now have an analytics folder that contains your analytical objects.

The resulting folder structure should look like this:

 root_folder
         └── analytics
         │       ├── datasets
         │       │       ├── dataset1.yaml
         │       │       ├── dataset2.yaml
         │       │       └── ...
         │       ├── metrics
         │       │       ├── metric1.yaml
         │       │       ├── metric2.yaml
         │       │       └── ...
         │       └── ...
         ├── gooddata.yaml
         ├── .env
         └── .gitignore

Create and Preview Metrics

Once you set up your environment, you can create new analytical objects using code.

Steps:

Create a new .yaml file in your metrics folder to represent your new metric. For example, pending_revenue.yaml.

Define a metric:

type: metric
id: pending_revenue

title: Pending revenue
description: The sum of orders in carts

maql: select SUM({fact/order_unit_price}) WHERE {label/order_status}= "In Cart"

Notice the Run button.

Click Run to run your metric.
You will be prompted whether you want to deploy your local environment.
Click Deploy only if you are sure you will not break anything.
Running metrics deploys local environment
When you run the metric, it automatically deploys your local environment to the workspace. This is because we assume your folder to be the source of truth. To avoid breaking your production, create a new development workspace.
Result:
Save the file.