Project guide

Creating a project

Before getting started, ensure that you meet the prerequisites for using SQLMesh.

To create a project from the command line, follow these steps:

Create a directory for your project:
```
mkdir my-project
```
Change directories into your new project:
```
cd my-project
```
From here, you can create your project structure from scratch, or SQLMesh can scaffold one for you. For the purposes of this guide, we'll show you how to scaffold your project so that you can get up and running quickly.
To scaffold a project, it is recommended that you use a python virtual environment by running the following commands:
```
python -m venv .env
```
```
source .env/bin/activate
```
```
pip install sqlmesh
```
Note: When using a python virtual environment, you must ensure that it is activated first. You should see (.env) in your command line; if you don't, run source .env/bin/activate from your project directory to activate your environment.
Once you have activated your environment, run the following command and SQLMesh will build out your project:
```
sqlmesh init
```
The following directories and files will be created that you can use to organize your SQLMesh project:
- config.py (database configuration file)
- ./models (SQL and Python models)
- ./audits (shared audits)
- ./tests (unit tests)
- ./macros

Editing an existing project

To edit an existing project, open the project file you wish to edit in your preferred editor.

If using CLI or Notebook, you can open a file in your project for editing by using the sqlmesh command with the -p variable, and pointing to your project's path as follows:

sqlmesh -p <your-project-path>

For more details, refer to CLI and Notebook.

Importing a project

dbt

To import a dbt project, use the sqlmesh init command with the dbt flag as follows:

sqlmesh init -t dbt