Quickstart & Installation

Get started with det in under 60 seconds.

📦 Requirements

Python: >= 3.14 (built for modern Python with prerelease support)
OS: Linux (Ubuntu, Debian, Fedora, Arch), macOS (Intel & Apple Silicon), Windows (10/11, Server)
Rust/Cargo: Not required (wheels are pre-compiled with Apache DataFusion binaries)

Install det-core directly from PyPI:

pip install det-core

For development or full format support (including Excel .xlsx and Delta Lake):

pip install "det-core[dev]"

Verify your installation:

det --version

Create and navigate to your project directory:
```
mkdir my_warehouse && cd my_warehouse
```
Install the complete NovaMart Global sample warehouse:
```
det sample-data
```
This unpacks:
- 14 Catalog connections across CSV, Parquet, JSON, NDJSON, XML, Excel, and Delta Lake
- 10 Production multi-stage ETL pipelines
- Source dataset files under source_data/
Launch the interactive Web IDE:
```
det gui
```
The local IDE will automatically launch in your default web browser at http://127.0.0.1:8000.

Run a single pipeline by name or folder/name:

det run Sample_Warehouse_Pipelines/fct_orders

Run with verbose execution output and step breakdown:

det run Sample_Warehouse_Pipelines/mart_customer_rfm_loyalty --verbose

Verify SQL syntax, column references, and dependency topologies before running:

det validate Sample_Warehouse_Pipelines/stg_customers_cleaned

Generate standalone, interactive HTML documentation complete with Mermaid DAG diagrams and data contracts:

det doc Sample_Warehouse_Pipelines/mart_customer_rfm_loyalty --format html

Maintained by Saurabh Rai (saurabhr2@gmail.com) and det-Labs • Released under the MIT License