Overview

Petaly is an open-source ETL/ELT (Extract, Load, "Transform") tool, created by and for data professionals! Our mission is to simplify data movement across different platforms with a tool that truly understands the needs of the data community.

Key Features

• Multiple Data Sources: Support for various endpoints:

  • PostgreSQL
  • MySQL
  • BigQuery
  • Redshift
  • Google Cloud Storage (GCS Bucket)
  • S3 Bucket
  • Local CSV files

• Features:

  • Source to target schema evaluation and mapping
  • CSV file load with column-type recognition
  • Target table structure generation
  • Configurable type mapping between different databases
  • Full table unload/load in CSV format

• User-Friendly: No programming knowledge required

• YAML/JSON Configuration: Easy pipeline setup

• Cloud Ready: Full support for AWS and GCP

[EXPERIMENTAL]:

Petaly went agentic!
The AI Agent can create and run pipeline using natural language prompts.
If you're interested in exploring, check out the experimental branch: petaly-ai-agent

Feedback is welcome!

Quick Start

1. Installation
2. Configuration
3. Create Pipeline
4. Run Pipeline

Requirements

System Requirements

• Python 3.10 - 3.12
• Operating System:
  • Linux
  • MacOS

Note: Petaly may work on other operating systems and Python versions, but these haven't been tested yet.

Installation

Basic Installation

# Create and activate virtual environment
mkdir petaly
cd petaly
python3 -m venv .venv
source .venv/bin/activate

# Install Petaly
python3 -m pip install petaly

Cloud Provider Support

GCP Support

# Install with GCP support
python3 -m pip install petaly[gcp]

Prerequisites:

1. Install Google Cloud SDK
2. Configure access to your Google Project
3. Set up service account authentication

AWS Support

# Install with AWS support
python3 -m pip install petaly[aws]

Prerequisites:

1. Install AWS CLI
2. Configure AWS credentials

Full Installation

# Install all features including AWS, GCP
python3 -m pip install petaly[all]

From Source

# Clone the repository
git clone https://github.com/petaly-labs/petaly.git
cd petaly

# Create and activate virtual environment
python3 -m venv .venv
source .venv/bin/activate

# Install development dependencies
pip3 install -r requirements.txt

# Install in editable mode (recommended)
pip install -e .

# Alternative: Add src to PYTHONPATH
export PYTHONPATH=$PYTHONPATH:$(pwd)/src

Configuration

1. Initialize Configuration

# Create petaly.ini in default location (~/.petaly/petaly.ini)
python3 -m petaly init

# Or specify custom location
python3 -m petaly -c /absolute-path-to-your-config-dir/petaly.ini init

2. Set Environment Variable (Optional)

# Set the environment variable if the folder differs from the default location
export PETALY_CONFIG_DIR=/absolute-path-to-your-config-dir

# Alternative run command using the main config parameter: -c /absolute-path-to-your-config-dir/petaly.ini
python3 -m petaly -c /absolute-path-to-your-config-dir/petaly.ini [command]

3. Initialize Workspace

1. Configure petaly.ini:

[workspace_config]
pipeline_dir_path=/home/user/petaly/pipelines
logs_dir_path=/home/user/petaly/logs
output_dir_path=/home/user/petaly/output

[global_settings]
logging_mode=INFO
pipeline_format=yaml

2. Create workspace:

python3 -m petaly init --workspace

Create Pipeline

Initialize a new pipeline:

python3 -m petaly init -p my_pipeline

Follow the wizard to configure your pipeline. For detailed configuration options, see Pipeline Configuration Guide.

Run Pipeline

Execute your pipeline:

python3 -m petaly run -p my_pipeline

Run Specific Operations

# Extract data from source only
python3 -m petaly run -p my_pipeline --source_only

# Load data to target only
python3 -m petaly run -p my_pipeline --target_only

# Run specific objects
python3 -m petaly run -p my_pipeline -o object1,object2

Tutorial: CSV to PostgreSQL

Prerequisites

• Petaly installed and workspace initialized
• PostgreSQL server running

Steps

1. Initialize Pipeline

python3 -m petaly init -p csv_to_postgres

2. Download Test Data

# Download and extract test files
gunzip options.csv.gz
gunzip stocks.csv.gz

3. Configure Pipeline

• Use csv as source
• Use postgres as target
• Configure database connection details

4. Run Pipeline

python3 -m petaly run -p csv_to_postgres

Example Configuration

pipeline:
  pipeline_attributes:
    pipeline_name: csv_to_postgres
    is_enabled: true
  source_attributes:
    connector_type: csv
  target_attributes:
    connector_type: postgres
    database_user: root
    database_password: db-password
    database_host: localhost
    database_port: 5432
    database_name: petalydb
    database_schema: petaly_tutorial
  data_attributes:
    use_data_objects_spec: only
    object_default_settings:
      header: true
      columns_delimiter: ","
      columns_quote: none

Documentation

• Pipeline Configuration Guide
• Cloud Platform Guide
• Troubleshooting Guide
• Video Tutorials

Contributing

We welcome contributions! Please see our Contributing Guide for details.

License

Petaly is licensed under the Apache License 2.0. See the LICENSE file for details.