stagecraft 0.1.5

A Python library for building robust ETL pipelines with declarative stages and data flow management

stagecraft

A Python library for building robust ETL (Extract, Transform, Load) pipelines with declarative stages and powerful data flow management.

Features

• Pipeline Architecture: Build complex data pipelines using declarative stages and conditions
• Type-Safe Variables: Strongly-typed variable system with support for DataFrames, NumPy arrays, and serializable data
• Memory Management: Built-in memory tracking and optimization for data-intensive workflows
• Data Sources: Out-of-the-box support for CSV, JSON, and file-based data sources
• Conditional Execution: Flexible condition system for controlling stage execution
• Exception Handling: Comprehensive exception handling with custom wrappers
• Logging: Configurable logging system for pipeline monitoring
• Utility Functions: Rich set of utility functions for file operations, string manipulation, and more

Installation

pip install stagecraft

Quick Start

from stagecraft import (
    PipelineDefinition,
    PipelineRunner,
    ETLStage,
    DFVar,
)

# Define your pipeline stages
class LoadDataStage(ETLStage):
    def recipe(self, **kwargs):
        # Load your data
        pass

# Create pipeline definition
pipeline = PipelineDefinition(
    name="my_pipeline",
    stages=[LoadDataStage()]
)

# Run the pipeline
runner = PipelineRunner()
result = runner.run(pipeline)

Examples

Check out the examples/ directory for comprehensive, runnable examples:

• basic_pipeline.py - Simple end-to-end pipeline with CSV loading, transformation, and saving
• dataframe_pipeline.py - DataFrame operations with Pandera schema validation
• conditional_execution.py - Conditional stage execution with various condition types

Each example is self-contained and demonstrates best practices. See the examples/README.md for detailed documentation.

Core Components

Pipeline System

• PipelineDefinition: Define pipeline structure and stages
• PipelineRunner: Execute pipelines with context management
• ETLStage: Base class for creating custom pipeline stages
• PipelineContext: Manage pipeline state and variables

Variables

• DFVar: pandas DataFrame variables
• NDArrayVar: NumPy array variables
• SVar: Serializable variables for general Python objects

Data Sources

• CSVSource: Read data from CSV files
• JSONSource: Read data from JSON files
• FileSource: Read data from text files

Conditions

• AlwaysExecute: Unconditional execution
• AndCondition/OrCondition: Combine multiple conditions
• ConfigFlagCondition: Execute based on configuration flags
• VariableExistsCondition: Check variable presence
• CustomCondition: Define custom execution logic

Utilities

• File operations: read_file, write_file, append_file
• CSV operations: read_csv, write_csv, append_csv
• JSON operations: read_json, write_json, append_json
• String utilities: camel_to_snake_case, snake_to_camel_case, and more
• Time utilities: get_timestamp, get_current_date

Requirements

• Python 3.8+

Development

Install development dependencies:

pip install stagecraft[dev]

Run tests:

pytest

License

Apache-2.0 License - see LICENSE file for details

Contributing

This project is not accepting contributions at this time.