dbxparams 0.1.18

Manage Databricks notebook parameters as Python classes

dbxparams

A lightweight library for managing Databricks notebook parameters with type safety, defaults, and automatic widget creation.

---

✨ Features

• Auto-create and populate Databricks widgets
• Type-safe parameter casting (int, float, bool, date, datetime)
• Support for defaults dict and class defaults
• Custom error handling: MissingParameterError, InvalidTypeError
• Cleaner, more maintainable notebooks (no more repetitive dbutils.widgets.get!)

---

📦 Installation

pip install dbxparams

---

🧩 Load Parameters from JSON

You can now initialize your parameter class directly from a JSON file or raw JSON string using from_json():

from dbxparams import NotebookParams

class ConfigParams(NotebookParams):
    database_name: str
    storage_path: str

params = ConfigParams.from_json("config/settings.json", dbutils=dbutils)

print(params.database_name)
print(params.storage_path)

This is useful for storing environment settings, connection strings, or other reusable constants without hardcoding them into your notebooks.

---

🚀 Quick Start

from dbxparams import NotebookParams

class MyParams(NotebookParams):
    market: str              # required
    env: str = "dev"         # optional with default
    retries: int = 3         # optional with default

# Pass dbutils to auto-create widgets and populate values
params = MyParams(dbutils)

print(params.market)   # Read from widget
print(params.env)      # Uses default "dev" if not set
print(params.retries)  # Uses default 3

---

🔒 Error Handling

• MissingParameterError → Raised when a required parameter is missing
• InvalidTypeError → Raised when a value cannot be cast to the expected type

Example:

class TypedParams(NotebookParams):
    threshold: float
    active: bool

# If "threshold" widget is "not-a-float" → InvalidTypeError

---

📜 License

This project is licensed under the MIT License.

Copyright (c) 2025 Víctor Ferrón Álvarez https://vicferron.github.io/