dataflow.config.settings

Application settings configuration for the Haive framework.

This module defines Pydantic models for application settings, providing a type-safe and validated configuration system. Settings are automatically loaded from environment variables with sensible defaults.

The settings hierarchy includes: - AppSettings: Top-level application settings - APISettings: API-specific settings - AgentSettings: Agent-specific settings

Settings can be accessed using the get_settings() function, which returns a singleton instance of the AppSettings class.

Typical usage example:

from haive.dataflow.config.settings import get_settings

settings = get_settings()

# Access settings properties api_prefix = settings.api.prefix is_production = settings.is_production agent_timeout = settings.agent.default_timeout

# Use in application logic if settings.is_development:

print(f”Running in development mode with debug={settings.api.debug}”)

Classes

APISettings	API-specific settings for the Haive framework.
AgentSettings	Agent-specific settings for the Haive framework.
AppSettings	Application-wide settings for the Haive framework.

Functions

get_settings()

Get application settings.

Module Contents

class dataflow.config.settings.APISettings(/, **data)

Bases: pydantic.BaseModel

API-specific settings for the Haive framework.

This model defines settings specific to the API server, including debug mode, CORS configuration, rate limits, and other HTTP-related settings. Values are loaded from environment variables with sensible defaults.

Parameters:

data (Any)

debug

Whether the API is running in debug mode

title

The title of the API for OpenAPI documentation

prefix

URL prefix for all API endpoints

cors_origins

List of allowed origins for CORS

rate_limit

Maximum requests per minute per client

Create a new model by parsing and validating input data from keyword arguments.

Raises [ValidationError][pydantic_core.ValidationError] if the input data cannot be validated to form a valid model.

self is explicitly positional-only to allow self as a field name.

class dataflow.config.settings.AgentSettings(/, **data)

Bases: pydantic.BaseModel

Agent-specific settings for the Haive framework.

This model defines settings specific to agent execution, including timeouts, streaming configuration, and usage costs. Values are loaded from environment variables with sensible defaults.

Parameters:

data (Any)

default_timeout

Default timeout in seconds for agent execution

streaming_enabled

Whether streaming responses are enabled

credit_cost_per_1k_tokens

Cost in credits per 1000 tokens processed

Create a new model by parsing and validating input data from keyword arguments.

Raises [ValidationError][pydantic_core.ValidationError] if the input data cannot be validated to form a valid model.

self is explicitly positional-only to allow self as a field name.

class dataflow.config.settings.AppSettings(/, **data)

Bases: pydantic.BaseModel

Application-wide settings for the Haive framework.

This is the top-level settings model that includes all other setting categories as nested models. It provides properties for determining the current environment and accessing environment-specific settings.

Parameters:

data (Any)

environment

The current environment (development, staging, production)

api

API-specific settings

agent

Agent-specific settings

Create a new model by parsing and validating input data from keyword arguments.

Raises [ValidationError][pydantic_core.ValidationError] if the input data cannot be validated to form a valid model.

self is explicitly positional-only to allow self as a field name.

property is_development: bool

Check if running in development environment.

Return type:

bool

property is_production: bool

Check if running in production environment.

Return type:

bool

dataflow.config.settings.get_settings()

Get application settings.

Return type:

AppSettings