NeuralLog Python SDK

The Python SDK for NeuralLog provides a client library for interacting with the NeuralLog server from Python applications. It offers a simple, intuitive API for logging messages at different levels.

Requirements

• Python 3.8 or later
• requests library

Installation

pip install neurallog-sdk

Basic Usage

from neurallog_sdk import NeuralLog, NeuralLogConfig

# Configure the SDK
config = NeuralLogConfig(
    server_url="http://localhost:3030",
    namespace="default",
    async_enabled=True,
    batch_size=100,
    batch_interval_ms=5000
)
NeuralLog.configure(config)

# Get a logger
logger = NeuralLog.get_logger("my-application")

# Log a simple message
logger.info("Hello, world!")

# Log with structured data
data = {
    "user": "john.doe",
    "action": "login",
    "ip": "192.168.1.1"
}
logger.info("User logged in", data=data)

# Log an error with exception
try:
    # Some code that might throw an exception
    raise Exception("Something went wrong")
except Exception as e:
    logger.error("Failed to process request", exception=e)

Configuration Options

The Python SDK supports various configuration options:

config = NeuralLogConfig(
    # Required settings
    server_url="https://logs.example.com",
    namespace="production",

    # Optional settings
    api_key="your-api-key",
    batch_size=100,
    batch_interval_ms=5000,
    max_retries=3,
    retry_backoff_ms=1000,
    async_enabled=True,
    debug_enabled=False,

    # HTTP client settings
    timeout_ms=30000,
    max_connections=10,

    # Custom HTTP headers
    headers={
        "X-Custom-Header": "value"
    }
)

NeuralLog.configure(config)

Advanced Features

Context Data

# Set global context data for all loggers
NeuralLog.set_global_context({
    "application": "my-application",
    "environment": "production",
    "version": "1.0.0"
})

# Set logger-specific context data
logger.set_context({
    "service": "user-service",
    "instance": "user-service-1"
})

Batching

The SDK supports batching of log messages to improve performance:

config = NeuralLogConfig(
    server_url="http://localhost:3030",
    namespace="default",
    async_enabled=True,
    batch_size=100,
    batch_interval_ms=5000
)
NeuralLog.configure(config)

With batching enabled, log messages are queued and sent in batches when:

• The batch size is reached
• The batch interval elapses
• The flush method is called

Flushing

To ensure all pending log messages are sent, you can flush the logger:

# Flush a specific logger
logger.flush()

# Flush all loggers
NeuralLog.flush_all()

Framework Integration

The NeuralLog Python SDK can be integrated with popular Python logging frameworks. Framework-specific adapters will be added in future releases.

Building from Source

# Clone the repository
git clone https://github.com/NeuralLog/python-sdk.git
cd python-sdk

# Install in development mode
pip install -e .

# Run the tests
python -m unittest discover tests

Documentation

Detailed documentation is available in the docs directory:

• API Reference
• Configuration
• Architecture
• Examples

For integration guides and tutorials, visit the NeuralLog Documentation Site.

Contributing

Contributions are welcome! Please read our Contributing Guide for details on our code of conduct and the process for submitting pull requests.

Related NeuralLog Components

• NeuralLog Auth - Authentication and authorization
• NeuralLog Server - Core server functionality
• NeuralLog Web - Web interface components
• NeuralLog TypeScript Client SDK - TypeScript client SDK
• NeuralLog Java Client SDK - Java client SDK

License

MIT