odins-eye 1.5.4

Python client for the Brookmimir API

odins-eye

Python client library for the Brookmimir API.

Features

• Synchronous and Asynchronous clients - Choose between sync (OdinsEyeClient) or async (AsyncOdinsEyeClient) based on your needs
• Comprehensive API coverage - Access all Brookmimir API endpoints
• Type-safe - Full type hints with Pydantic validation
• Automatic retries - Built-in retry logic with exponential backoff
• Rate limiting handling - Automatic detection and handling of rate limits
• HTTP/2 support - Faster connections with HTTP/2
• Context manager support - Automatic resource cleanup
• Debug logging - Optional logging for troubleshooting
• Comprehensive error handling - Detailed error messages and exception types

Installation

Install from PyPI:

pip install odins-eye

Install with development dependencies:

pip install odins-eye[dev]

Quick Start

Synchronous Usage

from odins_eye import OdinsEyeClient

# Initialize the client
client = OdinsEyeClient(api_key="your-api-key")

try:
    # Check API status
    status = client.index()
    print(f"API Status: {status}")

    # Get user profile
    profile = client.profile()
    print(f"User: {profile['user']['name']}")

    # Check credit balance
    credits = client.credits()
    print(f"Balance: {credits['current_balance']}")
finally:
    client.close()

Asynchronous Usage

import asyncio
from odins_eye import AsyncOdinsEyeClient

async def main():
    async with AsyncOdinsEyeClient(api_key="your-api-key") as client:
        # Make concurrent requests
        status, profile, credits = await asyncio.gather(
            client.index(),
            client.profile(),
            client.credits()
        )
        print(f"API Status: {status}")
        print(f"User: {profile['user']['name']}")
        print(f"Credits: {credits['current_balance']}")

asyncio.run(main())

Using Context Managers (Recommended)

with OdinsEyeClient(api_key="your-api-key") as client:
    status = client.index()
    print(status)

# Async version
async with AsyncOdinsEyeClient(api_key="your-api-key") as client:
    status = await client.index()
    print(status)

API Methods

Status and Version

# Get API status
status = client.index()

# Get API version
version = client.version()

# Run network test
result = client.nettest()

User Profile and Credits

# Get user profile
profile = client.profile()
print(f"Name: {profile['user']['name']}, Age: {profile['user']['age']}")

# Check credit balance
credits = client.credits()
print(f"Balance: {credits['current_balance']}")

Document Retrieval

# Fetch a document by ID
document = client.document("doc-123")

Query Submission

# Submit a media search
query_result = client.query({
    "limit": 2,
    "include_actors": True,
    "include_faces": True,
    "include_ocr_snippets": True,
    "filters": {
        "actor_name": "Alice",
        "environment": "prod",
        "keyword": "beach",
        "media_type": "image",
        "min_face_count": 1,
        "source_domain": "news.example",
        "tags": ["vip"],
    },
})

# Submit an image-based search
query_result = client.query({
    "image_base64": "...",
    "top_k": 10,
    "min_score": 0.1,
    "filters": {
        "source_domain": "goodsoil.church"
    },
})

Face Search

# Basic face search from a local image
results = client.face_search("path/to/image.jpg")

# Face search with canonical parameters
results = client.face_search(
    "path/to/image.jpg",
    top_k=10,
    min_score=0.8,
)

# Embedding-based face search
results = client.face_search(
    embedding=[0.5] * 512,
    top_k=5,
)

Legacy aliases such as max_k and image_base64 are still accepted for compatibility.

Error Handling

The client provides detailed error information through exception types:

from odins_eye import OdinsEyeClient, OdinsEyeAPIError, OdinsEyeError

with OdinsEyeClient(api_key="your-api-key") as client:
    try:
        result = client.profile()
    except OdinsEyeAPIError as e:
        # API returned an error (4xx or 5xx)
        print(f"API Error: {e.message}")
        print(f"Status Code: {e.status_code}")
        if e.error:
            print(f"Error: {e.error}")
        if e.error_details:
            print(f"Details: {e.error_details}")
    except OdinsEyeError as e:
        # Client-side error (validation, network, etc.)
        print(f"Client Error: {e.message}")

Handling Rate Limits

Rate limiting is automatically handled with retries:

from odins_eye import OdinsEyeAPIError

try:
    result = client.profile()
except OdinsEyeAPIError as e:
    if e.status_code == 429:
        print("Rate limit exceeded")
        if e.rate_limit_reset:
            print(f"Resets at: {e.rate_limit_reset}")

Configuration

Custom Timeout

import httpx
from odins_eye import OdinsEyeClient

timeout = httpx.Timeout(30.0, connect=10.0)
client = OdinsEyeClient(api_key="your-api-key", timeout=timeout)

Custom Retry Configuration

from odins_eye import OdinsEyeClient, RetryConfig

retry_config = RetryConfig(
    max_retries=5,
    initial_delay=2.0,
    max_delay=120.0,
    exponential_base=2.0,
    retry_on_rate_limit=True
)

client = OdinsEyeClient(
    api_key="your-api-key",
    retry_config=retry_config
)

Enable Debug Logging

import logging
from odins_eye import OdinsEyeClient

# Configure logging
logging.basicConfig(level=logging.DEBUG)

# Enable client logging
client = OdinsEyeClient(
    api_key="your-api-key",
    enable_logging=True
)

Custom Headers

client = OdinsEyeClient(
    api_key="your-api-key",
    headers={
        "X-Custom-Header": "value"
    }
)

Environment Variables

For security, consider using environment variables for your API key:

import os
from odins_eye import OdinsEyeClient

api_key = os.getenv("BROOK_MIMIR_API_KEY")
client = OdinsEyeClient(api_key=api_key)

Requirements

• Python >= 3.9
• httpx[http2] >= 0.27.0
• pydantic >= 2.6.0

API Documentation

For detailed API documentation, visit the Brookmimir API docs.

Changelog

See CHANGELOG.md for release history and changes.

License

This project is licensed under the MIT License - see the LICENSE file for details.

Links

• PyPI Package: https://pypi.org/project/odins-eye/
• Brookmimir: https://brookmimir.com

Support

For support and questions:

• Contact Brookmimir support at https://brookmimir.com/support
• Review the API documentation