Skip to main content

No project description provided

Project description

LLM Context Providers

A Python package to manage and fetch context from various sources like Asana, Quip, OneNote, Outlook Calendar, etc. This package is designed to be easily extensible, allowing new context providers to be added with minimal changes to the core logic.

Features

  • Fetch context from multiple sources asynchronously or synchronously.
  • Easily extendable with new context providers.
  • Configurable through YAML configuration.
  • Handles retries and errors gracefully.

Installation

pip install llm-context-providers

Quick Start

Create an app.py file with the following content to quickly get started:

from llm_context_providers import ContextManager

config = {
    'global': {
        'timezone': 'America/Toronto'
    },
    'context_providers': {
        'asana': {
            'enabled': True,
            'project_id': 'your_asana_project_id',
            'personal_access_token': 'your_personal_access_token',
            'fields': {
                'Task': 'name',
                'Task ID': 'gid',
                'Created At': 'created_at',
                'Modified At': 'modified_at',
                'Completed': 'completed',
                'Assignee': 'assignee',
                'Due On': 'due_on',
                'Notes': 'notes'
            }
        }
    }
}

def main_sync(config):
    manager = ContextManager(config)
    manager.fetch_contexts()
    context = manager.get_combined_context()
    print("Sync Fetch Context (All):\n", context)

if __name__ == "__main__":
    main_sync(config)

Run the application:

python app.py

Full Functionality

For easier maintainability and to handle more complex configurations, it's recommended to use a config.yml file. Here is an example configuration:

global:
  timezone: "America/Toronto"

context_providers:
  asana:
    enabled: true
    project_id: "your_asana_project_id"
    fields:
      Task: "name"
      Task ID: "gid"
      Created At: "created_at"
      Modified At: "modified_at"
      Completed: "completed"
      Assignee: "assignee"
      Due On: "due_on"
      Notes: "notes"
  # Add other context providers here

Using the Configuration File

Update your app.py to load configuration from config.yml and demonstrate the full functionality, including asynchronous and specific context fetching:

import yaml
import asyncio
from llm_context_providers import ContextManager

def load_config(config_file='config.yml'):
    with open(config_file, 'r') as file:
        return yaml.safe_load(file)

async def main_async(config):
    manager = ContextManager(config)
    await manager.fetch_contexts_async()
    context = manager.get_combined_context()
    print("Async Fetch Context (All):\n", context)

async def main_async_specific(config):
    manager = ContextManager(config)
    await manager.fetch_contexts_async(providers=['asana'])
    context = manager.get_combined_context(providers=['asana'])
    print("Async Fetch Context (Specific):\n", context)

def main_sync(config):
    manager = ContextManager(config)
    manager.fetch_contexts()
    context = manager.get_combined_context()
    print("Sync Fetch Context (All):\n", context)

def main_sync_specific(config):
    manager = ContextManager(config)
    manager.fetch_contexts(providers=['asana'])
    context = manager.get_combined_context(providers=['asana'])
    print("Sync Fetch Context (Specific):\n", context)

if __name__ == "__main__":
    config = load_config()

    print("Testing Async Fetch (All):")
    asyncio.run(main_async(config))

    print("\nTesting Async Fetch (Specific):")
    asyncio.run(main_async_specific(config))

    print("\nTesting Sync Fetch (All):")
    main_sync(config)

    print("\nTesting Sync Fetch (Specific):")
    main_sync_specific(config)

Run the application:

python app.py

Extending with New Context Providers

To add a new context provider, create a new class that inherits from ContextProvider and implement the necessary methods:

from llm_context_providers import ContextProvider

class NewContextProvider(ContextProvider):
    def __init__(self, **kwargs):
        super().__init__()
        # Initialize with necessary credentials and configurations

    async def fetch_context_async(self, full_fetch: bool = False):
        # Implement async context fetching logic
        pass

    def fetch_context(self, full_fetch: bool = False):
        # Implement sync context fetching logic
        pass

    def get_context(self):
        # Return the context information
        pass

    async def index_context(self):
        # Implement indexing logic
        pass

    async def search_index(self, query: str):
        # Implement search logic
        pass

    def load_from_index(self, search_results):
        # Implement logic to load search results into context
        pass

Add the new context provider to your configuration file and ensure the package's ContextProvider class can recognize the new provider.

Contributing

Contributions are welcome! Please create a pull request or raise an issue to discuss your ideas.

License

This project is licensed under the MIT License.

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

llm_context_providers-0.1.1.tar.gz (7.9 kB view details)

Uploaded Source

Built Distribution

llm_context_providers-0.1.1-py3-none-any.whl (9.8 kB view details)

Uploaded Python 3

File details

Details for the file llm_context_providers-0.1.1.tar.gz.

File metadata

  • Download URL: llm_context_providers-0.1.1.tar.gz
  • Upload date:
  • Size: 7.9 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/5.1.1 CPython/3.12.4

File hashes

Hashes for llm_context_providers-0.1.1.tar.gz
Algorithm Hash digest
SHA256 0084ef9e7f133184e40a8acfb0b487c2e26b5d043fcbd8440a03697e91841b52
MD5 225e4c2585eb98b7f06b13f6ccb47941
BLAKE2b-256 082748892de42950502e72d9957e787f8c01490976997e17d5f99de5704de1e1

See more details on using hashes here.

File details

Details for the file llm_context_providers-0.1.1-py3-none-any.whl.

File metadata

File hashes

Hashes for llm_context_providers-0.1.1-py3-none-any.whl
Algorithm Hash digest
SHA256 edf1dfed6c766140c5f4d22d710edb6198f952f535388e2d57f1c930d404b1ce
MD5 ebe661550a5aaf791c12365f63f3542f
BLAKE2b-256 0d4d0ba32725ed5d6c4b9ce8bc191846fb36825d6c9061931b220ba88a98b89e

See more details on using hashes here.

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page