Skip to content

Latest commit

 

History

History
97 lines (77 loc) · 2.4 KB

README.md

File metadata and controls

97 lines (77 loc) · 2.4 KB

chainlit-azure

Azure Data Lake Storage Gen2 client for Chainlit.

Azure Data Lake Storage Client

This storage client enables integration with Azure Data Lake Storage Gen2 for file operations.

Key features:

  • File upload/delete operations
  • Presigned URL generation
  • SAS token support
  • Multiple authentication methods (account key, connection string, SAS token)

Setup Example

  1. Install required dependencies:
pip install chainlit-azure[azure-datalake]
  1. Configure in your Chainlit app:
# Security Recommendation:
# Use Azure Managed Identity in production
# For local development, use azure-identity DefaultAzureCredential

import os
from chainlit.data.sql_alchemy import SQLAlchemyDataLayer
from chainlit_azure import AzureStorageClient

@cl.data_layer
def get_data_layer():
    # Example with connection string credential
    storage_client = AzureStorageClient(
        account_url="https://<your_account>.dfs.core.windows.net",
        container="<your_container>",
        credential=os.environ["AZURE_STORAGE_CONNECTION_STRING"]
    )

    return SQLAlchemyDataLayer(
        conninfo=os.environ["DATABASE_URL"],
        storage_provider=storage_client
    )

Authentication Options

Connection String

AzureStorageClient(
    account_url="https://account.dfs.core.windows.net",
    container="container",
    credential="DefaultEndpointsProtocol=https;AccountName=...;AccountKey=..."
)

Account Key

from azure.identity import AzureNamedKeyCredential

credential = AzureNamedKeyCredential("<account_name>", "<account_key>")
AzureStorageClient(
    account_url="https://account.dfs.core.windows.net",
    container="container",
    credential=credential
)

SAS Token

AzureStorageClient(
    account_url="https://account.dfs.core.windows.net",
    container="container",
    credential=None,  # Not needed if using SAS
    sas_token="<your_sas_token>"
)

Dependencies

  • Required: azure-storage-file-datalake
  • Optional: azure-identity (for token-based authentication)

Notes

  1. SAS tokens can be generated automatically if:

    • Account key is provided
    • No existing SAS token is configured
    • Read permissions are required for presigned URLs

  2. Supported credential types:

    • Connection strings
    • Account keys
    • AzureNamedKeyCredential
    • AzureSasCredential
    • TokenCredential (Azure Active Directory)