v0.0.33

What's Changed

• Update .clabot by @eyurtsev in https://github.com/langchain-ai/langserve/pull/269
• Added deployment option to Azure in the documentation by @sowsan in https://github.com/langchain-ai/langserve/pull/266
• Bump overall test timeout by @eyurtsev in https://github.com/langchain-ai/langserve/pull/273
• Refactor in preparation to add ability to enable/disable individual endpoints by @eyurtsev in https://github.com/langchain-ai/langserve/pull/271
• Bump aiohttp in lock file by @eyurtsev in https://github.com/langchain-ai/langserve/pull/278
• Moving API Handler to its own file by @eyurtsev in https://github.com/langchain-ai/langserve/pull/277
• Add trailing / to URLs in README.md by @eyurtsev in https://github.com/langchain-ai/langserve/pull/282

Fixes

• Bug Fix: Use deep copy in RemoteClient astream log@eyurtsev in https://github.com/langchain-ai/langserve/pull/280
• Enforce trailing slash in remote runnable by @eyurtsev in https://github.com/langchain-ai/langserve/pull/287

Features

• Enable/disable endpoints by @eyurtsev in https://github.com/langchain-ai/langserve/pull/281

You can enable / disable which endpoints are exposed. Use enabled_endpoints if you want to make sure to never get a new endpoint when upgrading langserve to a newer verison. See README.md for more details.

# Enable
add_routes(app, chain, enabled_endpoints=("invoke", "batch", "config_hashes"))
# Disable
add_routes(app, chain, disabled_endpoints=["playground"]) 

• Document Enabling/Disabling Endpoints in README.md by @eyurtsev in https://github.com/langchain-ai/langserve/pull/289

Examples

• Add configurable retriever example by @eyurtsev in https://github.com/langchain-ai/langserve/pull/267

New Contributors

• @sowsan made their first contribution in https://github.com/langchain-ai/langserve/pull/266

Full Changelog: https://github.com/langchain-ai/langserve/compare/v0.0.32...v0.0.33