v0.3.5

What's Changed

New Features 🎉

• correct .yml file name by @ravi03071991 in #382
• fix: error in run command by @masci in #384
• fix: fix rabbitmq import errors by @masci in #386
• Prepare 0.3.5 by @masci in #387

New Contributors

• @ravi03071991 made their first contribution in #382

Full Changelog: v0.3.4...v0.3.5

v0.3.4

v0.3.4

v0.3.3

What's Changed

New Features 🎉

• Integrate the Solace PubSub+ broker into LlamaDeploy by @alimosaed in #345
• v0.3.3 by @logan-markewich in #377

New Contributors

• @alimosaed made their first contribution in #345

Full Changelog: v0.3.2...v0.3.3

v0.3.2

What's Changed

New Features 🎉

• Downgrading python-multipart dependency by @tituslhy in #375

New Contributors

• @tituslhy made their first contribution in #375

Full Changelog: v0.3.1...v0.3.2

v0.3.1

What's Changed

New Features 🎉

• build(deps): bump aiohttp from 3.10.10 to 3.10.11 by @dependabot in #369
• Update release.yml by @masci in #371
• Use the new SDK within the api server by @masci in #372
• [0.3.1] vbump for llama-index-core by @logan-markewich in #373

Full Changelog: v0.3.0...v0.3.1

v0.3.0

What's Changed

New Features 🎉

• feat: add apiserver support to Python SDK by @masci in #327
• refact: make collections in apiserver models lazy by @masci in #343
• chore: deprecate old clients in favor of Client by @masci in #352
• feat: make the kafka topic configurable by @masci in #353
• feat: make message_type configurable in Control Plane by @masci in #356
• refact: Make topic explicit in message queue API by @masci in #358
• refact: Use the Python SDK in the CLI implementation by @masci in #365

Bug Fixes 🐛

• fix: fix typing for wrapped async methods by @masci in #344

Documentation 📚

• docs: Python SDK by @masci in #355

Full Changelog: v0.2.4...v0.3.0

v0.2.4

v0.2.4

v0.2.3

v0.2.3

v0.2.1

v0.2.1

v0.2.0

v0.2.0 is out now, with the main improvement being the addition of streaming support!

Now, if you have a workflow that writes to the event stream like:

class ProgressEvent(Event):
    progress: str


# create a dummy workflow
class MyWorkflow(Workflow):
    @step()
    async def run_step(self, ctx: Context, ev: StartEvent) -> StopEvent:
        # Your workflow logic here
        arg1 = str(ev.get("arg1", ""))
        result = arg1 + "_result"

        # stream events as steps run
        ctx.write_event_to_stream(
            ProgressEvent(progress="I am doing something!")
        )

        return StopEvent(result=result)

You can stream the events using the client

# create a session
session = client.create_session()

# kick off run
task_id = session.run_nowait("streaming_workflow", arg1="hello_world")

# stream events -- the will yield a dict representing each event
for event in session.get_task_result_stream(task_id):
    print(event)

# get final result
result = session.get_task_result(task_id)
print(result)
# prints 'hello_world_result'