Enabling Context Persistence Across Multiple Workflow R...

At a glance

The community member posted a question about how to enable Context to persist over multiple runs of the same workflow. In the comments, another community member provided a solution involving serializing the context using the JsonSerializer or JsonPickleSerializer, and then resuming the workflow at a later point by deserializing the context. There is no explicitly marked answer, but the provided code snippet demonstrates how to persist the context across multiple workflow runs.

DDS

Hey all, how do we enable Context to persist over multiple runs of the same workflow?

2 comments

LLogan M

Plain Text

handler = workflow.run(...)
result = await handler

handler = workflow.run(ctx=handler.ctx, ....)
result = await handler

You can also serialize the context (the pickling-serializer is opt in, in case you are storing non-json serializable objects in the context)

Plain Text

from llama_index.core.workflow.context_serializers import JsonSerializer, JsonPickleSerializer

# serialize
serialized_context = handler.ctx.to_dict(serializer=JsonSerializer())

# resume at a later point
ctx = Context.from_dict(serialized_context, serializer=JsonSerializer())

DDS

thanks!

Add a reply

Find answers from the community

Enabling Context Persistence Across Multiple Workflow Runs