Problem Description

I want to stream chunked json which contains a list of tasks in the key 'tasks' want to return each item as they are streamed in that means when you run this the yield tasks must be shown between yield chunks.

You must not delete the print lines.
And you must only update the function that yields tasks.

More details here:
https://github.com/jxnl/openai_function_call/issues/31

Acceptance Criteria

1. PR to this repo that addresses the issue (notebook is fine)
   OR 2) Replit with a set of test cases

Technical Details

from pydantic import BaseModel
 
class Task(BaseModel):
    id: int
    title: str
 
# This is your existing generator that yields chunks of JSON string do not edit this
def json_chunks(json_string):
    for i in range(0, len(json_string), 5):  # replace 5 with the chunk size you want
         chunk = json_string[i:i+5]
         print("yield chunk:", chunk)
         yield chunk
 
def tasks_from_chunks(json_chunks: Generator[str, None, None]):
     # do something to get a single task_json
     task = Task.parse_raw(**task_json)
     print("yield task", task)
     yield task
     
json_string = '{"tasks":[{"id":1,"title":"task1"},{"id":2,"title":"task2"},{"id":3,"title":"task3"}]}'
 
for task in tasks_from_chunks(json_chunks(json_string)):
     print(task)

Link to Project

https://github.com/jxnl/openai_function_call/issues/31