Stream output until stop token appears

Problem Statement

You are tasked with building a stream processor that reads chunks of text from a generator. Your goal is to yield the processed text until you encounter a specific 'stop token' string.

Once the stop token is found, the function must terminate immediately. Any part of the stop token that was partially matched at the end of a chunk should not be emitted. If the stop token never appears, yield all data from the stream.

Implement the following function:

def stream_until_stop(data_generator, stop_token):
    """
    data_generator: A generator that yields strings (chunks).
    stop_token: The string sequence that triggers termination.
    Yields: Parts of the string until the stop_token is encountered.
    """

Examples

Example 1:

• Input: data_generator yields ['abc', 'defg', 'hi'], stop_token = 'ef'
• Output: 'abc', 'd'
• Explanation: The stop token 'ef' starts at the end of the first chunk 'abc' (none) and continues into 'defg'. The function yields 'abc' and 'd', then stops.

Example 2:

• Input: data_generator yields ['hello', ' world'], stop_token = 'xyz'
• Output: 'hello', ' world'

Constraints

• stop_token length will be at least 1.
• Chunks can be of arbitrary length.
• Your solution should use O(M) space, where M is the length of the stop_token.

Edge Cases

• The stop token is split across three or more chunks.
• The stop token is longer than a single chunk.
• The stream contains multiple occurrences of the stop token (stop at the first one).