Summarize Transcript Events in a Sliding Window

Problem Statement

You are processing a stream of transcript events where each event is represented by a tuple (user_id, length). As these events arrive, you need to maintain a sliding window of the last K events. After each new event is added to the window, return a result containing the current number of distinct user_ids and the maximum length within that window.

Implement the class TranscriptSummarizer:

• __init__(self, K: int): Initializes the object with the window size K.
• add_event(self, user_id: int, length: int) -> Tuple[int, int]: Adds the event to the window. If the window exceeds size K, the oldest event is removed. Returns a tuple (distinct_users, max_length).

Constraints

• 1 <= K <= 200,000
• Total number of events N <= 200,000
• user_id is an integer.
• length is a non-negative integer.

Example

# K = 3
s = TranscriptSummarizer(3)
s.add_event(1, 10) # Window: [(1, 10)], Output: (1, 10)
s.add_event(2, 20) # Window: [(1, 10), (2, 20)], Output: (2, 20)
s.add_event(1, 5)  # Window: [(1, 10), (2, 20), (1, 5)], Output: (2, 20)
s.add_event(3, 15) # Window: [(2, 20), (1, 5), (3, 15)], Output: (3, 20)

Edge Cases

• The window is not yet full (N < K).
• Repeated user_ids being added and removed.
• All length values are the same.