Problem

In a Meta event pipeline, duplicate records may arrive for the same event ID. Write a function that removes duplicates by event_id, keeping only the event with the largest timestamp for each ID.

Return the deduplicated events sorted by timestamp ascending. If two events have the same event_id and the same timestamp, keep the one that appears later in the input.

Input / Output

• Input: events, a list of dictionaries. Each dictionary has keys: "event_id" (string), "timestamp" (integer), and "payload" (string).
• Output: A list of dictionaries representing the deduplicated events, sorted by timestamp ascending.

Examples

Example 1

Input:
events = [
  {"event_id": "a1", "timestamp": 100, "payload": "view"},
  {"event_id": "b2", "timestamp": 90, "payload": "click"},
  {"event_id": "a1", "timestamp": 105, "payload": "view_retry"}
]
Output:
[
  {"event_id": "b2", "timestamp": 90, "payload": "click"},
  {"event_id": "a1", "timestamp": 105, "payload": "view_retry"}
]

a1 appears twice, so we keep the later timestamp 105.

Example 2

Input:
events = [
  {"event_id": "x", "timestamp": 5, "payload": "old"},
  {"event_id": "x", "timestamp": 5, "payload": "new"}
]
Output:
[
  {"event_id": "x", "timestamp": 5, "payload": "new"}
]

The timestamps are equal, so the later input record is kept.

Constraints

• 0 <= len(events) <= 10^5
• event_id is a non-empty string
• 0 <= timestamp <= 10^9
• Each event dictionary contains event_id, timestamp, and payload