Find Top Errors in Time Window

Problem Statement

You are given a list of log entries and a time window defined by [start_time, end_time]. Each log entry is represented as a dictionary with keys timestamp (integer), error_code (string), and severity (integer).

Your task is to return the top k most frequent error codes that occurred strictly within the time range start_time <= timestamp < end_time.

If two error codes have the same frequency, the one that is lexicographically smaller should come first. If the number of unique error codes in the window is less than k, return all of them.

Function Signature

def get_top_errors(logs, start_time, end_time, k):
    # logs: list of dicts {'timestamp': int, 'error_code': str, 'severity': int}
    # start_time: int
    # end_time: int
    # k: int
    # returns: list of str (error codes)

Examples

Example 1: Input: logs = [{'timestamp': 10, 'error_code': '404', 'severity': 1}, {'timestamp': 15, 'error_code': '500', 'severity': 2}, {'timestamp': 20, 'error_code': '404', 'severity': 1}] start_time = 10, end_time = 20, k = 1 Output: ['404'] Explanation: Only the logs at timestamp 10 and 15 are in the window [10, 20). '404' appears once, '500' appears once. Since '404' < '500', it wins.

Constraints

• Number of logs: 1 <= N <= 10^5
• timestamp values are within [0, 10^9]
• error_code strings are length 1-10
• Time complexity should ideally be O(N log N) or O(N + U log U) where U is the number of unique error codes.