Deduplicate Automated Test Results

Problem

Given a list of test execution records, return the final unique failures after deduplication. Each record is a dictionary with keys test_id, status, timestamp, and message. Keep only records where status == "FAIL", and if the same test_id appears multiple times as a failure, keep only the most recent one by timestamp. Return the result sorted by ascending test_id.

Constraints

• 1 <= len(records) <= 10^5
• records[i] contains keys: test_id, status, timestamp, message
• test_id and status are strings
• 0 <= timestamp <= 10^9
• status is one of "PASS", "FAIL", "SKIP"