Problem

A mobile app at PulseFit calls several REST endpoints repeatedly. To reduce unnecessary network requests, implement an in-memory cache that stores API responses by endpoint key and expires entries after a fixed time-to-live.

Write a function that processes a sequence of cache operations and returns the result of each get operation.

Formal Specification

Implement process_api_cache(operations, ttl) where:

• operations is a list of operations.
• Each operation is one of:
  • ["set", key, value, timestamp]
  • ["get", key, timestamp]
  • ["invalidate", key, timestamp]
• key and value are strings.
• timestamp is a non-negative integer.
• ttl is a positive integer.

For a get at time t, return the cached value only if the key exists and t - set_timestamp < ttl. Otherwise return "MISS". An invalidate removes the key immediately if present. Return a list containing the outputs of all get operations in order.

Examples

Example 1

• Input: operations = [["set", "/users", "A", 1], ["get", "/users", 2], ["get", "/users", 5]], ttl = 3
• Output: ["A", "MISS"]
• Explanation: The value is valid at time 2, but expired by time 5.

Example 2

• Input: operations = [["set", "/feed", "X", 10], ["invalidate", "/feed", 11], ["get", "/feed", 12]], ttl = 5
• Output: ["MISS"]
• Explanation: The key is explicitly removed before the read.

Constraints

• 1 <= len(operations) <= 10^5
• 1 <= ttl <= 10^9
• Timestamps are non-negative integers
• Operations are processed in the given order