"""
Input: Start and finish time of n jobs
Output: Schedule with maximum number of non overlapping jobs
The Strategy: At each step choose the job with earliest finish time
Algorithm Type: Greedy
Time Complexity: O(n*log(n))
"""
jobs = [
  (0, 2, 8),  # (job_id, start_time, finish_time)
  (1, 6, 10),
  (2, 1, 3),
  (3, 4, 7),
  (4, 3, 6),
  (5, 1, 2),
  (6, 8, 10),
  (7, 10, 15),
  (8, 12, 16),
  (9, 14, 16)
]


def get_opt_schedule(jobs):
  """
  Return the job_id's in the optimal jobs
  https://en.wikipedia.org/wiki/Interval_scheduling#Greedy_polynomial_solution

  >>> get_opt_schedule(jobs)
  [5, 4, 1, 7]
  """  
  opt_schedule = []
  sorted_jobs = sorted(jobs, key=lambda j: j[2])
  number_of_jobs = len(sorted_jobs)
  opt_schedule.append(sorted_jobs[0][0])
  for i in range(1, number_of_jobs):
    if sorted_jobs[i][1] >= jobs[opt_schedule[-1]][2]:
      opt_schedule.append(sorted_jobs[i][0])
  return opt_schedule


if __name__ == "__main__":
  print(get_opt_schedule(jobs))