Find the longest sequence

Question

Problem:

Find the longest sequence consisting only of 0's, and the longest sequence consisting only of 1's?

Solution:

import sys
longestSequenceForOne = None
startIndexForOne = None
longestSequenceForZero = None
startIndexForZero = None
def setLongestSequence(array, longestSequence, startIndex):
 global longestSequenceForOne
 global startIndexForOne
 global longestSequenceForZero
 global startIndexForZero
 if array[startIndex] == 1:
 if longestSequenceForOne == None or longestSequenceForOne < longestSequence:
 longestSequenceForOne = longestSequence
 startIndexForOne = startIndex
 elif array[startIndex] == 0:
 if longestSequenceForZero == None or longestSequenceForZero < longestSequence: 
 longestSequenceForZero = longestSequence
 startIndexForZero = startIndex
def findLongestSequence(array, size):
 """
 Find the longest sequence of 0's and 1's
 """
 longestSequence = 1
 beginIndex = None
 index = 1
 startIndex = 0
 sameElementCount = 1
 while (index < size):
 if array[index] == array[index-1]:
 sameElementCount += 1
 elif sameElementCount > longestSequence:
 longestSequence = sameElementCount
 startIndex = index - sameElementCount
 sameElementCount = 1
 setLongestSequence(array, longestSequence, startIndex)
 else:
 sameElementCount = 1
 index += 1
 if sameElementCount > longestSequence:
 longestSequence = sameElementCount
 startIndex = index - sameElementCount
 setLongestSequence(array, longestSequence, startIndex)
if __name__ == '__main__':
 size = int(input())
 if size < 2 or size > 100:
 sys.exit()
 array = [int(x) for x in input().split() if x == '1' or '0']
 if len(array) != size:
 sys.exit()
 findLongestSequence(array, size)
 print(startIndexForZero, longestSequenceForZero)
 print(startIndexForOne, longestSequenceForOne)

---

Correctness

Input
12
0 1 0 1 1 1 0 1 0 0 0 0
Output
8 4
3 3
Input
17
1 1 0 1 0 1 0 0 1 0 0 0 0 1 0 1 0
Output
9 4
0 2

---

Questions:

1. Does the code require better error handling?

2. Can this code get more pythonic?

• Zyx: If you want to just increment by 1 each time, then for index in range(size) is both more pythonic and faster than while (index<size). Also, if your array is large, you can significantly improve the run time by incrementing by the current longest sequence instead of by 1. Check out this question of mine - it's not exactly the same, but the general idea of the algorithm still applies.
• overexchange: @ZyTelevan Yup for index in range(1, len(array)) works
• depperm: you could use regex and shorten the code too

Answer 1

• Correctness

  Input
  1 1 1 1 0 0 0
  Output
  None, None
  0, 4
  

  The reason is obvious: the longestSequence does not distinguish between longest sequence of zeroes and longest sequence of ones, so 0 0 0 is not considered long enough.

• General

  • Avoid globals; prefer returning values.

  • There is no need to pass size; a list already knows its size. For the same reason, size = int(input()) is redundant and reduces usability.

• overexchange: longestSequence = 1 after setLongestSequence() worked