I got this problem from an interview with Microsoft.
Given an array of random integers,
write an algorithm in C that removes
duplicated numbers and return the unique numbers in the original
array.
E.g Input: {4, 8, 4, 1, 1, 2, 9}
Output: {4, 8, 1, 2, 9, ?, ?}
One caveat is that the expected algorithm should not required the array to be sorted first. And when an element has been removed, the following elements must be shifted forward as well. Anyway, value of elements at the tail of the array where elements were shifted forward are negligible.
Update: The result must be returned in the original array and helper data structure (e.g. hashtable) should not be used. However, I guess order preservation is not necessary.
Update2: For those who wonder why these impractical constraints, this was an interview question and all these constraints are discussed during the thinking process to see how I can come up with different ideas.
Question&Answers:
os 与恶龙缠斗过久,自身亦成为恶龙;凝视深渊过久,深渊将回以凝视…