Store the frequency of all the letters in an array in one scan(like counting sort). In the next pass remove the duplicates and appropriately shift . takes 2 O(n) passes i guess....
On Fri, Aug 5, 2011 at 4:50 PM, priya v <[email protected]> wrote: > > Remove duplicate alphabets from a string and compress it in the same > string. For > example, MISSISSIPPI becomes MISP. O (n2) is not acceptable. > For this problem is it a good idea to sort the array using a sorting > technique with efficiency O(nlogn) > and remove the duplicates? > > -- > You received this message because you are subscribed to the Google Groups > "Algorithm Geeks" group. > To post to this group, send email to [email protected]. > To unsubscribe from this group, send email to > [email protected]. > For more options, visit this group at > http://groups.google.com/group/algogeeks?hl=en. > -- S.Nishaanth, Computer Science and engineering, IIT Madras. -- You received this message because you are subscribed to the Google Groups "Algorithm Geeks" group. To post to this group, send email to [email protected]. To unsubscribe from this group, send email to [email protected]. For more options, visit this group at http://groups.google.com/group/algogeeks?hl=en.
