My previous argument was wrong. In the worst case, within a row, you may have to traverse the whole row, and rest other rows for just testing. Therefore O(m+n).
But if we dont traverse the other rows once we traversed the whole row, it can be O(n) in the best case. Sanju :) On Sat, Aug 20, 2011 at 3:04 AM, Sanjay Rajpal <srn...@gmail.com> wrote: > Got it in the worst case also it will be O(m+n) > Worst case will be > 00000001 > 00000011 > 00000111 > 00001111 > 00011111 > 00111111 > 01111111 > 11111111 > > at each step just make one comparison and one step towards left, which in > worst case is > m comparisons and n increments, so final solution is O(m+n). > > Correct me if I am wrong. > Sanju > :) > > > > On Sat, Aug 20, 2011 at 3:00 AM, Shashank Gupta > <s.sha2nkgu...@gmail.com>wrote: > >> In worst case it would be O(m*n).. >> >> >> >> On Sat, Aug 20, 2011 at 3:27 PM, shady <sinv...@gmail.com> wrote: >> >>> @Sanjay awesome solution >>> it won't be O(n^2) in worst case, it will be O(m+n) only >>> >>> On Sat, Aug 20, 2011 at 3:22 PM, Sanjay Rajpal <srn...@gmail.com> wrote: >>> >>>> Yes, thats right. >>>> I think we can do the following also : >>>> >>>> Lets us assume rows are sorted in increasing order. >>>> >>>> start from first row say i. Traverse the array from the end of the row >>>> towards the beginning till 0 occurs say at position j. >>>> now proceed to the next row i+1, check the value at i+1,j if it is 0, >>>> go to next row i+2,j >>>> else if its 1, then go to left till 0 occurs and store that index of 0 >>>> and follow to the next row. >>>> >>>> In the worst case, it will be O(n^2), but in general its a good approach >>>> i guess. what do u say guys ? >>>> >>>> Average Case O(m+n) ? >>>> >>>> >>>> Sanju >>>> :) >>>> >>>> >>>> >>>> On Sat, Aug 20, 2011 at 2:47 AM, shady <sinv...@gmail.com> wrote: >>>> >>>>> binary search on every row which will give solution in O(m*(logn)) >>>>> >>>>> On Sat, Aug 20, 2011 at 3:13 PM, Sanjay Rajpal <srn...@gmail.com>wrote: >>>>> >>>>>> Sorry I forgot to mention that. >>>>>> >>>>>> Sanju >>>>>> :) >>>>>> >>>>>> -- >>>>>> You received this message because you are subscribed to the Google >>>>>> Groups "Algorithm Geeks" group. >>>>>> To post to this group, send email to algogeeks@googlegroups.com. >>>>>> To unsubscribe from this group, send email to >>>>>> algogeeks+unsubscr...@googlegroups.com. >>>>>> For more options, visit this group at >>>>>> http://groups.google.com/group/algogeeks?hl=en. >>>>>> >>>>> >>>>> -- >>>>> You received this message because you are subscribed to the Google >>>>> Groups "Algorithm Geeks" group. >>>>> To post to this group, send email to algogeeks@googlegroups.com. >>>>> To unsubscribe from this group, send email to >>>>> algogeeks+unsubscr...@googlegroups.com. >>>>> For more options, visit this group at >>>>> http://groups.google.com/group/algogeeks?hl=en. >>>>> >>>> >>>> -- >>>> You received this message because you are subscribed to the Google >>>> Groups "Algorithm Geeks" group. >>>> To post to this group, send email to algogeeks@googlegroups.com. >>>> To unsubscribe from this group, send email to >>>> algogeeks+unsubscr...@googlegroups.com. >>>> For more options, visit this group at >>>> http://groups.google.com/group/algogeeks?hl=en. >>>> >>> >>> -- >>> You received this message because you are subscribed to the Google Groups >>> "Algorithm Geeks" group. >>> To post to this group, send email to algogeeks@googlegroups.com. >>> To unsubscribe from this group, send email to >>> algogeeks+unsubscr...@googlegroups.com. >>> For more options, visit this group at >>> http://groups.google.com/group/algogeeks?hl=en. >>> >> >> >> >> -- >> ----------------------------------------- >> $hashank Gupta >> ----------------------------------------- >> Let not a man guard his dignity, but let his dignity guard him. >> >> >> -- >> You received this message because you are subscribed to the Google Groups >> "Algorithm Geeks" group. >> To post to this group, send email to algogeeks@googlegroups.com. >> To unsubscribe from this group, send email to >> algogeeks+unsubscr...@googlegroups.com. >> For more options, visit this group at >> http://groups.google.com/group/algogeeks?hl=en. >> > > -- You received this message because you are subscribed to the Google Groups "Algorithm Geeks" group. To post to this group, send email to algogeeks@googlegroups.com. To unsubscribe from this group, send email to algogeeks+unsubscr...@googlegroups.com. For more options, visit this group at http://groups.google.com/group/algogeeks?hl=en.