Find the smallest number that is greater than a given number in a sorted list

Question:

Given a sorted list of numbers, I need to find the smallest number that is greater than a given number. Consider this list:


arr=[1,2,3,5,7,11,101,131,151,181,191,313,353,373,383]

Say the specified number is 320. Then, my method should return 353 as 353 is the smallest number greater than 320.

I am trying to use a slightly modified form of binary search; however on execution the program goes into infinite loop.


def modBinarySearch(arr,x):
    l=len(arr)
    mid=l/2
    if arr[mid]>=x and arr[mid-1]<x:
        return arr[mid]
    elif arr[mid]>x and arr[mid-1]>x:
        modBinarySearch(arr[mid:l],x)
    else: 
        modBinarySearch(arr[0:mid],x)

N=int(raw_input())
arr=[1,2,3,5,7,11,101,131,151,181,191,313,353,373,383]
print modBinarySearch(arr,N)

Can someone point out what I am doing wrong ?

Asked By: OneMoreError

||

Answers:

If the size of your lists is going to be 15, ditch the binary search altogether and use a sequential search.

You’ll find the code much easier to write and, unless you need to do it many millions of times per second, the sequential solution will be more than fast enough.

If you do need to stick with the binary search, your first step should be to actually return the results of your recursive calls.

Answered By: paxdiablo

There is a standard module, bisect, that does this already:

In [49]: arr[bisect.bisect(arr, 320)]
Out[49]: 353

I think this should be the go-to method for searching sorted lists. There are a few examples in the manual.

As to your implementation, there is a number of problems:

  1. Your recursion doesn’t handle small arrays correctly.
  2. The slicing done in the second branch is incorrect.
  3. Your function doesn’t return anything.
  4. Because arr is in ascending order, arr[mid]>x and arr[mid-1]>x is equivalent to arr[mid-1]>x, suggesting you didn’t write what you meant

Last but not least, recursion and all that slicing are completely unnecessary for this problem.

Answered By: NPE

IF the list is sorted:

x = range(20)
N= 15

for i in x:
    if i>N:
        print i
        break

Gives 16.

If using numpy:

x = np.arange(20)
N = 15
x[x>15][0]

Gives 16.

If you were looking for easy implementations, for your specific problem, let me get back on that.

Answered By: arynaq
def modBinarySearch(arr, n):
    m = len(arr) / 2

    if arr[m] >= n and arr[m - 1] < n:
        return arr[m]
    elif arr[m] > n and arr[m - 1] > n:
        return modBinarySearch(arr[:m], n)
    else:
        return modBinarySearch(arr[m:], n)


arr = [1, 2, 3, 5, 7, 11, 101, 131, 151, 181, 191, 313, 353, 373, 383]
n = 320
print(modBinarySearch(arr, n))
Answered By: Danil Speransky

python 3.2

next(i for  i in arr if i>320)
Answered By: raton

If arr[mid] and arr[mid-1], both are greater than your number, you should search in arr[0:mid], don’t you think?

 elif arr[mid]>x and arr[mid-1]>x:
    modBinarySearch(arr[0:mid],x)
else: 
    modBinarySearch(arr[mid:1],x)
Answered By: Chocolava

The bisect module is your best choice for this:

from bisect import bisect_left, bisect_right

arr=[1,2,3,5,7,11,101,131,151,181,191,313,353,373,383]


def find_lt(a, x):
    'Find rightmost value less than x'
    i = bisect_left(a, x)
    if i:
        return a[i-1]
    raise ValueError

def find_gt(a, x):
    'Find leftmost value greater than x'
    i = bisect_right(a, x)
    if i != len(a):
        return a[i]
    raise ValueError

print find_lt(arr,320)          
print find_gt(arr,320)  

prints

313
353     
Answered By: the wolf
def modBinarySearch(arr,x):
    l=len(arr)
    mid=l/2
    if arr[mid] >= x and arr[mid-1] < x:
        return arr[mid]
    elif arr[mid]>x and arr[mid-1]>x:
        num = modBinarySearch(arr[0:mid],x)
    else:
        num = modBinarySearch(arr[mid:l],x)
    return num

N=int(raw_input('Enter a number: '))
arr=[1, 2, 3, 5, 7, 11, 101, 131, 151, 181, 191, 313, 353, 373, 383]
print modBinarySearch(arr,N)
Answered By: Love Bisaria
Categories: questions Tags: , ,
Answers are sorted by their score. The answer accepted by the question owner as the best is marked with
at the top-right corner.