searching and sorting organized

src/searching_and_sorting/binary_search.py

@@ -1,56 +0,0 @@
-#!/usr/bin/env python
-
-__author__ = "bt3"
-
-
-
-def binary_search_rec(array, item, lo=0, hi = None):
-    '''
-    >>> binary_search_rec([2,3,5,6,8,10,15,23], 15)
-    (True, 6)
-    >>> binary_search_rec([2,3,5,6,8,10,15,23], 4)
-    False
-    '''
-    hi = hi or len(array)
-    if hi < lo :
-        return False
-
-    mid = (hi + lo)//2
-
-    if array[mid] == item:
-        return True, mid
-    elif array[mid] < item:
-        return binary_search_rec(array, item, mid + 1, hi)
-    else:
-        return binary_search_rec(array[:mid], item, lo, mid -1)
-
-
-
-def binary_search_iter(array, item):
-    '''
-    >>> binary_search_iter([2,3,5,6,8,10,15,23], 15)
-    (True, 6)
-    >>> binary_search_iter([2,3,5,6,8,10,15,23], 4)
-    False
-    '''
-    hi = len(array)
-    lo = 0
-
-    while lo < hi:
-        mid = (hi+lo)//2
-        if array[mid] == item:
-            return True, mid
-        elif array[mid] > item:
-            hi = mid
-        else:
-            lo  = mid + 1
-    return False
-
-
-
-
-
-if __name__ == '__main__':
-    import doctest
-    doctest.testmod()
-

src/searching_and_sorting/quick_sort.py

@@ -1,27 +0,0 @@
-#!/usr/bin/env python
-
-__author__ = "bt3"
-
-
-def qs(array):
-    '''
-    >>> qs([4,1,6,2,7,9,3])
-    [1, 2, 3, 4, 6, 7, 9]
-    '''
-    if len(array) < 2:
-        return array
-
-    piv = len(array)//2
-    piv_element = array[piv]
-    new_array = array[:piv] + array[piv+1:]
-
-    left  = [a for a in new_array if a <= piv_element]
-    right = [a for a in new_array if a > piv_element]
-
-
-    return qs(left) + [array[piv]] + qs(right)
-
-
-if __name__ == '__main__':
-    import doctest
-    doctest.testmod()

src/searching_and_sorting/searching/binary_search.py

@@ -1,49 +1,56 @@
-#!/usr/bin/python
+#!/usr/bin/env python
 
-__author__ = "Mari Wahl"
+__author__ = "bt3"
-__email__ = "marina.w4hl@gmail.com"
 
 
 
-def binary_search(seq, key):
+def binary_search_rec(array, item, lo=0, hi = None):
-    ''' binary search iterative algorithm '''
+    '''
-    ''' observe that the index is returned '''
+    >>> binary_search_rec([2,3,5,6,8,10,15,23], 15)
-    hi = len(seq)
+    (True, 6)
+    >>> binary_search_rec([2,3,5,6,8,10,15,23], 4)
+    False
+    '''
+    hi = hi or len(array)
+    if hi < lo :
+        return False
+
+    mid = (hi + lo)//2
+
+    if array[mid] == item:
+        return True, mid
+    elif array[mid] < item:
+        return binary_search_rec(array, item, mid + 1, hi)
+    else:
+        return binary_search_rec(array[:mid], item, lo, mid -1)
+
+
+
+def binary_search_iter(array, item):
+    '''
+    >>> binary_search_iter([2,3,5,6,8,10,15,23], 15)
+    (True, 6)
+    >>> binary_search_iter([2,3,5,6,8,10,15,23], 4)
+    False
+    '''
+    hi = len(array)
     lo = 0
+
     while lo < hi:
         mid = (hi+lo)//2
-        if seq[mid] == key:
+        if array[mid] == item:
-            return mid
+            return True, mid
-        elif key < seq[mid]:
+        elif array[mid] > item:
             hi = mid
         else:
             lo  = mid + 1
+    return False
 
 
-def binary_search_rec(seq, key, lo=0, hi=None):
-    ''' binary search recursive algorithm '''
-    hi = hi or len(seq)
-    if hi < lo: return None
-    mid = (hi + lo) // 2
-    if seq[mid] == key:
-        return mid
-    elif seq[mid] < key:
-        return binary_search_rec(seq, key, mid + 1, hi)
-    else:
-        return binary_search_rec(seq, key, lo, mid - 1)
-
-
-def test_binary_search():
-    seq = [1,2,5,6,7,10,12,12,14,15]
-    key = 6
-    assert(binary_search(seq, key) == 3)
-    assert(binary_search_rec(seq, key) == 3)
-    print('Tests passed!')
-
-
-if __name__ == '__main__':
-    test_binary_search()
-
 
 
 
+if __name__ == '__main__':
+    import doctest
+    doctest.testmod()
+

src/searching_and_sorting/searching/binary_search_matrix.py

@@ -1,75 +1,89 @@
-#!/usr/bin/python
+#!/usr/bin/env python
 
-__author__ = "Mari Wahl"
+__author__ = "bt3"
-__email__ = "marina.w4hl@gmail.com"
 
 
-''' Searches an element in a matrix where in every row, the values are increasing from left to right, but the last number in a row is smaller than the first number in the next row.
+def binary_search_matrix_rec(m, key, lo=0, hi=None):
-
+    '''
-    (1) The naive brute force solution (sequential search)  scan all numbers and cost O(nm).  However, since the numbers are already sorted, the matrix can be viewed as a 1D sorted array.  The binary search algorithm is suitable. The efficiency is O(logmn).
-
     >>> m = [[1, 2, 3], [4, 5, 6], [7, 8, 9]]
     >>> binary_search_matrix_rec(m, 6)
     (1, 2)
     >>> binary_search_matrix_rec(m, 12)
-    >>> binary_search_matrix_iter(m, 6)
-    (1, 2)
-    >>> binary_search_matrix_iter(m, 12)
-    >>> binary_search_matrix_iter(m, 1)
-    (0, 0)
-
-    (2) Another solution is "discarding" arrays in the way. The efficiency is O(logm).
-    >>> m = [[1, 2, 3], [4, 5, 6], [7, 8, 9]]
-    >>> searching_matrix(m, 6)
-    (1, 2)
-    >>> searching_matrix(m, 12)
-
     '''
+    if not m:
+        return None
 
-
-def binary_search_matrix_rec(m, key, lo=0, hi=None):
-    if not m: return None
     rows = len(m)
     cols = len(m[0])
     hi = hi or rows*cols
-    if hi > lo: # -----> REMEMBER THIS OR INDEX WILL EXPLODE!!!!!!!!
+
+    if hi > lo:
+
         mid = (hi + lo)//2
         row = mid//cols
         col = mid%cols
         item = m[row][col]
-        if key == item: return row, col
+
-        elif key < item: return binary_search_matrix_rec(m, key, lo, mid-1)
+        if key == item:
-        else: return binary_search_matrix_rec(m, key, mid+1, hi)
+            return row, col
+        elif key < item:
+            return binary_search_matrix_rec(m, key, lo, mid-1)
+        else:
+            return binary_search_matrix_rec(m, key, mid+1, hi)
+
     return None
 
 
 
 def binary_search_matrix_iter(m, key):
-    if not m: return None
+    '''
+
+    '''
+
+    if not m:
+        return None
     rows = len(m)
     cols = len(m[0])
     lo, hi = 0, rows*cols
+
     while lo < hi:
         mid = (hi + lo)//2
         row = mid//rows
         col = mid%rows
         item = m[row][col]
-        if key == item: return (row, col)
+        if key == item:
-        elif key < item: hi = mid
+            return (row, col)
-        else: lo = mid +1
+        elif key < item:
+            hi = mid
+        else:
+            lo = mid +1
+
     return None
 
 
 def searching_matrix(m, key):
-    if not m: return None
+    '''
+    >>> m = [[1, 2, 3], [4, 5, 6], [7, 8, 9]]
+    >>> searching_matrix(m, 6)
+    (1, 2)
+    >>> searching_matrix(m, 12)
+    '''
+
+    if not m:
+        return None
     rows = len(m)
     cols = len(m[0])
     i, j = 0, cols -1
+
     while i < rows and j > 0:
         item = m[i][j]
-        if key == item: return (i, j)
+        if key == item:
-        elif key < item: j -= 1
+            return (i, j)
-        else: i += 1
+        elif key < item:
+            j -= 1
+        else:
+            i += 1
+
     return None
 
 

src/searching_and_sorting/searching/find_item_rotated_sorted_array.py

@@ -1,21 +1,27 @@
-#!/usr/bin/python
+#!/usr/bin/env python
 
-__author__ = "Mari Wahl"
+__author__ = "bt3"
-__email__ = "marina.w4hl@gmail.com"
 
 
 
-''' Given a sorted array that was rotated, find an item with binary search:
+'''
+Given a sorted array that was rotated, find an item with binary search:
 '''
 
 def find_element_rot_array(seq, key, lo=0, hi=None):
+
     hi = hi or len(seq)
-    if hi <= lo: return None # base case: <= for odd and even numbers!
+    if hi <= lo:
+        return None # base case: <= for odd and even numbers!
+
     mid = (hi + lo) // 2
-    if key == seq[mid]: return mid
+
+    if key == seq[mid]:
+        return mid
 
     # if left is ordered --> we work here
     if seq[lo] <= seq[mid]:
+
         # now, is the key there?
         if key < seq[mid] and key >= seq[lo]:
             return find_element_rot_array(seq, key, lo, mid)
@@ -25,6 +31,7 @@ def find_element_rot_array(seq, key, lo=0, hi=None):
 
     # right is ordered --> we work here
     else:
+
         # now, is the key there?
         if key > seq[mid] and key <= seq[hi-1]: # stupid hi-1!!!
             return find_element_rot_array(seq, key, mid+1, hi)

src/searching_and_sorting/searching/find_max_unimodal_array.py

@@ -1,15 +1,15 @@
-#!/usr/bin/python
+#!/usr/bin/env python
-
-__author__ = "Mari Wahl"
-__email__ = "marina.w4hl@gmail.com"
-
 
+__author__ = "bt3"
 
 def find_max_unimodal_array(A):
-    if len(A) <= 2 : return None
+    if len(A) <= 2 :
+        return None
     left = 0
     right = len(A)-1
+
     while right > left +1:
+
         mid = (left + right)//2
         if A[mid] > A[mid-1] and A[mid] > A[mid+1]:
             return A[mid]
@@ -17,6 +17,7 @@ def find_max_unimodal_array(A):
             left = mid
         else:
             right = mid
+
     return None
 
 

src/searching_and_sorting/searching/find_sqrt_bin_search.py

@@ -1,14 +1,11 @@
-#!/usr/bin/python
+#!/usr/bin/env python
-
-__author__ = "Mari Wahl"
-__email__ = "marina.w4hl@gmail.com"
-
 
+__author__ = "bt3"
 
+''' implement square root using binary search '''
 
 
 def find_sqrt_bin_search(n, error=0.001):
-    ''' implement square root using binary search '''
     lower = n < 1 and n or 1
     upper = n < 1 and 1 or n
     mid = lower + (upper - lower) / 2.0

src/searching_and_sorting/searching/find_str_array_with_empty_str.py

@@ -1,20 +1,22 @@
-#!/usr/bin/python
+#!/usr/bin/env python
 
-__author__ = "Mari Wahl"
+__author__ = "bt3"
-__email__ = "marina.w4hl@gmail.com"
 
 
-''' Given a sorted an array with empty strings, we use binary search to find some string (since
+''' Given a sorted an array with empty strings,
-    the list is sorted):
+    we use binary search to find some string (since the list is sorted):
-    --> we deal with the empty strings with strip and then run to left and right, or move
+    --> we deal with the empty strings with strip and then run to left
-    mid to the closed non-empty str (remember that the index must be conserved):
+    and right, or move mid to the closed non-empty str (remember that
+    the index must be conserved):
 '''
 
 
 def find_str_array_with_empty_str(seq, s1):
-    if not seq or not s1: return None
+    if not seq or not s1:
+        return None
     hi = len(seq)
     lo = 0
+
     while hi > lo:
         mid = (hi+lo)//2
 
@@ -32,9 +34,12 @@ def find_str_array_with_empty_str(seq, s1):
                 right += 1
                 left -= 1
 
-        if s1 == seq[mid] == s1: return mid
+        if s1 == seq[mid] == s1:
-        elif s1 < seq[mid]: hi = mid
+            return mid
-        else: lo = mid + 1
+        elif s1 < seq[mid]:
+            hi = mid
+        else:
+            lo = mid + 1
 
 
 

src/searching_and_sorting/searching/find_time_occurence_list.py

@@ -1,8 +1,6 @@
-#!/usr/bin/python
+#!/usr/bin/env python
-
-__author__ = "Mari Wahl"
-__email__ = "marina.w4hl@gmail.com"
 
+__author__ = "bt3"
 
 def binary_serch_counting(lst1, k, lo=0, hi=None):
     if hi is None: hi = len(lst1)
@@ -19,10 +17,12 @@ def binary_serch_counting(lst1, k, lo=0, hi=None):
 
 
 def find_time_occurrence_list(seq, k):
-    """ find how many times a k element appears in a sorted list. One way of doing this is using
+    """ find how many times a k element appears in a sorted list.
-    collections.OrderedDict to no mess with the sorting, and add entries for every count. This
+    One way of doing this is using collections.OrderedDict to no
-    should be O(n). It has a O(1) space complexity since the size of the dict is fixed.
+    mess with the sorting, and add entries for every count. This
-    Another way, since the array is sorted, it to use binary search, since this is only O(logn).
+    should be O(n). It has a O(1) space complexity since the size of
+    the dict is fixed. Another way, since the array is sorted, it to
+    use binary search, since this is only O(logn).
     """
     index_some_k = binary_serch_counting(seq, k)
     count = 1

src/searching_and_sorting/searching/interserction_two_arrays.py

@@ -1,80 +0,0 @@
-#!/usr/bin/python
-
-__author__ = "Mari Wahl"
-__email__ = "marina.w4hl@gmail.com"
-
-
-
-''' using sets '''
-
-def intersection_two_arrays_sets(seq1, seq2):
-    ''' find the intersection of two arrays using set proprieties '''
-    set1 = set(seq1)
-    set2 = set(seq2)
-    return set1.intersection(set2)  #same as list(set1 & set2
-
-
-
-''' using merge sort '''
-
-def intersection_two_arrays_ms(seq1, seq2):
-    ''' find the intersection of two arrays using merge sort '''
-    res = []
-    while seq1 and seq2:
-        if seq1[-1] == seq2[-1]:
-            res.append(seq1.pop())
-            seq2.pop()
-        elif seq1[-1] > seq2[-1]:
-            seq1.pop()
-        else:
-            seq2.pop()
-    res.reverse()
-    return res
-
-
-
-
-''' using binary search '''
-
-def binary_search(seq, key, lo=0, hi=None):
-    ''' binary search iterative algorithm '''
-    hi = hi or len(seq)
-    while lo < hi:
-        mid = (hi+lo) // 2
-        if seq[mid] == key:
-            return True
-        elif key > seq[mid]:
-            lo = mid + 1
-        else:
-            hi = mid
-    return None
-
-def intersection_two_arrays_bs(seq1, seq2):
-    ''' if A small and B is too large, we can do a binary search on each entry in B '''
-    ''' only works if sorted and the small sequence has not larger nmbers!!!'''
-    if len(seq1) > len(seq2): seq, key = seq1, seq2
-    else: seq, key = seq2, seq1
-
-    intersec = []
-    for item in key:
-        if binary_search(seq, item):
-            intersec.append(item)
-    return intersec
-
-
-
-def test_intersection_two_arrays(module_name='this module'):
-    seq1 = [1,2,3,5,7,8]
-    seq2 = [3,5,6]
-
-    assert(set(intersection_two_arrays_sets(seq1,seq2)) == set([3,5]))
-    assert(intersection_two_arrays_bs(seq1,seq2) == [3,5])
-    assert(intersection_two_arrays_ms(seq1,seq2) == [3,5])
-
-    s = 'Tests in {name} have {con}!'
-    print(s.format(name=module_name, con='passed'))
-
-
-if __name__ == '__main__':
-    test_intersection_two_arrays()
-

src/searching_and_sorting/searching/ordered_sequential_search.py

@@ -1,8 +1,6 @@
-#!/usr/bin/python
+#!/usr/bin/env python
-
-__author__ = "Mari Wahl"
-__email__ = "marina.w4hl@gmail.com"
 
+__author__ = "bt3"
 
 
 

src/searching_and_sorting/searching/quick_select.py

@@ -1,12 +1,10 @@
+#!/usr/bin/env python
 
-#!/usr/bin/python
+__author__ = "bt3"
 
-__author__ = "Mari Wahl"
-__email__ = "marina.w4hl@gmail.com"
 
 import random
 
-
 ''' The simplest way...'''
 def quickSelect(seq, k):
     # this part is the same as quick sort

src/searching_and_sorting/searching/searching_in_a_matrix.py

@@ -1,8 +1,6 @@
-#!/usr/bin/python
+#!/usr/bin/env python
-
-__author__ = "Mari Wahl"
-__email__ = "marina.w4hl@gmail.com"
 
+__author__ = "bt3"
 
 
 import numpy

src/searching_and_sorting/searching/sequential_search.py

@@ -1,8 +1,6 @@
-#!/usr/bin/python
+#!/usr/bin/env python
-
-__author__ = "Mari Wahl"
-__email__ = "marina.w4hl@gmail.com"
 
+__author__ = "bt3"
 
 
 def sequential_search(seq, n):

src/searching_and_sorting/sorting/bubble_sort.py

@@ -7,8 +7,6 @@ def bubble_sort(seq):
     """
     Implementation of bubble sort.
     O(n2) and thus highly ineffective.
-    :param seq: the sequence to be sorted.
-    :return: the sorted sequence.
     """
     size = len(seq) -1
     for num in range(size, 0, -1):
@@ -24,9 +22,6 @@ def test_bubble_sort(module_name='this module'):
     seq = [4, 5, 2, 1, 6, 2, 7, 10, 13, 8]
     assert(bubble_sort(seq) == sorted(seq))
 
-    s = 'Tests in {name} have {con}!'
-    print(s.format(name=module_name, con='passed'))
-
 
 if __name__ == '__main__':
     test_bubble_sort()

src/searching_and_sorting/sorting/count_sort.py

@@ -19,7 +19,6 @@ def count_sort_dict(a):
 def test_count_sort():
     seq = [3, 5, 2, 6, 8, 1, 0, 3, 5, 6, 2, 5, 4, 1, 5, 3]
     assert(count_sort_dict(seq) == sorted(seq))
-    print('Tests passed!')
 
 
 if __name__ == '__main__':

src/searching_and_sorting/sorting/gnome_sort.py

@@ -18,7 +18,6 @@ def gnome_sort(seq):
 def test_gnome_sort():
     seq = [3, 5, 2, 6, 8, 1, 0, 3, 5, 6, 2, 5, 4, 1, 5, 3]
     assert(gnome_sort(seq) == sorted(seq))
-    print('Tests passed!')
 
 
 if __name__ == '__main__':

src/searching_and_sorting/sorting/insertion_sort.py

@@ -31,7 +31,6 @@ def test_insertion_sort():
     seq = [3, 5, 2, 6, 8, 1, 0, 3, 5, 6, 2, 5, 4, 1, 5, 3]
     assert(insertion_sort(seq) == sorted(seq))
     assert(insertion_sort_rec(seq) == sorted(seq))
-    print('Tests passed!')
 
 
 if __name__ == '__main__':

src/searching_and_sorting/sorting/merge_and_sort_two_arrays.py

@@ -2,7 +2,6 @@
 
 __author__ = "bt3"
 
-
 '''
 You have two arrays with N integers in them. Merge those arrays using a
 recursive algorithm so that the integers in the final array are sorted.

src/searching_and_sorting/sorting/merge_sort.py

@@ -3,56 +3,12 @@
 __author__ = "bt3"
 
 
-
-
-''' Some examples of how to implement Merge Sort in Python.
-    --> RUNTIME: WORST/BEST/AVERAGE Is O(nlogn)
-    --> space complexity is O(n) for arrays
-    --> in general not in place, good for large arrays
-    --> In the case of two arrays: we can merge two arrays using the merge function from the merge sort
-    --> we can do this for files too, merging each two
-
-        1) If we can modify the arrays (pop) we can use:
-            def merge(left, right):
-                if not left or not right: return left or right # nothing to be merged
-                result = []
-                while left and right:
-                    if left[-1] >= right[-1]:
-                        result.append(left.pop())
-                    else:
-                        result.append(right.pop())
-                result.reverse()
-                return (left or right) + result
-
-
-        2) If we can't modify or we want to in place, we need two pointers:
-        >>> l1 = [1, 2, 3, 4, 5, 6, 7]
-        >>> l2 = [2, 4, 5, 8]
-        >>> merge(l1, l2)
-        [1, 2, 2, 3, 4, 4, 5, 5, 6, 7, 8]
-
-
-        3) For example, in the case we have a big array filled 0s in the end, and another array with the size of the number of 0s:
-        >>> l1 = [1, 2, 3, 4, 5, 6, 7, 0, 0, 0, 0]
-        >>> l2 = [2, 4, 5, 8]
-        >>> merge_two_arrays_inplace(l1, l2)
-        [1, 2, 2, 3, 4, 4, 5, 5, 6, 7, 8]
-
-
-        4) If we want to merge sorted files (and we have plenty of RAM to load all files):
-        >>> list_files = ['1.dat', '2.dat', '3.dat']
-        >>> merge_files(list_files)
-        [1, 1, 2, 3, 3, 3, 4, 5, 5, 5, 6, 7, 8]
-'''
-
-
-
-
-"""
-    The typical example...
-"""
-
 def merge_sort(seq):
+    '''
+    >>> seq = [3, 5, 2, 6, 8, 1, 0, 3, 5, 6, 2]
+    >>> merge_sort(seq)
+    [0, 1, 2, 2, 3, 3, 5, 5, 6, 6, 8]
+    '''
     if len(seq) < 2:
         return seq
     mid = len(seq)//2
@@ -73,22 +29,21 @@ def merge_sort(seq):
 
 
 
-
+# separating the merge part in another function
-'''
-    We could also divide this sort into two parts, separating
-    the merge part in another function
-'''
-
 def merge_sort_sep(seq):
+    '''
+    >>> seq = [3, 5, 2, 6, 8, 1, 0, 3, 5, 6, 2]
+    >>> merge_sort_sep(seq)
+    [0, 1, 2, 2, 3, 3, 5, 5, 6, 6, 8]
+    '''
     if len(seq) < 2 :
-        return seq    # base case
+        return seq
     mid = len(seq)//2
     left = merge_sort(seq[:mid])
     right = merge_sort(seq[mid:])   # notice that mid is included!
     return merge(left, right)      # merge iteratively
 
 
-
 def merge(left, right):
     if not left or not right:
         return left or right # nothing to be merged
@@ -106,43 +61,6 @@ def merge(left, right):
     return result
 
 
-
-
-''' The following merge functions is O(2n) and
-    illustrate many features in Python that '''
-def merge_2n(left, right):
-    if not left or not right: return left or right # nothing to be merged
-    result = []
-    while left and right:
-        if left[-1] >= right[-1]:
-            result.append(left.pop())
-        else:
-            result.append(right.pop())
-    result.reverse()
-    return (left or right) + result
-
-
-
-''' Merge two arrays in place '''
-def merge_two_arrays_inplace(l1, l2):
-    if not l1 or not l2: return l1 or l2 # nothing to be merged
-    p2 = len(l2) - 1
-    p1 = len(l1) - len(l2) - 1
-    p12 = len(l1) - 1
-    while p2 >= 0 and p1 >= 0:
-        item_to_be_merged = l2[p2]
-        item_bigger_array = l1[p1]
-        if item_to_be_merged < item_bigger_array:
-            l1[p12] = item_bigger_array
-            p1 -= 1
-        else:
-            l1[p12] = item_to_be_merged
-            p2 -= 1
-        p12 -= 1
-    return l1
-
-
-
 ''' Merge sort for files '''
 def merge_files(list_files):
     result = []
@@ -159,16 +77,6 @@ def merge_files(list_files):
     return final
 
 
-
-
-
-def test_merge_sort():
-    seq = [3, 5, 2, 6, 8, 1, 0, 3, 5, 6, 2]
-    seq_sorted = sorted(seq)
-    assert(merge_sort(seq) == seq_sorted)
-    assert(merge_sort_sep(seq) == seq_sorted)
-    print('Tests passed!')
-
-
 if __name__ == '__main__':
-    test_merge_sort()
+    import doctest
+    doctest.testmod()

src/searching_and_sorting/sorting/quick_sort.py

@@ -3,38 +3,23 @@
 __author__ = "bt3"
 
 
-''' Some examples of how to implement Quick Sort in Python
+def qs(array):
-    --> RUNTIME: BEST/AVERAGE Is O(nlogn), WORST is O(n2)
-    --> the first example is not in place, the second is in place
-    --> test with two element arrays, identical values
-
-    Quick sort in place:
-    1) select pivot as the index = 0
-    2) start pointer1 at index = 1 and pointer2 in the last element
-    3) while pointer1 < pointer2:
-        if value in pointer1 <= pivot
-            swap value in pointer1 with value in pointer2 and advanced pointer2
-        else
-            advance pointer1
-    4) now the array is like this:
-        [pivot, larger than pivot, smaller than pivot]
-    5) swap the pivot where pointer 1 stop
-    6) do recursively for [smaller] + [pivot] + [larger]
     '''
+    >>> qs([4,1,6,2,7,9,3])
+    [1, 2, 3, 4, 6, 7, 9]
+    '''
+    if len(array) < 2:
+        return array
+
+    piv = len(array)//2
+    piv_element = array[piv]
+    new_array = array[:piv] + array[piv+1:]
+
+    left  = [a for a in new_array if a <= piv_element]
+    right = [a for a in new_array if a > piv_element]
 
 
-
+    return qs(left) + [array[piv]] + qs(right)
-
-def quick_sort(seq):
-    if len(seq) < 2 : return seq
-    mid = len(seq)//2
-    pi = seq[mid]
-    seq = seq[:mid] + seq[mid+1:]
-    left = quick_sort([x for x in seq if x <= pi])    # REMEMBER TO INCLUDE X  (OR IN RIGHT)
-    right = quick_sort([x for x in seq if x > pi])
-    return left + [pi] + right
-
-
 
 
 """ we can also divide them into two functions """
@@ -45,39 +30,20 @@ def partition(seq):
     return lo, pi, hi
 
 def quick_sort_divided(seq):
-    if len(seq) < 2: return seq
+    '''
+    >>> quick_sort_divided([4,1,6,2,7,9,3])
+    [1, 2, 3, 4, 6, 7, 9]
+    '''
+    if len(seq) < 2:
+        return seq
     lo, pi, hi = partition(seq)
     return quick_sort_divided(lo) + [pi] + quick_sort_divided(hi)
 
 
-''' quick_sort in place '''
-def quick_sort_in(seq):
-    if len(seq) < 2 : return seq
-    if len(seq) == 2 and seq[0] > seq[1]:
-        seq[0], seq[1] = seq[1], seq[0]   # problems when only 2 elements because of swap
-    pivot = seq[0]  # start at the ends because we don't know how many elements
-    p1, p2 = 1, len(seq) -1   # set pointers at both ends
-    while p1 < p2: # must be < or out of range
-        if seq[p1] <= pivot: # must be <= because of pivot swap
-                seq[p1], seq[p2] = seq[p2], seq[p1]
-                p2 -= 1
-        else:
-            p1 += 1
-    seq[0], seq[p1] = seq[p1], pivot
-    return quick_sort_in(seq[p1+1:]) + [seq[p1]] + quick_sort_in(seq[:p1])
 
 
 
 
-def test_quick_sort():
-    seq = [3, 5, 2, 6, 8, 1, 0, 3, 5, 6, 2]
-    assert(quick_sort(seq) == sorted(seq))
-    assert(quick_sort_divided(seq) == sorted(seq))
-    print('Tests passed!')
-
-
 if __name__ == '__main__':
-    test_quick_sort()
+    import doctest
-
+    doctest.testmod()
-
-

src/searching_and_sorting/sorting/selection_sort.py

@@ -17,7 +17,6 @@ def selection_sort(seq):
 def test_selection_sort():
     seq = [3, 5, 2, 6, 8, 1, 0, 3, 5, 6, 2]
     assert(selection_sort(seq) == sorted(seq))
-    print('Tests passed!')
 
 
 if __name__ == '__main__':

src/searching_and_sorting/sorting/sort_anagrams_together.py

@@ -3,33 +3,31 @@
 __author__ = "bt3"
 
 
-''' A method to sort an array so that all the anagrams are together. Since we only
+''' A method to sort an array so that all the anagrams are together.
-    want the anagrams to be grouped, we can use a dictionary for this task. This
+    Since we only want the anagrams to be grouped, we can use a
-    algorithm is O(n).
+    dictionary for this task. This algorithm is O(n).
-    >>> l1 = ['hat', 'ball', 'tha', 'cut', 'labl', 'hta', 'cool', 'cuy', 'uct']
-    >>> sort_anagrams_together(l1)
-    ['cut', 'uct', 'cool', 'ball', 'labl', 'hat', 'tha', 'hta', 'cuy']
 '''
 
 from collections import defaultdict
+
 def sort_anagrams_together(l1):
+    '''
+    >>> l1 = ['hat', 'ball', 'tha', 'cut', 'labl', 'hta', 'cool', 'cuy', 'uct']
+    >>> sort_anagrams_together(l1)
+    ['cuy', 'cut', 'uct', 'cool', 'ball', 'labl', 'hat', 'tha', 'hta']
+    '''
     result = []
 
-    # step 1 save the anagrams together
+    dict_aux = defaultdict(list)
-    dict_aux = defaultdict(list) # rememebr to indicate the type
     for word in l1:
-        key = ''.join(sorted(word)) # need to sort the strings and join it
+        key = ''.join(sorted(word))
-        dict_aux[key].append(word) # because only sorted give a list of each char
+        dict_aux[key].append(word)
 
-    # step 2 print the anagrams. Note that if you want everything
-    # sorted you would have to sort the keys and insert the angrams after that
     for key in dict_aux:
         result.extend(dict_aux[key])
 
     return result
 
-
-
 if __name__ == '__main__':
     import doctest
     doctest.testmod()