Added parallel merge sort (#177)

Author: czgdp1807

pydatastructs/linear_data_structures/__init__.py

@@ -2,7 +2,8 @@
 
 from . import (
     arrays,
-    linked_lists
+    linked_lists,
+    algorithms
 )
 
 from .arrays import (
@@ -18,3 +19,8 @@
     DoublyCircularLinkedList
 )
 __all__.extend(linked_lists.__all__)
+
+from .algorithms import (
+    merge_sort_parallel
+)
+__all__.extend(algorithms.__all__)

pydatastructs/linear_data_structures/algorithms.py

@@ -0,0 +1,95 @@
+from pydatastructs.linear_data_structures.arrays import (
+    OneDimensionalArray, DynamicArray)
+from pydatastructs.utils.misc_util import _check_type
+from concurrent.futures import ThreadPoolExecutor
+from math import log, floor
+
+__all__ = [
+    'merge_sort_parallel'
+]
+
+def _merge(array, sl, el, sr, er, end):
+    l, r = [],  []
+    for i in range(sl, el + 1):
+        if (i <= end and
+            array[i] is not None):
+            l.append(array[i])
+            array[i] = None
+    for i in range(sr, er + 1):
+        if (i <= end and
+            array[i] is not None):
+            r.append(array[i])
+            array[i] = None
+    i, j, k = 0, 0, sl
+    while i < len(l) and j < len(r):
+        if l[i] <= r[j]:
+            array[k] = l[i]
+            i += 1
+        else:
+            array[k] = r[j]
+            j += 1
+        k += 1
+
+    while i < len(l):
+        array[k] = l[i]
+        i += 1
+        k += 1
+
+    while j < len(r):
+        array[k] = r[j]
+        j += 1
+        k += 1
+
+def merge_sort_parallel(array, num_threads, **kwargs):
+    """
+    Implements parallel merge sort.
+
+    Parameters
+    ==========
+
+    array: Array
+        The array which is to be sorted.
+    num_threads: int
+        The maximum number of threads
+        to be used for sorting.
+    start: int
+        The starting index of the portion
+        which is to be sorted.
+        Optional, by default 0
+    end: int
+        The ending index of the portion which
+        is to be sorted.
+        Optional, by default the index
+        of the last position filled.
+
+    Examples
+    ========
+
+    >>> from pydatastructs import OneDimensionalArray, merge_sort_parallel
+    >>> arr = OneDimensionalArray(int,[3, 2, 1])
+    >>> merge_sort_parallel(arr, 3)
+    >>> [arr[0], arr[1], arr[2]]
+    [1, 2, 3]
+
+    References
+    ==========
+
+    .. [1] https://en.wikipedia.org/wiki/Merge_sort
+    """
+    start = kwargs.get('start', 0)
+    end = kwargs.get('end', array._size - 1)
+    for size in range(floor(log(end - start + 1, 2)) + 1):
+        pow_2 = 2**size
+        with ThreadPoolExecutor(max_workers=num_threads) as Executor:
+            i = start
+            while i <= end:
+                Executor.submit(
+                    _merge,
+                    array,
+                    i, i + pow_2 - 1,
+                    i + pow_2, i + 2*pow_2 - 1,
+                    end).result()
+                i = i + 2*pow_2
+
+    if _check_type(array, DynamicArray):
+        array._modify(force=True)

pydatastructs/linear_data_structures/arrays.py

@@ -209,12 +209,12 @@ def __new__(cls, dtype=NoneType, *args, **kwargs):
         obj._last_pos_filled = obj._num - 1
         return obj
 
-    def _modify(self):
+    def _modify(self, force=False):
         """
         Contracts the array if Num(T)/Size(T) falls
         below load factor.
         """
-        if self._num/self._size < self._load_factor:
+        if (self._num/self._size < self._load_factor) or force:
             arr_new = OneDimensionalArray(self._dtype, 2*self._num + 1)
             j = 0
             for i in range(self._last_pos_filled + 1):
@@ -231,14 +231,12 @@ def append(self, el):
             for i in range(self._last_pos_filled + 1):
                 arr_new[i] = self[i]
             arr_new[self._last_pos_filled + 1] = el
-            self._last_pos_filled += 1
             self._size = arr_new._size
-            self._num += 1
             self._data = arr_new._data
         else:
             self[self._last_pos_filled + 1] = el
-            self._last_pos_filled += 1
-            self._num += 1
+        self._last_pos_filled += 1
+        self._num += 1
         self._modify()
 
     def delete(self, idx):

pydatastructs/linear_data_structures/tests/test_algorithm.py

@@ -0,0 +1,33 @@
+from pydatastructs import (
+    merge_sort_parallel, DynamicOneDimensionalArray,
+    OneDimensionalArray)
+import random
+
+def test_merge_sort_parallel():
+
+    random.seed(1000)
+
+    n = random.randint(10, 20)
+    arr = DynamicOneDimensionalArray(int, 0)
+    for _ in range(n):
+        arr.append(random.randint(1, 1000))
+    for _ in range(n//3):
+        arr.delete(random.randint(0, n//2))
+    expected_arr = [686, 779, 102, 134, 362,
+                    448, 480, 548, 228, 688,
+                    247, 373, 696, None, None,
+                    None, None, None, None,
+                    None, None, None, None,
+                    None, None, None, None]
+    merge_sort_parallel(arr, 5, start=2, end=10)
+    assert arr._data == expected_arr
+
+    n = random.randint(10, 20)
+    arr = OneDimensionalArray(int, n)
+    for i in range(n):
+        arr[i] = random.randint(1, 1000)
+    expected_arr = [42, 695, 147, 500, 768,
+                    998, 473, 732, 728, 426,
+                    709, 910]
+    merge_sort_parallel(arr, 5, start=2, end=5)
+    assert arr._data == expected_arr