Some usage of Python heapq:

1. Getting nlargest, nsmallest from a list etc.

from heapq import nlargest, nsmallest
l = [('a',10), ('b',4)]
print(nlargest(2, l, key=lambda x: x[1]))

1. Heap as it is.

import heapq as heap  # deafult get min heap formation, for max heap just negate all values
print("Min heap test")
lsvalue = [3, -4, -1, 15, 6, 79, -1, 0, 3, 44, 9]
pq = lsvalue[:]
heap.heapify (pq)
heap.heappush (pq, -100)  # adding new val in heap
for _ in range (0, len (pq)):
print (heap.heappop (pq), end='  ')

# For max heap
print("\nMax heap test")
pq = [-x for x in lsvalue]
heap.heapify (pq)
heap.heappush (pq, -(-100))  # negating original input(-100) too

for _ in range (0, len (pq)):
print ((-1) * heap.heappop (pq), end='  ') # -1 * to get original vals

1. Customized value/Node based max heap.

import heapq
class Node:
def __init__(self, val, count):
self.val,self.count = val, count
def __lt__(self, other):
return self.count > other.count  # change < for min heap

lsnode = []
for i in range(10):
lsnode.append(Node(i,i))
pq = lsnode[:6] # creating another variable not to overrirde else  work on lsnode
heapq.heapify(pq)
for i in range(len(pq)):
print(heapq.heappop(pq).count)

1. Important APIs.

</pre> [‘heappush’, ‘heappop’, ‘heapify’, ‘heapreplace: pop then push’, ‘merge: merges 2 sorted list’, ‘nlargest’, ‘nsmallest’, ‘heappushpop’] </pre>