迄今为止见过的最方便的huffman编码，效率很高

一个外国人写的，很具有研究价值

Constructing a Huffman Tree according to the number of times each symbol appears in the data stream:

1)　　Create an array of N nodes, representing N possible symbols (ranging between 0 and N-1).

2)　　Set the value of each node to the number of times that its symbol appears in the data stream.

3)　　Create a Minimum-Heap of N nodes.

4)　　Add every node whose value is greater than zero to the heap.

5)　　Extract the best two nodes in the heap.

6)　　Create a parent node whose children are the two extracted nodes.

7)　　Add the parent node to the heap.

8)　　Repeat the previous three steps N-1 times (until only 1 node remains in the heap).

9)　　Extract the last node in the heap.

The array created in the first step stores the leaves of the tree, and is used in order to encode the data stream.

The node extracted in the last step is in fact the root of the tree, and is used in order to decode the data stream.-So far seen the most convenient huffman coding, efficient write a foreigner, I have research value Constructing a Huffman Tree according to the number of times each symbol appears in the data stream: 1) Create an array of N nodes , representing N possible symbols (ranging between 0 and N-1) .2) Set the value of each node to the number of times that its symbol appears in the data stream.3) Create a Minimum-Heap of N nodes.4) Add every node whose value is greater than zero to the heap.5) Extract the best two nodes in the heap.6) Create a parent node whose children are the two extracted nodes.7) Add the parent node to the heap.8) Repeat the previous three steps N-1 times (until only 1 node remains in the heap) .9) Extract the last node in the heap.The array created in the first step stores the leaves of the tree, and is used in order to encode the data stream.The node extracted in the last step is in fact the root of the tree, and is used in order to decode the data stream.