Huffuman編碼與譯碼C語言實現
一、Huffman編碼
Huffuman編碼的過程主要分為兩步，第一步根據字元出現的權值構建Huffuman樹，第二步，遍歷Huffuman樹找出對應字元的編碼
1. 構建Huffuman樹程式碼如下：
typedef struct iNode {
int weight;
int parent, lchild, rchild;
} huffman_tree, *phuffman_tree;
int sum_weight(int *weight, int len)//計算權值之和
{
int sum = 0;
for(int i = 0; i < len; i++)
sum += weight[i];
return sum;
}
void select(huffman_tree *ht, int len, int &s1, int &s2)//搜尋ht陣列中parent為0的最小兩個節點，分別存入s1，s2，len為要搜尋的長度
{
s1 = s2 = 0;
for(int i = 1; i <= len; i++) {
if(ht[i].parent == 0) {
if(ht[s1].weight >= ht[i].weight) {
s1 = i;
s2 = s1;
}
}
}
}
void make_tree(int *weight, huffman_tree* ht, int len)//weight是字元的權重，ht是儲存二叉樹的節點陣列
{
int m = 2*len - 1;
int s1 = 0, s2 = 0;
ht[0] = {sun_weight(weight, len) + 1, 0, 0, 0};//0號節點權設定為總權值加1便於select時設定比較初值
for(int i = 1; i <= len; i++)
ht[i] = {weight[i - 1], 0, 0, 0};
for( ; i <= m; i++)
ht[i] = {0, 0, 0, 0};
for(i = n + 1; i <= m; i++) {
select(ht, i - 1, s1, s2);
ht[s1].parent = i;
ht[s2].parent = i;
ht[i].lchild = s1;
ht[i].rchild = s2;
ht[i].weight = ht[s1].weight + ht[i].weight;
}
}
1. 構建Huffuman編碼
這裡有兩種構建方法，第一種是由葉子到根的構建，程式碼如下：
vodi make_code_from_leaf(huffman_tree *ht, int len, huffman_code *hc)//從葉子結點到根的編碼過程
{
char code[len];
//int start = len;
//code[len - 1] = "\0";
for (int i = 1; i <= len; i++) {
int start = len;
code[--start] = "\0";
int next = i;
while (ht[next].parent ! = 0) {
if (ht[ht[next].parent].lchild = next)
code[--start] = "0";
else
code[--start] = "1";
next = ht[next].parent;
}
strncpy(hc[i], &code[start], len - start);
}
}
從根節點開始遍歷求huffuman編碼的程式碼如下（注意weight這裡發揮標記的作用：
void make_code_from_root(huffman_tree *ht, int len, huffman_code *hc)
{
char code[len];
int clen = 0;
int p = 2*len - 1;
for ( int i = 1; i <= p ; i++)
ht[i].weight = 0;
while (p) {
if(ht[p].weight == 0) {//左節點遍歷
ht[p].weight = 1;
if(ht[p].lchild != 0) {
p = ht[p].lchild;
code[clen++] = "0";
}
else if(ht[p].rchild == 0) {
code[clen++] = "\0";
hc[p] = (char *)malloc(clen + 1);
strcpy([hc[p], code);
}
}
else if (ht[p].weight == 1) {//右子樹遍歷
ht[p].weight = 2;
if (ht[p].rchild != 0) {
code[clen++] = "1";
p = ht[p].rchild;
}
}
else { //到達葉子結點回溯另一分支
ht[p].weight = 0;
p = ht[p].parent;
clen--;
}
}
}
void make_code_from_root(huffman_tree *ht, int len, huffman_code *hc)
{
char code[len];
int clen = 0;
int p = 2*len - 1;
for ( int i = 1; i <= p ; i++)
ht[i].weight = 0;
while (p) {
if(ht[p].weight == 0) {
ht[p].weight = 1;
if(ht[p].lchild != 0) {
p = ht[p].lchild;
code[clen++] = "0";
}
else if(ht[p].rchild == 0) {
code[clen++] = "\0";
hc[p] = (char *)malloc(clen + 1);
strcpy([hc[p], code);
}
}
else if (ht[p].weight == 1) {
ht[p].weight = 2;
if (ht[p].rchild != 0) {
code[clen++] = "1";
p = ht[p].rchild;
}
}
else {
ht[p].weight = 0;
p = ht[p].parent;
clen--;
}
}
}
void make_code_from_root(huffman_tree *ht, int len, huffman_code *hc)
{
char code[len];
int clen = 0;
int p = 2*len - 1;
for ( int i = 1; i <= p ; i++)
ht[i].weight = 0;
while (p) {
if(ht[p].weight == 0) {
ht[p].weight = 1;
if(ht[p].lchild != 0) {
p = ht[p].lchild;
code[clen++] = "0";
}
else if(ht[p].rchild == 0) {
code[clen++] = "\0";
hc[p] = (char *)malloc(clen + 1);
strcpy([hc[p], code);
}
}
else if (ht[p].weight == 1) {
ht[p].weight = 2;
if (ht[p].rchild != 0) {
code[clen++] = "1";
p = ht[p].rchild;
}
}
else {
ht[p].weight = 0;
p = ht[p].parent;
clen--;
}
}
}
一、Huffman譯碼
根據Huffuman編碼規則的規則可知，如果所有左子樹代表“0”，右子樹代表“1”則，可以用搜索碼中的“0”來作為字元的分割碼，但是要注意最後的全1碼的識別其程式碼如下：
int select_hc(huffman_code *hc, huffman_code s_hc, int len)//根據獲得的編碼與Huffuman編碼表進行比較獲取該碼字在碼錶中的位置
{
for (int i = 0; i < len; i++)
if (strcmp(hc[i], s_hc) == 0)
return i;
return -1;
}
/*******************encode function**********************
********************input : code is 0 an 1 serial*************************
********************hc is huffman code table, chars is the encoded characters, hc_len is char sum, len is code length****/
void decode(char *code, huffuman_code *hc, char *chars, char *encode_char, int hc_len, int len)//
{
char temp_code[10];
for (int i = 0, j = 0, k = 0; i < len; i++) {
if (code[i] == "0") {
temp_code[k++] = "0";
temp_code[k++] = "\0";
int select_pos = select_hc(hc, temp_code, hc_len);
if (select_pps >= 0)
encode_char[j++] = chars[select_pos];
else
printf("error code : %s invalid", temp_code);
k = 0;
}
else {
temp_code[k++] = "1";
if (strcmp(temp_code, hc[hc_len - 1]) == 0) {//注意處理“1”
temp_code[k++] = "\0";
encode_char[j++] = chars[hc_len - 1];
k = 0;
}
}
}
}