suanfa/public/c/ch4/huffman/halfall.c

#include <stdio.h>
#include <stdlib.h>
#include <string.h>

#define MAX_TREE_HT 256

// 定义 Huffman 树的节点结构
typedef struct Node {
    int weight;           // 节点的权重，即字符的频率
    char data;            // 存储字符
    struct Node *left, *right;  // 左右子节点
} Node;

// 比较函数，用于 qsort 按照权重排序
int compare(const void* a, const void* b) {
    return (*(Node**)a)->weight - (*(Node**)b)->weight;
}

// 创建一个新的节点
Node* newNode(char data, int weight) {
    Node* temp = (Node*)malloc(sizeof(Node));    
    temp->left = temp->right = NULL;
    temp->data = data;
    temp->weight = weight;
    return temp;
}

// 构建 Huffman 树
Node* buildHuffmanTree(char data[], int weight[], int size) {
    // 创建一个指向节点的数组
    Node** nodeArr = (Node**)malloc(size * sizeof(Node*));
    // 将每个字符及其权重转化为节点并存入数组
    for (int i = 0; i < size; ++i) {
        nodeArr[i] = newNode(data[i], weight[i]);
    }
    // 使用 qsort 按照权重排序节点
    qsort(nodeArr, size, sizeof(Node*), compare);
    // 构建 Huffman 树
    while (size > 1) {
        // 取出两个最小的节点
        Node* left = nodeArr[0];
        Node* right = nodeArr[1];
        // 创建一个新的父节点，权重为两个子节点的权重之和
        Node* parent = newNode('\0', left->weight + right->weight);
        parent->left = left;
        parent->right = right;
        // 将父节点插入到 heap 中
        nodeArr[0] = parent;
        nodeArr[1] = nodeArr[size - 1];  // 将最后一个节点放到第二个位置
        size--;
        // 重新排序数组
        qsort(nodeArr, size, sizeof(Node*), compare);
    }
    Node* root = nodeArr[0];
    free(nodeArr);
    return root;
}

// 打印 Huffman 编码
void printHuffmanCodes(Node* root, int arr[], int top, char* codes[]) {
    if (root->left) {
        arr[top] = 0;
        printHuffmanCodes(root->left, arr, top + 1, codes);
    }

    if (root->right) {
        arr[top] = 1;
        printHuffmanCodes(root->right, arr, top + 1, codes);
    }

    // 如果是叶节点，打印字符和对应的编码
    if (!root->left && !root->right) {
        codes[root->data] = (char*)malloc((top + 1) * sizeof(char));
        if (!codes[root->data]) {
            fprintf(stderr, "Memory allocation failed\n");
            exit(1);
        }
        for (int i = 0; i < top; ++i)
            codes[root->data][i] = '0' + arr[i];
        codes[root->data][top] = '\0';
        printf("%c: %s\n", root->data, codes[root->data]);
    }
}

// 获取字符的 Huffman 编码
const char* getHuffmanCode(Node* root, char ch, int arr[], int top, char* codes[]) {
    if (codes[ch])
        return codes[ch];

    if (root->left) {
        arr[top] = 0;
        const char* code = getHuffmanCode(root->left, ch, arr, top + 1, codes);
        if (code)
            return code;
    }

    if (root->right) {
        arr[top] = 1;
        const char* code = getHuffmanCode(root->right, ch, arr, top + 1, codes);
        if (code)
            return code;
    }

    // 如果是叶节点，检查是否是我们要找的字符
    if (!root->left && !root->right && root->data == ch) {
        codes[root->data] = (char*)malloc((top + 1) * sizeof(char));
        if (!codes[root->data]) {
            fprintf(stderr, "Memory allocation failed\n");
            exit(1);
        }
        for (int i = 0; i < top; ++i)
            codes[root->data][i] = '0' + arr[i];
        codes[root->data][top] = '\0';
        return codes[root->data];
    }

    return NULL;
}

// 编码函数：使用 Huffman 树编码文本
void encode(Node* root, const char* str, char* encodedStr, char* codes[]) {
    int arr[MAX_TREE_HT], top = 0;
    printf("Huffman Codes:\n");
    printHuffmanCodes(root, arr, top, codes);

    printf("\nEncoded Text: ");
    for (int i = 0; str[i] != '\0'; i++) {
        const char* code = getHuffmanCode(root, str[i], arr, 0, codes);
        if (code) {
            printf("%s", code);
            strcat(encodedStr, code);  // 将每个字符的编码拼接到最终的编码字符串中
        } else {
            fprintf(stderr, "Character '%c' not found in Huffman tree\n", str[i]);
            exit(1);
        }
    }
    printf("\n");
}

// 解码函数：从 Huffman 树解码编码文本
void decode(Node* root, const char* encodedStr) {
    Node* current = root;  // 从根节点开始
    printf("\nDecoded Text: ");
    
    for (int i = 0; encodedStr[i] != '\0'; i++) {
        // 根据编码字符串的每个字符决定树的遍历方向
        if (encodedStr[i] == '0') {
            current = current->left;  // 向左子节点移动
        } else if (encodedStr[i] == '1') {
            current = current->right; // 向右子节点移动
        }

        // 如果到达叶节点，输出字符并返回根节点
        if (!current->left && !current->right) {
            printf("%c", current->data);
            current = root;  // 重置为根节点，准备解码下一个字符
        }
    }
    printf("\n");
}

// 释放 Huffman 树的内存
void freeHuffmanTree(Node* node) {
    if (node == NULL)
        return;

    freeHuffmanTree(node->left);
    freeHuffmanTree(node->right);
    free(node);
}

// 释放 Huffman 编码的内存
void freeHuffmanCodes(char* codes[]) {
    for (int i = 0; i < 256; i++) {
        if (codes[i]) {
            free(codes[i]);
            codes[i] = NULL;
        }
    }
}

int main() {
    const char* text = "hello huffman coding";
    
    // 计算每个字符的频率
    int freq[256] = {0};
    for (int i = 0; text[i] != '\0'; i++) {
        freq[(unsigned char)text[i]]++;
    }

    // 构建字符和频率的数组
    char data[256];
    int frequencies[256];
    int size = 0;
    for (int i = 0; i < 256; i++) {
        if (freq[i] > 0) {
            data[size] = (char)i;
            frequencies[size] = freq[i];
            size++;
        }
    }

    // 构建 Huffman 树
    Node* root = buildHuffmanTree(data, frequencies, size);

    // 初始化 Huffman 编码数组
    char* codes[256] = {NULL};

    // 编码
    char encodedText[MAX_TREE_HT * strlen(text)];
    memset(encodedText, 0, sizeof(encodedText));
    encode(root, text, encodedText, codes);

    // 解码
    decode(root, encodedText);

    // 释放 Huffman 树的内存
    freeHuffmanTree(root);

    // 释放 Huffman 编码的内存
    freeHuffmanCodes(codes);

    return 0;
}