数据结构PTA 案例5-1.4 字符串关键字的散列映射

题目

给定一系列由大写英文字母组成的字符串关键字和素数P，用移位法定义的散列函数H(Key)将关键字Key中的最后3个字符映射为整数，每个字符占5位；再用除留余数法将整数映射到长度为P的散列表中。例如将字符串AZDEG插入长度为1009的散列表中，我们首先将26个大写英文字母顺序映射到整数0~25；再通过移位将其映射为3×32^​2+4×32+6=3206；然后根据表长得到，即是该字符串的散列映射位置。

发生冲突时请用平方探测法解决。

输入格式:
输入第一行首先给出两个正整数N（≤500）和P（≥2N的最小素数），分别为待插入的关键字总数、以及散列表的长度。第二行给出N个字符串关键字，每个长度不超过8位，其间以空格分隔。

输出格式:
在一行内输出每个字符串关键字在散列表中的位置。数字间以空格分隔，但行末尾不得有多余空格。

输入样例1:

4 11
HELLO ANNK ZOE LOLI

输出样例1:

3 10 4 0

输入样例2:

6 11
LLO ANNA NNK ZOJ INNK AAA

输出样例2:

3 0 10 9 6 1

解法

思路

1. 因为此题的输入数据是字符串，所以需要定义数据类型为EleType。用EleType a即可存储输入数据。

#define MAXL 8
typedef char EleType[MAXL+1];

2. 整个哈希表涉及的操作有：创建表，查找关键词，插入关键词。这里的散列函数HashFunction比整数型的要复杂一些，具体计算原理见下一点
3. HashFunction中应该首先计算key的字符个数，然后计算其最后三个字符相对于字符A的差值，再逐步进行移位操作（当然严格地来说做一个字符个数的判断，少于三个字符的key直接抛弃，但是因为题目会默认大于等于3，所以这里也没判断了），最后对Pos求余。
   
4. 凡是涉及比较，用strcmp，凡是涉及赋值，用strcpy

实现一
实现一是把每个字符串的长度也存在哈希表里了，但是这个不是必要的。

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

#define MAXL 8

typedef char EleType[MAXL+1];
typedef enum{legitimate, empty, deleted} EntryType;

typedef struct HashEntry Cell;
struct HashEntry
{
    EleType Data;
    EntryType Info;
    int Length;
};

typedef struct TblNode *HashTable;
struct TblNode
{
    int TableSize;
    Cell *Cells;
};

HashTable CreateTable(int TableSize)
{
    HashTable H = (HashTable)malloc(sizeof(struct TblNode));
    H->TableSize = TableSize;

    H->Cells = (Cell *)malloc(H->TableSize*sizeof(Cell));
    int i;
    for(i=0; i<H->TableSize; i++)
    {
        H->Cells[i].Data[0] = '\0';
        H->Cells[i].Info = empty;
        H->Cells[i].Length = 0;
    }
    return H;
}

int HashFunction(EleType key, int*Length, int TableSize)
{
    int i,tmpLength;
    unsigned int Pos = 0;
    tmpLength = 0;
    while(key[tmpLength] != '\0')
        tmpLength++;
    *Length = tmpLength;
    if(tmpLength>2)
    {
        for(i=tmpLength-3; i<tmpLength; i++)
            Pos = (Pos<<5) + (key[i]-'A');
        Pos = Pos % TableSize;
    }
    return Pos;
}

int SqFind(HashTable H, EleType key, int *Length)
{
    int CurPos, NewPos;
    Length = 0;     //Length is necessary?
    CurPos = NewPos = HashFunction(key, &Length, H->TableSize);

    int count = 0;
    while(H->Cells[NewPos].Info!=empty && strcmp(H->Cells[NewPos].Data, key))
    {
        if(++count %2)
        {
            NewPos = CurPos + (count+1)*(count+1)/4;
            if(NewPos >= H->TableSize)
                NewPos = NewPos % H->TableSize;

        }
        else
        {
            NewPos = CurPos - (count)*(count)/4;
            while(NewPos < 0)
                NewPos = NewPos + H->TableSize;
        }
    }

    return NewPos;
}

void Insert(HashTable H, EleType key)
{
    int StrLen = 0;
    int Pos = SqFind(H, key, &StrLen);
    if(H->Cells[Pos].Info != legitimate)
    {
        strcpy(H->Cells[Pos].Data, key);
        H->Cells[Pos].Info = legitimate;
        H->Cells[Pos].Length = StrLen;
    }

}

int main()
{
    int N,P;
    scanf("%d %d", &N, &P);
    HashTable H = CreateTable(P);

    EleType key;
    int i,Pos,len,flag;
    flag = 0;
    for(i=0; i<N; i++)
    {
        scanf("%s", key);
        Insert(H, key);
        Pos = SqFind(H, key, &len);
        if(flag == 0)
        {
            printf("%d", Pos);
            flag = 1;
        }
        else
            printf(" %d", Pos);
    }


    return 0;
}

实现二
实现二在哈希表里就不考虑字符串的长度了

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

#define MAXL 8

typedef char EleType[MAXL+1];
typedef enum{legitimate, empty, deleted} EntryType;

typedef struct HashEntry Cell;
struct HashEntry
{
    EleType Data;
    EntryType Info;
};

typedef struct TblNode *HashTable;
struct TblNode
{
    int TableSize;
    Cell *Cells;
};

HashTable CreateTable(int TableSize)
{
    HashTable H = (HashTable)malloc(sizeof(struct TblNode));
    H->TableSize = TableSize;

    H->Cells = (Cell *)malloc(H->TableSize*sizeof(Cell));
    int i;
    for(i=0; i<H->TableSize; i++)
    {
        H->Cells[i].Data[0] = '\0';
        H->Cells[i].Info = empty;
    }
    return H;
}

int HashFunction(EleType key, int TableSize)
{
    int i,tmpLength;
    unsigned int Pos = 0;
    tmpLength = 0;
    while(key[tmpLength] != '\0')
        tmpLength++;
    if(tmpLength>2)
    {
        for(i=tmpLength-3; i<tmpLength; i++)
            Pos = (Pos<<5) + (key[i]-'A');
        Pos = Pos % TableSize;
    }
    return Pos;
}

int SqFind(HashTable H, EleType key)
{
    int CurPos, NewPos;
    CurPos = NewPos = HashFunction(key, H->TableSize);

    int count = 0;
    while(H->Cells[NewPos].Info!=empty && strcmp(H->Cells[NewPos].Data, key))
    {
        if(++count %2)
        {
            NewPos = CurPos + (count+1)*(count+1)/4;
            if(NewPos >= H->TableSize)
                NewPos = NewPos % H->TableSize;

        }
        else
        {
            NewPos = CurPos - (count)*(count)/4;
            while(NewPos < 0)
                NewPos = NewPos + H->TableSize;
        }
    }

    return NewPos;
}

void Insert(HashTable H, EleType key)
{
    int Pos = SqFind(H, key);
    if(H->Cells[Pos].Info != legitimate)
    {
        strcpy(H->Cells[Pos].Data, key);
        H->Cells[Pos].Info = legitimate;
    }

}

int main()
{
    int N,P;
    scanf("%d %d", &N, &P);
    HashTable H = CreateTable(P);

    EleType key;
    int i,Pos,len,flag;
    flag = 0;
    for(i=0; i<N; i++)
    {
        scanf("%s", key);
        Insert(H, key);
        Pos = SqFind(H, key);
        if(flag == 0)
        {
            printf("%d", Pos);
            flag = 1;
        }
        else
            printf(" %d", Pos);
    }


    return 0;
}