I have a problem with hash

I'm programming a hash with alphabetics, use as a dictionary, but I i have his problem:

When I add a new word after I had a hash with some word, old hash is deleted and it's creat a new hash, but I don't want so, I want it has the same hash because i want to save it:

And, this is my code:

Code:

#include<conio.h> #include<stdio.h> #include<malloc.h> #include<string.h> #include <stdlib.h> #define B 26 #define maxlegth 30 typedef char datatype; typedef struct Node { datatype Tu[maxlegth]; datatype Nghia[maxlegth]; Node*Next; }; typedef Node*Position; typedef Position Dictionary[B]; //Thiet ke ham bam int H(datatype X[maxlegth]) { int chudau; char*s; s=strlwr(X); chudau=int(s[0])-97; return chudau%B; } //Tao bang bam rong; void MakenullSet(Dictionary*D) { for(int i=0;i<B;i++) (*D)[i]=NULL; } //Kiem tra thanh vien trong bang bam; int MemberSet(datatype X[maxlegth],Dictionary D) { Position P; P=D[H(X)]; int Found=0; while((P!=NULL) && (Found!=1)) { //so sanh chuoi if(strcmp(P->Tu,X)==0) Found=1; else P=P->Next; } return Found; } //Them phan tu vao bang bam void InsertSet(datatype X[maxlegth],datatype Y[maxlegth],Dictionary *D) { int Bucket; Position P; Bucket=H(X); P=(*D)[Bucket]; (*D)[Bucket]=(Node*)malloc(sizeof(Node)); //Khoi tao chuoi memset((*D)[Bucket]->Tu,'\0',maxlegth); memset((*D)[Bucket]->Nghia,'\0',maxlegth); //Chep chuoi strcpy((*D)[Bucket]->Tu,X); strcpy((*D)[Bucket]->Nghia,Y); (*D)[Bucket]->Next=P; } //Nhap du lieu cho bang bam tu ban phim void ReadSet(Dictionary*D) { int i; int n; datatype X[maxlegth]; datatype Y[maxlegth]; MakenullSet(D); printf("\n--------------------------- Number of word to add < Number only>: "); scanf("%d",&n); for(i=1;i<=n;i++) { flushall(); // Nhap chuoi ham gets printf("\nWord number %d: ",i); gets(X); printf("\nMeaning: "); gets(Y); InsertSet(X,Y,D); } } //In du lieu ra man hinh void PrintSet(Dictionary D) { Position P; for(int i=0;i<B;i++) { P=D[i]; printf("Bucket %c: ",char(i+65)); while(P!=NULL) { printf("%s",P->Tu); printf(" - "); printf("%s",P->Nghia); P=P->Next; } printf("\n"); } printf("\n"); } void FindSet(datatype F[maxlegth],Dictionary D) { Position P; F=strlwr(F); int Found=0; for(int i=0;i<B;i++) { P=D[i]; while(P!=NULL && Found==0) { if(strcmp(F,P->Tu)==0) { printf("Meaning: %s",P->Nghia); Found=1; } else { P=P->Next; } } } if(Found==0) { printf("\nWe don't see this word."); } printf("\n"); } void showmenu() { printf("\n\t-------------------- COMPUTER DICTIONARY -------------------"); printf("\n[1]. Add WORDS."); printf("\n[2]. Find a word."); printf("\n[3]. Display all."); printf("\n[4]. Save and Close."); printf("\n\nEnter Choice: "); } void prog() { clrscr(); int menuchoice; Dictionary D; datatype X[maxlegth]; datatype F[maxlegth]; while(1) { clrscr(); showmenu(); scanf("%d",&menuchoice); switch(menuchoice) { case 1:printf("\n\nAdd content of hash with %d Bucket: \n",B); ReadSet(&D); break; case 2:flushall(); printf("\nAdd word which you want to find: "); gets(F); FindSet(F,D); getch(); break; case 3:printf("\n\nDictionary: \n"); PrintSet(D); getch(); break; case 4:exit(1); break; default:printf("Enter Again"); prog(); break; } } } int main() { clrscr(); showmenu(); prog(); return 0; }

What do you mean it deletes the old hash? It's supposed to generate a different hash for each word. That's the whole point of hashing. Oh, and get rid of malloc.h. malloc is included in stdlib.h, as are its related functions.

Quzah.

So, what should I do to have a same hash?

Why do you think you should have the same hash? Give an example. The only time you should end up with the same hash is if you're trying to hash the same value. Or, in the case of a collision, which you generally seek to avoid.

Quzah.

hi, now I knows how I have this problem. I used MakenullSet(D) in ReadSet function, so it deleted my hash ^_^

Quote:

It's supposed to generate a different hash for each word.

I thought it is something like map with unique key.
Hash table, otherwise, can has several entries with the same hash value. The point here is to search the specific velue not through the whole array, but only through the part having the same hash value.

If you would have read my second post, you'd have seen that your post has already been addressed, even before you posted it.

The main problem here, is that by "hash", the OP actually meant that their entire hash table was being erased. I assumed they meant that they were trying to get the hash of a word, and then tried to get the hash for a second word, and couldn't figure out why the two were different.

Quote:

When I add a new word after I had a hash with some word, old hash is deleted and it's creat a new hash, but I don't want so, I want it has the same hash because i want to save it

See? That doesn't say anything about the entire table getting nuked. You'd only guess that after reading their final post, where they've already figured out what they're doing wrong.

As to hash tables having several entries with the same hash value, I already covered that in my second post. But that's not what they were talking about anyway, apparently.

Quzah.

Code:

//Thiet ke ham bam int H(datatype X[maxlegth]) { int chudau; char*s; s=strlwr(X); chudau=int(s[0])-97; return chudau%B; }

Not a very good hash function there, is it.

Looks like a strange dialect of C. Didn't know you could write Vietnamese in the Roman alphabet. :rolleyes:

I could write Vienamese with Roman Alphabet but I order them with English, so, must use 26 character.

About you, what's hash function do you think best?

Letting the entire hash depend on only the first character will cause problems. Say I input "ada", "abc" and "algol". You would hash all of them to the same value.

You could try en.wikipedia.org/wiki/Hash_table for better types of functions (maybe they have one in Vietnamese too). But if this is for some homework assignment or something where you're not supposed to care, I guess it doesn't really matter.

More importantly, get rid of all that nonstandard junk in your code. Replace getc() with getchar(), remove <malloc.h>, use fflush instead of flushall(), etc.