gistfile1.c

#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#define WORDLIM 128
#define REPEAT for(;;)


struct word {
	char *cont;
	char *wsp;
	int ctr;
};

char peekchar(void)
{
    char c;

    c = getchar();
    if(c != EOF) ungetc(c, stdin);      /* puts it back */
    return c;
}

struct word *getword(){
	char cont[WORDLIM];
	char wsp[WORDLIM];
	cont[0] = '\0';
	wsp[0] = '\0';
	if (peekchar() == EOF) {return NULL;}

	REPEAT{
		char c = getchar();
		char buf[2];
		buf[0]=c;
		buf[1]='\0';
		if (c == '\n' || c == ' ' || c == EOF){
			if (c != EOF)
				strcat(wsp, buf);
			if (peekchar() != '\n' && peekchar() != ' '){
                        	struct word *toret;
				toret = malloc(sizeof(struct word));
                        	toret->cont = malloc(strlen(cont) + 1);
				strcpy(toret->cont, cont);
                        	toret->wsp = malloc(strlen(wsp) + 1);
				strcpy(toret->wsp, wsp);
				toret->ctr = -1;
                        	return toret;
                	}
			continue;
		}
		else {
                        strcat(cont, buf);
			continue;
                }
	}
	printf("PANIC PANIC PANIC THIS IS GOING WROOOOONG!!!!!\n");
}

void numbrify(struct word **wordlist){
	int oc = 0;
	int roc = oc;
	struct word *w;
	while ((w = wordlist[oc]) != NULL){
		int ic;
		if (w->ctr == -1){
			for (ic = oc + 1; wordlist[ic] != NULL; ic++){
				if (!strcmp(wordlist[ic]->cont, w->cont)){
					//printf("**found match between %s and %s**\n", wordlist[ic]->cont, w->cont);
					wordlist[ic]->ctr = roc;
				}
			}
			if (w->cont[0]!='\0')
				roc++;
		}
		oc++;
	}
}

int main(void){
	struct word *wlist[4096];
	int i = 0;
	struct word *w;
	for (i = 0; (w = getword()) != NULL; i++){
		wlist[i]=w;
	}
	wlist[i+1] = NULL;
	numbrify(wlist);
	i = 0;
	for (i = 0; wlist[i]!=NULL; i++){
		if (wlist[i]->ctr == -1) printf("%s%s", wlist[i]->cont, wlist[i]->wsp);
		else printf("%d%s", wlist[i]->ctr, wlist[i]->wsp);
		//printf("'%s'\t'%s'\t%d\n", wlist[i]->cont, wlist[i]->wsp, wlist[i]->ctr);
	}
	return 0;
}