#include<stdio.h>
#include<unistd.h>
#include<stdlib.h>
#include<fcntl.h>
#include<ctype.h>
#include<string.h>
#include<sys/mman.h>
#include "tree.h"

/* On FreeBSD, we could just use the system's tree instead:
#include<sys/tree.h>
*/

struct tree_node {
	char *word;   // This is the key
	unsigned int count;   // This is the value
	RB_ENTRY(tree_node) tree_entry;
};
int tncompare(struct tree_node *a, struct tree_node *b){
	return strcmp(a->word, b->word);	
}
RB_HEAD(word_tree, tree_node);
RB_GENERATE_STATIC(word_tree, tree_node, tree_entry, tncompare);
struct word_tree wtree;

void new_word(char *start, char *stop){
	/* Instead of printing it out, can we put it in a tree? */
	char *word_place = malloc(1 + stop - start);
	for(int i = 0; start + i < stop; i++)
		word_place[i] = tolower(start[i]);
	word_place[stop - start] = 0;
	struct tree_node *existing_node = RB_FIND(word_tree, &wtree, (struct tree_node*)&word_place);	
	if(existing_node){
		existing_node->count++;
		free(word_place);
	} else {
		struct tree_node *nn = malloc(sizeof(struct tree_node));
		nn->count = 1;
		nn->word = word_place;
		RB_INSERT(word_tree, &wtree, nn);
	}
}

int main(int argc, char ** argv){
	if(argc < 2) {
		printf("Usage:  %s filename\n", argv[0]);
		return 1;
	}
	int fd = open(argv[1], O_RDONLY);
	if(fd < 0){
		perror(argv[1]);
		return 2;
	}
	ssize_t length = lseek(fd, 0, SEEK_END);
	printf("Opened file %s (%ld bytes)\n", argv[1], length);
	char *file_contents = mmap(0, length, PROT_READ, MAP_SHARED, fd, 0);

	char inword = 0;
	char *wordstart = 0;
	for(size_t i = 0; i < length; i++){
		if(!inword && isalpha(file_contents[i])){
			inword = 1;
			wordstart = &file_contents[i];
		} else if( inword && !isalpha(file_contents[i])){
			inword = 0;
			new_word(wordstart, &file_contents[i]);
		}
	}
	munmap(file_contents, length);
	close(fd);

	struct tree_node *i;
	RB_FOREACH(i, word_tree, &wtree){
		printf("%s: %u\n", i->word, i->count);
	}




	return 0;
}