symbol* new_node(symbol *n1,symbol *n2){
/* new_node() -
given two symbols from the symbol table-tree,
combines them into one Internal Node and then
returns a pointer to the new internal node.
*/
// Init
symbol *s;
s=init_symbol(INTERNAL_NODE);
// Set left and right children
s->left=n1;
s->right=n2;
// Set parents of children
n1->parent=s;
n2->parent=s;
// Set frequency
s->frequency = n1->frequency + n2->frequency;
return s;
}
symbol* add_symbol_to_table (char* buffer, int size) {
symbol* sym = NULL;
#ifdef DEBUG
printf ("adding symbol with buffer: %s\n", buffer);
#endif
// Find or create the symbol as needed
if (symbols->size > 0) {
sym = get_symbol(buffer);
}
if (sym == NULL) { // Symbol wasn't found
// Actually create the symbol from the buffer
sym = init_symbol();
set_symbol_name(sym, buffer, size);
}
else {
// Don't need to update table, return early since sym has been found
return sym;
}
// Adjust the symbolTable pointers
if (symbols->size == 0) {
symbols->first = sym;
symbols->last = sym;
}
else {
symbols->last->next = sym; // Point the previous last's next to the new last
sym->prev = symbols->last;
symbols->last = sym;
}
symbols->size++;
return sym;
}
void update_frequency(symbol *r, int c){
/*
update_frequency() - given the root of a symbol table-tree and a character,
increases the frequency of that character in the symbol table-tree,
creating nodes as necessary.
*/
symbol *s;
if (!exists_symbol(r,c))
s=init_symbol(c);
else
s=remove_symbol(r,c);
s->frequency++;
insert_symbol(r,s);
}
int main(int argc, char *argv[]){
// 1. Initialization
int c;
symbol *root;
root=init_symbol(ROOT_NODE);
root->frequency=NON_EXISTENT;
root->parent=root;
// 1.a. init: file pointers
FILE *inputfile;
FILE *outputfile;
inputfile=NULL;
outputfile=NULL;
if (argc >=2)
inputfile = fopen(argv[1],"r");
if (argc >=3)
outputfile = fopen(argv[2],"w");
if (inputfile == NULL) {
fprintf(stderr, "Can't open input file. \n");
exit(1);
}
if (outputfile == NULL) {
outputfile=stdout;
}
// 2. count the frequency of each character in inputfile
while((c=fgetc(inputfile))!=EOF){
update_frequency(root,c);
}
update_frequency(root,-1);
// 3. Make the huffman tree
while(count_parentless_nodes(root) > 1){
symbol *node1,*node2, *newnode;
node1 = get_rarest_parentless_node(root);
node1->parent=node1; //temporarily
node2 = get_rarest_parentless_node(root);
node2->parent=node2; //temporarily
newnode = new_node(node1, node2);
insert_symbol(root,newnode);
}
// 4. Output symbol mappings
for(c=-1;c<256;c++){
char *encodingstring;
encodingstring=get_encoding(root,c);
int freq=get_frequency(root,c);
if (freq > 0){
/*if (c <=127 && c>=32 )
fprintf(outputfile,"%c\t%d\t%s\n", c, freq, encodingstring);
else */
fprintf(outputfile,"%d\t%d\t%s\n", c, freq, encodingstring);
}
free(encodingstring);
}
fprintf(outputfile,"\n");
// 5. Output encoding of inputfile
rewind(inputfile);
while((c=fgetc(inputfile))!=EOF){
char *encodingstring;
encodingstring=get_encoding(root,c);
fprintf(outputfile, "%s", encodingstring);
free(encodingstring);
}
char *encodingstring;
encodingstring=get_encoding(root,EOF);
fprintf(outputfile, "%s", encodingstring);
free(encodingstring);
fprintf(outputfile,"\n");
free_symbol(root);
// 6. Close file pointers
fclose(inputfile);
fclose(outputfile);
}
