void index_update(HashTable *index, char *line_content)
{
//delimiting each line by word
const char delim[2] = " ";
char *word = strtok(line_content, delim);
//getting total file count
char *file_count_converted_to_string = strtok(NULL, delim);
int file_count = 0;
char *tmp;
*&file_count = strtod(file_count_converted_to_string, &tmp); //converting
int doc_id;
int freq;
for (int i = 0; i < file_count; i++) { //iterating through the documents that have the word specified
//breaking up the line into delimited sections by a space
char *doc_id_converted_to_string = strtok(NULL, delim);
//converting doc_id to a digit
*&doc_id = strtod(doc_id_converted_to_string, &tmp);
char *freq_converted_to_string = strtok(NULL, delim);
*&freq = strtod(freq_converted_to_string, &tmp);
//updating the index to rebuild it into its former glory-self
for (int j = 0; j < freq; j++) {
insert_to_index(doc_id, word, index);
}
}
}
/* -----------------------------------------------------------------------
Function Name: initial_index()
Functionality: Crawler data in directory specified is scanned and each word
inserted into the Hashtable Index
Input <--- directory where crawler data lives (html from crawled pages) and
the Hashtable Index
Outputs ---> void (updated index)
----------------------------------------------------------------------- */
void initial_index(char *argv[], HashTable *Index)
{
int doc_id;
char **results_filenames = NULL;
int num_files = GetFilenamesInDir(argv[1], &results_filenames);
//going through each file, allocating memory, parsing html and updating the index with word
for (doc_id = 1; doc_id < num_files; doc_id++) {
char *file_name = calloc(1, floor(log10(abs(doc_id))) + 2);
snprintf(file_name, floor(log10(abs(doc_id))) + 2, "%d", doc_id);
// printf("results_filename is: %s", file_name);
char *file_path = calloc(1, strlen(argv[1]) + strlen(file_name) + 4);
snprintf(file_path, strlen(argv[1]) + strlen(file_name) + 4, "./%s/%s", argv[1], file_name);
// printf("path is: %s \n", file_path);
if (IsFile(file_path)) {
char *content = file_parsing(file_path); //parsing out first two line_contents of each file, so only html left over
//normalizing each word in each file and updating the index hashtable with it
if (content != NULL) {
int pos = 0;
char *word;
while ((pos = GetNextWord(content, pos, &word)) > 0) {
NormalizeWord(word);
insert_to_index(doc_id, word, Index);
}
}
free(content);
}
free(file_name);
free(file_path);
}
free(results_filenames);
}
void page_node::insert(bt_key* key, RID rid, bt_key* itr) {
if (is_leaf_node()) insert_to_leaf(key, rid, itr);
else insert_to_index(key, rid, itr);
}