int main ( int argc, char **argv )
{
/*для int*/
int arr[MAX_LEN];
int i;
for (i = 0; i < MAX_LEN; ++i )
arr[i] = i;
int data = 4;
int pos;
find_int(arr, MAX_LEN, data, &pos );
printf ( "find_method %d pos %d\n", data+1, pos );
pos = -1;
FIND( arr, MAX_LEN, data, &pos );
printf ( "find_macros %d pos %d\n", data+1, pos );
/*для char*/
char arrf[MAX_LEN];
for (i = 0; i < MAX_LEN; ++i )
arrf[i] = i;
pos = -1;
char dataf = 40;
find_float(arrf, MAX_LEN, dataf, &pos );
printf ( "find_method %d pos %d\n", dataf, pos );
pos = -1;
FIND( arrf, MAX_LEN, dataf, &pos );
printf ( "find_macros %d pos %d\n", dataf, pos );
}
/*
* Binary search an array for an integer
* obj: pointer to an array of integers
or array of fixed length structs containing an integer (in this case the obj pointer should point to the integer in the first struct)
* struct_len: length of one object in the array in bytes
* i: search integer
* min: minimal index to search
* max: maximal index to search
*/
int find_int(void *objs, int struct_len, int i, int min, int max) {
int middle = (min + max) / 2;
int cmp_val = *((int *) (((char *) objs) + middle * struct_len));
if (cmp_val == i) {
// string found
return middle;
} else if (min != max) {
if (cmp_val > i && min != middle) {
// continue search in left half
return find_int(objs, struct_len, i, min, middle - 1);
} else if (max != middle) {
// continue search in right half
return find_int(objs, struct_len, i, middle + 1, max);
}
}
// finished searching
return -1;
}
int
JSONConfiguration::GetInt(IN const string &name)
{
return find_int(_rootValue.get_obj(),name);
}
/*
* Parses a file and adds its words to the index
*/
void parse_file_for_index(index_p index, char *file) {
// open file or print error message
FILE *f = fopen(file, "r");
if (!f) {
printf("Cannot open %s!\nIndex not updated.\n", file);
return;
}
// document id = index of document in list of all documents in filebase (alphabetically ordered)
int doc_id = find_str(&index->documents[0].name, sizeof(indexed_document_t), file, 0, index->nr_docs-1);
if (doc_id < 0) {
printf("Error: %s is not in the filebase!\n", file);
return;
}
char *l;
while ((l = read_line(f))) {
// turn non alpha characters into spaces
nonalpha_to_space(l);
char *word = strtok(l, " ");
while (word) {
// ignore stopwords
if (is_stopword(word)) {
word = strtok(NULL, " ");
continue;
}
char *word_stem = stem(word);
if (!strlen(word_stem)) {
word = strtok(NULL, " ");
continue;
}
// insert document into index / add new stem to index
indexed_word_p w = index->words; // current word
indexed_word_p p = NULL; // previous word
int flag = 0;
while (w && !flag) {
int cmp = strcmp(w->stem, word_stem);
if (!cmp) {
// stem is already indexed
flag = 1;
break;
} else if (0 < cmp) {
// stem not indexed yet
flag = 2;
break;
}
p = w;
w = w->next;
}
if (flag == 1) {
// stem indexed, add document to list
int i;
for (i = 0; i < w->nr_docs; i++) {
if (w->documents[i].id == doc_id) {
// document is already indexed for this stem
flag = 0;
break;
} else if (w->documents[i].id > doc_id) {
break;
}
}
// only add document to list if it's not already in the list
if (flag) {
w = (indexed_word_p) realloc(w, sizeof(indexed_word_t) + sizeof(doc_t) * (w->nr_docs + 1));
// update pointer to this group (needed after realloc)
if (!p) {
index->words = w;
} else {
p->next = w;
}
// insert document in list
memmove(&w->documents[i+1], &w->documents[i], sizeof(doc_t) * (w->nr_docs - i));
w->documents[i].id = doc_id;
w->documents[i].tf = 1;
w->nr_docs++;
} else {
// increase counter for number of occurances of this word in this document
w->documents[i].tf++;
}
free(word_stem);
} else {
// stem is not indexed, add it to index
w = (indexed_word_p) malloc(sizeof(indexed_word_t) + sizeof(doc_t));
w->stem = word_stem;
w->nr_docs = 1;
w->documents[0].id = doc_id;
w->documents[0].tf = 1;
index->nr_words++;
// insert this word in linked list
if (!p) {
w->next = index->words;
index->words = w;
} else {
w->next = p->next;
p->next = w;
}
}
// increase counter for total number of words in this document
index->documents[doc_id].nr_words++;
// get next word
word = strtok(NULL, " ");
}
free(l);
}
fclose(f);
// finalize computation of TF
indexed_word_p w = index->words;
while (w) {
int i = find_int(&w->documents[0].id, sizeof(doc_t), doc_id, 0, w->nr_docs - 1);
if (i >= 0) {
w->documents[i].tf /= index->documents[doc_id].nr_words;
}
w = w->next;
}
}
