int test_scaling_accuracy(const char *bloom_file, const char *words_file)
{
FILE *fp;
char word[256];
scaling_bloom_t *bloom;
int i;
struct stats results = { 0 };
printf("\n* test scaling accuracy\n");
if ((fp = fopen(bloom_file, "r"))) {
fclose(fp);
remove(bloom_file);
}
if (!(bloom = new_scaling_bloom(CAPACITY, ERROR_RATE, bloom_file))) {
fprintf(stderr, "ERROR: Could not create bloom filter\n");
return TEST_FAIL;
}
if (!(fp = fopen(words_file, "r"))) {
fprintf(stderr, "ERROR: Could not open words file\n");
return TEST_FAIL;
}
for (i = 0; fgets(word, sizeof(word), fp); i++) {
if (i % 2 == 0) {
chomp_line(word);
scaling_bloom_add(bloom, word, strlen(word), i);
}
}
fseek(fp, 0, SEEK_SET);
for (i = 0; fgets(word, sizeof(word), fp); i++) {
if (i % 2 == 1) {
chomp_line(word);
bloom_score(scaling_bloom_check(bloom, word, strlen(word)), 0, &results, word);
}
}
fclose(fp);
printf("Elements added: %6d" "\n"
"Elements checked: %6d" "\n"
"Total size: %d KiB" "\n\n",
(i + 1) / 2, i / 2,
(int) bloom->num_bytes / 1024);
free_scaling_bloom(bloom);
return print_results(&results);
}
static int Dablooms_init(Dablooms *self, PyObject *args, PyObject *kwds)
{
double error_rate;
const char *filepath;
unsigned int capacity;
static char *kwlist[] = {"capacity", "error_rate", "filepath", NULL};
if (! PyArg_ParseTupleAndKeywords(args, kwds, "|ids", kwlist,
&capacity, &error_rate, &filepath)) {
return -1;
}
self->filter = new_scaling_bloom(capacity, error_rate, filepath);
return 0;
}
