db: smarter string pointer encoding (include length field)

So we don't need explicit null terminator in most cases saving
space. It will also speed up comparisons as getting string blob is
now constant time (no strlen needed).

1 files changed, 3 insertions, 5 deletions

diff --git a/squark-filter.c b/squark-filter.c
index 8973d04..e47cbf5 100644
--- a/squark-filter.c
+++ b/squark-filter.c
@@ -112,12 +112,10 @@ static blob_t url_classify(struct url_info *url, struct sqdb *db)
 	void *cmph;
 	struct sqdb_index_entry *indx;
 	uint32_t *categories;
-	char *strings;
 	cmph_uint32 i = -1, previ;
 
 	cmph = sqdb_section_get(db, SQDB_SECTION_INDEX_MPH, NULL);
 	indx = sqdb_section_get(db, SQDB_SECTION_INDEX, NULL);
-	strings = sqdb_section_get(db, SQDB_SECTION_STRINGS, NULL);
 
 	/* search for most qualified domain match; do first lookup
 	 * with two domain components */
@@ -132,14 +130,14 @@ static blob_t url_classify(struct url_info *url, struct sqdb *db)
 
 		previ = i;
 		i = cmph_search_packed(cmph, key.ptr, key.len);
-		if (blob_cmp(got, BLOB_STR(&strings[indx[i].component])) != 0) {
+		if (blob_cmp(got, sqdb_get_string_literal(db, indx[i].component)) != 0) {
 			/* the subdomain did no longer match, use 
 			 * parents classification */
 			i = previ;
 			goto parent_dns_match;
 		}
 		if (!blob_is_null(tld)) {
-			if (blob_cmp(tld, BLOB_STR(&strings[indx[indx[i].parent].component])) != 0) {
+			if (blob_cmp(tld, sqdb_get_string_literal(db, indx[indx[i].parent].component)) != 0) {
 				/* top level domain did not match */
 				i = -1;
 				goto parent_dns_match;
@@ -163,7 +161,7 @@ parent_dns_match:
 
 	categories = sqdb_section_get(db, SQDB_SECTION_CATEGORIES, NULL);
 	printf("%d\n", indx[i].category);
-	return BLOB_STR(&strings[categories[indx[i].category]]);
+	return sqdb_get_string_literal(db, categories[indx[i].category]);
 }
 
 int main(int argc, char **argv)