1
0
mirror of git://jb55.com/damus synced 2024-09-05 21:03:51 +00:00

nostrdb/search: fix infinite loop when parsing some notes

Our word parser gets stuck on some notes with utf8 chars. Make sure we
are always advancing so we don't get stuck.

Signed-off-by: William Casarin <jb55@jb55.com>
This commit is contained in:
William Casarin 2023-11-29 12:01:17 -08:00
parent 0e0c53145f
commit b74bde5cc4

View File

@ -2218,6 +2218,7 @@ static inline int consume_until_boundary(struct cursor *cur) {
// TODO: We should work towards // TODO: We should work towards
// handling all UTF-8 characters. // handling all UTF-8 characters.
//printf("Invalid UTF-8 code point: %x\n", c); //printf("Invalid UTF-8 code point: %x\n", c);
return 0;
} }
} }
} }
@ -2305,6 +2306,14 @@ static int ndb_parse_words(struct cursor *cur, void *ctx, ndb_word_parser_fn fn)
if (word_len == 0 && cur->p >= cur->end) if (word_len == 0 && cur->p >= cur->end)
break; break;
if (word_len == 0) {
if (!cursor_skip(cur, 1))
break;
continue;
}
//ndb_debug("writing word index '%.*s'\n", word_len, word);
if (!fn(ctx, word, word_len, words)) if (!fn(ctx, word, word_len, words))
continue; continue;