diff options
-rw-r--r-- | utils/training/gen_ngram.cpp | 3 |
1 files changed, 3 insertions, 0 deletions
diff --git a/utils/training/gen_ngram.cpp b/utils/training/gen_ngram.cpp index 3203cd1..918c9c8 100644 --- a/utils/training/gen_ngram.cpp +++ b/utils/training/gen_ngram.cpp @@ -100,6 +100,9 @@ int main(int argc, char * argv[]){ glong phrase_len = 0; utf16_t * phrase = g_utf8_to_utf16(linebuf, -1, NULL, &phrase_len, NULL); + if ( phrase_len == 0 ) + continue; + phrase_token_t token = 0; int result = g_phrases->search( phrase_len, phrase, token); if ( ! (result & SEARCH_OK) ) |