use onigenc_wb_is_break_position()

diff --git a/src/regenc.h b/src/regenc.h
index 8a3397d4d29444865f1f40e98f499bc9385eaf5b..c980d590e80648a55edb8ac562b2750b3faf8e8b 100644 (file)
--- a/src/regenc.h
+++ b/src/regenc.h
@@ -114,6 +114,7 @@ struct PropertyNameCtype {
 /* #define USE_CRNL_AS_LINE_TERMINATOR */
 #define USE_UNICODE_PROPERTIES
 #define USE_UNICODE_EXTENDED_GRAPHEME_CLUSTER
+#define USE_UNICODE_WORD_BREAK
 /* #define USE_UNICODE_CASE_FOLD_TURKISH_AZERI */
 /* #define USE_UNICODE_ALL_LINE_TERMINATORS */  /* see Unicode.org UTS #18 */
 
@@ -174,7 +175,7 @@ extern int onigenc_mb4_is_code_ctype P_((OnigEncoding enc, OnigCodePoint code, u
 extern struct PropertyNameCtype* onigenc_euc_jp_lookup_property_name P_((register const char *str, register size_t len));
 extern struct PropertyNameCtype* onigenc_sjis_lookup_property_name P_((register const char *str, register size_t len));
 
-/* in enc/unicode.c */
+/* in unicode.c */
 extern int onigenc_unicode_is_code_ctype P_((OnigCodePoint code, unsigned int ctype));
 extern int onigenc_utf16_32_get_ctype_code_range P_((OnigCtype ctype, OnigCodePoint *sb_out, const OnigCodePoint* ranges[]));
 extern int onigenc_unicode_ctype_code_range P_((OnigCtype ctype, const OnigCodePoint* ranges[]));
@@ -182,7 +183,9 @@ extern int onigenc_unicode_get_case_fold_codes_by_str P_((OnigEncoding enc, Onig
 extern int onigenc_unicode_mbc_case_fold P_((OnigEncoding enc, OnigCaseFoldType flag, const UChar** pp, const UChar* end, UChar* fold));
 extern int onigenc_unicode_apply_all_case_fold P_((OnigCaseFoldType flag, OnigApplyAllCaseFoldFunc f, void* arg));
 extern int onigenc_egcb_is_break_position P_((OnigEncoding enc, UChar* p, UChar* prev, const UChar* start, const UChar* end));
-
+#ifdef USE_UNICODE_WORD_BREAK
+extern int onigenc_wb_is_break_position P_((OnigEncoding enc, UChar* p, UChar* prev, const UChar* start, const UChar* end));
+#endif
 
 #define UTF16_IS_SURROGATE_FIRST(c)    (((c) & 0xfc) == 0xd8)
 #define UTF16_IS_SURROGATE_SECOND(c)   (((c) & 0xfc) == 0xdc)

diff --git a/src/regexec.c b/src/regexec.c
index bad45670fdaaa3b93720ccf3f2427490ea2d1c74..2863faa04f5426f3c34cfb49ea3027c92740bf78 100644 (file)
--- a/src/regexec.c
+++ b/src/regexec.c
@@ -3259,6 +3259,11 @@ match_at(regex_t* reg, const UChar* str, const UChar* end,
         case EXTENDED_GRAPHEME_CLUSTER_BOUNDARY:
           is_break = onigenc_egcb_is_break_position(encode, s, sprev, str, end);
           break;
+#ifdef USE_UNICODE_WORD_BREAK
+        case WORD_BOUNDARY:
+          is_break = onigenc_wb_is_break_position(encode, s, sprev, str, end);
+          break;
+#endif
         default:
           goto bytecode_error;
           break;