diff options
| author | ksaito <ksaito@b2dd03c8-39d4-4d8f-98ff-823fe69b080e> | 2004-11-28 03:09:13 +0000 |
|---|---|---|
| committer | ksaito <ksaito@b2dd03c8-39d4-4d8f-98ff-823fe69b080e> | 2004-11-28 03:09:13 +0000 |
| commit | 09542e3269f5b915b45eb5c03fcb8c61452d751e (patch) | |
| tree | 91eb489927947f75436f5bd366896772c6b0c71c /regparse.c | |
| parent | 5fb312bd18a09ac50e8c7d707cb87893609ce9fc (diff) | |
* regparse.c, test/ruby/test_regexp.rb: fixed problem with UTF-8 characters that have U+00FE or invalid characters.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@7398 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
Diffstat (limited to 'regparse.c')
| -rw-r--r-- | regparse.c | 20 |
1 files changed, 9 insertions, 11 deletions
diff --git a/regparse.c b/regparse.c index e6fea8e68a..16792ee9eb 100644 --- a/regparse.c +++ b/regparse.c @@ -3631,6 +3631,9 @@ next_state_val(CClassNode* cc, OnigCodePoint *vs, OnigCodePoint v, case CCS_RANGE: if (intype == *type) { if (intype == CCV_SB) { + if (*vs > 0xff || v > 0xff) + return ONIGERR_INVALID_WIDE_CHAR_VALUE; + if (*vs > v) { if (IS_SYNTAX_BV(env->syntax, ONIG_SYN_ALLOW_EMPTY_RANGE_IN_CC)) goto ccs_range_end; @@ -3646,14 +3649,8 @@ next_state_val(CClassNode* cc, OnigCodePoint *vs, OnigCodePoint v, } else { #if 0 - if (intype == CCV_CODE_POINT && *type == CCV_SB && - ONIGENC_IS_CONTINUOUS_SB_MB(env->enc)) { - bitset_set_range(cc->bs, (int )*vs, 0x7f); - r = add_code_range(&(cc->mbuf), env, (OnigCodePoint )0x80, v); - if (r < 0) return r; - } -#else if (intype == CCV_CODE_POINT && *type == CCV_SB) { +#endif if (*vs > v) { if (IS_SYNTAX_BV(env->syntax, ONIG_SYN_ALLOW_EMPTY_RANGE_IN_CC)) goto ccs_range_end; @@ -3663,10 +3660,11 @@ next_state_val(CClassNode* cc, OnigCodePoint *vs, OnigCodePoint v, bitset_set_range(cc->bs, (int )*vs, (int )(v < 0xff ? v : 0xff)); r = add_code_range(&(cc->mbuf), env, (OnigCodePoint )*vs, v); if (r < 0) return r; +#if 0 } -#endif else return ONIGERR_MISMATCH_CODE_LENGTH_IN_CLASS_RANGE; +#endif } ccs_range_end: *state = CCS_COMPLETE; @@ -3826,7 +3824,7 @@ parse_char_class(Node** np, OnigToken* tok, UChar** src, UChar* end, break; case TK_CODE_POINT: - v = tok->u.code; + v = (OnigCodePoint )tok->u.code; in_israw = 1; val_entry: len = ONIGENC_CODE_TO_MBCLEN(env->enc, v); @@ -3952,7 +3950,7 @@ parse_char_class(Node** np, OnigToken* tok, UChar** src, UChar* end, case TK_CC_AND: /* && */ { if (state == CCS_VALUE) { - r = next_state_val(cc, &vs, 0, &val_israw, 0, CCV_SB, + r = next_state_val(cc, &vs, 0, &val_israw, 0, val_type, &val_type, &state, env); if (r != 0) goto err; } @@ -3992,7 +3990,7 @@ parse_char_class(Node** np, OnigToken* tok, UChar** src, UChar* end, } if (state == CCS_VALUE) { - r = next_state_val(cc, &vs, 0, &val_israw, 0, CCV_SB, + r = next_state_val(cc, &vs, 0, &val_israw, 0, val_type, &val_type, &state, env); if (r != 0) goto err; } |