merge from develop

kkos · Nov 7, 2016 · 8961d8a · 8961d8a
2 parents d5e9cc2 + 700bbec
commit 8961d8a
Show file tree

Hide file tree

Showing 17 changed files with 286 additions and 41 deletions.
diff --git a/CMakeLists.txt b/CMakeLists.txt
@@ -4,7 +4,7 @@ cmake_minimum_required(VERSION 2.8)
 project(oniguruma C)
 
 set(PACKAGE onig)
-set(PACKAGE_VERSION "6.1.1")
+set(PACKAGE_VERSION "6.1.2")
 
 set(USE_COMBINATION_EXPLOSION_CHECK 0)
 set(USE_CRNL_AS_LINE_TERMINATOR 0)

diff --git a/HISTORY b/HISTORY
@@ -1,5 +1,13 @@
 History
 
+2016/11/07: Version 6.1.2
+
+2016/10/25: allow word bound, word begin and word end in look-behind.
+2016/10/19: add ONIG_OPTION_CHECK_VALIDITY_OF_STRING option.
+2016/10/16: fix use after free node.
+2016/10/10: fix memory leaks after parsing regexp error.
+2016/09/22: implement many of is_valid_mbc_string().
+
 2016/09/02: Version 6.1.1
 
 2016/08/31: fix segfault /W.?{888}{888}{888}\x00/ (found by libfuzzer)
@@ -1721,7 +1729,7 @@ History
 2003/03/12: [spec] change named backref and subexp call format.
  backref: \k<name>, call: \g<name> (thanks akr)
 2003/03/11: [inst] add regparse.[ch] in win32/Makefile.
-2003/03/11: [bug] if UNALIGNED_WORD_ACCESS isn't set
+2003/03/11: [bug] if UNALIGNED_WORD_ACCESS isn't set,
  then compile error in unset_addr_list_fix(). (thanks knu)
 2003/03/10: [impl] divide regcomp.c to regcomp.c, regparse.c and regparse.h.
 2003/03/10: [bug] should handle multi-byte code name in fetch_name().

diff --git a/README.md b/README.md
@@ -20,6 +20,12 @@ Supported character encodings:
 * CP1251: contributed by Byte
 
 
+New feature of version 6.1.2
+--------------------------
+
+* allow word bound, word begin and word end in look-behind.
+* NEW option: ONIG_OPTION_CHECK_VALIDITY_OF_STRING
+
 New feature of version 6.1
 --------------------------
 

diff --git a/configure.ac b/configure.ac
@@ -1,5 +1,5 @@
 dnl Process this file with autoconf to produce a configure script.
-AC_INIT(onig, 6.1.1)
+AC_INIT(onig, 6.1.2)
 
 AC_CONFIG_MACRO_DIR([m4])
 

diff --git a/dist.info b/dist.info
@@ -1,7 +1,7 @@
 --- This file is part of LuaDist project
 
 name = "onig"
-version = "6.1.1"
+version = "6.1.2"
 
 desc = "Oniguruma is a regular expressions library."
 author = "K.Kosako"

diff --git a/index.html b/index.html
@@ -8,17 +8,17 @@
 <h1>Oniguruma</h1> (<a href="index_ja.html">Japanese</a>)
 
 <p>
-(c) K.Kosako, updated at: 2016/08/31
+(c) K.Kosako, updated at: 2016/11/07
 </p>
 
 <dl>
 <font color="orange">
 <dt><b>What's new</b>
 </font>
 <ul>
+<li>2016/11/07: Version 6.1.2 released.</li>
 <li>2016/09/02: Version 6.1.1 released.</li>
 <li>2016/08/29: Version 6.1.0 released.</li>
-<li>2016/05/09: Version 6.0.0 released.</li>
 <li>2014/12/12: Version 5.9.6 released.</li>
 </ul>
 </dl>

diff --git a/index_ja.html b/index_ja.html
@@ -8,17 +8,17 @@
 <h1>鬼車</h1>
 
 <p>
-(c) K.Kosako, 最終更新: 2016/08/31
+(c) K.Kosako, 最終更新: 2016/11/07
 </p>
 
 <dl>
 <font color="orange">
 <dt><b>更新情報</b>
 </font>
 <ul>
+<li>2016/11/07: Version 6.1.2 リリース</li>
 <li>2016/09/02: Version 6.1.1 リリース</li>
 <li>2016/08/29: Version 6.1.0 リリース</li>
-<li>2016/05/09: Version 6.0.0 リリース</li>
 <li>2014/12/12: Version 5.9.6 リリース</li>
 </ul>
 </dl>

diff --git a/src/big5.c b/src/big5.c
@@ -55,9 +55,28 @@ big5_mbc_enc_len(const UChar* p)
 }
 
 static int
-is_valid_mbc_string(const UChar* s, const UChar* end)
+is_valid_mbc_string(const UChar* p, const UChar* end)
 {
- return onigenc_length_check_is_valid_mbc_string(ONIG_ENCODING_BIG5, s, end);
+ while (p < end) {
+ if (*p < 0x80) {
+ p++;
+ }
+ else if (*p < 0xa1) {
+ return FALSE;
+ }
+ else if (*p < 0xff) {
+ p++;
+ if (p >= end) return FALSE;
+ if (*p < 0x40) return FALSE;
+ if (*p > 0x7e && *p < 0xa1) return FALSE;
+ if (*p == 0xff) return FALSE;
+ p++;
+ }
+ else
+ return FALSE;
+ }
+
+ return TRUE;
 }
 
 static OnigCodePoint

diff --git a/src/euc_jp.c b/src/euc_jp.c
@@ -57,9 +57,39 @@ mbc_enc_len(const UChar* p)
 }
 
 static int
-is_valid_mbc_string(const UChar* s, const UChar* end)
+is_valid_mbc_string(const UChar* p, const UChar* end)
 {
- return onigenc_length_check_is_valid_mbc_string(ONIG_ENCODING_EUC_JP, s, end);
+ while (p < end) {
+ if (*p < 0x80) {
+ p++;
+ }
+ else if (*p > 0xa0) {
+ if (*p == 0xff) return FALSE;
+ p++;
+ if (p >= end) return FALSE;
+ if (*p < 0xa1 || *p == 0xff) return FALSE;
+ p++;
+ }
+ else if (*p == 0x8e) {
+ p++;
+ if (p >= end) return FALSE;
+ if (*p < 0xa1 || *p > 0xdf) return FALSE;
+ p++;
+ }
+ else if (*p == 0x8f) {
+ p++;
+ if (p >= end) return FALSE;
+ if (*p < 0xa1 || *p == 0xff) return FALSE;
+ p++;
+ if (p >= end) return FALSE;
+ if (*p < 0xa1 || *p == 0xff) return FALSE;
+ p++;
+ }
+ else
+ return FALSE;
+ }
+
+ return TRUE;
 }
 
 static OnigCodePoint

diff --git a/src/euc_kr.c b/src/euc_kr.c
@@ -55,9 +55,27 @@ euckr_mbc_enc_len(const UChar* p)
 }
 
 static int
-is_valid_mbc_string(const UChar* s, const UChar* end)
+is_valid_mbc_string(const UChar* p, const UChar* end)
 {
- return onigenc_length_check_is_valid_mbc_string(ONIG_ENCODING_EUC_KR, s, end);
+ while (p < end) {
+ if (*p < 0x80) {
+ p++;
+ }
+ else if (*p < 0xa1) {
+ return FALSE;
+ }
+ else if (*p < 0xff) {
+ p++;
+ if (p >= end) return FALSE;
+ if (*p < 0xa1 || *p == 0xff)
+ return FALSE;
+ p++;
+ }
+ else
+ return FALSE;
+ }
+
+ return TRUE;
 }
 
 static OnigCodePoint

diff --git a/src/euc_tw.c b/src/euc_tw.c
@@ -55,9 +55,42 @@ euctw_mbc_enc_len(const UChar* p)
 }
 
 static int
-is_valid_mbc_string(const UChar* s, const UChar* end)
+is_valid_mbc_string(const UChar* p, const UChar* end)
 {
- return onigenc_length_check_is_valid_mbc_string(ONIG_ENCODING_EUC_TW, s, end);
+ while (p < end) {
+ if (*p < 0x80) {
+ p++;
+ }
+ else if (*p < 0xa1) {
+ if (*p == 0x8e) {
+ p++;
+ if (p >= end) return FALSE;
+ if (*p < 0xa1 || *p > 0xb0) return FALSE;
+ p++;
+ if (p >= end) return FALSE;
+ if (*p < 0xa1 || *p == 0xff)
+ return FALSE;
+ p++;
+ if (p >= end) return FALSE;
+ if (*p < 0xa1 || *p == 0xff)
+ return FALSE;
+ p++;
+ }
+ else
+ return FALSE;
+ }
+ else if (*p < 0xff) {
+ p++;
+ if (p >= end) return FALSE;
+ if (*p < 0xa1 || *p == 0xff)
+ return FALSE;
+ p++;
+ }
+ else
+ return FALSE;
+ }
+
+ return TRUE;
 }
 
 static OnigCodePoint

diff --git a/src/gb18030.c b/src/gb18030.c
@@ -76,9 +76,43 @@ gb18030_mbc_enc_len(const UChar* p)
 }
 
 static int
-is_valid_mbc_string(const UChar* s, const UChar* end)
+is_valid_mbc_string(const UChar* p, const UChar* end)
 {
- return onigenc_length_check_is_valid_mbc_string(ONIG_ENCODING_GB18030, s, end);
+ while (p < end) {
+ if (*p < 0x80) {
+ p++;
+ }
+ else if (*p == 0x80 || *p == 0xff) {
+ return FALSE;
+ }
+ else {
+ p++;
+ if (p >= end) return FALSE;
+ if (*p < 0x40) {
+ if (*p < 0x30 || *p > 0x39)
+ return FALSE;
+
+ p++;
+ if (p >= end) return FALSE;
+ if (*p < 0x81 || *p == 0xff) return FALSE;
+
+ p++;
+ if (p >= end) return FALSE;
+ if (*p < 0x30 || *p > 0x39)
+ return FALSE;
+
+ p++;
+ }
+ else if (*p == 0x7f || *p == 0xff) {
+ return FALSE;
+ }
+ else {
+ p++;
+ }
+ }
+ }
+
+ return TRUE;
 }
 
 static OnigCodePoint

diff --git a/src/oniguruma.h b/src/oniguruma.h
@@ -36,7 +36,7 @@ extern "C" {
 #define ONIGURUMA
 #define ONIGURUMA_VERSION_MAJOR 6
 #define ONIGURUMA_VERSION_MINOR 1
-#define ONIGURUMA_VERSION_TEENY 1
+#define ONIGURUMA_VERSION_TEENY 2
 
 #ifdef __cplusplus
 # ifndef HAVE_PROTOTYPES
@@ -364,7 +364,7 @@ int onigenc_strlen_null P_((OnigEncoding enc, const OnigUChar* p));
 ONIG_EXTERN
 int onigenc_str_bytelen_null P_((OnigEncoding enc, const OnigUChar* p));
 ONIG_EXTERN
-int onigenc_is_valid_mbc_string P_((OnigEncoding enc, const UChar* s, const UChar* end));
+int onigenc_is_valid_mbc_string P_((OnigEncoding enc, const OnigUChar* s, const OnigUChar* end));
 
 
 
@@ -398,7 +398,8 @@ typedef unsigned int OnigOptionType;
 #define ONIG_OPTION_NOTBOL (ONIG_OPTION_CAPTURE_GROUP << 1)
 #define ONIG_OPTION_NOTEOL (ONIG_OPTION_NOTBOL << 1)
 #define ONIG_OPTION_POSIX_REGION (ONIG_OPTION_NOTEOL << 1)
-#define ONIG_OPTION_MAXBIT ONIG_OPTION_POSIX_REGION /* limit */
+#define ONIG_OPTION_CHECK_VALIDITY_OF_STRING (ONIG_OPTION_POSIX_REGION << 1)
+#define ONIG_OPTION_MAXBIT ONIG_OPTION_CHECK_VALIDITY_OF_STRING /* limit */
 
 #define ONIG_OPTION_ON(options,regopt) ((options) |= (regopt))
 #define ONIG_OPTION_OFF(options,regopt) ((options) &= ~(regopt))
@@ -742,7 +743,7 @@ void onig_free P_((OnigRegex));
 ONIG_EXTERN
 void onig_free_body P_((OnigRegex));
 ONIG_EXTERN
-int onig_scan(regex_t* reg, const UChar* str, const UChar* end, OnigRegion* region, OnigOptionType option, int (*scan_callback)(int, int, OnigRegion*, void*), void* callback_arg);
+int onig_scan(regex_t* reg, const OnigUChar* str, const OnigUChar* end, OnigRegion* region, OnigOptionType option, int (*scan_callback)(int, int, OnigRegion*, void*), void* callback_arg);
 ONIG_EXTERN
 int onig_search P_((OnigRegex, const OnigUChar* str, const OnigUChar* end, const OnigUChar* start, const OnigUChar* range, OnigRegion* region, OnigOptionType option));
 ONIG_EXTERN