Manual merge

mysql-test/r/func_test.result: Auto merged mysql-test/r/subselect.result: Auto merged mysql-test/r/user_var.result: Auto merged sql/item_cmpfunc.cc: Auto merged sql/item_cmpfunc.h: Auto merged sql/item_func.cc: Auto merged sql/item_func.h: Auto merged sql/item_row.cc: Auto merged sql/item_row.h: Auto merged sql/item_strfunc.h: Auto merged sql/sql_list.h: Auto merged sql/sql_select.h: Auto merged
author: unknown <igor@rurik.mysql.com> 2004-02-19 01:09:54 -0800
committer: unknown <igor@rurik.mysql.com> 2004-02-19 01:09:54 -0800
commit: f4353d48ccc39527540daf709a1167462b39738c (patch)
tree: aa627335f475f099e0d352b39af4e24d6e1ac189 /strings
parent: c88e5213d4f8acbbba33d293a9573042818f0aec (diff)
parent: 9ce4695da0bfe74e840dcf708da10cac9740e137 (diff)
download: mariadb-git-f4353d48ccc39527540daf709a1167462b39738c.tar.gz
20 files changed, 375 insertions, 24 deletions
diff --git a/strings/ctype-big5.c b/strings/ctype-big5.c
index 85f066c4eba..574156a99ed 100644
--- a/strings/ctype-big5.c
+++ b/strings/ctype-big5.c
@@ -377,7 +377,7 @@ static my_bool my_like_range_big5(CHARSET_INFO *cs __attribute__((unused)),
       *min_length= (uint) (min_str-min_org);
       *max_length= res_length;
       do {
-	*min_str++ = '\0';		/* Because if key compression */
+	*min_str++ = ' ';		/* Because if key compression */
 	*max_str++ = max_sort_char;
       } while (min_str != min_end);
       return 0;
@@ -6245,6 +6245,7 @@ static MY_CHARSET_HANDLER my_charset_big5_handler=
   mbcharlen_big5,
   my_numchars_mb,
   my_charpos_mb,
+  my_wellformedlen_mb,
   my_lengthsp_8bit,
   my_mb_wc_big5,	/* mb_wc       */
   my_wc_mb_big5,	/* wc_mb       */
@@ -6281,6 +6282,7 @@ CHARSET_INFO my_charset_big5_chinese_ci=
     "",
     "",
     1,			/* strxfrm_multiply */
+    1,			/* mbminlen   */
     2,			/* mbmaxlen   */
     0,
     &my_charset_big5_handler,
@@ -6304,6 +6306,7 @@ CHARSET_INFO my_charset_big5_bin=
     "",
     "",
     1,			/* strxfrm_multiply */
+    1,			/* mbminlen   */
     2,			/* mbmaxlen   */
     0,
     &my_charset_big5_handler,
diff --git a/strings/ctype-bin.c b/strings/ctype-bin.c
index 6f28c43b2c6..fc22938d46e 100644
--- a/strings/ctype-bin.c
+++ b/strings/ctype-bin.c
@@ -294,7 +294,7 @@ uint my_instr_bin(CHARSET_INFO *cs __attribute__((unused)),
     end= (const uchar*) b+b_length-s_length+1;
     search_end= (const uchar*) s + s_length;
 
-skipp:
+skip:
     while (str != end)
     {
       if ( (*str++) == (*search))
@@ -306,7 +306,7 @@ skipp:
 
 	while (j != search_end)
 	  if ((*i++) != (*j++))
-            goto skipp;
+            goto skip;
 
         if (nmatch > 0)
 	{
@@ -347,6 +347,7 @@ static MY_CHARSET_HANDLER my_charset_handler=
     my_mbcharlen_8bit,		/* mbcharlen     */
     my_numchars_8bit,
     my_charpos_8bit,
+    my_wellformedlen_8bit,
     my_lengthsp_8bit,
     my_mb_wc_bin,
     my_wc_mb_bin,
@@ -381,6 +382,7 @@ CHARSET_INFO my_charset_bin =
     NULL,			/* tab_from_uni  */
     "","",
     1,				/* strxfrm_multiply */
+    1,				/* mbminlen      */
     1,				/* mbmaxlen      */
     (char) 255,			/* max_sort_char */
     &my_charset_handler,
diff --git a/strings/ctype-czech.c b/strings/ctype-czech.c
index b2e4f1886ed..1a07a5eba7e 100644
--- a/strings/ctype-czech.c
+++ b/strings/ctype-czech.c
@@ -631,6 +631,7 @@ CHARSET_INFO my_charset_latin2_czech_ci =
     idx_uni_8859_2,	/* tab_from_uni */
     "","",
     4,			/* strxfrm_multiply */
+    1,			/* mbminlen   */
     1,			/* mbmaxlen  */
     0,
     &my_charset_8bit_handler,
diff --git a/strings/ctype-euc_kr.c b/strings/ctype-euc_kr.c
index addd7803680..278e8529e83 100644
--- a/strings/ctype-euc_kr.c
+++ b/strings/ctype-euc_kr.c
@@ -8653,6 +8653,7 @@ static MY_CHARSET_HANDLER my_charset_handler=
   mbcharlen_euc_kr,
   my_numchars_mb,
   my_charpos_mb,
+  my_wellformedlen_mb,
   my_lengthsp_8bit,
   my_mb_wc_euc_kr,	/* mb_wc   */
   my_wc_mb_euc_kr,	/* wc_mb   */
@@ -8689,6 +8690,7 @@ CHARSET_INFO my_charset_euckr_korean_ci=
     "",
     "",
     1,			/* strxfrm_multiply */
+    1,			/* mbminlen   */
     2,			/* mbmaxlen   */
     0,
     &my_charset_handler,
@@ -8712,6 +8714,7 @@ CHARSET_INFO my_charset_euckr_bin=
     "",
     "",
     1,			/* strxfrm_multiply */
+    1,			/* mbminlen   */
     2,			/* mbmaxlen   */
     0,
     &my_charset_handler,
diff --git a/strings/ctype-extra.c b/strings/ctype-extra.c
index 55bfa09ea5f..0085d264416 100644
--- a/strings/ctype-extra.c
+++ b/strings/ctype-extra.c
@@ -34,6 +34,7 @@ CHARSET_INFO compiled_charsets[] = {
     0,
     0,
     0,
+    0,
     NULL,
     NULL
   }
diff --git a/strings/ctype-gb2312.c b/strings/ctype-gb2312.c
index b84ddc9081b..722f00f0f7a 100644
--- a/strings/ctype-gb2312.c
+++ b/strings/ctype-gb2312.c
@@ -5704,6 +5704,7 @@ static MY_CHARSET_HANDLER my_charset_handler=
   mbcharlen_gb2312,
   my_numchars_mb,
   my_charpos_mb,
+  my_wellformedlen_mb,
   my_lengthsp_8bit,
   my_mb_wc_gb2312,	/* mb_wc      */
   my_wc_mb_gb2312,	/* wc_mb      */
@@ -5740,6 +5741,7 @@ CHARSET_INFO my_charset_gb2312_chinese_ci=
     "",
     "",
     1,			/* strxfrm_multiply */
+    1,			/* mbminlen   */
     2,			/* mbmaxlen   */
     0,
     &my_charset_handler,
@@ -5762,6 +5764,7 @@ CHARSET_INFO my_charset_gb2312_bin=
     "",
     "",
     1,			/* strxfrm_multiply */
+    1,			/* mbminlen   */
     2,			/* mbmaxlen   */
     0,
     &my_charset_handler,
diff --git a/strings/ctype-gbk.c b/strings/ctype-gbk.c
index 585dc66be4c..9e71a18e531 100644
--- a/strings/ctype-gbk.c
+++ b/strings/ctype-gbk.c
@@ -9900,6 +9900,7 @@ static MY_CHARSET_HANDLER my_charset_handler=
   mbcharlen_gbk,
   my_numchars_mb,
   my_charpos_mb,
+  my_wellformedlen_mb,
   my_lengthsp_8bit,
   my_mb_wc_gbk,
   my_wc_mb_gbk,
@@ -9936,6 +9937,7 @@ CHARSET_INFO my_charset_gbk_chinese_ci=
     "",
     "",
     1,			/* strxfrm_multiply */
+    1,			/* mbminlen   */
     2,			/* mbmaxlen */
     0,
     &my_charset_handler,
@@ -9958,6 +9960,7 @@ CHARSET_INFO my_charset_gbk_bin=
     "",
     "",
     1,			/* strxfrm_multiply */
+    1,			/* mbminlen   */
     2,			/* mbmaxlen */
     0,
     &my_charset_handler,
diff --git a/strings/ctype-latin1.c b/strings/ctype-latin1.c
index 15798abb85b..933737b5f61 100644
--- a/strings/ctype-latin1.c
+++ b/strings/ctype-latin1.c
@@ -180,6 +180,7 @@ static MY_CHARSET_HANDLER my_charset_handler=
     my_mbcharlen_8bit,
     my_numchars_8bit,
     my_charpos_8bit,
+    my_wellformedlen_8bit,
     my_lengthsp_8bit,
     my_mb_wc_latin1,
     my_wc_mb_latin1,
@@ -215,6 +216,7 @@ CHARSET_INFO my_charset_latin1=
     NULL,		/* tab_from_uni */
     "","",
     1,			/* strxfrm_multiply */
+    1,			/* mbminlen   */
     1,			/* mbmaxlen  */
     0,
     &my_charset_handler,
@@ -410,6 +412,7 @@ CHARSET_INFO my_charset_latin1_german2_ci=
   NULL,					/* tab_from_uni */
   "","",
   2,					/* strxfrm_multiply */
+  1,					/* mbminlen   */
   1,					/* mbmaxlen  */
   0,
   &my_charset_handler,
@@ -433,6 +436,7 @@ CHARSET_INFO my_charset_latin1_bin=
   "",
   "",
   1,					/* strxfrm_multiply */
+  1,					/* mbminlen   */
   1,					/* mbmaxlen  */
   0,
   &my_charset_handler,
diff --git a/strings/ctype-mb.c b/strings/ctype-mb.c
index 271c56b8a0a..46f3e2f4fc3 100644
--- a/strings/ctype-mb.c
+++ b/strings/ctype-mb.c
@@ -271,9 +271,28 @@ uint my_charpos_mb(CHARSET_INFO *cs __attribute__((unused)),
     b+= (mblen= my_ismbchar(cs,b,e)) ? mblen : 1;
     pos--;
   }
-  return b-b0;
+  return pos ? e+2-b0 : b-b0;
 }
 
+uint my_wellformedlen_mb(CHARSET_INFO *cs,
+			 const char *b, const char *e, uint pos)
+{
+  my_wc_t wc;
+  int mblen;
+  const char *b0= b;
+  
+  while (pos)
+  {
+    if ((mblen= cs->cset->mb_wc(cs, &wc, b, e)) <0)
+      break;
+    b+= mblen;
+    pos--;
+  }
+  return b - b0;
+}
+
+
+
 uint my_instr_mb(CHARSET_INFO *cs,
                  const char *b, uint b_length, 
                  const char *s, uint s_length,
diff --git a/strings/ctype-simple.c b/strings/ctype-simple.c
index 6f77d6f3e16..233251e16a8 100644
--- a/strings/ctype-simple.c
+++ b/strings/ctype-simple.c
@@ -1020,6 +1020,15 @@ uint my_charpos_8bit(CHARSET_INFO *cs __attribute__((unused)),
   return pos;
 }
 
+uint my_wellformedlen_8bit(CHARSET_INFO *cs __attribute__((unused)),
+			const char *start,
+			const char *end,
+			uint nchars)
+{
+  uint nbytes= (uint) (end-start);
+  return nbytes < nchars ? nbytes : nchars;
+}
+
 uint my_lengthsp_8bit(CHARSET_INFO *cs __attribute__((unused)),
 		      const char *ptr, uint length)
 {
@@ -1055,7 +1064,7 @@ uint my_instr_simple(CHARSET_INFO *cs,
     end= (const uchar*) b+b_length-s_length+1;
     search_end= (const uchar*) s + s_length;
     
-skipp:
+skip:
     while (str != end)
     {
       if (cs->sort_order[*str++] == cs->sort_order[*search])
@@ -1067,7 +1076,7 @@ skipp:
 	
 	while (j != search_end)
 	  if (cs->sort_order[*i++] != cs->sort_order[*j++]) 
-            goto skipp;
+            goto skip;
         
 	if (nmatch > 0)
 	{
@@ -1096,6 +1105,7 @@ MY_CHARSET_HANDLER my_charset_8bit_handler=
     my_mbcharlen_8bit,		/* mbcharlen     */
     my_numchars_8bit,
     my_charpos_8bit,
+    my_wellformedlen_8bit,
     my_lengthsp_8bit,
     my_mb_wc_8bit,
     my_wc_mb_8bit,
diff --git a/strings/ctype-sjis.c b/strings/ctype-sjis.c
index a84fbd16e5d..feff0fff227 100644
--- a/strings/ctype-sjis.c
+++ b/strings/ctype-sjis.c
@@ -4489,6 +4489,7 @@ static MY_CHARSET_HANDLER my_charset_handler=
   mbcharlen_sjis,
   my_numchars_mb,
   my_charpos_mb,
+  my_wellformedlen_mb,
   my_lengthsp_8bit,
   my_mb_wc_sjis,	/* mb_wc */
   my_wc_mb_sjis,	/* wc_mb */
@@ -4525,6 +4526,7 @@ CHARSET_INFO my_charset_sjis_japanese_ci=
     "",
     "",
     1,			/* strxfrm_multiply */
+    1,			/* mbminlen   */
     2,			/* mbmaxlen */
     0,
     &my_charset_handler,
@@ -4547,6 +4549,7 @@ CHARSET_INFO my_charset_sjis_bin=
     "",
     "",
     1,			/* strxfrm_multiply */
+    1,			/* mbminlen   */
     2,			/* mbmaxlen */
     0,
     &my_charset_handler,
diff --git a/strings/ctype-tis620.c b/strings/ctype-tis620.c
index a2edc33b3d2..d577b964405 100644
--- a/strings/ctype-tis620.c
+++ b/strings/ctype-tis620.c
@@ -701,6 +701,207 @@ void ThNormalize(uchar* ptr, uint field_length, const uchar* from, uint length)
 }
 #endif /* NOT_NEEDED */
 
+static unsigned short cs_to_uni[256]={
+0x0000,0x0001,0x0002,0x0003,0x0004,0x0005,0x0006,0x0007,
+0x0008,0x0009,0x000A,0x000B,0x000C,0x000D,0x000E,0x000F,
+0x0010,0x0011,0x0012,0x0013,0x0014,0x0015,0x0016,0x0017,
+0x0018,0x0019,0x001A,0x001B,0x001C,0x001D,0x001E,0x001F,
+0x0020,0x0021,0x0022,0x0023,0x0024,0x0025,0x0026,0x0027,
+0x0028,0x0029,0x002A,0x002B,0x002C,0x002D,0x002E,0x002F,
+0x0030,0x0031,0x0032,0x0033,0x0034,0x0035,0x0036,0x0037,
+0x0038,0x0039,0x003A,0x003B,0x003C,0x003D,0x003E,0x003F,
+0x0040,0x0041,0x0042,0x0043,0x0044,0x0045,0x0046,0x0047,
+0x0048,0x0049,0x004A,0x004B,0x004C,0x004D,0x004E,0x004F,
+0x0050,0x0051,0x0052,0x0053,0x0054,0x0055,0x0056,0x0057,
+0x0058,0x0059,0x005A,0x005B,0x005C,0x005D,0x005E,0x005F,
+0x0060,0x0061,0x0062,0x0063,0x0064,0x0065,0x0066,0x0067,
+0x0068,0x0069,0x006A,0x006B,0x006C,0x006D,0x006E,0x006F,
+0x0070,0x0071,0x0072,0x0073,0x0074,0x0075,0x0076,0x0077,
+0x0078,0x0079,0x007A,0x007B,0x007C,0x007D,0x007E,0x007F,
+0x0080,0x0081,0x0082,0x0083,0x0084,0x0085,0x0086,0x0087,
+0x0088,0x0089,0x008A,0x008B,0x008C,0x008D,0x008E,0x008F,
+0x0090,0x0091,0x0092,0x0093,0x0094,0x0095,0x0096,0x0097,
+0x0098,0x0099,0x009A,0x009B,0x009C,0x009D,0x009E,0x009F,
+0xFFFD,0x0E01,0x0E02,0x0E03,0x0E04,0x0E05,0x0E06,0x0E07,
+0x0E08,0x0E09,0x0E0A,0x0E0B,0x0E0C,0x0E0D,0x0E0E,0x0E0F,
+0x0E10,0x0E11,0x0E12,0x0E13,0x0E14,0x0E15,0x0E16,0x0E17,
+0x0E18,0x0E19,0x0E1A,0x0E1B,0x0E1C,0x0E1D,0x0E1E,0x0E1F,
+0x0E20,0x0E21,0x0E22,0x0E23,0x0E24,0x0E25,0x0E26,0x0E27,
+0x0E28,0x0E29,0x0E2A,0x0E2B,0x0E2C,0x0E2D,0x0E2E,0x0E2F,
+0x0E30,0x0E31,0x0E32,0x0E33,0x0E34,0x0E35,0x0E36,0x0E37,
+0x0E38,0x0E39,0x0E3A,0xFFFD,0xFFFD,0xFFFD,0xFFFD,0x0E3F,
+0x0E40,0x0E41,0x0E42,0x0E43,0x0E44,0x0E45,0x0E46,0x0E47,
+0x0E48,0x0E49,0x0E4A,0x0E4B,0x0E4C,0x0E4D,0x0E4E,0x0E4F,
+0x0E50,0x0E51,0x0E52,0x0E53,0x0E54,0x0E55,0x0E56,0x0E57,
+0x0E58,0x0E59,0x0E5A,0x0E5B,0xFFFD,0xFFFD,0xFFFD,0xFFFD
+};
+static unsigned char pl00[256]={
+0x0000,0x0001,0x0002,0x0003,0x0004,0x0005,0x0006,0x0007,
+0x0008,0x0009,0x000A,0x000B,0x000C,0x000D,0x000E,0x000F,
+0x0010,0x0011,0x0012,0x0013,0x0014,0x0015,0x0016,0x0017,
+0x0018,0x0019,0x001A,0x001B,0x001C,0x001D,0x001E,0x001F,
+0x0020,0x0021,0x0022,0x0023,0x0024,0x0025,0x0026,0x0027,
+0x0028,0x0029,0x002A,0x002B,0x002C,0x002D,0x002E,0x002F,
+0x0030,0x0031,0x0032,0x0033,0x0034,0x0035,0x0036,0x0037,
+0x0038,0x0039,0x003A,0x003B,0x003C,0x003D,0x003E,0x003F,
+0x0040,0x0041,0x0042,0x0043,0x0044,0x0045,0x0046,0x0047,
+0x0048,0x0049,0x004A,0x004B,0x004C,0x004D,0x004E,0x004F,
+0x0050,0x0051,0x0052,0x0053,0x0054,0x0055,0x0056,0x0057,
+0x0058,0x0059,0x005A,0x005B,0x005C,0x005D,0x005E,0x005F,
+0x0060,0x0061,0x0062,0x0063,0x0064,0x0065,0x0066,0x0067,
+0x0068,0x0069,0x006A,0x006B,0x006C,0x006D,0x006E,0x006F,
+0x0070,0x0071,0x0072,0x0073,0x0074,0x0075,0x0076,0x0077,
+0x0078,0x0079,0x007A,0x007B,0x007C,0x007D,0x007E,0x007F,
+0x0080,0x0081,0x0082,0x0083,0x0084,0x0085,0x0086,0x0087,
+0x0088,0x0089,0x008A,0x008B,0x008C,0x008D,0x008E,0x008F,
+0x0090,0x0091,0x0092,0x0093,0x0094,0x0095,0x0096,0x0097,
+0x0098,0x0099,0x009A,0x009B,0x009C,0x009D,0x009E,0x009F,
+0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,
+0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,
+0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,
+0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,
+0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,
+0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,
+0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,
+0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,
+0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,
+0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,
+0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,
+0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,0x0000
+};
+static unsigned char pl0E[256]={
+0x0000,0x00A1,0x00A2,0x00A3,0x00A4,0x00A5,0x00A6,0x00A7,
+0x00A8,0x00A9,0x00AA,0x00AB,0x00AC,0x00AD,0x00AE,0x00AF,
+0x00B0,0x00B1,0x00B2,0x00B3,0x00B4,0x00B5,0x00B6,0x00B7,
+0x00B8,0x00B9,0x00BA,0x00BB,0x00BC,0x00BD,0x00BE,0x00BF,
+0x00C0,0x00C1,0x00C2,0x00C3,0x00C4,0x00C5,0x00C6,0x00C7,
+0x00C8,0x00C9,0x00CA,0x00CB,0x00CC,0x00CD,0x00CE,0x00CF,
+0x00D0,0x00D1,0x00D2,0x00D3,0x00D4,0x00D5,0x00D6,0x00D7,
+0x00D8,0x00D9,0x00DA,0x0000,0x0000,0x0000,0x0000,0x00DF,
+0x00E0,0x00E1,0x00E2,0x00E3,0x00E4,0x00E5,0x00E6,0x00E7,
+0x00E8,0x00E9,0x00EA,0x00EB,0x00EC,0x00ED,0x00EE,0x00EF,
+0x00F0,0x00F1,0x00F2,0x00F3,0x00F4,0x00F5,0x00F6,0x00F7,
+0x00F8,0x00F9,0x00FA,0x00FB,0x0000,0x0000,0x0000,0x0000,
+0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,
+0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,
+0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,
+0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,
+0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,
+0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,
+0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,
+0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,
+0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,
+0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,
+0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,
+0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,
+0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,
+0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,
+0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,
+0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,
+0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,
+0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,
+0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,
+0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,0x0000
+};
+static unsigned char plFF[256]={
+0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,
+0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,
+0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,
+0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,
+0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,
+0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,
+0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,
+0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,
+0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,
+0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,
+0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,
+0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,
+0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,
+0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,
+0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,
+0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,
+0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,
+0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,
+0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,
+0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,
+0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,
+0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,
+0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,
+0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,
+0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,
+0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,
+0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,
+0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,
+0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,
+0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,
+0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,0x0000,
+0x0000,0x0000,0x0000,0x0000,0x0000,0x00FF,0x0000,0x0000
+};
+static unsigned char *uni_to_cs[256]={
+pl00,NULL,NULL,NULL,NULL,NULL,NULL,NULL,
+NULL,NULL,NULL,NULL,NULL,NULL,pl0E,NULL,
+NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL,
+NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL,
+NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL,
+NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL,
+NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL,
+NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL,
+NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL,
+NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL,
+NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL,
+NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL,
+NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL,
+NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL,
+NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL,
+NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL,
+NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL,
+NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL,
+NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL,
+NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL,
+NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL,
+NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL,
+NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL,
+NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL,
+NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL,
+NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL,
+NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL,
+NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL,
+NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL,
+NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL,
+NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL,
+NULL,NULL,NULL,NULL,NULL,NULL,NULL,plFF
+};
+
+
+static
+int my_mb_wc_tis620(CHARSET_INFO *cs  __attribute__((unused)),
+		  my_wc_t *wc,
+		  const unsigned char *str,
+		  const unsigned char *end __attribute__((unused)))
+{
+  if (str >= end)
+    return MY_CS_TOOFEW(0);
+  
+  *wc=cs_to_uni[*str];
+  return (!wc[0] && str[0]) ? MY_CS_ILSEQ : 1;
+}
+
+static
+int my_wc_mb_tis620(CHARSET_INFO *cs  __attribute__((unused)),
+		  my_wc_t wc,
+		  unsigned char *str,
+		  unsigned char *end __attribute__((unused)))
+{
+  unsigned char *pl;
+  
+  if (str >= end)
+    return MY_CS_TOOSMALL;
+  
+  pl= uni_to_cs[(wc>>8) & 0xFF];
+  str[0]= pl ? pl[wc & 0xFF] : '\0';
+  return (!str[0] && wc) ? MY_CS_ILUNI : 1;
+}
+
 
 static MY_COLLATION_HANDLER my_collation_ci_handler =
 {
@@ -720,9 +921,10 @@ static MY_CHARSET_HANDLER my_charset_handler=
     my_mbcharlen_8bit,	/* mbcharlen */
     my_numchars_8bit,
     my_charpos_8bit,
+    my_wellformedlen_8bit,
     my_lengthsp_8bit,
-    my_mb_wc_8bit,	/* mb_wc     */
-    my_wc_mb_8bit,	/* wc_mb     */
+    my_mb_wc_tis620,	/* mb_wc     */
+    my_wc_mb_tis620,	/* wc_mb     */
     my_caseup_str_8bit,
     my_casedn_str_8bit,
     my_caseup_8bit,
@@ -757,6 +959,7 @@ CHARSET_INFO my_charset_tis620_thai_ci=
     "",
     "",
     4,			/* strxfrm_multiply */
+    1,			/* mbminlen   */
     1,			/* mbmaxlen  */
     0,
     &my_charset_handler,
@@ -779,6 +982,7 @@ CHARSET_INFO my_charset_tis620_bin=
     "",
     "",
     1,			/* strxfrm_multiply */
+    1,			/* mbminlen   */
     1,			/* mbmaxlen  */
     0,
     &my_charset_handler,
diff --git a/strings/ctype-ucs2.c b/strings/ctype-ucs2.c
index beb803a69f2..bb74e0cf56b 100644
--- a/strings/ctype-ucs2.c
+++ b/strings/ctype-ucs2.c
@@ -1002,6 +1002,17 @@ uint my_charpos_ucs2(CHARSET_INFO *cs __attribute__((unused)),
 }
 
 static
+uint my_wellformedlen_ucs2(CHARSET_INFO *cs __attribute__((unused)),
+		     const char *b,
+		     const char *e,
+		     uint nchars)
+{
+  uint nbytes= (e-b) & ~ (uint)1;
+  nchars*= 2;
+  return nbytes < nchars ? nbytes : nchars;
+}
+
+static
 void my_fill_ucs2(CHARSET_INFO *cs __attribute__((unused)),
 		   char *s, uint l, int fill)
 {
@@ -1044,14 +1055,13 @@ int my_wildcmp_ucs2(CHARSET_INFO *cs,
       scan= my_ucs2_uni(cs,&w_wc, (const uchar*)wildstr, (const uchar*)wildend);
       if (scan <= 0)
         return 1;
-      wildstr+= scan;
       
       if (w_wc ==  (my_wc_t)escape)
       {
+        wildstr+= scan;
         scan= my_ucs2_uni(cs,&w_wc, (const uchar*)wildstr, (const uchar*)wildend);
         if (scan <= 0)
           return 1;
-        wildstr+= scan;
       }
       
       if (w_wc == (my_wc_t)w_many)
@@ -1060,6 +1070,7 @@ int my_wildcmp_ucs2(CHARSET_INFO *cs,
         break;
       }
       
+      wildstr+= scan;
       scan= my_ucs2_uni(cs, &s_wc, (const uchar*)str, (const uchar*)str_end);
       if (scan <=0)
         return 1;
@@ -1095,13 +1106,16 @@ int my_wildcmp_ucs2(CHARSET_INFO *cs,
         scan= my_ucs2_uni(cs,&w_wc, (const uchar*)wildstr, (const uchar*)wildend);
         if (scan <= 0)
           return 1;
-        wildstr+= scan;
         
 	if (w_wc == (my_wc_t)w_many)
+	{
+	  wildstr+= scan;
 	  continue;
+	} 
 	
 	if (w_wc == (my_wc_t)w_one)
 	{
+	  wildstr+= scan;
 	  scan= my_ucs2_uni(cs, &s_wc, (const uchar*)str, (const uchar*)str_end);
           if (scan <=0)
             return 1;
@@ -1120,17 +1134,16 @@ int my_wildcmp_ucs2(CHARSET_INFO *cs,
       scan= my_ucs2_uni(cs,&w_wc, (const uchar*)wildstr, (const uchar*)wildend);
       if (scan <= 0)
         return 1;
-      wildstr+= scan;
       
       if (w_wc ==  (my_wc_t)escape)
       {
+        wildstr+= scan;
         scan= my_ucs2_uni(cs,&w_wc, (const uchar*)wildstr, (const uchar*)wildend);
         if (scan <= 0)
           return 1;
-        wildstr+= scan;
       }
       
-      do
+      while (1)
       {
         /* Skip until the first character from wildstr is found */
         while (str != str_end)
@@ -1138,8 +1151,6 @@ int my_wildcmp_ucs2(CHARSET_INFO *cs,
           scan= my_ucs2_uni(cs,&s_wc, (const uchar*)str, (const uchar*)str_end);
           if (scan <= 0)
             return 1;
-          str+= scan;
-          
           if (weights)
           {
             plane=(s_wc>>8) & 0xFF;
@@ -1150,17 +1161,19 @@ int my_wildcmp_ucs2(CHARSET_INFO *cs,
           
           if (s_wc == w_wc)
             break;
+          str+= scan;
         }
         if (str == str_end)
           return -1;
         
         result= my_wildcmp_ucs2(cs,str,str_end,wildstr,wildend,escape,
                                 w_one,w_many,weights);
+        
         if (result <= 0)
           return result;
         
-      } while (str != str_end && w_wc != (my_wc_t)w_many);
-      return -1;
+        str+= scan;
+      } 
     }
   }
   return (str != str_end ? 1 : 0);
@@ -1285,6 +1298,7 @@ static MY_CHARSET_HANDLER my_charset_ucs2_handler=
     my_mbcharlen_ucs2,	/* mbcharlen    */
     my_numchars_ucs2,
     my_charpos_ucs2,
+    my_wellformedlen_ucs2,
     my_lengthsp_ucs2,
     my_ucs2_uni,	/* mb_wc        */
     my_uni_ucs2,	/* wc_mb        */
@@ -1322,6 +1336,7 @@ CHARSET_INFO my_charset_ucs2_general_ci=
     "",
     "",
     1,			/* strxfrm_multiply */
+    2,			/* mbminlen     */
     2,			/* mbmaxlen     */
     0,
     &my_charset_ucs2_handler,
@@ -1345,6 +1360,7 @@ CHARSET_INFO my_charset_ucs2_bin=
     "",
     "",
     1,			/* strxfrm_multiply */
+    2,			/* mbminlen     */
     2,			/* mbmaxlen     */
     0,
     &my_charset_ucs2_handler,
diff --git a/strings/ctype-ujis.c b/strings/ctype-ujis.c
index 2815b70351b..f27ddcf3e30 100644
--- a/strings/ctype-ujis.c
+++ b/strings/ctype-ujis.c
@@ -8444,6 +8444,7 @@ static MY_CHARSET_HANDLER my_charset_handler=
     mbcharlen_ujis,
     my_numchars_mb,
     my_charpos_mb,
+    my_wellformedlen_mb,
     my_lengthsp_8bit,
     my_mb_wc_euc_jp,	 /* mb_wc       */
     my_wc_mb_euc_jp,	 /* wc_mb       */
@@ -8480,6 +8481,7 @@ CHARSET_INFO my_charset_ujis_japanese_ci=
     NULL,		/* tab_from_uni */
     "","",
     1,			/* strxfrm_multiply */
+    1,			/* mbminlen     */
     3,			/* mbmaxlen     */
     0,
     &my_charset_handler,
@@ -8502,6 +8504,7 @@ CHARSET_INFO my_charset_ujis_bin=
     NULL,		/* tab_from_uni */
     "","",
     1,			/* strxfrm_multiply */
+    1,			/* mbminlen     */
     3,			/* mbmaxlen     */
     0,
     &my_charset_handler,
diff --git a/strings/ctype-utf8.c b/strings/ctype-utf8.c
index b5716c53ea2..ef9719bf040 100644
--- a/strings/ctype-utf8.c
+++ b/strings/ctype-utf8.c
@@ -1969,6 +1969,7 @@ static MY_CHARSET_HANDLER my_charset_handler=
     my_mbcharlen_utf8,
     my_numchars_mb,
     my_charpos_mb,
+    my_wellformedlen_mb,
     my_lengthsp_8bit,
     my_utf8_uni,
     my_uni_utf8,
@@ -2006,6 +2007,7 @@ CHARSET_INFO my_charset_utf8_general_ci=
     "",
     "",
     1,			/* strxfrm_multiply */
+    1,			/* mbminlen     */
     3,			/* mbmaxlen     */
     0,
     &my_charset_handler,
@@ -2029,6 +2031,7 @@ CHARSET_INFO my_charset_utf8_bin=
     "",
     "",
     1,			/* strxfrm_multiply */
+    1,			/* mbminlen     */
     3,			/* mbmaxlen     */
     0,
     &my_charset_handler,
diff --git a/strings/ctype-win1250ch.c b/strings/ctype-win1250ch.c
index 60a5737009f..d3b5c9d1796 100644
--- a/strings/ctype-win1250ch.c
+++ b/strings/ctype-win1250ch.c
@@ -671,6 +671,7 @@ CHARSET_INFO my_charset_cp1250_czech_ci =
     idx_uni_cp1250,		/* tab_from_uni */
     "","",
     2,				/* strxfrm_multiply */
+    1,				/* mbminlen  */
     1,				/* mbmaxlen  */
     0,
     &my_charset_8bit_handler,
diff --git a/strings/dump_map.c b/strings/dump_map.c
new file mode 100644
index 00000000000..f999160d0d4
--- /dev/null
+++ b/strings/dump_map.c
@@ -0,0 +1,72 @@
+#include <stdio.h>
+#include <string.h>
+
+static void print_short_array(unsigned short *a)
+{
+  int i;
+  printf("{\n");
+  for (i=0; i<=0xFF; i++)
+  {
+    printf("0x%04X%s%s",(int)a[i],i<0xFF?",":"",(i+1) % 8 ? "" :"\n");
+  }
+  printf("};\n");
+  
+}
+
+
+
+int main(void)
+{
+  char str[160];
+  unsigned short touni[256];
+  unsigned short fromuni[65536];
+  unsigned short fromstat[256];
+  int i;
+  
+  bzero((void*)touni,sizeof(touni));
+  bzero((void*)fromuni,sizeof(fromuni));
+  bzero((void*)fromstat,sizeof(fromstat));
+  
+  while (fgets(str,sizeof(str),stdin))
+  {
+    unsigned int c,u;
+    
+    if ((str[0]=='#') || (2!=sscanf(str,"%x%x",&c,&u)))
+      continue;
+    if (c>0xFF || u>0xFFFF)
+      continue;
+    
+    touni[c]= u;
+    fromuni[u]= c;
+  }
+  
+  printf("unsigned short cs_to_uni[256]=");
+  print_short_array(touni);
+  
+  for (i=0;i<=0xFF;i++)
+  {
+    fromstat[touni[i]>>8]++;
+  }
+  
+  for (i=0;i<=256;i++)
+  {
+    if (fromstat[i])
+    { 
+      printf("unsigned char pl%02X[256]=",i);
+      print_short_array(fromuni+i*256);
+    }
+  }
+  
+  printf("unsigned short *uni_to_cs[256]={\n");
+  for (i=0;i<=255;i++)
+  {
+    if (fromstat[i])
+      printf("pl%02X",i);
+    else
+      printf("NULL");
+    printf("%s%s",i<255?",":"",((i+1) % 8) ? "":"\n");
+  }
+  printf("};\n");
+  
+  return 0;
+}
diff --git a/strings/r_strinstr.c b/strings/r_strinstr.c
index 76d310a3fda..fdd6e010a22 100644
--- a/strings/r_strinstr.c
+++ b/strings/r_strinstr.c
@@ -35,7 +35,7 @@ uint r_strinstr(reg1 my_string str,int from, reg4 my_string search)
   /* pointer to the last char of search */
   my_string	search_end = search + len - 1;
 
- skipp:
+ skip:
   while (start >= str)		/* Cant be != because the first char */
   {
     if (*start-- == *search_end)
@@ -43,7 +43,7 @@ uint r_strinstr(reg1 my_string str,int from, reg4 my_string search)
       i = start; j = search_end - 1;
       while (j >= search && start > str)
 	if (*i-- != *j--)
-	  goto skipp;
+	  goto skip;
       return (uint) ((start - len) - str + 3);
     }
   }
diff --git a/strings/strinstr.c b/strings/strinstr.c
index 1c814d19d47..457581af950 100644
--- a/strings/strinstr.c
+++ b/strings/strinstr.c
@@ -34,14 +34,14 @@ uint strinstr(reg1 const char *str,reg4 const char *search)
   reg2 my_string i,j;
   my_string start = (my_string) str;
 
- skipp:
+ skip:
   while (*str != '\0')
   {
     if (*str++ == *search)
     {
       i=(my_string) str; j= (my_string) search+1;
       while (*j)
-	if (*i++ != *j++) goto skipp;
+	if (*i++ != *j++) goto skip;
       return ((uint) (str - start));
     }
   }
diff --git a/strings/strstr.c b/strings/strstr.c
index ca845568ddb..bd65ef654b1 100644
--- a/strings/strstr.c
+++ b/strings/strstr.c
@@ -37,12 +37,12 @@ char *strstr(register const char *str,const char *search)
  register char *i,*j;
  register char first= *search;
 
-skipp:
+skip:
   while (*str != '\0') {
     if (*str++ == first) {
       i=(char*) str; j=(char*) search+1;
       while (*j)
-	if (*i++ != *j++) goto skipp;
+	if (*i++ != *j++) goto skip;
       return ((char*) str-1);
     }
   }
author	unknown <igor@rurik.mysql.com>	2004-02-19 01:09:54 -0800
committer	unknown <igor@rurik.mysql.com>	2004-02-19 01:09:54 -0800
commit	f4353d48ccc39527540daf709a1167462b39738c (patch)
tree	aa627335f475f099e0d352b39af4e24d6e1ac189 /strings
parent	c88e5213d4f8acbbba33d293a9573042818f0aec (diff)
parent	9ce4695da0bfe74e840dcf708da10cac9740e137 (diff)
download	mariadb-git-f4353d48ccc39527540daf709a1167462b39738c.tar.gz