Bug#15375 Unassigned multibyte codes are broken

  into parts when converting to Unicode.
m_ctype.h:
  Reorganizing mb_wc return codes to be able
  to return "an unassigned N-byte-long character".
sql_string.cc:
  Adding code to detect and properly handle
  unassigned characters (i.e. the those character
  which are correctly formed according to the 
  character specifications, but don't have Unicode
  mapping).
Many files:
  Fixing conversion function to return new codes.
ctype_ujis.test, ctype_gbk.test, ctype_big5.test:
  Adding a test case.
ctype_ujis.result, ctype_gbk.result, ctype_big5.result:
  Fixing results accordingly.


include/m_ctype.h:
  Reorganizing mb_wc return codes to be able
  to return "an unassigned N-byte long character".
  Bug#15375 Unassigned multibyte codes are broken into parts when converting to Unicode.
mysql-test/r/ctype_big5.result:
  Fixing results accordingly.
mysql-test/r/ctype_gbk.result:
  Fixing results accordingly.
mysql-test/r/ctype_ujis.result:
  Fixing results accordingly.
mysql-test/t/ctype_big5.test:
  Adding a test case.
mysql-test/t/ctype_gbk.test:
  Adding a test case.
mysql-test/t/ctype_ujis.test:
  Adding a test case.
sql/sql_string.cc:
  Adding code to detect and properly hanlde
  unassigned characters (i.e. the those character
  which are correctly formed according to the 
  character specifications, but don't have Unicode
  mapping).
strings/ctype-big5.c:
  Fixing conversion function to return new codes.
strings/ctype-bin.c:
  Fixing conversion function to return new codes.
strings/ctype-cp932.c:
  Fixing conversion function to return new codes.
strings/ctype-euc_kr.c:
  Fixing conversion function to return new codes.
strings/ctype-gb2312.c:
  Fixing conversion function to return new codes.
strings/ctype-gbk.c:
  Fixing conversion function to return new codes.
strings/ctype-latin1.c:
  Fixing conversion function to return new codes.
strings/ctype-simple.c:
  Fixing conversion function to return new codes.
strings/ctype-sjis.c:
  Fixing conversion function to return new codes.
strings/ctype-tis620.c:
  Fixing conversion function to return new codes.
strings/ctype-ucs2.c:
  Fixing conversion function to return new codes.
strings/ctype-ujis.c:
  Fixing conversion function to return new codes.
strings/ctype-utf8.c:
  Fixing conversion function to return new codes.

13 files changed, 51 insertions, 51 deletions

diff --git a/strings/ctype-big5.c b/strings/ctype-big5.c
index e15554fa576..89a40b15288 100644
--- a/strings/ctype-big5.c
+++ b/strings/ctype-big5.c
@@ -6259,7 +6259,7 @@ my_mb_wc_big5(CHARSET_INFO *cs __attribute__((unused)),
   int hi=s[0];
   
   if (s >= e)
-    return MY_CS_TOOFEW(0);
+    return MY_CS_TOOSMALL;
   
   if (hi<0x80)
   {
@@ -6268,10 +6268,10 @@ my_mb_wc_big5(CHARSET_INFO *cs __attribute__((unused)),
   }
   
   if (s+2>e)
-    return MY_CS_TOOFEW(0);
+    return MY_CS_TOOSMALL2;
 
   if (!(pwc[0]=func_big5_uni_onechar((hi<<8)+s[1])))
-    return MY_CS_ILSEQ;
+    return -2;
   
   return 2;
 }
diff --git a/strings/ctype-bin.c b/strings/ctype-bin.c
index 1ac79ac9ca0..7d0932e77bc 100644
--- a/strings/ctype-bin.c
+++ b/strings/ctype-bin.c
@@ -220,7 +220,7 @@ static int my_mb_wc_bin(CHARSET_INFO *cs __attribute__((unused)),
 			const unsigned char *end __attribute__((unused)))
 {
   if (str >= end)
-    return MY_CS_TOOFEW(0);
+    return MY_CS_TOOSMALL;
   
   *wc=str[0];
   return 1;
diff --git a/strings/ctype-cp932.c b/strings/ctype-cp932.c
index 63f95a28037..e8c62b0315e 100644
--- a/strings/ctype-cp932.c
+++ b/strings/ctype-cp932.c
@@ -5355,7 +5355,7 @@ my_mb_wc_cp932(CHARSET_INFO *cs  __attribute__((unused)),
   int hi=s[0];
   
   if (s >= e)
-    return MY_CS_TOOFEW(0);
+    return MY_CS_TOOSMALL;
   
   if (hi < 0x80)
   {
@@ -5370,10 +5370,10 @@ my_mb_wc_cp932(CHARSET_INFO *cs  __attribute__((unused)),
   }
   
   if (s+2>e)
-    return MY_CS_TOOFEW(0);
+    return MY_CS_TOOSMALL2;
   
   if (!(pwc[0]=func_cp932_uni_onechar((hi<<8)+s[1])))
-    return MY_CS_ILSEQ;
+    return -2;
   
   return 2;
 }
diff --git a/strings/ctype-euc_kr.c b/strings/ctype-euc_kr.c
index f15e97de5be..8ebcb34c0ab 100644
--- a/strings/ctype-euc_kr.c
+++ b/strings/ctype-euc_kr.c
@@ -8601,7 +8601,7 @@ my_wc_mb_euc_kr(CHARSET_INFO *cs __attribute__((unused)),
     return MY_CS_ILUNI;
   
   if (s+2>e)
-    return MY_CS_TOOSMALL;
+    return MY_CS_TOOSMALL2;
   
   s[0]=code>>8;
   s[1]=code&0xFF;
@@ -8617,7 +8617,7 @@ my_mb_wc_euc_kr(CHARSET_INFO *cs __attribute__((unused)),
   int hi=s[0];
   
   if (s >= e)
-    return MY_CS_TOOFEW(0);
+    return MY_CS_TOOSMALL;
   
   if (hi<0x80)
   {
@@ -8626,10 +8626,10 @@ my_mb_wc_euc_kr(CHARSET_INFO *cs __attribute__((unused)),
   }
   
   if (s+2>e)
-    return MY_CS_TOOFEW(0);
+    return MY_CS_TOOSMALL2;
   
   if (!(pwc[0]=func_ksc5601_uni_onechar((hi<<8)+s[1])))
-    return MY_CS_ILSEQ;
+    return -2;
   
   return 2;
 }
diff --git a/strings/ctype-gb2312.c b/strings/ctype-gb2312.c
index 0cbad2d1c55..4009b032dd8 100644
--- a/strings/ctype-gb2312.c
+++ b/strings/ctype-gb2312.c
@@ -5651,7 +5651,7 @@ my_wc_mb_gb2312(CHARSET_INFO *cs  __attribute__((unused)),
     return MY_CS_ILUNI;
   
   if (s+2>e)
-    return MY_CS_TOOSMALL;
+    return MY_CS_TOOSMALL2;
   
   code|=0x8080;
   s[0]=code>>8;
@@ -5668,7 +5668,7 @@ my_mb_wc_gb2312(CHARSET_INFO *cs  __attribute__((unused)),
   hi=(int) s[0];
   
   if (s >= e)
-    return MY_CS_TOOFEW(0);
+    return MY_CS_TOOSMALL;
   
   if (hi<0x80)
   {
@@ -5677,10 +5677,10 @@ my_mb_wc_gb2312(CHARSET_INFO *cs  __attribute__((unused)),
   }
   
   if (s+2>e)
-    return MY_CS_TOOFEW(0);
+    return MY_CS_TOOSMALL2;
   
   if (!(pwc[0]=func_gb2312_uni_onechar(((hi<<8)+s[1])&0x7F7F)))
-    return MY_CS_ILSEQ;
+    return -2;
   
   return 2;
 }
diff --git a/strings/ctype-gbk.c b/strings/ctype-gbk.c
index b5b86984794..a58c99fa1d6 100644
--- a/strings/ctype-gbk.c
+++ b/strings/ctype-gbk.c
@@ -9889,7 +9889,7 @@ my_wc_mb_gbk(CHARSET_INFO *cs  __attribute__((unused)),
     return MY_CS_ILUNI;
   
   if (s+2>e)
-    return MY_CS_TOOSMALL;
+    return MY_CS_TOOSMALL2;
     
   s[0]=code>>8;
   s[1]=code&0xFF;
@@ -9903,7 +9903,7 @@ my_mb_wc_gbk(CHARSET_INFO *cs __attribute__((unused)),
   int hi;
   
   if (s >= e)
-    return MY_CS_TOOFEW(0);
+    return MY_CS_TOOSMALL;
   
   hi=s[0];
   
@@ -9914,10 +9914,10 @@ my_mb_wc_gbk(CHARSET_INFO *cs __attribute__((unused)),
   }
   
   if (s+2>e)
-    return MY_CS_TOOFEW(0);
+    return MY_CS_TOOSMALL2;
     
   if (!(pwc[0]=func_gbk_uni_onechar( (hi<<8) + s[1])))
-    return MY_CS_ILSEQ;
+    return -2;
   
   return 2;
   
diff --git a/strings/ctype-latin1.c b/strings/ctype-latin1.c
index e5deba885e7..5d3fc75f337 100644
--- a/strings/ctype-latin1.c
+++ b/strings/ctype-latin1.c
@@ -363,10 +363,10 @@ int my_mb_wc_latin1(CHARSET_INFO *cs  __attribute__((unused)),
 		    const unsigned char *end __attribute__((unused)))
 {
   if (str >= end)
-    return MY_CS_TOOFEW(0);
+    return MY_CS_TOOSMALL;
   
   *wc=cs_to_uni[*str];
-  return (!wc[0] && str[0]) ? MY_CS_ILSEQ : 1;
+  return (!wc[0] && str[0]) ? -1 : 1;
 }
 
 static
diff --git a/strings/ctype-simple.c b/strings/ctype-simple.c
index 095b2f3a4ac..ccdfb5936b7 100644
--- a/strings/ctype-simple.c
+++ b/strings/ctype-simple.c
@@ -207,10 +207,10 @@ int my_mb_wc_8bit(CHARSET_INFO *cs,my_wc_t *wc,
 		  const unsigned char *end __attribute__((unused)))
 {
   if (str >= end)
-    return MY_CS_TOOFEW(0);
+    return MY_CS_TOOSMALL;
   
   *wc=cs->tab_to_uni[*str];
-  return (!wc[0] && str[0]) ? MY_CS_ILSEQ : 1;
+  return (!wc[0] && str[0]) ? -1 : 1;
 }
 
 int my_wc_mb_8bit(CHARSET_INFO *cs,my_wc_t wc,
diff --git a/strings/ctype-sjis.c b/strings/ctype-sjis.c
index da79f1796b8..38a9c9a6428 100644
--- a/strings/ctype-sjis.c
+++ b/strings/ctype-sjis.c
@@ -4501,7 +4501,7 @@ my_wc_mb_sjis(CHARSET_INFO *cs  __attribute__((unused)),
 
 mb:
   if (s+2>e)
-    return MY_CS_TOOSMALL;
+    return MY_CS_TOOSMALL2;
   
   s[0]=code>>8;
   s[1]=code&0xFF;
@@ -4515,7 +4515,7 @@ my_mb_wc_sjis(CHARSET_INFO *cs  __attribute__((unused)),
   int hi=s[0];
   
   if (s >= e)
-    return MY_CS_TOOFEW(0);
+    return MY_CS_TOOSMALL;
   
   if (hi < 0x80)
   {
@@ -4530,10 +4530,10 @@ my_mb_wc_sjis(CHARSET_INFO *cs  __attribute__((unused)),
   }
   
   if (s+2>e)
-    return MY_CS_TOOFEW(0);
+    return MY_CS_TOOSMALL2;
   
   if (!(pwc[0]=func_sjis_uni_onechar((hi<<8)+s[1])))
-    return MY_CS_ILSEQ;
+    return -2;
   
   return 2;
 }
diff --git a/strings/ctype-tis620.c b/strings/ctype-tis620.c
index 35cf10f1bcc..35c712b65e8 100644
--- a/strings/ctype-tis620.c
+++ b/strings/ctype-tis620.c
@@ -820,10 +820,10 @@ int my_mb_wc_tis620(CHARSET_INFO *cs  __attribute__((unused)),
 		  const unsigned char *end __attribute__((unused)))
 {
   if (str >= end)
-    return MY_CS_TOOFEW(0);
+    return MY_CS_TOOSMALL;
   
   *wc=cs_to_uni[*str];
-  return (!wc[0] && str[0]) ? MY_CS_ILSEQ : 1;
+  return (!wc[0] && str[0]) ? -1 : 1;
 }
 
 static
diff --git a/strings/ctype-ucs2.c b/strings/ctype-ucs2.c
index ad07fd9903c..2dd906d7953 100644
--- a/strings/ctype-ucs2.c
+++ b/strings/ctype-ucs2.c
@@ -95,7 +95,7 @@ static int my_ucs2_uni(CHARSET_INFO *cs __attribute__((unused)),
 		       my_wc_t * pwc, const uchar *s, const uchar *e)
 {
   if (s+2 > e) /* Need 2 characters */
-    return MY_CS_TOOFEW(0);
+    return MY_CS_TOOSMALL2;
   
   *pwc= ((unsigned char)s[0]) * 256  + ((unsigned char)s[1]);
   return 2;
@@ -105,7 +105,7 @@ static int my_uni_ucs2(CHARSET_INFO *cs __attribute__((unused)) ,
 		       my_wc_t wc, uchar *r, uchar *e)
 {
   if ( r+2 > e ) 
-    return MY_CS_TOOSMALL;
+    return MY_CS_TOOSMALL2;
   
   r[0]= (uchar) (wc >> 8);
   r[1]= (uchar) (wc & 0xFF);
diff --git a/strings/ctype-ujis.c b/strings/ctype-ujis.c
index b3bba85968e..e7a5cc97867 100644
--- a/strings/ctype-ujis.c
+++ b/strings/ctype-ujis.c
@@ -242,7 +242,7 @@ my_mb_wc_jisx0201(CHARSET_INFO *cs  __attribute__((unused)),
 		  const uchar *e __attribute__((unused)))
 {
   wc[0]=tab_jisx0201_uni[*s];
-  return (!wc[0] && s[0]) ? MY_CS_ILSEQ : 1;
+  return (!wc[0] && s[0]) ? -1 : 1;
 }
 
 
@@ -8341,7 +8341,7 @@ my_mb_wc_euc_jp(CHARSET_INFO *cs,my_wc_t *pwc, const uchar *s, const uchar *e)
   int c1,c2,c3;
   
   if (s >= e)
-    return MY_CS_TOOFEW(0);
+    return MY_CS_TOOSMALL;
   
   c1=s[0];
   
@@ -8353,7 +8353,7 @@ my_mb_wc_euc_jp(CHARSET_INFO *cs,my_wc_t *pwc, const uchar *s, const uchar *e)
   }
   
   if (s+2>e)
-    return MY_CS_TOOFEW(0);
+    return MY_CS_TOOSMALL2;
     
   c2=s[1];
   
@@ -8368,7 +8368,7 @@ my_mb_wc_euc_jp(CHARSET_INFO *cs,my_wc_t *pwc, const uchar *s, const uchar *e)
     {
       pwc[0]=my_jisx0208_uni_onechar( ((c1-0x80) << 8) + (c2-0x80));
       if (!pwc[0])
-        return MY_CS_ILSEQ;
+        return -2;
     }
     else
     {
@@ -8388,7 +8388,7 @@ my_mb_wc_euc_jp(CHARSET_INFO *cs,my_wc_t *pwc, const uchar *s, const uchar *e)
     
     ret = my_mb_wc_jisx0201(cs,pwc,s+1,e);
     if (ret!=1)
-      return ret;
+      return -2;
     return 2;
   }
   
@@ -8399,7 +8399,7 @@ my_mb_wc_euc_jp(CHARSET_INFO *cs,my_wc_t *pwc, const uchar *s, const uchar *e)
       return MY_CS_ILSEQ;
     
     if (s+3>e)
-      return MY_CS_TOOFEW(0);
+      return MY_CS_TOOSMALL3;
     
     c3=s[2];
     if (c3 < 0xA1 || c3>=0xFF)
@@ -8408,8 +8408,8 @@ my_mb_wc_euc_jp(CHARSET_INFO *cs,my_wc_t *pwc, const uchar *s, const uchar *e)
     if (c2<0xF5)
     {
       pwc[0]=my_jisx0212_uni_onechar((c2-0x80)*256 + (c3-0x80));
-      if (!pwc)
-        return MY_CS_ILSEQ;
+      if (!pwc[0])
+        return -3;
     }
     else
     {
@@ -8440,7 +8440,7 @@ my_wc_mb_euc_jp(CHARSET_INFO *c,my_wc_t wc, unsigned char *s, unsigned char *e)
   if ((jp=my_uni_jisx0208_onechar(wc)))
   {
     if (s+2>e)
-      return MY_CS_TOOSMALL;
+      return MY_CS_TOOSMALL2;
       
     jp+=0x8080;
     s[0]=jp>>8;
@@ -8452,7 +8452,7 @@ my_wc_mb_euc_jp(CHARSET_INFO *c,my_wc_t wc, unsigned char *s, unsigned char *e)
   if (my_wc_mb_jisx0201(c,wc,s,e) == 1)
   {
     if (s+2>e)
-      return MY_CS_TOOSMALL;
+      return MY_CS_TOOSMALL2;
     s[1]= s[0];
     s[0]= 0x8E;
     return 2;
@@ -8462,7 +8462,7 @@ my_wc_mb_euc_jp(CHARSET_INFO *c,my_wc_t wc, unsigned char *s, unsigned char *e)
   if ((jp=my_uni_jisx0212_onechar(wc)))
   {
     if (s+3>e)
-      return MY_CS_TOOSMALL;
+      return MY_CS_TOOSMALL3;
       
     jp+=0x8080;
     s[0]=0x8F;
@@ -8476,7 +8476,7 @@ my_wc_mb_euc_jp(CHARSET_INFO *c,my_wc_t wc, unsigned char *s, unsigned char *e)
   if (wc>=0xE000 && wc<0xE3AC)
   { 
     if (s+2>e)
-      return MY_CS_TOOSMALL;
+      return MY_CS_TOOSMALL2;
       
     c1=((unsigned)(wc-0xE000)/94)+0xF5;
     s[0]=c1;
@@ -8490,7 +8490,7 @@ my_wc_mb_euc_jp(CHARSET_INFO *c,my_wc_t wc, unsigned char *s, unsigned char *e)
   if (wc>=0xE3AC && wc<0xE758)
   {
     if (s+3>e)
-      return MY_CS_TOOSMALL;
+      return MY_CS_TOOSMALL3;
       
     s[0]=0x8F;
     c1=((unsigned)(wc-0xE3AC)/94)+0xF5;
diff --git a/strings/ctype-utf8.c b/strings/ctype-utf8.c
index b3e78ce27e9..94e8e6ba797 100644
--- a/strings/ctype-utf8.c
+++ b/strings/ctype-utf8.c
@@ -1765,7 +1765,7 @@ static int my_utf8_uni(CHARSET_INFO *cs __attribute__((unused)),
   unsigned char c;
 
   if (s >= e)
-    return MY_CS_TOOFEW(0);
+    return MY_CS_TOOSMALL;
 
   c= s[0];
   if (c < 0x80)
@@ -1778,7 +1778,7 @@ static int my_utf8_uni(CHARSET_INFO *cs __attribute__((unused)),
   else if (c < 0xe0)
   {
     if (s+2 > e) /* We need 2 characters */
-      return MY_CS_TOOFEW(0);
+      return MY_CS_TOOSMALL2;
 
     if (!((s[1] ^ 0x80) < 0x40))
       return MY_CS_ILSEQ;
@@ -1789,7 +1789,7 @@ static int my_utf8_uni(CHARSET_INFO *cs __attribute__((unused)),
   else if (c < 0xf0)
   {
     if (s+3 > e) /* We need 3 characters */
-      return MY_CS_TOOFEW(0);
+      return MY_CS_TOOSMALL3;
 
     if (!((s[1] ^ 0x80) < 0x40 && (s[2] ^ 0x80) < 0x40 && (c >= 0xe1 || s[1] >= 0xa0)))
       return MY_CS_ILSEQ;
@@ -1804,7 +1804,7 @@ static int my_utf8_uni(CHARSET_INFO *cs __attribute__((unused)),
   else if (c < 0xf8 && sizeof(my_wc_t)*8 >= 32)
   {
     if (s+4 > e) /* We need 4 characters */
-      return MY_CS_TOOFEW(0);
+      return MY_CS_TOOSMALL4;
 
     if (!((s[1] ^ 0x80) < 0x40 &&
           (s[2] ^ 0x80) < 0x40 &&
@@ -1822,7 +1822,7 @@ static int my_utf8_uni(CHARSET_INFO *cs __attribute__((unused)),
    else if (c < 0xfc && sizeof(my_wc_t)*8 >= 32)
   {
     if (s+5 >e) /* We need 5 characters */
-      return MY_CS_TOOFEW(0);
+      return MY_CS_TOOSMALL5;
 
     if (!((s[1] ^ 0x80) < 0x40 &&
           (s[2] ^ 0x80) < 0x40 &&
@@ -1841,7 +1841,7 @@ static int my_utf8_uni(CHARSET_INFO *cs __attribute__((unused)),
   else if (c < 0xfe && sizeof(my_wc_t)*8 >= 32)
   {
     if ( s+6 >e ) /* We need 6 characters */
-      return MY_CS_TOOFEW(0);
+      return MY_CS_TOOSMALL6;
 
     if (!((s[1] ^ 0x80) < 0x40   &&
           (s[2] ^ 0x80) < 0x40   &&
@@ -1892,7 +1892,7 @@ static int my_uni_utf8 (CHARSET_INFO *cs __attribute__((unused)) ,
     Because of it (r+count > e), not (r+count-1 >e )
    */
   if ( r+count > e )
-    return MY_CS_TOOSMALL;
+    return MY_CS_TOOSMALLN(count);
 
   switch (count) {
     /* Fall through all cases!!! */