charset<->unicode conversion tables and routines

some warnings fixes

include/m_ctype.h

@@ -42,7 +42,11 @@ typedef struct unicase_info_st {
 #define MY_CS_TOOSMALL	-1
 #define MY_CS_TOOFEW(n)	(-1-(n))
 
-                
+typedef struct my_uni_idx_st {
+  uint16 from;
+  uint16 to;
+  uchar  *tab;
+} MY_UNI_IDX;
 
 
 typedef struct charset_info_st
@@ -53,7 +57,10 @@ typedef struct charset_info_st
     uchar    *to_lower;
     uchar    *to_upper;
     uchar    *sort_order;
+    uint16      *tab_to_uni;
+    MY_UNI_IDX  *tab_from_uni;
     
+    /* Collation routines */
     uint      strxfrm_multiply;
     int     (*strnncoll)(struct charset_info_st *,
                          const uchar *, uint, const uchar *, uint);
@@ -63,11 +70,18 @@ typedef struct charset_info_st
                           const char *, uint, pchar, uint,
                           char *, char *, uint *, uint *);
     
+    /* Multibyte routines */
     uint      mbmaxlen;
     int     (*ismbchar)(struct charset_info_st *, const char *, const char *);
     my_bool (*ismbhead)(struct charset_info_st *, uint);
     int     (*mbcharlen)(struct charset_info_st *, uint);
     
+    /* Unicode convertion */
+    int (*mb_wc)(struct charset_info_st *cs,my_wc_t *wc,
+    		 const unsigned char *s,const unsigned char *e);
+    int (*wc_mb)(struct charset_info_st *cs,my_wc_t wc,
+    		 unsigned char *s,unsigned char *e);
+    
     /* Functions for case convertion */
     void    (*caseup_str)(struct charset_info_st *, char *);
     void    (*casedn_str)(struct charset_info_st *, char *);
@@ -106,6 +120,12 @@ extern void my_casedn_8bit(CHARSET_INFO *, char *, uint);
 extern int my_strcasecmp_8bit(CHARSET_INFO * cs, const char *, const char *);
 extern int my_strncasecmp_8bit(CHARSET_INFO * cs, const char *, const char *, uint);
 
+int my_mb_wc_8bit(CHARSET_INFO *cs,my_wc_t *wc,
+		  const unsigned char *s,const unsigned char *e);
+int my_wc_mb_8bit(CHARSET_INFO *cs,my_wc_t wc,
+		  unsigned char *s, unsigned char *e);
+
+
 #ifdef USE_MB
 /* Functions for multibyte charsets */
 extern void my_caseup_str_mb(CHARSET_INFO *, char *);
@@ -229,7 +249,7 @@ int my_strcasecmp_utf8(CHARSET_INFO *cs, const char *s, const char *t);
 int my_strncasecmp_utf8(CHARSET_INFO *cs, const char *s,const char *t,uint l);
 
 int my_utf8_uni (CHARSET_INFO *cs, my_wc_t *p, const uchar *s, const uchar *e);
-int my_uni_utf8 (CHARSET_INFO *cs, my_wc_t pwc  , uchar *b, uchar *e);
+int my_uni_utf8 (CHARSET_INFO *cs, my_wc_t wc, uchar *b, uchar *e);
 #endif
 
 #define	_U	01	/* Upper case */

mysys/hash.c

@@ -114,7 +114,8 @@ static uint hash_rec_mask(HASH *hash,HASH_LINK *pos,uint buffmax,
 
 	/* Calc hashvalue for a key */
 
-static uint calc_hashnr(CHARSET_INFO *cs, const byte *key,uint length)
+static uint calc_hashnr(CHARSET_INFO *cs __attribute__((unused)), 
+			const byte *key,uint length)
 {
   register uint nr=1, nr2=4;
   while (length--)

strings/ctype-big5.c

@@ -215,7 +215,7 @@ static uint16 big5strokexfrm(uint16 i)
   return 0xA140;
 }
 
-int my_strnncoll_big5(CHARSET_INFO *cs, 
+int my_strnncoll_big5(CHARSET_INFO *cs __attribute__((unused)), 
                       const uchar * s1, uint len1, 
                       const uchar * s2, uint len2)
 {
@@ -239,7 +239,7 @@ int my_strnncoll_big5(CHARSET_INFO *cs,
   return (int) (len1-len2);
 }
 
-int my_strnxfrm_big5(CHARSET_INFO *cs,
+int my_strnxfrm_big5(CHARSET_INFO *cs __attribute__((unused)),
                      uchar * dest, uint len, 
                      const uchar * src, uint srclen)
 {
@@ -328,7 +328,7 @@ int my_strxfrm_big5(uchar * dest, const uchar * src, int len)
 #define wild_one '_'
 #define wild_many '%'
 
-my_bool my_like_range_big5(CHARSET_INFO *cs,
+my_bool my_like_range_big5(CHARSET_INFO *cs __attribute__((unused)),
 		           const char *ptr,uint ptr_length,pchar escape,
 		           uint res_length, char *min_str,char *max_str,
 		           uint *min_length,uint *max_length)
@@ -378,17 +378,18 @@ my_bool my_like_range_big5(CHARSET_INFO *cs,
   return 0;
 }
 
-int ismbchar_big5(CHARSET_INFO *cs,const char* p, const char *e)
+int ismbchar_big5(CHARSET_INFO *cs __attribute__((unused)),
+                  const char* p, const char *e)
 {
   return (isbig5head(*(p)) && (e)-(p)>1 && isbig5tail(*((p)+1))? 2: 0);
 }
 
-my_bool ismbhead_big5(CHARSET_INFO *cs, uint c)
+my_bool ismbhead_big5(CHARSET_INFO *cs __attribute__((unused)), uint c)
 {
   return isbig5head(c);
 }
 
-int mbcharlen_big5(CHARSET_INFO *cs, uint c)
+int mbcharlen_big5(CHARSET_INFO *cs __attribute__((unused)), uint c)
 {
   return (isbig5head(c)? 2: 0);
 }

strings/ctype-czech.c

@@ -282,7 +282,7 @@ int my_strxfrm_czech(uchar * dest, const uchar * src, int len)
 
 /* Function strnncoll, actually strcoll, with Czech sorting, which expect
    the length of the strings being specified */
-int my_strnncoll_czech(CHARSET_INFO *cs,
+int my_strnncoll_czech(CHARSET_INFO *cs __attribute__((unused)),
                        const uchar * s1, uint len1, 
                        const uchar * s2, uint len2)
 	{
@@ -308,7 +308,7 @@ int my_strnncoll_czech(CHARSET_INFO *cs,
 
 /* Function strnxfrm, actually strxfrm, with Czech sorting, which expect
    the length of the strings being specified */
-int my_strnxfrm_czech(CHARSET_INFO *cs, 
+int my_strnxfrm_czech(CHARSET_INFO *cs __attribute__((unused)), 
                       uchar * dest, uint len,
                       const uchar * src, uint srclen)
 	{
@@ -378,7 +378,7 @@ int my_strnxfrm_czech(CHARSET_INFO *cs,
 
 #define EXAMPLE
 
-my_bool my_like_range_czech(CHARSET_INFO *cs,
+my_bool my_like_range_czech(CHARSET_INFO *cs __attribute__((unused)),
                             const char *ptr,uint ptr_length,pchar escape,
 		            uint res_length, char *min_str,char *max_str,
 		            uint *min_length,uint *max_length)

strings/ctype-euc_kr.c

@@ -183,19 +183,20 @@ uchar NEAR sort_order_euc_kr[]=
 #define iseuc_kr(c)     ((0xa1<=(uchar)(c) && (uchar)(c)<=0xfe))
 
 
-int ismbchar_euc_kr(CHARSET_INFO *cs,const char* p, const char *e)
+int ismbchar_euc_kr(CHARSET_INFO *cs __attribute__((unused)),
+		    const char* p, const char *e)
 {
   return ((*(uchar*)(p)<0x80)? 0:\
           iseuc_kr(*(p)) && (e)-(p)>1 && iseuc_kr(*((p)+1))? 2:\
           0);
 }
 
-my_bool ismbhead_euc_kr(CHARSET_INFO *cs,uint c)
+my_bool ismbhead_euc_kr(CHARSET_INFO *cs __attribute__((unused)),uint c)
 {
   return (iseuc_kr(c));
 }
 
-int mbcharlen_euc_kr(CHARSET_INFO *cs,uint c)
+int mbcharlen_euc_kr(CHARSET_INFO *cs __attribute__((unused)),uint c)
 {
   return (iseuc_kr(c) ? 2 : 0);
 }

strings/ctype-gb2312.c

@@ -166,17 +166,18 @@ uchar NEAR sort_order_gb2312[]=
 #define isgb2312tail(c) (0xa1<=(uchar)(c) && (uchar)(c)<=0xfe)
 
 
-int ismbchar_gb2312(CHARSET_INFO *cs,const char* p, const char *e)
+int ismbchar_gb2312(CHARSET_INFO *cs __attribute__((unused)),
+		    const char* p, const char *e)
 {
   return (isgb2312head(*(p)) && (e)-(p)>1 && isgb2312tail(*((p)+1))? 2: 0);
 }
 
-my_bool ismbhead_gb2312(CHARSET_INFO *cs,uint c)
+my_bool ismbhead_gb2312(CHARSET_INFO *cs __attribute__((unused)),uint c)
 {
   return isgb2312head(c);
 }
 
-int mbcharlen_gb2312(CHARSET_INFO *cs,uint c)
+int mbcharlen_gb2312(CHARSET_INFO *cs __attribute__((unused)),uint c)
 {
   return (isgb2312head(c)? 2:0);
 }

strings/ctype-gbk.c

@@ -2582,7 +2582,7 @@ static uint16 gbksortorder(uint16 i)
 }
 
 
-int my_strnncoll_gbk(CHARSET_INFO *cs,
+int my_strnncoll_gbk(CHARSET_INFO *cs __attribute__((unused)),
                      const uchar * s1, uint len1, 
                      const uchar * s2, uint len2)
 {
@@ -2609,7 +2609,7 @@ int my_strnncoll_gbk(CHARSET_INFO *cs,
 }
 
 
-int my_strnxfrm_gbk(CHARSET_INFO *cs, 
+int my_strnxfrm_gbk(CHARSET_INFO *cs __attribute__((unused)),
                     uchar * dest, uint len,
                     const uchar * src, uint srclen)
 {
@@ -2653,7 +2653,7 @@ int my_strnxfrm_gbk(CHARSET_INFO *cs,
 #define wild_one '_'
 #define wild_many '%'
 
-extern my_bool my_like_range_gbk(CHARSET_INFO *cs,
+extern my_bool my_like_range_gbk(CHARSET_INFO *cs __attribute__((unused)),
                                  const char *ptr,uint ptr_length,pchar escape,
                                  uint res_length, char *min_str,char *max_str,
                                  uint *min_length,uint *max_length)
@@ -2704,17 +2704,18 @@ extern my_bool my_like_range_gbk(CHARSET_INFO *cs,
 }
 
 
-int ismbchar_gbk(CHARSET_INFO *cs,const char* p, const char *e)
+int ismbchar_gbk(CHARSET_INFO *cs __attribute__((unused)),
+		 const char* p, const char *e)
 {
   return (isgbkhead(*(p)) && (e)-(p)>1 && isgbktail(*((p)+1))? 2: 0);
 }
 
-my_bool ismbhead_gbk(CHARSET_INFO *cs,uint c)
+my_bool ismbhead_gbk(CHARSET_INFO *cs __attribute__((unused)),uint c)
 {
   return isgbkhead(c);
 }
 
-int mbcharlen_gbk(CHARSET_INFO *cs,uint c)
+int mbcharlen_gbk(CHARSET_INFO *cs __attribute__((unused)),uint c)
 {
   return (isgbkhead(c)? 2:0);
 }

strings/ctype-latin1_de.c

@@ -164,7 +164,7 @@ uchar sort_order_latin1_de[] = {
   }
 
 
-int my_strnncoll_latin1_de(CHARSET_INFO *cs,
+int my_strnncoll_latin1_de(CHARSET_INFO *cs __attribute__((unused)),
                            const uchar * s1, uint len1,
                            const uchar * s2, uint len2)
 {
@@ -238,7 +238,7 @@ int my_strnncoll_latin1_de(CHARSET_INFO *cs,
 }
 
 
-int my_strnxfrm_latin1_de(CHARSET_INFO *cs,
+int my_strnxfrm_latin1_de(CHARSET_INFO *cs __attribute__((unused)),
                           uchar * dest, uint len,
                           const uchar * src, uint srclen)
 {
@@ -303,7 +303,7 @@ int my_strnxfrm_latin1_de(CHARSET_INFO *cs,
 #define wild_one '_'
 #define wild_many '%'
 
-my_bool my_like_range_latin1_de(CHARSET_INFO *cs,
+my_bool my_like_range_latin1_de(CHARSET_INFO *cs __attribute__((unused)),
 				const char *ptr, uint ptr_length,
 				pchar escape, uint res_length,
 				char *min_str, char *max_str,

strings/ctype-simple.c

@@ -85,3 +85,26 @@ int my_strncasecmp_8bit(CHARSET_INFO * cs,
  while (len-- != 0 && my_toupper(cs,(uchar)*s++) == my_toupper(cs,(uchar)*t++)) ;
    return (int) len+1;
 }
+
+int my_mb_wc_8bit(CHARSET_INFO *cs,my_wc_t *wc,
+		  const unsigned char *str,
+		  const unsigned char *end __attribute__((unused)))
+{
+  *wc=cs->tab_to_uni[*str];
+  return (!wc[0] && str[0]) ? MY_CS_ILSEQ : 1;
+}
+
+int my_wc_mb_8bit(CHARSET_INFO *cs,my_wc_t wc,
+		  unsigned char *s,
+		  unsigned char *e __attribute__((unused)))
+{
+  MY_UNI_IDX *idx;
+
+  for(idx=cs->tab_from_uni; idx->tab ; idx++){
+    if(idx->from<=wc && idx->to>=wc){
+      s[0]=idx->tab[wc-idx->from];
+      return (!s[0] && wc) ? MY_CS_ILUNI : 1;
+    }
+  }
+  return MY_CS_ILUNI;
+}

strings/ctype-sjis.c

@@ -183,17 +183,18 @@ uchar NEAR sort_order_sjis[]=
                        (0x80<=(c) && (c)<=0xfc))
 
 
-int ismbchar_sjis(CHARSET_INFO *cs,const char* p, const char *e)
+int ismbchar_sjis(CHARSET_INFO *cs __attribute__((unused)),
+		  const char* p, const char *e)
 {
   return (issjishead((uchar) *p) && (e-p)>1 && issjistail((uchar)p[1]) ? 2: 0);
 }
 
-my_bool ismbhead_sjis(CHARSET_INFO *cs,uint c)
+my_bool ismbhead_sjis(CHARSET_INFO *cs __attribute__((unused)),uint c)
 {
   return issjishead((uchar) c);
 }
 
-int mbcharlen_sjis(CHARSET_INFO *cs,uint c)
+int mbcharlen_sjis(CHARSET_INFO *cs __attribute__((unused)),uint c)
 {
   return (issjishead((uchar) c) ? 2: 0);
 }
@@ -201,7 +202,7 @@ int mbcharlen_sjis(CHARSET_INFO *cs,uint c)
 
 #define sjiscode(c,d)	((((uint) (uchar)(c)) << 8) | (uint) (uchar) (d))
 
-int my_strnncoll_sjis(CHARSET_INFO *cs,
+int my_strnncoll_sjis(CHARSET_INFO *cs __attribute__((unused)),
                       const uchar *s1, uint len1,
                       const uchar *s2, uint len2)
 {
@@ -226,7 +227,7 @@ int my_strnncoll_sjis(CHARSET_INFO *cs,
   return len1 - len2;
 }
 
-int my_strnxfrm_sjis(CHARSET_INFO *cs,
+int my_strnxfrm_sjis(CHARSET_INFO *cs __attribute__((unused)),
                      uchar *dest, uint len,
                      const uchar *src, uint srclen)
 {
@@ -265,7 +266,7 @@ int my_strnxfrm_sjis(CHARSET_INFO *cs,
 #define wild_one '_'
 #define wild_many '%'
 
-my_bool my_like_range_sjis(CHARSET_INFO *cs,
+my_bool my_like_range_sjis(CHARSET_INFO *cs __attribute__((unused)),
                            const char *ptr,uint ptr_length,pchar escape,
                            uint res_length, char *min_str,char *max_str,
                            uint *min_length,uint *max_length)

strings/ctype-tis620.c

@@ -537,7 +537,7 @@ static uchar* thai2sortable(const uchar * tstr,uint len)
    Arg: 2 Strings and it compare length
    Ret: strcmp result
 */
-int my_strnncoll_tis620(CHARSET_INFO *cs,
+int my_strnncoll_tis620(CHARSET_INFO *cs __attribute__((unused)),
                         const uchar * s1, uint len1, 
                         const uchar * s2, uint len2)
 {
@@ -555,7 +555,7 @@ int my_strnncoll_tis620(CHARSET_INFO *cs,
    Arg: Destination buffer, source string, dest length and source length
    Ret: Conveted string size
 */
-int my_strnxfrm_tis620(CHARSET_INFO *cs,
+int my_strnxfrm_tis620(CHARSET_INFO *cs __attribute__((unused)),
                        uchar * dest, uint len,
                        const uchar * src, uint srclen)
 {
@@ -613,7 +613,7 @@ int my_strxfrm_tis620(uchar * dest, const uchar * src, int len)
 #define wild_one '_'
 #define wild_many '%'
 
-my_bool my_like_range_tis620(CHARSET_INFO *cs,
+my_bool my_like_range_tis620(CHARSET_INFO *cs __attribute__((unused)),
                        const char *ptr, uint ptr_length, pchar escape,
 		       uint res_length, char *min_str, char *max_str,
 		       uint *min_length, uint *max_length)

strings/ctype-ujis.c

@@ -183,7 +183,8 @@ uchar NEAR sort_order_ujis[]=
 #define isujis_ss3(c) (((c)&0xff) == 0x8f)
 
 
-int ismbchar_ujis(CHARSET_INFO *cs,const char* p, const char *e)
+int ismbchar_ujis(CHARSET_INFO *cs __attribute__((unused)),
+		  const char* p, const char *e)
 {
   return ((*(uchar*)(p)<0x80)? 0:\
     isujis(*(p)) && (e)-(p)>1 && isujis(*((p)+1))? 2:\
@@ -192,12 +193,12 @@ int ismbchar_ujis(CHARSET_INFO *cs,const char* p, const char *e)
     0);
 }
 
-my_bool ismbhead_ujis(CHARSET_INFO *cs,uint c)
+my_bool ismbhead_ujis(CHARSET_INFO *cs __attribute__((unused)),uint c)
 {
   return (isujis(c) || isujis_ss2(c) || isujis_ss3(c));
 }
 
-int mbcharlen_ujis(CHARSET_INFO *cs,uint c)
+int mbcharlen_ujis(CHARSET_INFO *cs __attribute__((unused)),uint c)
 {
   return (isujis(c)? 2: isujis_ss2(c)? 2: isujis_ss3(c)? 3: 0);
 }

strings/ctype-utf8.c

@@ -1571,7 +1571,7 @@ uchar to_upper_utf8[] = {
 };
 
 
-int my_utf8_uni (__attribute__((unused)) CHARSET_INFO *cs, 
+int my_utf8_uni (CHARSET_INFO *cs __attribute__((unused)) , 
                  my_wc_t * pwc, const uchar *s, const uchar *e)
 {
   unsigned char c = s[0];
@@ -1673,7 +1673,7 @@ int my_utf8_uni (__attribute__((unused)) CHARSET_INFO *cs,
     return MY_CS_ILSEQ;
 }
 
-int my_uni_utf8 (__attribute__((unused)) CHARSET_INFO *cs, 
+int my_uni_utf8 (CHARSET_INFO *cs __attribute__((unused)) ,
                  my_wc_t wc, uchar *r, uchar *e)
 {
   int count;
@@ -1879,12 +1879,12 @@ int my_ismbchar_utf8(CHARSET_INFO *cs,const char *b, const char *e)
   return (res>1) ? res : 0;
 }
 
-my_bool my_ismbhead_utf8(__attribute__((unused)) CHARSET_INFO *cs, uint ch)
+my_bool my_ismbhead_utf8(CHARSET_INFO *cs __attribute__((unused)) , uint ch)
 {
   return ( ch >= 0xc2 );
 }
 
-int my_mbcharlen_utf8(__attribute__((unused)) CHARSET_INFO *cs, uint c)
+int my_mbcharlen_utf8(CHARSET_INFO *cs  __attribute__((unused)) , uint c)
 {
   if (c < 0x80)
     return 1;