New ctype functions/macros to support many charsets at a time

15 files changed, 53 insertions, 252 deletions

diff --git a/mysys/charset.c b/mysys/charset.c
index 93e8b697904..1d84bb67d68 100644
--- a/mysys/charset.c
+++ b/mysys/charset.c
@@ -68,7 +68,7 @@ static my_bool get_word(struct simpleconfig_buf_st *fb, char *buf)
 
   for (;;)
   {
-    while (isspace(*endptr))
+    while (my_isspace(system_charset_info, *endptr))
       ++endptr;
     if (*endptr && *endptr != '#')		/* Not comment */
       break;					/* Found something */
@@ -77,7 +77,7 @@ static my_bool get_word(struct simpleconfig_buf_st *fb, char *buf)
     endptr = fb->buf;
   }
 
-  while (!isspace(*endptr))
+  while (!my_isspace(system_charset_info, *endptr))
     *buf++= *endptr++;
   *buf=0;
   fb->p = endptr;
@@ -540,11 +540,9 @@ void _print_csinfo(CHARSET_INFO *cs)
   printf("to_lower:\n"); _print_array(cs->to_lower, 256);
   printf("to_upper:\n"); _print_array(cs->to_upper, 256);
   printf("sort_order:\n"); _print_array(cs->sort_order, 256);
-  printf("collate:    %3s (%d, %p, %p, %p, %p, %p)\n",
+  printf("collate:    %3s (%d, %p, %p, %p)\n",
          cs->strxfrm_multiply ? "yes" : "no",
          cs->strxfrm_multiply,
-         cs->strcoll,
-         cs->strxfrm,
          cs->strnncoll,
          cs->strnxfrm,
          cs->like_range);
diff --git a/mysys/default.c b/mysys/default.c
index d3c1b3815eb..ad34d4b949d 100644
--- a/mysys/default.c
+++ b/mysys/default.c
@@ -247,7 +247,7 @@ static my_bool search_default_file(DYNAMIC_ARRAY *args, MEM_ROOT *alloc,
   {
     line++;
     /* Ignore comment and empty lines */
-    for (ptr=buff ; isspace(*ptr) ; ptr++ ) ;
+    for (ptr=buff ; my_isspace(system_charset_info,*ptr) ; ptr++ ) ;
     if (*ptr == '#' || *ptr == ';' || !*ptr)
       continue;
     if (*ptr == '[')				/* Group name */
@@ -260,7 +260,7 @@ static my_bool search_default_file(DYNAMIC_ARRAY *args, MEM_ROOT *alloc,
 		name,line);
 	goto err;
       }
-      for ( ; isspace(end[-1]) ; end--) ;	/* Remove end space */
+      for ( ; my_isspace(system_charset_info,end[-1]) ; end--) ;/* Remove end space */
       end[0]=0;
       read_values=find_type(ptr,group,3) > 0;
       continue;
@@ -276,7 +276,7 @@ static my_bool search_default_file(DYNAMIC_ARRAY *args, MEM_ROOT *alloc,
       continue;
     if (!(end=value=strchr(ptr,'=')))
       end=strend(ptr);				/* Option without argument */
-    for ( ; isspace(end[-1]) ; end--) ;
+    for ( ; my_isspace(system_charset_info,end[-1]) ; end--) ;
     if (!value)
     {
       if (!(tmp=alloc_root(alloc,(uint) (end-ptr)+3)))
@@ -289,9 +289,9 @@ static my_bool search_default_file(DYNAMIC_ARRAY *args, MEM_ROOT *alloc,
     {
       /* Remove pre- and end space */
       char *value_end;
-      for (value++ ; isspace(*value); value++) ;
+      for (value++ ; my_isspace(system_charset_info,*value); value++) ;
       value_end=strend(value);
-      for ( ; isspace(value_end[-1]) ; value_end--) ;
+      for ( ; my_isspace(system_charset_info,value_end[-1]) ; value_end--) ;
       if (value_end < value)			/* Empty string */
 	value_end=value;
       if (!(tmp=alloc_root(alloc,(uint) (end-ptr)+3 +
diff --git a/mysys/getvar.c b/mysys/getvar.c
index bb470423c57..b5cc123f4b4 100644
--- a/mysys/getvar.c
+++ b/mysys/getvar.c
@@ -58,14 +58,15 @@ my_bool set_changeable_var(my_string str,CHANGEABLE_VAR *vars)
       longlong num;
 
       /* Skip end space from variable */
-      for (var_end=end ; end > str && isspace(var_end[-1]) ; var_end--) ;
+      for (var_end=end ; end > str && my_isspace(system_charset_info, var_end[-1]) ; var_end--) ;
       length=(uint) (var_end-str);
       /* Skip start space from argument */
-      for (end++ ; isspace(*end) ; end++) ;
+      for (end++ ; my_isspace(system_charset_info, *end) ; end++) ;
 
       for (var=vars,found=0 ; (name=var->name) ; var++)
       {
-	if (!my_casecmp(name,str,length))
+        /* BAR TODO: remove system_charset_info */
+	if (!my_strncasecmp(system_charset_info,name,str,length))
 	{
 	  found=var; found_count++;
 	  if (!name[length])
@@ -93,7 +94,7 @@ my_bool set_changeable_var(my_string str,CHANGEABLE_VAR *vars)
 	num*=1024L*1024L;
       else if (endchar == 'g' || endchar == 'G')
 	num*=1024L*1024L*1024L;
-      else if (!isdigit(endchar))
+      else if (!my_isdigit(system_charset_info, endchar))
       {
 	fprintf(stderr,"Unknown prefix used for variable value '%s'\n",str);
 	DBUG_RETURN(1);
diff --git a/mysys/hash.c b/mysys/hash.c
index 0b326ebc508..f1ce5052e4f 100644
--- a/mysys/hash.c
+++ b/mysys/hash.c
@@ -129,7 +129,9 @@ static uint calc_hashnr_caseup(const byte *key,uint length)
   register uint nr=1, nr2=4;
   while (length--)
   {
-    nr^= (((nr & 63)+nr2)*((uint) (uchar) toupper(*key++)))+ (nr << 8);
+    /* BAR TODO: remove default_charset_info */
+    nr^= (((nr & 63)+nr2)*
+         ((uint) (uchar) my_toupper(default_charset_info, *key++)))+ (nr << 8);
     nr2+=3;
   }
   return((uint) nr);
@@ -271,9 +273,10 @@ static int hashcmp(HASH *hash,HASH_LINK *pos,const byte *key,uint length)
 {
   uint rec_keylength;
   byte *rec_key=hash_key(hash,pos->data,&rec_keylength,1);
+  /* BAR TODO: remove default_charset_info */
   return (length && length != rec_keylength) ||
     (hash->flags & HASH_CASE_INSENSITIVE ?
-     my_casecmp(rec_key,key,rec_keylength) :
+     my_strncasecmp(default_charset_info, rec_key,key,rec_keylength) :
      memcmp(rec_key,key,rec_keylength));
 }
 
diff --git a/mysys/mf_casecnv.c b/mysys/mf_casecnv.c
index 0302688ad1f..510a9b08f76 100644
--- a/mysys/mf_casecnv.c
+++ b/mysys/mf_casecnv.c
@@ -28,234 +28,50 @@
 #include <m_string.h>
 #endif
 
-	/* string to uppercase */
-
-void caseup_str(my_string str)
-{
-#ifdef USE_MB
-  if (use_mb(default_charset_info))
-  {
-    register uint32 l;
-    register char *end=str+strlen(str);
-    while (*str)
-    {
-      if ((l=my_ismbchar(default_charset_info, str,end))) str+=l;
-      else *str=toupper(*str),++str;
-    }
-  }
-  else
-#endif
-    while ((*str = toupper(*str)) != 0)
-      str++;
-} /* caseup_str */
-
-	/* string to lowercase */
-
-void casedn_str(my_string str)
-{
-#ifdef USE_MB
-  if (use_mb(default_charset_info))
-  {
-    register uint32 l;
-    register char *end=str+strlen(str);
-    while (*str)
-    {
-      if ((l=my_ismbchar(default_charset_info, str,end))) str+=l;
-      else *str=tolower(*str),++str;
-    }
-  }
-  else
-#endif
-    while ((*str= tolower(*str)) != 0)
-      str++;
-} /* casedn_str */
-
-
-	/* to uppercase */
-
-void caseup(my_string str, uint length)
-{
-#ifdef USE_MB
-  if (use_mb(default_charset_info))
-  {
-    register uint32 l;
-    register char *end=str+length;
-    while (str<end)
-    {
-      if ((l=my_ismbchar(default_charset_info, str,end))) str+=l;
-      else *str=toupper(*str),++str;
-    }
-  }
-  else
-#endif
-    for ( ; length>0 ; length--, str++)
-      *str= toupper(*str);
-} /* caseup */
-
-	/* to lowercase */
-
-void casedn(my_string str, uint length)
-{
-#ifdef USE_MB
-  if (use_mb(default_charset_info))
-  {
-    register uint32 l;
-    register char *end=str+length;
-    while (str<end)
-    {
-      if ((l=my_ismbchar(default_charset_info, str,end))) str+=l;
-      else *str=tolower(*str),++str;
-    }
-  }
-  else
-#endif
-    for ( ; length>0 ; length--, str++)
-      *str= tolower(*str);
-} /* casedn */
 
 	/* to sort-string that can be compared to get text in order */
 
-void case_sort(my_string str, uint length)
+void case_sort(CHARSET_INFO *cs, my_string str, uint length)
 {
   for ( ; length>0 ; length--, str++)
-    *str= (char) my_sort_order[(uchar) *str];
+    *str= (char) cs->sort_order[(uchar) *str];
 } /* case_sort */
 
-	/* find string in another with no case_sensivity */
-
-/* ToDo: This function should be modified to support multibyte charset.
-         However it is not used untill 3.23.5.
-	 Wei He (hewei@mail.ied.ac.cn)
-*/
-
-my_string my_strcasestr(const char *str, const char *search)
-{
- uchar *i,*j,*pos;
-
- pos=(uchar*) str;
-skipp:
- while (*pos != '\0')
- {
-   if (toupper((uchar) *pos++) == toupper((uchar) *search))
-   {
-     i=(uchar*) pos; j=(uchar*) search+1;
-     while (*j)
-       if (toupper(*i++) != toupper(*j++)) goto skipp;
-     return ((char*) pos-1);
-   }
- }
- return ((my_string) 0);
-} /* strcstr */
-
-
-	/* compare strings without regarding to case */
-
-int my_strcasecmp(const char *s, const char *t)
-{
-#ifdef USE_MB
-  if (use_mb(default_charset_info))
-  {
-    register uint32 l;
-    register const char *end=s+strlen(s);
-    while (s<end)
-    {
-      if ((l=my_ismbchar(default_charset_info, s,end)))
-      {
-        while (l--)
-          if (*s++ != *t++) return 1;
-      }
-      else if (my_ismbhead(default_charset_info, *t)) return 1;
-      else if (toupper((uchar) *s++) != toupper((uchar) *t++)) return 1;
-    }
-    return *t;
-  }
-  else
-#endif
-  {
-    while (toupper((uchar) *s) == toupper((uchar) *t++))
-      if (!*s++) return 0;
-    return ((int) toupper((uchar) s[0]) - (int) toupper((uchar) t[-1]));
-  }
-}
-
-
-int my_casecmp(const char *s, const char *t, uint len)
-{
-#ifdef USE_MB
-  if (use_mb(default_charset_info))
-  {
-    register uint32 l;
-    register const char *end=s+len;
-    while (s<end)
-    {
-      if ((l=my_ismbchar(default_charset_info, s,end)))
-      {
-        while (l--)
-          if (*s++ != *t++) return 1;
-      }
-      else if (my_ismbhead(default_charset_info, *t)) return 1;
-      else if (toupper((uchar) *s++) != toupper((uchar) *t++)) return 1;
-    }
-    return 0;
-  }
-  else
-#endif
-  {
-    while (len-- != 0 && toupper(*s++) == toupper(*t++)) ;
-    return (int) len+1;
-  }
-}
-
-
-int my_strsortcmp(const char *s, const char *t)
-{
-#ifdef USE_STRCOLL
-  if (use_strcoll(default_charset_info))
-    return my_strcoll(default_charset_info, (uchar *)s, (uchar *)t);
-  else
-#endif
-  {
-    while (my_sort_order[(uchar) *s] == my_sort_order[(uchar) *t++])
-      if (!*s++) return 0;
-    return ((int) my_sort_order[(uchar) s[0]] -
-            (int) my_sort_order[(uchar) t[-1]]);
-  }
-}
 
-int my_sortcmp(const char *s, const char *t, uint len)
+int my_sortcmp(CHARSET_INFO *cs, const char *s, const char *t, uint len)
 {
 #ifdef USE_STRCOLL
-  if (use_strcoll(default_charset_info))
-    return my_strnncoll(default_charset_info,
-                        (uchar *)s, len, (uchar *)t, len);
+  if (use_strcoll(cs))
+    return my_strnncoll(cs,(uchar *)s, len, (uchar *)t, len);
   else
 #endif
   {
     while (len--)
     {
-      if (my_sort_order[(uchar) *s++] != my_sort_order[(uchar) *t++])
-        return ((int) my_sort_order[(uchar) s[-1]] -
-                (int) my_sort_order[(uchar) t[-1]]);
+      if (cs->sort_order[(uchar) *s++] != cs->sort_order[(uchar) *t++])
+        return ((int) cs->sort_order[(uchar) s[-1]] -
+                (int) cs->sort_order[(uchar) t[-1]]);
     }
     return 0;
   }
 }
 
-int my_sortncmp(const char *s, uint s_len, const char *t, uint t_len)
+int my_sortncmp(CHARSET_INFO *cs,
+                const char *s, uint s_len, 
+                const char *t, uint t_len)
 {
 #ifdef USE_STRCOLL
-  if (use_strcoll(default_charset_info))
-    return my_strnncoll(default_charset_info,
-                        (uchar *)s, s_len, (uchar *)t, t_len);
+  if (use_strcoll(cs))
+    return my_strnncoll(cs, (uchar *)s, s_len, (uchar *)t, t_len);
   else
 #endif
   {
     uint len= min(s_len,t_len);
     while (len--)
     {
-      if (my_sort_order[(uchar) *s++] != my_sort_order[(uchar) *t++])
-        return ((int) my_sort_order[(uchar) s[-1]] -
-                (int) my_sort_order[(uchar) t[-1]]);
+      if (cs->sort_order[(uchar) *s++] != cs->sort_order[(uchar) *t++])
+        return ((int) cs->sort_order[(uchar) s[-1]] -
+                (int) cs->sort_order[(uchar) t[-1]]);
     }
     return (int) (s_len - t_len);
   }
diff --git a/mysys/mf_dirname.c b/mysys/mf_dirname.c
index 6523abcdb12..320917c8f3e 100644
--- a/mysys/mf_dirname.c
+++ b/mysys/mf_dirname.c
@@ -108,11 +108,5 @@ char *convert_dirname(char *to, const char *from, const char *from_end)
     *to++=FN_LIBCHAR;
     *to=0;
   }
-#ifdef FN_UPPER_CASE
-  caseup_str(to_org);
-#endif
-#ifdef FN_LOWER_CASE
-  casedn_str(to_org);
-#endif
   return to;					/* Pointer to end of dir */
 } /* convert_dirname */
diff --git a/mysys/mf_format.c b/mysys/mf_format.c
index ab1904da162..e68f7df9d83 100644
--- a/mysys/mf_format.c
+++ b/mysys/mf_format.c
@@ -91,12 +91,6 @@ my_string fn_format(my_string to, const char *name, const char *dir,
       name=buff;
     }
     pos=strmake(strmov(to,dev),name,length);
-#ifdef FN_UPPER_CASE
-    caseup_str(to);
-#endif
-#ifdef FN_LOWER_CASE
-    casedn_str(to);
-#endif
     (void) strmov(pos,ext);			/* Don't convert extension */
   }
   /*
diff --git a/mysys/mf_iocache2.c b/mysys/mf_iocache2.c
index 127bf20e8fe..147850a10bc 100644
--- a/mysys/mf_iocache2.c
+++ b/mysys/mf_iocache2.c
@@ -249,7 +249,7 @@ uint my_b_vprintf(IO_CACHE *info, const char* fmt, va_list args)
       /* Found one '%' */
     }
     /* Skipp if max size is used (to be compatible with printf) */
-    while (isdigit(*fmt) || *fmt == '.' || *fmt == '-')
+    while (my_isdigit(system_charset_info, *fmt) || *fmt == '.' || *fmt == '-')
       fmt++;
     if (*fmt == 's')				/* String parameter */
     {
diff --git a/mysys/mf_soundex.c b/mysys/mf_soundex.c
index 4f7aa7da601..8ecd95137b0 100644
--- a/mysys/mf_soundex.c
+++ b/mysys/mf_soundex.c
@@ -34,12 +34,12 @@
 #include <m_ctype.h>
 #include "my_static.h"
 
-static char get_scode(char **ptr,pbool remove_garbage);
+static char get_scode(CHARSET_INFO * cs, char **ptr,pbool remove_garbage);
 
 		/* outputed string is 4 byte long */
 		/* out_pntr can be == in_pntr */
 
-void soundex(register my_string out_pntr, my_string in_pntr,
+void soundex(CHARSET_INFO * cs,register my_string out_pntr, my_string in_pntr,
 	     pbool remove_garbage)
 {
   char ch,last_ch;
@@ -47,11 +47,11 @@ void soundex(register my_string out_pntr, my_string in_pntr,
 
   if (remove_garbage)
   {
-    while (*in_pntr && isspace(*in_pntr))	/* Skipp pre-space */
+    while (*in_pntr && my_isspace(cs,*in_pntr))	/* Skipp pre-space */
       in_pntr++;
   }
-  *out_pntr++ = toupper(*in_pntr);	/* Copy first letter		 */
-  last_ch = get_scode(&in_pntr,0);	/* code of the first letter	 */
+  *out_pntr++ = my_toupper(cs,*in_pntr);/* Copy first letter		 */
+  last_ch = get_scode(cs,&in_pntr,0);	/* code of the first letter	 */
 					/* for the first 'double-letter  */
 					/* check.			 */
   end=out_pntr+3;			/* Loop on input letters until	 */
@@ -59,7 +59,7 @@ void soundex(register my_string out_pntr, my_string in_pntr,
 					/* letter code count = 3	 */
 
   in_pntr++;
-  while (out_pntr < end && (ch = get_scode(&in_pntr,remove_garbage)) != 0)
+  while (out_pntr < end && (ch = get_scode(cs,&in_pntr,remove_garbage)) != 0)
   {
     in_pntr++;
     if ((ch != '0') && (ch != last_ch)) /* if not skipped or double */
@@ -81,19 +81,19 @@ void soundex(register my_string out_pntr, my_string in_pntr,
     else return 0
     */
 
-static char get_scode(char **ptr, pbool remove_garbage)
+static char get_scode(CHARSET_INFO * cs,char **ptr, pbool remove_garbage)
 {
   uchar ch;
 
   if (remove_garbage)
   {
-    while (**ptr && !isalpha(**ptr))
+    while (**ptr && !my_isalpha(cs,**ptr))
       (*ptr)++;
   }
-  ch=toupper(**ptr);
+  ch=my_toupper(cs,**ptr);
   if (ch < 'A' || ch > 'Z')
   {
-    if (isalpha(ch))			/* If exetended alfa (country spec) */
+    if (my_isalpha(cs,ch))		/* If exetended alfa (country spec) */
       return '0';			/* threat as vokal */
     return 0;				/* Can't map */
   }
diff --git a/mysys/mf_wfile.c b/mysys/mf_wfile.c
index e9e12c72755..3c10de1f238 100644
--- a/mysys/mf_wfile.c
+++ b/mysys/mf_wfile.c
@@ -54,13 +54,6 @@ WF_PACK *wf_comp(my_string str)
   for (pos=str ; *pos ; pos++)
     ant+= test(*pos == ' ' || *pos == ',');
 
-#ifdef FN_UPPER_CASE
-    caseup(str,(int) (pos-str));
-#endif
-#ifdef FN_LOWER_CASE
-    casedn(str,(int) (pos-str));
-#endif
-
   if ((ret= (WF_PACK*) my_malloc((uint) ant*(sizeof(my_string*)+2)+
 				 sizeof(WF_PACK)+ (uint) strlen(str)+1,
 				 MYF(MY_WME)))
diff --git a/mysys/my_error.c b/mysys/my_error.c
index 61f8c16aba5..fd0cdcc2289 100644
--- a/mysys/my_error.c
+++ b/mysys/my_error.c
@@ -62,7 +62,7 @@ int my_error(int nr,myf MyFlags, ...)
     else
     {
       /* Skipp if max size is used (to be compatible with printf) */
-      while (isdigit(*tpos) || *tpos == '.' || *tpos == '-')
+      while (my_isdigit(system_charset_info, *tpos) || *tpos == '.' || *tpos == '-')
 	tpos++;
       if (*tpos == 'l')				/* Skipp 'l' argument */
 	tpos++;
diff --git a/mysys/my_getwd.c b/mysys/my_getwd.c
index 22c6a4dbc2b..240b17ef87f 100644
--- a/mysys/my_getwd.c
+++ b/mysys/my_getwd.c
@@ -109,7 +109,8 @@ int my_setwd(const char *dir, myf MyFlags)
     uint drive,drives;
 
     pos++;				/* Skipp FN_DEVCHAR */
-    drive=(uint) (toupper(dir[0])-'A'+1); drives= (uint) -1;
+    drive=(uint) (my_toupper(system_charset_info,dir[0])-'A'+1);
+    drives= (uint) -1;
     if ((pos-(byte*) dir) == 2 && drive > 0 && drive < 32)
     {
 #ifdef OS2
diff --git a/mysys/my_init.c b/mysys/my_init.c
index c4e6132aa9c..faf4e7f83f0 100644
--- a/mysys/my_init.c
+++ b/mysys/my_init.c
@@ -17,7 +17,6 @@
 #include "mysys_priv.h"
 #include "my_static.h"
 #include "mysys_err.h"
-#include "m_ctype.h"
 #include <m_string.h>
 #include <m_ctype.h>
 #ifdef THREAD
@@ -50,7 +49,7 @@ my_bool my_init_done=0;
 static ulong atoi_octal(const char *str)
 {
   long int tmp;
-  while (*str && isspace(*str))
+  while (*str && my_isspace(system_charset_info, *str))
     str++;
   str2int(str,
 	  (*str == '0' ? 8 : 10),		/* Octalt or decimalt */
diff --git a/mysys/my_vsnprintf.c b/mysys/my_vsnprintf.c
index 0fe6eff24a6..519faa1de03 100644
--- a/mysys/my_vsnprintf.c
+++ b/mysys/my_vsnprintf.c
@@ -41,7 +41,7 @@ int my_vsnprintf(char *to, size_t n, const char* fmt, va_list ap)
     }
     /* Skipp if max size is used (to be compatible with printf) */
     fmt++;
-    while (isdigit(*fmt) || *fmt == '.' || *fmt == '-')
+    while (my_isdigit(system_charset_info,*fmt) || *fmt == '.' || *fmt == '-')
       fmt++;
     if (*fmt == 'l')
       fmt++;
diff --git a/mysys/typelib.c b/mysys/typelib.c
index e2c8eade5c8..e524f903b5d 100644
--- a/mysys/typelib.c
+++ b/mysys/typelib.c
@@ -47,7 +47,9 @@ int find_type(my_string x, TYPELIB *typelib, uint full_name)
   find=0;
   for (pos=0 ; (j=typelib->type_names[pos]) ; pos++)
   {
-    for (i=x ; *i && toupper(*i) == toupper(*j) ; i++, j++) ;
+    for (i=x ; 
+    	*i && my_toupper(system_charset_info,*i) == 
+    		my_toupper(system_charset_info,*j) ; i++, j++) ;
     if (! *j)
     {
       while (*i == ' ')