Charset loading has been moved to /string library

include/m_ctype.h

@@ -27,6 +27,13 @@ extern "C" {
 #endif
 
 
+#define MY_CS_NAME_SIZE			32
+#define MY_CS_CTYPE_TABLE_SIZE		257
+#define MY_CS_TO_LOWER_TABLE_SIZE	256
+#define MY_CS_TO_UPPER_TABLE_SIZE	256
+#define MY_CS_SORT_ORDER_TABLE_SIZE	256
+#define MY_CS_TO_UNI_TABLE_SIZE		256
+
 #define CHARSET_DIR	"charsets/"
 
 #define my_wc_t ulong
@@ -145,7 +152,8 @@ extern CHARSET_INFO *default_charset_info;
 extern CHARSET_INFO *system_charset_info;
 extern CHARSET_INFO *all_charsets[256];
 extern my_bool init_compiled_charsets(myf flags);
-
+extern my_bool my_parse_charset_xml(const char *bug, uint len,
+				    int (*add)(CHARSET_INFO *cs));
 
 /* declarations for simple charsets */
 extern int  my_strnxfrm_simple(CHARSET_INFO *, uchar *, uint, const uchar *, uint); 

mysys/charset.c

@@ -22,217 +22,6 @@
 #include <my_xml.h>
 
 
-#define CS_MAX_NM_LEN	      32
-#define CTYPE_TABLE_SIZE      257
-#define TO_LOWER_TABLE_SIZE   256
-#define TO_UPPER_TABLE_SIZE   256
-#define SORT_ORDER_TABLE_SIZE 256
-#define TO_UNI_TABLE_SIZE     256
-
-
-static char *mstr(char *str,const char *src,uint l1,uint l2)
-{
-  l1= l1<l2 ? l1 : l2;
-  memcpy(str,src,l1);
-  str[l1]='\0';
-  return str;
-}
-
-
-struct my_cs_file_section_st
-{
-  int        state;
-  const char *str;
-};
-
-#define _CS_MISC	1
-#define _CS_ID		2
-#define _CS_CSNAME	3
-#define _CS_FAMILY	4
-#define _CS_ORDER	5
-#define _CS_COLNAME	6
-#define _CS_FLAG	7
-#define _CS_CHARSET	8
-#define _CS_COLLATION	9
-#define _CS_UPPERMAP	10
-#define _CS_LOWERMAP	11
-#define _CS_UNIMAP	12
-#define _CS_COLLMAP	13
-#define _CS_CTYPEMAP	14
-
-static struct my_cs_file_section_st sec[] =
-{
-  {_CS_MISC,		"xml"},
-  {_CS_MISC,		"xml.version"},
-  {_CS_MISC,		"xml.encoding"},
-  {_CS_MISC,		"charsets"},
-  {_CS_MISC,		"charsets.max-id"},
-  {_CS_MISC,		"charsets.description"},
-  {_CS_CHARSET,		"charsets.charset"},
-  {_CS_CSNAME,		"charsets.charset.name"},
-  {_CS_FAMILY,		"charsets.charset.family"},
-  {_CS_MISC,		"charsets.charset.alias"},
-  {_CS_MISC,		"charsets.charset.ctype"},
-  {_CS_CTYPEMAP,	"charsets.charset.ctype.map"},
-  {_CS_MISC,		"charsets.charset.upper"},
-  {_CS_UPPERMAP,	"charsets.charset.upper.map"},
-  {_CS_MISC,		"charsets.charset.lower"},
-  {_CS_LOWERMAP,	"charsets.charset.lower.map"},
-  {_CS_MISC,		"charsets.charset.unicode"},
-  {_CS_UNIMAP,		"charsets.charset.unicode.map"},
-  {_CS_COLLATION,	"charsets.charset.collation"},
-  {_CS_COLNAME,		"charsets.charset.collation.name"},
-  {_CS_ID,		"charsets.charset.collation.id"},
-  {_CS_ORDER,		"charsets.charset.collation.order"},
-  {_CS_FLAG,		"charsets.charset.collation.flag"},
-  {_CS_COLLMAP,		"charsets.charset.collation.map"},
-  {0,	NULL}
-};
-
-static struct my_cs_file_section_st * cs_file_sec(const char *attr, uint len)
-{
-  struct my_cs_file_section_st *s;
-  for (s=sec; s->str; s++)
-  {
-    if (!strncmp(attr,s->str,len))
-      return s;
-  }
-  return NULL;
-}
-
-struct my_cs_file_info 
-{
-  char   csname[CS_MAX_NM_LEN];
-  char   name[CS_MAX_NM_LEN];
-  uchar  ctype[CTYPE_TABLE_SIZE];
-  uchar  to_lower[TO_LOWER_TABLE_SIZE];
-  uchar  to_upper[TO_UPPER_TABLE_SIZE];
-  uchar  sort_order[SORT_ORDER_TABLE_SIZE];
-  uint16 tab_to_uni[TO_UNI_TABLE_SIZE];
-  CHARSET_INFO cs;
-  myf myflags;
-  int (*add_collation)(CHARSET_INFO *cs);
-};
-
-
-
-static int fill_uchar(uchar *a,uint size,const char *str, uint len)
-{
-  uint i= 0;
-  const char *s, *b, *e=str+len;
-  
-  for (s=str ; s < e ; i++)
-  { 
-    for ( ; (s < e) && strchr(" \t\r\n",s[0]); s++) ;
-    b=s;
-    for ( ; (s < e) && !strchr(" \t\r\n",s[0]); s++) ;
-    if (s == b || i > size)
-      break;
-    a[i]= my_strntoul(my_charset_latin1,b,s-b,NULL,16);
-  }
-  return 0;
-}
-
-static int fill_uint16(uint16 *a,uint size,const char *str, uint len)
-{
-  uint i= 0;
-  const char *s, *b, *e=str+len;
-  for (s=str ; s < e ; i++)
-  { 
-    for ( ; (s < e) && strchr(" \t\r\n",s[0]); s++) ;
-    b=s;
-    for ( ; (s < e) && !strchr(" \t\r\n",s[0]); s++) ;
-    if (s == b || i > size)
-      break;
-    a[i]= my_strntol(my_charset_latin1,b,s-b,NULL,16);
-  }
-  return 0;
-}
-
-
-static int cs_enter(MY_XML_PARSER *st,const char *attr, uint len)
-{
-  struct my_cs_file_info *i= (struct my_cs_file_info *)st->user_data;
-  struct my_cs_file_section_st *s= cs_file_sec(attr,len);
-  
-  if ( s && (s->state == _CS_CHARSET))
-  {
-    bzero(&i->cs,sizeof(i->cs));
-  }
-  return MY_XML_OK;
-}
-
-
-static int cs_leave(MY_XML_PARSER *st,const char *attr, uint len)
-{
-  struct my_cs_file_info *i= (struct my_cs_file_info *)st->user_data;
-  struct my_cs_file_section_st *s= cs_file_sec(attr,len);
-  int    state= s ? s->state : 0;
-  int    rc;
-  
-  switch(state){
-  case _CS_COLLATION:
-    rc= i->add_collation ? i->add_collation(&i->cs) : MY_XML_OK;
-    break;
-  default:
-    rc=MY_XML_OK;
-  }
-  return rc;
-}
-
-
-static int cs_value(MY_XML_PARSER *st,const char *attr, uint len)
-{
-  struct my_cs_file_info *i= (struct my_cs_file_info *)st->user_data;
-  struct my_cs_file_section_st *s;
-  int    state= (s=cs_file_sec(st->attr,strlen(st->attr))) ? s->state : 0;
-  
-#ifndef DBUG_OFF
-  if(0){
-    char   str[1024];
-    mstr(str,attr,len,sizeof(str)-1);
-    printf("VALUE %d %s='%s'\n",state,st->attr,str);
-  }
-#endif
-  
-  switch (state) {
-  case _CS_ID:
-    i->cs.number= my_strntoul(my_charset_latin1,attr,len,(char**)NULL,0);
-    break;
-  case _CS_COLNAME:
-    i->cs.name=mstr(i->name,attr,len,CS_MAX_NM_LEN-1);
-    break;
-  case _CS_CSNAME:
-    i->cs.csname=mstr(i->csname,attr,len,CS_MAX_NM_LEN-1);
-    break;
-  case _CS_FLAG:
-    if (!strncmp("primary",attr,len))
-      i->cs.state|= MY_CS_PRIMARY;
-    break;
-  case _CS_UPPERMAP:
-    fill_uchar(i->to_upper,TO_UPPER_TABLE_SIZE,attr,len);
-    i->cs.to_upper=i->to_upper;
-    break;
-  case _CS_LOWERMAP:
-    fill_uchar(i->to_lower,TO_LOWER_TABLE_SIZE,attr,len);
-    i->cs.to_lower=i->to_lower;
-    break;
-  case _CS_UNIMAP:
-    fill_uint16(i->tab_to_uni,TO_UNI_TABLE_SIZE,attr,len);
-    i->cs.tab_to_uni=i->tab_to_uni;
-    break;
-  case _CS_COLLMAP:
-    fill_uchar(i->sort_order,SORT_ORDER_TABLE_SIZE,attr,len);
-    i->cs.sort_order=i->sort_order;
-    break;
-  case _CS_CTYPEMAP:
-    fill_uchar(i->ctype,CTYPE_TABLE_SIZE,attr,len);
-    i->cs.ctype=i->ctype;
-    break;
-  }
-  return MY_XML_OK;
-}
-
 
 static void set_max_sort_char(CHARSET_INFO *cs)
 {
@@ -385,23 +174,23 @@ static void simple_cs_copy_data(CHARSET_INFO *to, CHARSET_INFO *from)
   
   if (from->ctype)
     to->ctype= (uchar*) my_once_memdup((char*) from->ctype,
-				       CTYPE_TABLE_SIZE, MYF(MY_WME));
+				       MY_CS_CTYPE_TABLE_SIZE, MYF(MY_WME));
   if (from->to_lower)
     to->to_lower= (uchar*) my_once_memdup((char*) from->to_lower,
-					  TO_LOWER_TABLE_SIZE, MYF(MY_WME));
+					  MY_CS_TO_LOWER_TABLE_SIZE, MYF(MY_WME));
   if (from->to_upper)
     to->to_upper= (uchar*) my_once_memdup((char*) from->to_upper,
-					  TO_UPPER_TABLE_SIZE, MYF(MY_WME));
+					  MY_CS_TO_UPPER_TABLE_SIZE, MYF(MY_WME));
   if (from->sort_order)
   {
     to->sort_order= (uchar*) my_once_memdup((char*) from->sort_order,
-					    SORT_ORDER_TABLE_SIZE,
+					    MY_CS_SORT_ORDER_TABLE_SIZE,
 					    MYF(MY_WME));
     set_max_sort_char(to);
   }
   if (from->tab_to_uni)
   {
-    uint sz=TO_UNI_TABLE_SIZE*sizeof(uint16);
+    uint sz= MY_CS_TO_UNI_TABLE_SIZE*sizeof(uint16);
     to->tab_to_uni= (uint16*)  my_once_memdup((char*)from->tab_to_uni, sz,
 					     MYF(MY_WME));
     create_fromuni(to);
@@ -448,24 +237,6 @@ static int add_collation(CHARSET_INFO *cs)
 }
 
 
-static my_bool my_parse_charset_xml(const char *buf, uint len)
-{
-  MY_XML_PARSER p;
-  struct my_cs_file_info i;
-  my_bool rc;
-  
-  my_xml_parser_create(&p);
-  my_xml_set_enter_handler(&p,cs_enter);
-  my_xml_set_value_handler(&p,cs_value);
-  my_xml_set_leave_handler(&p,cs_leave);
-  i.add_collation= add_collation;
-  my_xml_set_user_data(&p,(void*)&i);
-  rc= (my_xml_parse(&p,buf,len) == MY_XML_OK) ? FALSE : TRUE;
-  my_xml_parser_free(&p);
-  return rc;
-}
-
-
 #define MAX_BUF 1024*16
 #define MY_CHARSET_INDEX "Index.xml"
 
@@ -490,7 +261,7 @@ static my_bool my_read_charset_file(const char *filename, myf myflags)
   len=read(fd,buf,MAX_BUF);
   my_close(fd,myflags);
   
-  if (my_parse_charset_xml(buf,len))
+  if (my_parse_charset_xml(buf,len,add_collation))
   {
 #ifdef NOT_YET
     printf("ERROR at line %d pos %d '%s'\n",

strings/ctype.c

@@ -16,6 +16,7 @@
 
 #include <my_global.h>
 #include <m_ctype.h>
+#include <my_xml.h>
 #ifndef SCO
 #include <m_string.h>
 #endif
@@ -3895,6 +3896,228 @@ static CHARSET_INFO compiled_charsets[] = {
 };
 
 
+static char *mstr(char *str,const char *src,uint l1,uint l2)
+{
+  l1= l1<l2 ? l1 : l2;
+  memcpy(str,src,l1);
+  str[l1]='\0';
+  return str;
+}
+
+
+struct my_cs_file_section_st
+{
+  int        state;
+  const char *str;
+};
+
+#define _CS_MISC	1
+#define _CS_ID		2
+#define _CS_CSNAME	3
+#define _CS_FAMILY	4
+#define _CS_ORDER	5
+#define _CS_COLNAME	6
+#define _CS_FLAG	7
+#define _CS_CHARSET	8
+#define _CS_COLLATION	9
+#define _CS_UPPERMAP	10
+#define _CS_LOWERMAP	11
+#define _CS_UNIMAP	12
+#define _CS_COLLMAP	13
+#define _CS_CTYPEMAP	14
+
+static struct my_cs_file_section_st sec[] =
+{
+  {_CS_MISC,		"xml"},
+  {_CS_MISC,		"xml.version"},
+  {_CS_MISC,		"xml.encoding"},
+  {_CS_MISC,		"charsets"},
+  {_CS_MISC,		"charsets.max-id"},
+  {_CS_MISC,		"charsets.description"},
+  {_CS_CHARSET,		"charsets.charset"},
+  {_CS_CSNAME,		"charsets.charset.name"},
+  {_CS_FAMILY,		"charsets.charset.family"},
+  {_CS_MISC,		"charsets.charset.alias"},
+  {_CS_MISC,		"charsets.charset.ctype"},
+  {_CS_CTYPEMAP,	"charsets.charset.ctype.map"},
+  {_CS_MISC,		"charsets.charset.upper"},
+  {_CS_UPPERMAP,	"charsets.charset.upper.map"},
+  {_CS_MISC,		"charsets.charset.lower"},
+  {_CS_LOWERMAP,	"charsets.charset.lower.map"},
+  {_CS_MISC,		"charsets.charset.unicode"},
+  {_CS_UNIMAP,		"charsets.charset.unicode.map"},
+  {_CS_COLLATION,	"charsets.charset.collation"},
+  {_CS_COLNAME,		"charsets.charset.collation.name"},
+  {_CS_ID,		"charsets.charset.collation.id"},
+  {_CS_ORDER,		"charsets.charset.collation.order"},
+  {_CS_FLAG,		"charsets.charset.collation.flag"},
+  {_CS_COLLMAP,		"charsets.charset.collation.map"},
+  {0,	NULL}
+};
+
+static struct my_cs_file_section_st * cs_file_sec(const char *attr, uint len)
+{
+  struct my_cs_file_section_st *s;
+  for (s=sec; s->str; s++)
+  {
+    if (!strncmp(attr,s->str,len))
+      return s;
+  }
+  return NULL;
+}
+
+typedef struct my_cs_file_info
+{
+  char   csname[MY_CS_NAME_SIZE];
+  char   name[MY_CS_NAME_SIZE];
+  uchar  ctype[MY_CS_CTYPE_TABLE_SIZE];
+  uchar  to_lower[MY_CS_TO_LOWER_TABLE_SIZE];
+  uchar  to_upper[MY_CS_TO_UPPER_TABLE_SIZE];
+  uchar  sort_order[MY_CS_SORT_ORDER_TABLE_SIZE];
+  uint16 tab_to_uni[MY_CS_TO_UNI_TABLE_SIZE];
+  CHARSET_INFO cs;
+  int (*add_collation)(CHARSET_INFO *cs);
+} MY_CHARSET_LOADER;
+
+
+
+static int fill_uchar(uchar *a,uint size,const char *str, uint len)
+{
+  uint i= 0;
+  const char *s, *b, *e=str+len;
+  
+  for (s=str ; s < e ; i++)
+  { 
+    for ( ; (s < e) && strchr(" \t\r\n",s[0]); s++) ;
+    b=s;
+    for ( ; (s < e) && !strchr(" \t\r\n",s[0]); s++) ;
+    if (s == b || i > size)
+      break;
+    a[i]= my_strntoul(my_charset_latin1,b,s-b,NULL,16);
+  }
+  return 0;
+}
+
+static int fill_uint16(uint16 *a,uint size,const char *str, uint len)
+{
+  uint i= 0;
+  const char *s, *b, *e=str+len;
+  for (s=str ; s < e ; i++)
+  { 
+    for ( ; (s < e) && strchr(" \t\r\n",s[0]); s++) ;
+    b=s;
+    for ( ; (s < e) && !strchr(" \t\r\n",s[0]); s++) ;
+    if (s == b || i > size)
+      break;
+    a[i]= my_strntol(my_charset_latin1,b,s-b,NULL,16);
+  }
+  return 0;
+}
+
+
+static int cs_enter(MY_XML_PARSER *st,const char *attr, uint len)
+{
+  struct my_cs_file_info *i= (struct my_cs_file_info *)st->user_data;
+  struct my_cs_file_section_st *s= cs_file_sec(attr,len);
+  
+  if ( s && (s->state == _CS_CHARSET))
+  {
+    bzero(&i->cs,sizeof(i->cs));
+  }
+  return MY_XML_OK;
+}
+
+
+static int cs_leave(MY_XML_PARSER *st,const char *attr, uint len)
+{
+  struct my_cs_file_info *i= (struct my_cs_file_info *)st->user_data;
+  struct my_cs_file_section_st *s= cs_file_sec(attr,len);
+  int    state= s ? s->state : 0;
+  int    rc;
+  
+  switch(state){
+  case _CS_COLLATION:
+    rc= i->add_collation ? i->add_collation(&i->cs) : MY_XML_OK;
+    break;
+  default:
+    rc=MY_XML_OK;
+  }
+  return rc;
+}
+
+
+static int cs_value(MY_XML_PARSER *st,const char *attr, uint len)
+{
+  struct my_cs_file_info *i= (struct my_cs_file_info *)st->user_data;
+  struct my_cs_file_section_st *s;
+  int    state= (s=cs_file_sec(st->attr,strlen(st->attr))) ? s->state : 0;
+  
+#ifndef DBUG_OFF
+  if(0){
+    char   str[1024];
+    mstr(str,attr,len,sizeof(str)-1);
+    printf("VALUE %d %s='%s'\n",state,st->attr,str);
+  }
+#endif
+  
+  switch (state) {
+  case _CS_ID:
+    i->cs.number= my_strntoul(my_charset_latin1,attr,len,(char**)NULL,0);
+    break;
+  case _CS_COLNAME:
+    i->cs.name=mstr(i->name,attr,len,MY_CS_NAME_SIZE-1);
+    break;
+  case _CS_CSNAME:
+    i->cs.csname=mstr(i->csname,attr,len,MY_CS_NAME_SIZE-1);
+    break;
+  case _CS_FLAG:
+    if (!strncmp("primary",attr,len))
+      i->cs.state|= MY_CS_PRIMARY;
+    break;
+  case _CS_UPPERMAP:
+    fill_uchar(i->to_upper,MY_CS_TO_UPPER_TABLE_SIZE,attr,len);
+    i->cs.to_upper=i->to_upper;
+    break;
+  case _CS_LOWERMAP:
+    fill_uchar(i->to_lower,MY_CS_TO_LOWER_TABLE_SIZE,attr,len);
+    i->cs.to_lower=i->to_lower;
+    break;
+  case _CS_UNIMAP:
+    fill_uint16(i->tab_to_uni,MY_CS_TO_UNI_TABLE_SIZE,attr,len);
+    i->cs.tab_to_uni=i->tab_to_uni;
+    break;
+  case _CS_COLLMAP:
+    fill_uchar(i->sort_order,MY_CS_SORT_ORDER_TABLE_SIZE,attr,len);
+    i->cs.sort_order=i->sort_order;
+    break;
+  case _CS_CTYPEMAP:
+    fill_uchar(i->ctype,MY_CS_CTYPE_TABLE_SIZE,attr,len);
+    i->cs.ctype=i->ctype;
+    break;
+  }
+  return MY_XML_OK;
+}
+
+
+my_bool my_parse_charset_xml(const char *buf, uint len, 
+				    int (*add_collation)(CHARSET_INFO *cs))
+{
+  MY_XML_PARSER p;
+  struct my_cs_file_info i;
+  my_bool rc;
+  
+  my_xml_parser_create(&p);
+  my_xml_set_enter_handler(&p,cs_enter);
+  my_xml_set_value_handler(&p,cs_value);
+  my_xml_set_leave_handler(&p,cs_leave);
+  i.add_collation= add_collation;
+  my_xml_set_user_data(&p,(void*)&i);
+  rc= (my_xml_parse(&p,buf,len) == MY_XML_OK) ? FALSE : TRUE;
+  my_xml_parser_free(&p);
+  return rc;
+}
+
+
 CHARSET_INFO *my_charset_latin1 = &compiled_charsets[0];
 CHARSET_INFO *all_charsets[256];
 CHARSET_INFO *default_charset_info = &compiled_charsets[0];