> Tatsuo Ishii wrote:

> > > > It was made to cope with encoding such as an Asian bloc in 7.2Beta2.
> > > >
> > > > Added ServerEncoding
> > > >         Korean (JOHAB), Thai (WIN874),
> > > >         Vietnamese (TCVN), Arabic (WIN1256)
> > > >
> > > > Added ClientEncoding
> > > >         Simplified Chinese (GBK), Korean (UHC)
> > > >
> > > >
> > > >
> http://www.sankyo-unyu.co.jp/Pool/postgresql-7.2b2.newencoding.diff.tar.gz
> > > > (608K)
> > >
> > > Looks good.  I need some people to review this for me.
> >
> > For me they look good too. The only missing part is a
> > documentation. I will ask him to write it up. If he couldn't, I will
> > do it for him.
> > > The diff is 3mb
> > > but appears to address only additions to multibyte.  I have attached a
> > > list of files it modifies.  Also, look at the sizes of the mb/
> > > directory.  It is getting large:
> > >
> > >   4       ./CVS
> > >   6       ./Unicode/CVS
> > >   3433    ./Unicode
> > >   6197    .
> >
> > Yes. We definitely need the on-the-fly encoding addition capability:
> > i.e. CREATE CHRACTER SET in the future...
> > --
> > Tatsuo Ishii
> >
> >

Address chainge.

http://www.sankyo-unyu.co.jp/Pool/postgresql-7.2.newencoding.diff.gz

Add PsqlODBC and document ...etc patch.

Eiji Tokuya

3 files changed, 383 insertions, 122 deletions

diff --git a/src/interfaces/odbc/connection.c b/src/interfaces/odbc/connection.c
index 66ef5e9ae4f..389da5698ae 100644
--- a/src/interfaces/odbc/connection.c
+++ b/src/interfaces/odbc/connection.c
@@ -944,6 +944,9 @@ another_version_retry:
 	CC_send_settings(self);
 	CC_lookup_lo(self);			/* a hack to get the oid of our large
 								 * object oid type */
+#ifdef MULTIBYTE

+	CC_lookup_characterset(self);

+#endif

 	CC_lookup_pg_version(self); /* Get PostgreSQL version for SQLGetInfo
 								 * use */
 
@@ -1247,9 +1250,27 @@ CC_send_query(ConnectionClass *self, char *query, QueryInfo *qi)
 				if (QR_command_successful(res))
 					QR_set_status(res, PGRES_NONFATAL_ERROR);
 				QR_set_notice(res, cmdbuffer);	/* will dup this string */
-
-				mylog("~~~ NOTICE: '%s'\n", cmdbuffer);
-				qlog("NOTICE from backend during send_query: '%s'\n", cmdbuffer);
+#ifdef MULTIBYTE

+				if (strstr(cmdbuffer,"encoding is"))

+				{

+					if (strstr(cmdbuffer,"Current client encoding is"))

+						strcpy(PG_CCSS, cmdbuffer + 36);

+					if (strstr(cmdbuffer,"Current server encoding is"))

+						strcpy(PG_SCSS, cmdbuffer + 36);

+					mylog("~~~ NOTICE: '%s'\n", cmdbuffer);

+					qlog("NOTICE from backend during send_query: '%s'\n ClientEncoding = %s\n ServerEncoding = %s\n", cmdbuffer, PG_CCSS, PG_SCSS);

+

+				}

+				else

+				{

+

+					mylog("~~~ NOTICE: '%s'\n", cmdbuffer);

+					qlog("NOTICE from backend during send_query: '%s'\n", cmdbuffer);

+				}

+#else

+				mylog("~~~ NOTICE: '%s'\n", cmdbuffer);

+				qlog("NOTICE from backend during send_query: '%s'\n", cmdbuffer);

+#endif
 				while (msg_truncated)
 					msg_truncated = SOCK_get_string(sock, cmdbuffer, ERROR_MSG_LENGTH);
 
diff --git a/src/interfaces/odbc/multibyte.c b/src/interfaces/odbc/multibyte.c
index d10c5754455..464bf244326 100644
--- a/src/interfaces/odbc/multibyte.c
+++ b/src/interfaces/odbc/multibyte.c
@@ -1,138 +1,330 @@
 /*--------
  * Module :			multibyte.c
  *
- * Description:		Mlutibyte related additional function.
+ * Description:		New Mlutibyte related additional function.
  *
  *					Create 2001-03-03 Eiji Tokuya
+ *					New Create 2001-09-16 Eiji Tokuya
  *--------
  */
 
 #include "multibyte.h"
+#include "connection.h"
+#include "pgapifunc.h"
 #include <string.h>
+#include <ctype.h>
+#include <stdio.h>
+#include <stdlib.h>
 
-int			multibyte_client_encoding;	/* Multibyte Client Encoding. */
-int			multibyte_status;	/* Multibyte Odds and ends character. */
+int PG_CCST;				/* Client Charcter Status */
 
+int PG_SCSC;				/* Server Charcter Set (code) */
+int PG_CCSC;				/* Client Charcter Set (code) */
+unsigned char *PG_SCSS;	/* Server Charcter Set (string) */
+unsigned char *PG_CCSS;	/* Client Charcter Set (string) */
 
-unsigned char *
-multibyte_strchr(const unsigned char *s, unsigned char c)
+pg_CS CS_Table[] =
 {
-	int			mb_st = 0,
-				i = 0;
+	{ "SQL_ASCII",	SQL_ASCII },
+	{ "EUC_JP",	EUC_JP },
+	{ "EUC_CN",	EUC_CN },
+	{ "EUC_KR",	EUC_KR },
+	{ "EUC_TW",	EUC_TW },
+	{ "JOHAB", JOHAB },
+	{ "UNICODE",	UTF8 },
+	{ "MULE_INTERNAL",MULE_INTERNAL },
+	{ "LATIN1",	LATIN1 },
+	{ "LATIN2",	LATIN2 },
+	{ "LATIN3",	LATIN3 },
+	{ "LATIN4",	LATIN4 },
+	{ "LATIN5",	LATIN5 },
+	{ "LATIN6", LATIN6 },
+	{ "LATIN7", LATIN7 },
+	{ "LATIN8", LATIN8 },
+	{ "LATIN9", LATIN9 },
+	{ "LATIN10", LATIN10 },
+	{ "WIN1256", WIN1256 },
+	{ "TCVN", TCVN },
+	{ "WIN874", WIN874 },
+	{ "KOI8",	KOI8R },
+	{ "WIN",	WIN1251 },
+	{ "ALT",	ALT },
+	{ "ISO_8859_5", ISO_8859_5 },
+	{ "ISO_8859_6", ISO_8859_6 },
+	{ "ISO_8859_7", ISO_8859_7 },
+	{ "ISO_8859_8", ISO_8859_8 },
 
-	while (!(mb_st == 0 && (s[i] == c || s[i] == 0)))
-	{
-		if (s[i] == 0)
-			return (0);
-		switch (multibyte_client_encoding)
-		{
-			case SJIS:
-				{
-					if (mb_st < 2 && s[i] > 0x80 && !(s[i] > 0x9f && s[i] < 0xe0))
-						mb_st = 2;
-					else if (mb_st == 2)
-						mb_st = 1;
-					else
-						mb_st = 0;
-				}
-				break;
 
-/* Chinese Big5 Support. */
-			case BIG5:
-				{
-					if (mb_st < 2 && s[i] > 0xA0)
-						mb_st = 2;
-					else if (mb_st == 2)
-						mb_st = 1;
-					else
-						mb_st = 0;
-				}
-				break;
-			default:
-				mb_st = 0;
-		}
+	{ "SJIS",	SJIS },
+	{ "BIG5",	BIG5 },
+	{ "GBK", GBK },
+	{ "UHC", UHC },
+	{ "WIN1250",	WIN1250 },
+	{ "OTHER", OTHER }
+};
+
+int
+pg_ismb(int characterset_code)
+{
+	int i=0,MB_CHARACTERSET[]={EUC_JP,EUC_CN,EUC_KR,EUC_TW,UTF8,MULE_INTERNAL,SJIS,BIG5,GBK,UHC,JOHAB};
+
+	while (MB_CHARACTERSET[i] != characterset_code || OTHER != MB_CHARACTERSET[i] )
+	{
 		i++;
 	}
-#ifdef _DEBUG
-	qlog("i = %d\n", i);
-#endif
-	return (char *) (s + i);
+	return (MB_CHARACTERSET[i]);
 }
 
-
-void
-multibyte_init(void)
+int
+pg_CS_code(const unsigned char *characterset_string)
 {
-	multibyte_status = 0;
+	int i = 0, c;
+	for(i = 0; CS_Table[i].code != OTHER; i++)
+	{
+		if (strstr(characterset_string,CS_Table[i].name))
+			c = CS_Table[i].code;
+	}
+	return (c);
 }
 
-
 unsigned char *
-check_client_encoding(unsigned char *str)
+pg_CS_name(const int characterset_code)
 {
-	if (strstr(str, "%27SJIS%27") ||
-		strstr(str, "%27Shift_JIS%27") ||
-		strstr(str, "'SJIS'") ||
-		strstr(str, "'sjis'") ||
-		strstr(str, "'Shift_JIS'"))
-	{
-		multibyte_client_encoding = SJIS;
-		return ("SJIS");
-	}
-	if (strstr(str, "%27BIG5%27") ||
-		strstr(str, "%27Big5%27") ||
-		strstr(str, "'BIG5'") ||
-		strstr(str, "'big5'") ||
-		strstr(str, "'Big5'"))
+	int i = 0;
+	for (i = 0; CS_Table[i].code != OTHER; i++)
 	{
-		multibyte_client_encoding = BIG5;
-		return ("BIG5");
+		if (CS_Table[i].code == characterset_code)
+			return CS_Table[i].name;
 	}
 	return ("OTHER");
 }
 
-
-/*--------
- * Multibyte Status Function.
- *	Input	char
- *	Output	0	: 1 Byte Character.
- *			1	: MultibyteCharacter Last Byte.
- *			N	: MultibyteCharacter Fast or Middle Byte.
- *--------
- */
 int
-multibyte_char_check(unsigned char s)
+pg_CS_stat(int stat,unsigned int character,int characterset_code)
 {
-	switch (multibyte_client_encoding)
+	if (character == 0)
+		stat = 0;
+	switch (characterset_code)
 	{
-			/* Japanese Shift-JIS(CP932) Support. */
-		case SJIS:
+		case UTF8:
 			{
-				if (multibyte_status < 2 && s > 0x80 && !(s > 0x9f && s < 0xE0))
-					multibyte_status = 2;
-				else if (multibyte_status == 2)
-					multibyte_status = 1;
+				if (stat < 2 &&
+					character >= 0x80)
+				{
+					if (character >= 0xfc)
+						stat = 6;
+					else if (character >= 0xf8)
+						stat = 5;
+					else if (character >= 0xf0)
+						stat = 4;
+					else if (character >= 0xe0)
+						stat = 3;
+					else if (character >= 0xc0)
+						stat = 2;
+				}
+				else if (stat > 2 &&
+					character > 0x7f)
+					stat--;
 				else
-					multibyte_status = 0;
+					stat=0;
 			}
 			break;
-
-			/* Chinese Big5(CP950) Support. */
+/* Shift-JIS Support. */
+			case SJIS:
+			{
+				if (stat < 2 &&
+					character > 0x80 &&
+					!(character > 0x9f &&
+					character < 0xe0))
+					stat = 2;
+				else if (stat == 2)
+					stat = 1;
+				else
+					stat = 0;
+			}
+			break;
+/* Chinese Big5 Support. */
 		case BIG5:
 			{
-				if (multibyte_status < 2 && s > 0xA0)
-					multibyte_status = 2;
-				else if (multibyte_status == 2)
-					multibyte_status = 1;
+				if (stat < 2 &&
+					character > 0xA0)
+					stat = 2;
+				else if (stat == 2)
+					stat = 1;
+				else
+					stat = 0;
+			}
+			break;
+/* Chinese GBK Support. */
+		case GBK:
+			{
+				if (stat < 2 &&
+					character > 0x7F)
+					stat = 2;
+				else if (stat == 2)
+					stat = 1;
+				else
+					stat = 0;
+			}
+			break;
+
+/* Korian UHC Support. */
+		case UHC:
+			{
+				if (stat < 2 &&
+					character > 0x7F)
+					stat = 2;
+				else if (stat == 2)
+					stat = 1;
+				else
+					stat = 0;
+			}
+			break;
+
+/* EUC_JP Support */
+		case EUC_JP:
+			{
+				if (stat < 3 && 
+					character == 0x8f)	/* JIS X 0212 */
+					stat = 3;
 				else
-					multibyte_status = 0;
+				if (stat != 2 && 
+					(character == 0x8e ||
+					character > 0xa0))	/* Half Katakana HighByte & Kanji HighByte */
+					stat = 2;
+				else if (stat == 2)
+					stat = 1;
+				else
+					stat = 0;
+			}
+			break;
+
+/* EUC_CN, EUC_KR, JOHAB Support */
+		case EUC_CN:
+		case EUC_KR:
+		case JOHAB:
+			{
+				if (stat < 2 &&
+					character > 0xa0)
+					stat = 2;
+				else if (stat == 2)
+					stat = 1;
+				else
+					stat = 0;
+			}
+			break;
+		case EUC_TW:
+			{
+				if (stat < 4 &&
+					character == 0x8e)
+					stat = 4;
+				else if (stat == 4 &&
+					character > 0xa0)
+					stat = 3;
+				else if (stat == 3 ||
+					stat < 2 &&
+					character > 0xa0)
+					stat = 2;
+				else if (stat == 2)
+					stat = 1;
+				else
+					stat = 0;
 			}
 			break;
 		default:
-			multibyte_status = 0;
+			{
+				stat = 0;
+			}
+			break;
+	}
+	return stat;
+}
+
+
+unsigned char *
+pg_mbschr(const unsigned char *string, unsigned int character)
+{
+	int			mb_st = 0;
+	unsigned char *s;
+	s = (unsigned char *) string;
+
+	for(;;) 
+	{
+		mb_st = pg_CS_stat(mb_st, (unsigned char) *s,PG_CCSC);
+		if (mb_st == 0 && (*s == character || *s == 0))
+			break;
+		else
+			s++;
+	}
+	return (s);
+}
+
+int
+pg_mbslen(const unsigned char *string)
+{
+	unsigned char *s;
+	int len, cs_stat;
+	for (len = 0, cs_stat = 0, s = (unsigned char *) string; *s != 0; s++)
+	{
+		cs_stat = pg_CS_stat(cs_stat,(unsigned int) *s, PG_CCSC);
+		if (cs_stat < 2)
+			len++;
+	}
+	return len;
+}
+
+unsigned char *
+pg_mbsinc(const unsigned char *current )
+{
+	int mb_stat = 0;
+	if (*current != 0)
+	{
+		mb_stat = (int) pg_CS_stat(mb_stat, *current, PG_CCSC);
+		if (mb_stat == 0)
+			mb_stat = 1;
+		return ((unsigned char *) current + mb_stat);
+	}
+	else
+		return NULL;
+}
+
+void
+CC_lookup_characterset(ConnectionClass *self)
+{
+	HSTMT		hstmt;
+	StatementClass *stmt;
+	RETCODE		result;
+	static char *func = "CC_lookup_characterset";
+
+	mylog("%s: entering...\n", func);
+	PG_SCSS = malloc(MAX_CHARACTERSET_NAME);
+	PG_CCSS = malloc(MAX_CHARACTERSET_NAME);
+
+	result = PGAPI_AllocStmt(self, &hstmt);
+	if ((result != SQL_SUCCESS) && (result != SQL_SUCCESS_WITH_INFO))
+		return;
+	stmt = (StatementClass *) hstmt;
+
+	result = PGAPI_ExecDirect(hstmt, "Show Client_Encoding", SQL_NTS);
+	if ((result != SQL_SUCCESS) && (result != SQL_SUCCESS_WITH_INFO))
+	{
+		PGAPI_FreeStmt(hstmt, SQL_DROP);
+		return;
+	}
+	result = PGAPI_AllocStmt(self, &hstmt);
+	if ((result != SQL_SUCCESS) && (result != SQL_SUCCESS_WITH_INFO))
+		return;
+	stmt = (StatementClass *) hstmt;
+
+	result = PGAPI_ExecDirect(hstmt, "Show Server_Encoding", SQL_NTS);
+	if ((result != SQL_SUCCESS) && (result != SQL_SUCCESS_WITH_INFO))
+	{
+		PGAPI_FreeStmt(hstmt, SQL_DROP);
+		return;
 	}
-#ifdef _DEBUG
-	qlog("multibyte_client_encoding = %d   s = 0x%02X   multibyte_stat = %d\n", multibyte_client_encoding, s, multibyte_status);
-#endif
-	return (multibyte_status);
+
+	strcpy(PG_SCSS , pg_CS_name(PG_SCSC = pg_CS_code(PG_SCSS)));
+	strcpy(PG_CCSS , pg_CS_name(PG_CCSC = pg_CS_code(PG_CCSS)));
+
+	qlog("    [ Server encoding = '%s' (code = %d), Client encoding = '%s' (code = %d) ]\n", PG_SCSS, PG_SCSC, PG_CCSS, PG_CCSC);
 }
diff --git a/src/interfaces/odbc/multibyte.h b/src/interfaces/odbc/multibyte.h
index c171c68fbd3..b6d487b634d 100644
--- a/src/interfaces/odbc/multibyte.h
+++ b/src/interfaces/odbc/multibyte.h
@@ -10,30 +10,78 @@
 #define EUC_JP				1	/* EUC for Japanese */
 #define EUC_CN				2	/* EUC for Chinese */
 #define EUC_KR				3	/* EUC for Korean */
-#define EUC_TW				4	/* EUC for Taiwan */
-#define UNICODE				5	/* Unicode UTF-8 */
-#define MULE_INTERNAL		6	/* Mule internal code */
-#define LATIN1				7	/* ISO-8859 Latin 1 */
-#define LATIN2				8	/* ISO-8859 Latin 2 */
-#define LATIN3				9	/* ISO-8859 Latin 3 */
-#define LATIN4				10	/* ISO-8859 Latin 4 */
-#define LATIN5				11	/* ISO-8859 Latin 5 */
-#define LATIN6				12	/* ISO-8859 Latin 6 */
-#define LATIN7				13	/* ISO-8859 Latin 7 */
-#define LATIN8				14	/* ISO-8859 Latin 8 */
-#define LATIN9				15	/* ISO-8859 Latin 9 */
-#define KOI8				16	/* KOI8-R/U */
-#define WIN					17	/* windows-1251 */
-#define ALT					18	/* Alternativny Variant (MS-DOS CP866) */
-#define SJIS				32	/* Shift JIS */
-#define BIG5				33	/* Big5 */
-#define WIN1250				34	/* windows-1250 */
+#define EUC_TW				4	/* EUC for Taiwan */

+#define JOHAB				5
+#define UTF8				6	/* Unicode UTF-8 */
+#define MULE_INTERNAL		7	/* Mule internal code */
+#define LATIN1				8	/* ISO-8859 Latin 1 */
+#define LATIN2				9	/* ISO-8859 Latin 2 */
+#define LATIN3				10	/* ISO-8859 Latin 3 */
+#define LATIN4				11	/* ISO-8859 Latin 4 */
+#define LATIN5				12	/* ISO-8859 Latin 5 */
+#define LATIN6				13	/* ISO-8859 Latin 6 */
+#define LATIN7				14	/* ISO-8859 Latin 7 */
+#define LATIN8				15	/* ISO-8859 Latin 8 */
+#define LATIN9				16	/* ISO-8859 Latin 9 */

+#define LATIN10				17	/* ISO-8859 Latin 10 */

+#define WIN1256				18	/* Arabic Windows */

+#define TCVN				19	/* Vietnamese Windows */

+#define WIN874				20	/* Thai Windows */

+#define KOI8R				21	/* KOI8-R/U */

+#define WIN1251				22	/* windows-1251 */
+#define ALT					23	/* Alternativny Variant (MS-DOS CP866) */
+#define ISO_8859_5			24	/* ISO-8859-5 */

+#define ISO_8859_6			25	/* ISO-8859-6 */

+#define ISO_8859_7			26	/* ISO-8859-7 */

+#define ISO_8859_8			27	/* ISO-8859-8 */

+

+#define SJIS				28	/* Shift JIS */
+#define BIG5				29	/* Big5 */

+#define GBK					30	/* GBK */

+#define UHC					31  /* UHC */

+#define WIN1250				32	/* windows-1250 */

+#define OTHER				-1

+

+#define MAX_CHARACTERSET_NAME	24

+#define MAX_CHARACTER_LEN	6

 
-
-extern int	multibyte_client_encoding;	/* Multibyte client encoding. */
-extern int	multibyte_status;	/* Multibyte charcter status. */
-
-void		multibyte_init(void);
-unsigned char *check_client_encoding(unsigned char *str);
-int			multibyte_char_check(unsigned char s);
-unsigned char *multibyte_strchr(const unsigned char *s, unsigned char c);
+/* OLD Type */

+// extern int	multibyte_client_encoding;	/* Multibyte client encoding. */
+// extern int	multibyte_status;	/* Multibyte charcter status. */
+//
+// void		multibyte_init(void);
+// unsigned char *check_client_encoding(unsigned char *sql_string);
+// int			multibyte_char_check(unsigned char s);
+// unsigned char *multibyte_strchr(const unsigned char *string, unsigned int c);

+

+/* New Type */

+

+extern int PG_CCST;				/* Client Character StaTus */

+

+extern int PG_SCSC;				/* Server Character Set (Code) */

+extern int PG_CCSC;				/* Client Character Set (Code) */

+extern unsigned char *PG_SCSS;	/* Server Character Set (String) */

+extern unsigned char *PG_CCSS;	/* Client Character Set (String) */

+

+extern void CC_lookup_characterset(ConnectionClass *self);

+

+extern int pg_CS_stat(int stat,unsigned int charcter,int characterset_code);

+extern int pg_CS_code(const unsigned char *stat_string);

+extern unsigned char *pg_CS_name(const int code);

+

+typedef struct pg_CS

+{

+	unsigned char *name;

+	int code;

+}pg_CS;

+extern pg_CS CS_Table[];

+

+extern int pg_mbslen(const unsigned char *string);

+extern unsigned char *pg_mbschr(const unsigned char *string, unsigned int character);

+extern unsigned char *pg_mbsinc( const unsigned char *current );

+

+/* Old Type Compatible */

+#define multibyte_init() (PG_CCST = 0)

+#define multibyte_char_check(X) pg_CS_stat(PG_CCST, (unsigned int) X, PG_CCSC)

+#define multibyte_strchr(X,Y) pg_mbschr(X,Y)

+#define check_client_encoding(X) pg_CS_name(PG_CCSC = pg_CS_code(X))