Hi, here are the patches to enhance existing MB handling. This time

I have implemented a framework of encoding translation between the backend and the frontend. Also I have added a new variable setting command: SET CLIENT_ENCODING TO 'encoding'; Other features include: Latin1 support more 8 bit cleaness See doc/README.mb for more details. Note that the pacthes are against May 30 snapshot. Tatsuo Ishii
2025-11-12 05:01:15 +03:00 · 1998-06-16 07:29:54 +00:00
parent 0d8e7f6381
commit cb7cbc16fa
37 changed files with 1115 additions and 341 deletions
--- a/src/backend/access/common/Makefile
+++ b/src/backend/access/common/Makefile
@@ -4,7 +4,7 @@
 #    Makefile for access/common
 #
 # IDENTIFICATION
-#    $Header: /cvsroot/pgsql/src/backend/access/common/Makefile,v 1.10 1998/04/06 00:20:44 momjian Exp $
+#    $Header: /cvsroot/pgsql/src/backend/access/common/Makefile,v 1.11 1998/06/16 07:29:18 momjian Exp $
 #
 #-------------------------------------------------------------------------

@@ -13,6 +13,10 @@ include ../../../Makefile.global

 CFLAGS+=-I../..

+ifdef MB
+CFLAGS+= -DMB=$(MB)
+endif
+
 OBJS = heaptuple.o heapvalid.o indextuple.o indexvalid.o printtup.o \
       scankey.o tupdesc.o  

--- a/src/backend/access/common/printtup.c
+++ b/src/backend/access/common/printtup.c
@@ -8,7 +8,7 @@
 *
 *
 * IDENTIFICATION
- *	  $Header: /cvsroot/pgsql/src/backend/access/common/printtup.c,v 1.28 1998/05/14 17:18:12 momjian Exp $
+ *	  $Header: /cvsroot/pgsql/src/backend/access/common/printtup.c,v 1.29 1998/06/16 07:29:18 momjian Exp $
 *
 *-------------------------------------------------------------------------
 */
@@ -23,6 +23,10 @@
 #include <libpq/libpq.h>
 #include <utils/syscache.h>

+#ifdef MB
+#include <commands/variable.h>
+#endif
+
 /* ----------------------------------------------------------------
 *		printtup / debugtup support
 * ----------------------------------------------------------------
@@ -80,6 +84,9 @@ printtup(HeapTuple tuple, TupleDesc typeinfo)
 	Datum		attr;
 	bool		isnull;
 	Oid			typoutput;
+#ifdef MB
+	unsigned char *p;
+#endif

 	/* ----------------
 	 *	tell the frontend to expect new tuple data
@@ -125,8 +132,14 @@ printtup(HeapTuple tuple, TupleDesc typeinfo)
 			outputstr = fmgr(typoutput, attr,
 							 gettypelem(typeinfo->attrs[i]->atttypid),
 							 typeinfo->attrs[i]->atttypmod);
+#ifdef MB
+			p = pg_server_to_client(outputstr, strlen(outputstr));
+			pq_putint(strlen(p) + VARHDRSZ, VARHDRSZ);
+			pq_putnchar(p, strlen(p));
+#else
 			pq_putint(strlen(outputstr) + VARHDRSZ, VARHDRSZ);
 			pq_putnchar(outputstr, strlen(outputstr));
+#endif
 			pfree(outputstr);
 		}
 	}
@@ -268,8 +281,12 @@ printtup_internal(HeapTuple tuple, TupleDesc typeinfo)
 				/* variable length, assume a varlena structure */
 				len = VARSIZE(attr) - VARHDRSZ;

+#ifdef MB
+				pq_putncharlen(VARDATA(attr), len);
+#else
 				pq_putint(len, VARHDRSZ);
 				pq_putnchar(VARDATA(attr), len);
+#endif
 #ifdef IPORTAL_DEBUG
 				{
 					char	   *d = VARDATA(attr);
--- a/src/backend/commands/Makefile
+++ b/src/backend/commands/Makefile
@@ -4,7 +4,7 @@
 #    Makefile for commands
 #
 # IDENTIFICATION
-#    $Header: /cvsroot/pgsql/src/backend/commands/Makefile,v 1.12 1998/04/06 00:22:19 momjian Exp $
+#    $Header: /cvsroot/pgsql/src/backend/commands/Makefile,v 1.13 1998/06/16 07:29:20 momjian Exp $
 #
 #-------------------------------------------------------------------------

@@ -13,11 +13,19 @@ include ../../Makefile.global

 CFLAGS += -I..

+ifdef MB
+CFLAGS += -DMB=$(MB)
+endif
+
 OBJS = async.o creatinh.o command.o copy.o defind.o define.o \
       remove.o rename.o vacuum.o version.o view.o cluster.o \
       recipe.o explain.o sequence.o trigger.o user.o proclang.o \
       dbcommands.o variable.o

+ifdef MB
+OBJS += mbutils.o
+endif
+
 all: SUBSYS.o

 SUBSYS.o: $(OBJS)
--- a/src/backend/commands/variable.c
+++ b/src/backend/commands/variable.c
@@ -2,7 +2,7 @@
 * Routines for handling of 'SET var TO',
 *	'SHOW var' and 'RESET var' statements.
 *
- * $Id: variable.c,v 1.6 1998/06/15 19:28:17 momjian Exp $
+ * $Id: variable.c,v 1.7 1998/06/16 07:29:21 momjian Exp $
 *
 */

@@ -15,6 +15,9 @@
 #include "commands/variable.h"
 #include "utils/builtins.h"
 #include "optimizer/internal.h"
+#ifdef MB
+#include "regex/pg_wchar.h"
+#endif

 extern Cost _cpu_page_wight_;
 extern Cost _cpu_index_page_wight_;
@@ -519,6 +522,54 @@ reset_timezone()
 	return TRUE;
 }	/* reset_timezone() */

+#ifdef MB
+/*-----------------------------------------------------------------------*/
+bool
+parse_client_encoding(const char *value)
+{
+  int encoding;
+
+  encoding = pg_valid_client_encoding(value);
+  if (encoding < 0) {
+    elog(ERROR, "Client encoding %s is not supported", value);
+  } else {    
+    if (pg_set_client_encoding(encoding)) {
+      elog(ERROR, "Conversion between %s and %s is not supported",
+	   value, pg_encoding_to_char(MB));
+    }
+  }
+  return TRUE;
+}
+
+bool
+show_client_encoding()
+{
+  elog(NOTICE, "Current client encoding is %s",
+       pg_encoding_to_char(pg_get_client_encoding()));
+  return TRUE;
+}
+
+bool
+reset_client_encoding()
+{
+  int encoding;
+  char *env = getenv("PGCLIENTENCODING");
+
+  if (env) {
+    encoding = pg_char_to_encoding(env);
+    if (encoding < 0) {
+      encoding = MB;
+    }
+  } else {
+    encoding = MB;
+  }
+  pg_set_client_encoding(encoding);
+  return TRUE;
+}
+
+/*-----------------------------------------------------------------------*/
+#endif
+
 /*-----------------------------------------------------------------------*/
 struct VariableParsers
 {
@@ -547,6 +598,11 @@ struct VariableParsers
 	{
 		"r_plans", parse_r_plans, show_r_plans, reset_r_plans
 	},
+#ifdef MB
+	{
+		"client_encoding", parse_client_encoding, show_client_encoding, reset_client_encoding
+	},
+#endif
 	{
 		NULL, NULL, NULL, NULL
 	}
--- a/src/backend/libpq/Makefile
+++ b/src/backend/libpq/Makefile
@@ -4,7 +4,7 @@
 #    Makefile for libpq subsystem (backend half of libpq interface)
 #
 # IDENTIFICATION
-#    $Header: /cvsroot/pgsql/src/backend/libpq/Makefile,v 1.11 1998/04/06 00:22:39 momjian Exp $
+#    $Header: /cvsroot/pgsql/src/backend/libpq/Makefile,v 1.12 1998/06/16 07:29:22 momjian Exp $
 #
 #-------------------------------------------------------------------------

@@ -19,6 +19,10 @@ CFLAGS+= $(KRBFLAGS)
 LDFLAGS+= $(KRBLIBS)
 endif

+ifdef MB
+CFLAGS+= -DMB=$(MB)
+endif
+
 OBJS = be-dumpdata.o be-fsstubs.o be-pqexec.o pqcomprim.o\
       auth.o hba.o crypt.o pqcomm.o portal.o util.o portalbuf.o pqpacket.o pqsignal.o \
       password.o
--- a/src/backend/libpq/pqcomm.c
+++ b/src/backend/libpq/pqcomm.c
@@ -7,7 +7,7 @@
 *
 *
 * IDENTIFICATION
- *	  $Header: /cvsroot/pgsql/src/backend/libpq/pqcomm.c,v 1.44 1998/06/15 19:28:27 momjian Exp $
+ *	  $Header: /cvsroot/pgsql/src/backend/libpq/pqcomm.c,v 1.45 1998/06/16 07:29:23 momjian Exp $
 *
 *-------------------------------------------------------------------------
 */
@@ -23,6 +23,9 @@
 *		pq_putstr		- send a null terminated string to connection
 *		pq_putnchar		- send n characters to connection
 *		pq_putint		- send an integer to connection
+ *		pq_putncharlen		- send n characters to connection
+ *					  (also send an int header indicating
+ *					   the length)
 *		pq_getinaddr	- initialize address from host and port number
 *		pq_getinserv	- initialize address from host and service name
 *		pq_connect		- create remote input / output connection
@@ -66,6 +69,9 @@
 #include "libpq/auth.h"
 #include "libpq/libpq.h"		/* where the declarations go */
 #include "storage/ipc.h"
+#ifdef MB
+#include "commands/variable.h"
+#endif

 /* ----------------
 *		declarations
@@ -180,6 +186,14 @@ pq_getstr(char *s, int maxlen)
 {
 	int			c = '\0';

+#ifdef MB
+	unsigned char *p, *ps;
+	int len;
+
+	ps = s;
+	len = maxlen;
+#endif
+
 	if (Pfin == (FILE *) NULL)
 	{
 /*		elog(DEBUG, "Input descriptor is null"); */
@@ -190,6 +204,13 @@ pq_getstr(char *s, int maxlen)
 		*s++ = c;
 	*s = '\0';

+#ifdef MB
+	p = pg_client_to_server(ps, len);
+	if (ps != p) {	/* actual conversion has been done? */
+	  strcpy(ps, p);
+	}
+#endif
+	
 	/* -----------------
 	 *	   If EOF reached let caller know.
 	 *	   (This will only happen if we hit EOF before the string
@@ -325,7 +346,14 @@ pq_getint(int b)
 void
 pq_putstr(char *s)
 {
+#ifdef MB
+        unsigned char *p;
+
+        p = pg_server_to_client(s, strlen(s));
+	if (pqPutString(p, Pfout))
+#else
 	if (pqPutString(s, Pfout))
+#endif
 	{
 		sprintf(PQerrormsg,
 				"FATAL: pq_putstr: fputs() failed: errno=%d\n", errno);
@@ -788,3 +816,17 @@ StreamOpen(char *hostName, short portName, Port *port)

 	return (STATUS_OK);
 }
+
+#ifdef MB
+void
+pq_putncharlen(char *s, int n)
+{
+  unsigned char *p;
+  int len;
+
+  p = pg_server_to_client(s, n);
+  len = strlen(p);
+  pq_putint(len, sizeof(int));
+  pq_putnchar(p, len);
+}
+#endif
--- a/src/backend/parser/scan.c
+++ b/src/backend/parser/scan.c
@@ -1,7 +1,7 @@
 /* A lexical scanner generated by flex */

 /* Scanner skeleton version:
- * $Header: /cvsroot/pgsql/src/backend/parser/Attic/scan.c,v 1.21 1998/06/15 19:28:56 momjian Exp $
+ * $Header: /cvsroot/pgsql/src/backend/parser/Attic/scan.c,v 1.22 1998/06/16 07:29:25 momjian Exp $
 */

 #define FLEX_SCANNER
@@ -555,7 +555,7 @@ char *yytext;
 *
 *
 * IDENTIFICATION
- *	  $Header: /cvsroot/pgsql/src/backend/parser/Attic/scan.c,v 1.21 1998/06/15 19:28:56 momjian Exp $
+ *	  $Header: /cvsroot/pgsql/src/backend/parser/Attic/scan.c,v 1.22 1998/06/16 07:29:25 momjian Exp $
 *
 *-------------------------------------------------------------------------
 */
@@ -1178,7 +1178,8 @@ YY_RULE_SETUP

 					BEGIN(xm);
 					for(i = 0; yytext[i]; i++)
-						if (isupper(yytext[i]))
+						if (isascii((unsigned char)yytext[i]) &&
+							isupper(yytext[i]))
 							yytext[i] = tolower(yytext[i]);

 					keyword = ScanKeywordLookup((char*)yytext);
@@ -1194,7 +1195,7 @@ YY_RULE_SETUP
 	YY_BREAK
 case 34:
 YY_RULE_SETUP
-#line 336 "scan.l"
+#line 337 "scan.l"
 {
 					char* endptr;

@@ -1216,7 +1217,7 @@ YY_RULE_SETUP
 	YY_BREAK
 case 35:
 YY_RULE_SETUP
-#line 354 "scan.l"
+#line 355 "scan.l"
 {
 					char* endptr;

@@ -1231,7 +1232,7 @@ YY_RULE_SETUP
 	YY_BREAK
 case 36:
 YY_RULE_SETUP
-#line 365 "scan.l"
+#line 366 "scan.l"
 {
 					char* endptr;

@@ -1252,7 +1253,7 @@ YY_RULE_SETUP
 	YY_BREAK
 case 37:
 YY_RULE_SETUP
-#line 382 "scan.l"
+#line 383 "scan.l"
 {
 					char* endptr;

@@ -1266,13 +1267,14 @@ YY_RULE_SETUP
 	YY_BREAK
 case 38:
 YY_RULE_SETUP
-#line 394 "scan.l"
+#line 395 "scan.l"
 {
 					int i;
 					ScanKeyword		*keyword;

 					for(i = 0; yytext[i]; i++)
-						if (isupper(yytext[i]))
+						if (isascii((unsigned char)yytext[i]) &&
+							isupper(yytext[i]))
 							yytext[i] = tolower(yytext[i]);

 					keyword = ScanKeywordLookup((char*)yytext);
@@ -1288,20 +1290,20 @@ YY_RULE_SETUP
 	YY_BREAK
 case 39:
 YY_RULE_SETUP
-#line 412 "scan.l"
+#line 414 "scan.l"
 { /* ignore */ }
 	YY_BREAK
 case 40:
 YY_RULE_SETUP
-#line 414 "scan.l"
+#line 416 "scan.l"
 { return (yytext[0]); }
 	YY_BREAK
 case 41:
 YY_RULE_SETUP
-#line 416 "scan.l"
+#line 418 "scan.l"
 ECHO;
 	YY_BREAK
-#line 1305 "lex.yy.c"
+#line 1307 "lex.yy.c"
 			case YY_STATE_EOF(INITIAL):
 			case YY_STATE_EOF(xb):
 			case YY_STATE_EOF(xc):
@@ -2187,7 +2189,7 @@ int main()
 	return 0;
 	}
 #endif
-#line 416 "scan.l"
+#line 418 "scan.l"


 void yyerror(char message[])
--- a/src/backend/parser/scan.l
+++ b/src/backend/parser/scan.l
@@ -8,7 +8,7 @@
 *
 *
 * IDENTIFICATION
- *	  $Header: /cvsroot/pgsql/src/backend/parser/scan.l,v 1.39 1998/05/09 23:15:20 thomas Exp $
+ *	  $Header: /cvsroot/pgsql/src/backend/parser/scan.l,v 1.40 1998/06/16 07:29:27 momjian Exp $
 *
 *-------------------------------------------------------------------------
 */
@@ -320,7 +320,8 @@ other			.

 					BEGIN(xm);
 					for(i = 0; yytext[i]; i++)
-						if (isupper(yytext[i]))
+						if (isascii((unsigned char)yytext[i]) &&
+							isupper(yytext[i]))
 							yytext[i] = tolower(yytext[i]);

 					keyword = ScanKeywordLookup((char*)yytext);
@@ -396,7 +397,8 @@ other			.
 					ScanKeyword		*keyword;

 					for(i = 0; yytext[i]; i++)
-						if (isupper(yytext[i]))
+						if (isascii((unsigned char)yytext[i]) &&
+							isupper(yytext[i]))
 							yytext[i] = tolower(yytext[i]);

 					keyword = ScanKeywordLookup((char*)yytext);
--- a/src/backend/regex/engine.c
+++ b/src/backend/regex/engine.c
@@ -127,6 +127,9 @@ extern		"C"
 #  elif MB == UNICODE
 #    define NONCHAR(c)	((c) > USHRT_MAX)
 #    define NNONCHAR	(CODEMAX-USHRT_MAX)
+#  else	/* assume 1 byte code such as ISO8859-1 */
+#    define NONCHAR(c)	((c) > UCHAR_MAX)
+#    define NNONCHAR	(CODEMAX-UCHAR_MAX)
 #  endif
 #else
 #  define NONCHAR(c)		((c) > CHAR_MAX)
--- a/src/backend/regex/regcomp.c
+++ b/src/backend/regex/regcomp.c
@@ -1344,7 +1344,7 @@ cset	   *cs;

 	for (i = 0; i < css; i++)
 		if (CHIN(cs, i))
-			return ((char) i);
+			return (i);
 	assert(never);
 	return (0);					/* arbitrary */
 }
--- a/src/backend/regex/utils.c
+++ b/src/backend/regex/utils.c
@@ -1,164 +1,24 @@
 /*
 * misc conversion functions between pg_wchar and other encodings.
 * Tatsuo Ishii
- * $Id: utils.c,v 1.2 1998/04/27 17:07:53 scrappy Exp $
+ * $Id: utils.c,v 1.3 1998/06/16 07:29:29 momjian Exp $
 */
 #include <regex/pg_wchar.h>
-/*
- * convert EUC to pg_wchar (EUC process code)
- * caller should allocate enough space for "to"
- */
-static void pg_euc2wchar(const unsigned char *from, pg_wchar *to)
-{
-  while (*from) {
-    if (*from == SS2) {
-      from++;
-      *to = *from++;
-    } else if (*from == SS3) {
-      from++;
-      *to = *from++ << 8;
-      *to |= 0x3f & *from++;
-    } else if (*from & 0x80) {
-      *to = *from++ << 8;
-      *to |= *from++;
-    } else {
-      *to = *from++;
-    }
-    to++;
-  }
-  *to = 0;
-}
-
-static void pg_eucjp2wchar(const unsigned char *from, pg_wchar *to)
-{
-  pg_euc2wchar(from,to);
-}
-
-static void pg_euckr2wchar(const unsigned char *from, pg_wchar *to)
-{
-  pg_euc2wchar(from,to);
-}
-
-static void pg_eucch2wchar(const unsigned char *from, pg_wchar *to)
-{
-  while (*from) {
-    if (*from == SS2) {
-      from++;
-      *to = 0x3f00 & (*from++ << 8);
-      *to = *from++;
-    } else if (*from == SS3) {
-      from++;
-      *to = *from++ << 8;
-      *to |= 0x3f & *from++;
-    } else if (*from & 0x80) {
-      *to = *from++ << 8;
-      *to |= *from++;
-    } else {
-      *to = *from++;
-    }
-    to++;
-  }
-  *to = 0;
-}
-
-static void pg_euccn2wchar(const unsigned char *from, pg_wchar *to)
-{
-  while (*from) {
-    if (*from == SS2) {
-      from++;
-      *to = *from++ << 16;
-      *to |= *from++ << 8;
-      *to |= *from++;
-    } else if (*from == SS3) {
-      from++;
-      *to = *from++ << 8;
-      *to |= 0x3f & *from++;
-    } else if (*from & 0x80) {
-      *to = *from++ << 8;
-      *to |= *from++;
-    } else {
-      *to = *from++;
-    }
-    to++;
-  }
-  *to = 0;
-}

 /*
- * convert UTF-8 to pg_wchar (UCS-2)
- * caller should allocate enough space for "to"
+ * conversion to pg_wchar is done by "table driven."
+ * to add an encoding support, define mb2wchar_with_len(), mblen()
+ * for the particular encoding. Note that if the encoding is only
+ * supported in the client, you don't need to define 
+ * mb2wchar_with_len() function (SJIS is the case).
 */
-static void pg_utf2wchar(const unsigned char *from, pg_wchar *to)
-{
-  unsigned char c1,c2,c3;
-  while (*from) {
-    if ((*from & 0x80) == 0) {
-      *to = *from++;
-    } else if ((*from & 0xe0) == 0xc0) {
-      c1 = *from++ & 0x1f;
-      c2 = *from++ & 0x3f;
-      *to = c1 << 6;
-      *to |= c2;
-    } else if ((*from & 0xe0) == 0xe0) {
-      c1 = *from++ & 0x0f;
-      c2 = *from++ & 0x3f;
-      c3 = *from++ & 0x3f;
-      *to = c1 << 12;
-      *to |= c2 << 6;
-      *to |= c3;
-    }
-    to++;
-  }
-  *to = 0;
-}
+typedef struct {
+  void	(*mb2wchar_with_len)();	/* convert a multi-byte string to a wchar */
+  int	(*mblen)();		/* returns the length of a multi-byte word */
+} pg_wchar_tbl;

-/*
- * convert mule internal code to pg_wchar.
- * in this case pg_wchar consists of following 4 bytes:
- *
- * 0x00(unused)
- * 0x00(ASCII)|leading character (one of LC1, LC12, LC2 or LC22)
- * 0x00(ASCII,1 byte code)|other than 0x00(2 byte code)
- * the lowest byte of the code
- *
- * note that Type N (variable length byte encoding) cannot be represented by
- * this schema. sorry.
- * caller should allocate enough space for "to"
- */
-static void pg_mule2wchar(const unsigned char *from, pg_wchar *to)
-{
-  while (*from) {
-    if (IS_LC1(*from)) {
-      *to = *from++ << 16;
-      *to |= *from++;
-    } else if (IS_LCPRV1(*from)) {
-      from++;
-      *to = *from++ << 16;
-      *to |= *from++;
-    } else if (IS_LC2(*from)) {
-      *to = *from++ << 16;
-      *to |= *from++ << 8;
-      *to |= *from++;
-    } else if (IS_LCPRV2(*from)) {
-      from++;
-      *to = *from++ << 16;
-      *to |= *from++ << 8;
-      *to |= *from++;
-    } else {	/* assume ASCII */
-      *to = *from++;
-    }
-    to++;
-  }
-  *to = 0;
-}
-
-/*
- * convert EUC to pg_wchar (EUC process code)
- * caller should allocate enough space for "to"
- * len: length of from.
- * "from" not necessarily null terminated.
- */
-static void pg_euc2wchar_with_len(const unsigned char *from, pg_wchar *to, int len)
+static void pg_euc2wchar_with_len
+(const unsigned char *from, pg_wchar *to, int len)
 {
  while (*from && len > 0) {
    if (*from == SS2) {
@@ -184,19 +44,54 @@ static void pg_euc2wchar_with_len(const unsigned char *from, pg_wchar *to, int l
  *to = 0;
 }

+static int pg_euc_mblen(const unsigned char *s)
+{
+  int len;
+
+  if (*s == SS2) {
+    len = 2;
+  } else if (*s == SS3) {
+    len = 3;
+  } else if (*s & 0x80) {
+    len = 2;
+  } else {
+    len = 1;
+  }
+  return(len);
+}
+
+/*
+ * EUC_JP
+ */
 static void pg_eucjp2wchar_with_len
 (const unsigned char *from, pg_wchar *to, int len)
 {
  pg_euc2wchar_with_len(from,to,len);
 }

+static int pg_eucjp_mblen(const unsigned char *s)
+{
+  return(pg_euc_mblen(s));
+}
+
+/*
+ * EUC_KR
+ */
 static void pg_euckr2wchar_with_len
 (const unsigned char *from, pg_wchar *to, int len)
 {
  pg_euc2wchar_with_len(from,to,len);
 }

-static void pg_eucch2wchar_with_len
+static int pg_euckr_mblen(const unsigned char *s)
+{
+  return(pg_euc_mblen(s));
+}
+
+/*
+ * EUC_CN
+ */
+static void pg_euccn2wchar_with_len
 (const unsigned char *from, pg_wchar *to, int len)
 {
  while (*from && len > 0) {
@@ -224,7 +119,26 @@ static void pg_eucch2wchar_with_len
  *to = 0;
 }

-static void pg_euccn2wchar_with_len
+static int pg_euccn_mblen(const unsigned char *s)
+{
+  int len;
+
+  if (*s == SS2) {
+    len = 3;
+  } else if (*s == SS3) {
+    len = 3;
+  } else if (*s & 0x80) {
+    len = 2;
+  } else {
+    len = 1;
+  }
+  return(len);
+}
+
+/*
+ * EUC_TW
+ */
+static void pg_euctw2wchar_with_len
 (const unsigned char *from, pg_wchar *to, int len)
 {
  while (*from && len > 0) {
@@ -253,6 +167,22 @@ static void pg_euccn2wchar_with_len
  *to = 0;
 }

+static int pg_euctw_mblen(const unsigned char *s)
+{
+  int len;
+
+  if (*s == SS2) {
+    len = 4;
+  } else if (*s == SS3) {
+    len = 3;
+  } else if (*s & 0x80) {
+    len = 2;
+  } else {
+    len = 1;
+  }
+  return(len);
+}
+
 /*
 * convert UTF-8 to pg_wchar (UCS-2)
 * caller should allocate enough space for "to"
@@ -286,6 +216,20 @@ static void pg_utf2wchar_with_len(const unsigned char *from, pg_wchar *to, int l
  *to = 0;
 }

+static int pg_utf_mblen(const unsigned char *s)
+{
+  int len = 1;
+
+  if ((*s & 0x80) == 0) {
+    len = 1;
+  } else if ((*s & 0xe0) == 0xc0) {
+    len = 2;
+  } else if ((*s & 0xe0) == 0xe0) {
+    len = 3;
+  }
+  return(len);
+}
+
 /*
 * convert mule internal code to pg_wchar
 * caller should allocate enough space for "to"
@@ -324,78 +268,6 @@ static void pg_mule2wchar_with_len(const unsigned char *from, pg_wchar *to, int
  *to = 0;
 }

-static int pg_euc_mblen(const unsigned char *s)
-{
-  int len;
-
-  if (*s == SS2) {
-    len = 2;
-  } else if (*s == SS3) {
-    len = 3;
-  } else if (*s & 0x80) {
-    len = 2;
-  } else {
-    len = 1;
-  }
-  return(len);
-}
-
-static int pg_eucjp_mblen(const unsigned char *s)
-{
-  return(pg_euc_mblen(s));
-}
-
-static int pg_euckr_mblen(const unsigned char *s)
-{
-  return(pg_euc_mblen(s));
-}
-
-static int pg_eucch_mblen(const unsigned char *s)
-{
-  int len;
-
-  if (*s == SS2) {
-    len = 3;
-  } else if (*s == SS3) {
-    len = 3;
-  } else if (*s & 0x80) {
-    len = 2;
-  } else {
-    len = 1;
-  }
-  return(len);
-}
-
-static int pg_euccn_mblen(const unsigned char *s)
-{
-  int len;
-
-  if (*s == SS2) {
-    len = 4;
-  } else if (*s == SS3) {
-    len = 3;
-  } else if (*s & 0x80) {
-    len = 2;
-  } else {
-    len = 1;
-  }
-  return(len);
-}
-
-static int pg_utf_mblen(const unsigned char *s)
-{
-  int len = 1;
-
-  if ((*s & 0x80) == 0) {
-    len = 1;
-  } else if ((*s & 0xe0) == 0xc0) {
-    len = 2;
-  } else if ((*s & 0xe0) == 0xe0) {
-    len = 3;
-  }
-  return(len);
-}
-
 static int pg_mule_mblen(const unsigned char *s)
 {
  int len;
@@ -414,25 +286,71 @@ static int pg_mule_mblen(const unsigned char *s)
  return(len);
 }

-typedef struct {
-  void	(*mb2wchar)();		/* convert a multi-byte string to a wchar */
-  void	(*mb2wchar_with_len)();	/* convert a multi-byte string to a wchar 
-				   with a limited length */
-  int	(*mblen)();		/* returns the length of a multi-byte word */
-} pg_wchar_tbl;
+/*
+ * ISO8859-1
+ */
+static void pg_latin12wchar_with_len(const unsigned char *from, pg_wchar *to, int len)
+{
+  while (*from && len-- > 0) {
+    *to++ = *from++;
+  }
+  *to = 0;
+}
+
+static int pg_latin1_mblen(const unsigned char *s)
+{
+  return(1);
+}
+
+/*
+ * SJIS
+ */
+static int pg_sjis_mblen(const unsigned char *s)
+{
+  int len;
+
+  if (*s >= 0xa1 && *s <= 0xdf) {	/* 1 byte kana? */
+    len = 1;
+  } else if (*s > 0x7f) {	/* kanji? */
+    len = 2;
+  } else {	/* should be ASCII */
+    len = 1;
+  }
+  return(len);
+}

 static pg_wchar_tbl pg_wchar_table[] = {
-  {pg_eucjp2wchar, pg_eucjp2wchar_with_len, pg_eucjp_mblen},
-  {pg_eucch2wchar, pg_eucch2wchar_with_len, pg_eucch_mblen},
-  {pg_euckr2wchar, pg_euckr2wchar_with_len, pg_euckr_mblen},
-  {pg_euccn2wchar, pg_euccn2wchar_with_len, pg_euccn_mblen},
-  {pg_utf2wchar, pg_utf2wchar_with_len, pg_utf_mblen},
-  {pg_mule2wchar, pg_mule2wchar_with_len, pg_mule_mblen}};
+  {pg_eucjp2wchar_with_len, pg_eucjp_mblen},
+  {pg_euccn2wchar_with_len, pg_euccn_mblen},
+  {pg_euckr2wchar_with_len, pg_euckr_mblen},
+  {pg_euctw2wchar_with_len, pg_euctw_mblen},
+  {pg_utf2wchar_with_len, pg_utf_mblen},
+  {pg_mule2wchar_with_len, pg_mule_mblen},
+  {pg_latin12wchar_with_len, pg_latin1_mblen},
+  {0, 0},
+  {0, 0},
+  {0, 0},
+  {0, 0},
+  {0, 0},
+  {0, 0},
+  {0, 0},
+  {0, 0},
+  {0, 0},
+  {0, pg_sjis_mblen}
+};
+
+/*
+ *########################################################################
+ *
+ * Public functions
+ *
+ *########################################################################
+ */

 /* convert a multi-byte string to a wchar */
 void pg_mb2wchar(const unsigned char *from, pg_wchar *to)
 {
-  (*pg_wchar_table[MB].mb2wchar)(from,to);
+  (*pg_wchar_table[MB].mb2wchar_with_len)(from,to,strlen(from));
 }

 /* convert a multi-byte string to a wchar with a limited length */
@@ -447,6 +365,18 @@ int pg_mblen(const unsigned char *mbstr)
  return((*pg_wchar_table[MB].mblen)(mbstr));
 }

+/* returns the byte length of a multi-byte word for an encoding */
+int pg_encoding_mblen(int encoding, const unsigned char *mbstr)
+{
+  return((*pg_wchar_table[encoding].mblen)(mbstr));
+}
+
+/* returns the byte length of a word for mule internal code */
+int pg_mic_mblen(const unsigned char *mbstr)
+{
+  return(pg_mule_mblen(mbstr));
+}
+
 /* returns the length (counted as a wchar) of a multi-byte string */
 int pg_mbstrlen(const unsigned char *mbstr)
 {
--- a/src/backend/tcop/Makefile
+++ b/src/backend/tcop/Makefile
@@ -4,7 +4,7 @@
 #    Makefile for tcop
 #
 # IDENTIFICATION
-#    $Header: /cvsroot/pgsql/src/backend/tcop/Makefile,v 1.16 1998/04/06 00:26:05 momjian Exp $
+#    $Header: /cvsroot/pgsql/src/backend/tcop/Makefile,v 1.17 1998/06/16 07:29:30 momjian Exp $
 #
 #-------------------------------------------------------------------------

@@ -13,6 +13,10 @@ include ../../Makefile.global

 CFLAGS+= -I..

+ifdef MB
+CFLAGS+= -DMB=$(MB)
+endif
+
 ifeq ($(CC), gcc)
 CFLAGS+= -Wno-error
 endif
--- a/src/backend/tcop/postgres.c
+++ b/src/backend/tcop/postgres.c
@@ -7,7 +7,7 @@
 *
 *
 * IDENTIFICATION
- *	  $Header: /cvsroot/pgsql/src/backend/tcop/postgres.c,v 1.76 1998/06/15 19:29:27 momjian Exp $
+ *	  $Header: /cvsroot/pgsql/src/backend/tcop/postgres.c,v 1.77 1998/06/16 07:29:30 momjian Exp $
 *
 * NOTES
 *	  this is the "main" module of the postgres backend and
@@ -83,6 +83,10 @@
 #include "nodes/memnodes.h"
 #endif

+#ifdef MB
+#include "commands/variable.h"
+#endif
+
 /* ----------------
 *		global variables
 * ----------------
@@ -1270,6 +1274,19 @@ PostgresMain(int argc, char *argv[], int real_argc, char *real_argv[])

 	InitPostgres(DBName);

+#ifdef MB
+	/* set default client encoding */
+	if (!Quiet)
+	{
+		puts("\treset_client_encoding()..");
+	}
+	reset_client_encoding();
+	if (!Quiet)
+	{
+		puts("\treset_client_encoding() done.");
+	}
+#endif
+
 	/* ----------------
 	 *	if an exception is encountered, processing resumes here
 	 *	so we abort the current transaction and start a new one.
@@ -1308,7 +1325,7 @@ PostgresMain(int argc, char *argv[], int real_argc, char *real_argv[])
 	if (!IsUnderPostmaster)
 	{
 		puts("\nPOSTGRES backend interactive interface");
-		puts("$Revision: 1.76 $ $Date: 1998/06/15 19:29:27 $");
+		puts("$Revision: 1.77 $ $Date: 1998/06/16 07:29:30 $");
 	}

 	/* ----------------