diff --git a/doc/README.mb b/doc/README.mb
index 4d6c3288af1030d5d178e45e9bf4a6c83c6c0e21..379622eedfcdb1ba8c34619a0036d0acd9580f32 100644
--- a/doc/README.mb
+++ b/doc/README.mb
@@ -1,4 +1,4 @@
-postgresql 6.4 multi-byte (MB) support README Dec 16 1998
+postgresql 6.5 multi-byte (MB) support README Jan 26 1999
Tatsuo Ishii
t-ishii@sra.co.jp
@@ -113,6 +113,7 @@ Supported encodings for PGCLIENTENCODING are:
EUC_CN Chinese EUC
EUC_KR Korean EUC
EUC_TW Taiwan EUC
+ BIG5 Traditional chinese
MULE_INTERNAL Mule internal
LATIN1 ISO 8859-1 English and some European languages
LATIN2 ISO 8859-2 English and some European languages
@@ -169,6 +170,12 @@ Unicode: http://www.unicode.org/
5. History
+Jan 26, 1999
+ * Add support Big5 for fronend encoding
+ (you need to create a database with EUC_TW to use Big5)
+ * Add regression test case for EUC_TW
+ (contributed by Jonah Kuo <jonahkuo@mail.ttn.com.tw>)
+
Dec 15, 1998
* Bugs related to SQL_ASCII support fixed
diff --git a/doc/README.mb.jp b/doc/README.mb.jp
index 52ae09ec48a15b5b57100e0be817ad20ad69b37f..6030613ec9026c4306feec16913ac203ff42f742 100644
--- a/doc/README.mb.jp
+++ b/doc/README.mb.jp
@@ -1,4 +1,4 @@
-postgresql 6.4 multi-byte (MB) support README 1998/12/16 $B:n@.(B
+postgresql 6.5 multi-byte (MB) support README 1999/1/26 $B:n@.(B
$B@P0fC#IW(B
t-ishii@sra.co.jp
@@ -130,7 +130,7 @@ initdb/createdb/create database $B$K$*$1$k%(%s%3!<%G%#%s%0$N;XDj$K$D$$$F(B
$B4D6-JQ?t(B PGCLIENTENCODING $B$,@_Dj$5$l$F$$$k>l9g$O$=$NCM$,M%@h$5$l!"%P%C(B
$B%/%(%s%IB&$H0[$J$k%(%s%3!<%G%#%s%0$,;HMQ$G$-$^$9!#@_Dj2DG=$J%(%s%3!<(B
- $B%G%#%s%0$O!">e5-$K2C$(!"(BSJIS ($B%7%U%H(BJIS)$B$,;XDj$G$-$^$9!#(B
+ $B%G%#%s%0$O!">e5-$K2C$(!"(BSJIS ($B%7%U%H(BJIS)$B$H(B BiG5 $B$,;XDj$G$-$^$9!#(B
$B$A$J$_$K!"(BSJIS $B$O(B JISX0201 $B$N(B 1$B%P%$%H%+%J!"$$$o$f$k!VH>3Q%+%?(B
$B%+%J!W$b%5%]!<%H$7$F$$$^$9(B($B7h$7$F!VH>3Q%+%?%+%J!W$N;HMQ$r$*4+(B
@@ -206,6 +206,13 @@ initdb/createdb/create database $B$K$*$1$k%(%s%3!<%G%#%s%0$N;XDj$K$D$$$F(B
$B2~DjMzNr!'(B
+ 1999/1/26 Big5 $B%5%]!<%HDI2C(B
+ * Big5 $B$,%U%m%s%H%(%s%IB&$N%(%s%3!<%G%#%s%0$H$7$FMxMQ$G$-$k$h(B
+ $B$&$K$J$j$^$7$?!#$3$N>l9g!"%P%C%/%(%s%IB&$N%(%s%3!<%G%#%s%0$O(B
+ EUC_TW $B$^$?$O(B MULE_INTERNAL $B$H$7$^$9!#(B
+ * EUC_TW $B$N(B regression test $B%1!<%9$rDI2C(B
+ (contributed by Jonah Kuo <jonahkuo@mail.ttn.com.tw>)
+
1998/12/16 $BK\%I%-%e%a%s%H=$@5!#(B
* Makefile.custom $B$G(B MB=EUC_JP $B$J$I$H@_Dj$9$kJ}K!$O(B 6.4 $B0J9_(B
$B%5%]!<%H$5$l$F$$$J$$$N$G:o=|$7$?!#(B
diff --git a/src/backend/utils/mb/Makefile b/src/backend/utils/mb/Makefile
index 5b7cb745add50a22bd6bceda8d67847a3dcd467e..47180e8004d6536023d8576380c7993ba462b6a5 100644
--- a/src/backend/utils/mb/Makefile
+++ b/src/backend/utils/mb/Makefile
@@ -4,7 +4,7 @@
# Makefile for utils/mb
#
# IDENTIFICATION
-# $Header: /cvsroot/pgsql/src/backend/utils/mb/Makefile,v 1.2 1998/07/26 04:31:03 scrappy Exp $
+# $Header: /cvsroot/pgsql/src/backend/utils/mb/Makefile,v 1.3 1999/02/02 18:51:21 momjian Exp $
#
#-------------------------------------------------------------------------
@@ -16,7 +16,8 @@ ifdef MULTIBYTE
CFLAGS+= $(MBFLAGS)
endif
-OBJS = common.o conv.o mbutils.o wchar.o wstrcmp.o wstrncmp.o variable.o
+OBJS = common.o conv.o mbutils.o wchar.o wstrcmp.o wstrncmp.o variable.o \
+ big5.o
all: SUBSYS.o
diff --git a/src/backend/utils/mb/big5.c b/src/backend/utils/mb/big5.c
new file mode 100644
index 0000000000000000000000000000000000000000..105854d24799ee59b8742442f308461871dc5745
--- /dev/null
+++ b/src/backend/utils/mb/big5.c
@@ -0,0 +1,347 @@
+/*
+ * conversion between BIG5 and Mule Internal Code(CNS 116643-1992
+ * plane 1 and plane 2).
+ * This program is partially copied from lv(Multilingual file viewer)
+ * and slightly modified. lv is written and copyrighted by NARITA Tomio
+ * (nrt@web.ad.jp).
+ *
+ * 1999/1/15 Tatsuo Ishii
+ *
+ * $Id: big5.c,v 1.1 1999/02/02 18:51:22 momjian Exp $
+ */
+
+#include "mb/pg_wchar.h"
+
+typedef struct {
+ unsigned short code, peer;
+} codes_t;
+
+/* map Big5 Level 1 to CNS 11643-1992 Plane 1 */
+static codes_t big5Level1ToCnsPlane1[ 25 ] = { /* range */
+ { 0xA140, 0x2121 },
+ { 0xA1F6, 0x2258 },
+ { 0xA1F7, 0x2257 },
+ { 0xA1F8, 0x2259 },
+ { 0xA2AF, 0x2421 },
+ { 0xA3C0, 0x4221 },
+ { 0xa3e1, 0x0000 },
+ { 0xA440, 0x4421 },
+ { 0xACFE, 0x5753 },
+ { 0xacff, 0x0000 },
+ { 0xAD40, 0x5323 },
+ { 0xAFD0, 0x5754 },
+ { 0xBBC8, 0x6B51 },
+ { 0xBE52, 0x6B50 },
+ { 0xBE53, 0x6F5C },
+ { 0xC1AB, 0x7536 },
+ { 0xC2CB, 0x7535 },
+ { 0xC2CC, 0x7737 },
+ { 0xC361, 0x782E },
+ { 0xC3B9, 0x7865 },
+ { 0xC3BA, 0x7864 },
+ { 0xC3BB, 0x7866 },
+ { 0xC456, 0x782D },
+ { 0xC457, 0x7962 },
+ { 0xc67f, 0x0000 }
+};
+
+/* map CNS 11643-1992 Plane 1 to Big5 Level 1 */
+static codes_t cnsPlane1ToBig5Level1[ 26 ] = { /* range */
+ { 0x2121, 0xA140 },
+ { 0x2257, 0xA1F7 },
+ { 0x2258, 0xA1F6 },
+ { 0x2259, 0xA1F8 },
+ { 0x234f, 0x0000 },
+ { 0x2421, 0xA2AF },
+ { 0x2571, 0x0000 },
+ { 0x4221, 0xA3C0 },
+ { 0x4242, 0x0000 },
+ { 0x4421, 0xA440 },
+ { 0x5323, 0xAD40 },
+ { 0x5753, 0xACFE },
+ { 0x5754, 0xAFD0 },
+ { 0x6B50, 0xBE52 },
+ { 0x6B51, 0xBBC8 },
+ { 0x6F5C, 0xBE53 },
+ { 0x7535, 0xC2CB },
+ { 0x7536, 0xC1AB },
+ { 0x7737, 0xC2CC },
+ { 0x782D, 0xC456 },
+ { 0x782E, 0xC361 },
+ { 0x7864, 0xC3BA },
+ { 0x7865, 0xC3B9 },
+ { 0x7866, 0xC3BB },
+ { 0x7962, 0xC457 },
+ { 0x7d4c, 0x0000 }
+};
+
+/* map Big5 Level 2 to CNS 11643-1992 Plane 2 */
+static codes_t big5Level2ToCnsPlane2[ 48 ] = { /* range */
+ { 0xC940, 0x2121 },
+ { 0xc94a, 0x0000 },
+ { 0xC94B, 0x212B },
+ { 0xC96C, 0x214D },
+ { 0xC9BE, 0x214C },
+ { 0xC9BF, 0x217D },
+ { 0xC9ED, 0x224E },
+ { 0xCAF7, 0x224D },
+ { 0xCAF8, 0x2439 },
+ { 0xD77A, 0x3F6A },
+ { 0xD77B, 0x387E },
+ { 0xDBA7, 0x3F6B },
+ { 0xDDFC, 0x4176 },
+ { 0xDDFD, 0x4424 },
+ { 0xE8A3, 0x554C },
+ { 0xE976, 0x5723 },
+ { 0xEB5B, 0x5A29 },
+ { 0xEBF1, 0x554B },
+ { 0xEBF2, 0x5B3F },
+ { 0xECDE, 0x5722 },
+ { 0xECDF, 0x5C6A },
+ { 0xEDAA, 0x5D75 },
+ { 0xEEEB, 0x642F },
+ { 0xEEEC, 0x6039 },
+ { 0xF056, 0x5D74 },
+ { 0xF057, 0x6243 },
+ { 0xF0CB, 0x5A28 },
+ { 0xF0CC, 0x6337 },
+ { 0xF163, 0x6430 },
+ { 0xF16B, 0x6761 },
+ { 0xF16C, 0x6438 },
+ { 0xF268, 0x6934 },
+ { 0xF269, 0x6573 },
+ { 0xF2C3, 0x664E },
+ { 0xF375, 0x6762 },
+ { 0xF466, 0x6935 },
+ { 0xF4B5, 0x664D },
+ { 0xF4B6, 0x6962 },
+ { 0xF4FD, 0x6A4C },
+ { 0xF663, 0x6A4B },
+ { 0xF664, 0x6C52 },
+ { 0xF977, 0x7167 },
+ { 0xF9C4, 0x7166 },
+ { 0xF9C5, 0x7234 },
+ { 0xF9C6, 0x7240 },
+ { 0xF9C7, 0x7235 },
+ { 0xF9D2, 0x7241 },
+ { 0xf9d6, 0x0000 }
+};
+
+/* map CNS 11643-1992 Plane 2 to Big5 Level 2 */
+static codes_t cnsPlane2ToBig5Level2[ 49 ] = { /* range */
+ { 0x2121, 0xC940 },
+ { 0x212B, 0xC94B },
+ { 0x214C, 0xC9BE },
+ { 0x214D, 0xC96C },
+ { 0x217D, 0xC9BF },
+ { 0x224D, 0xCAF7 },
+ { 0x224E, 0xC9ED },
+ { 0x2439, 0xCAF8 },
+ { 0x387E, 0xD77B },
+ { 0x3F6A, 0xD77A },
+ { 0x3F6B, 0xDBA7 },
+ { 0x4424, 0x0000 },
+ { 0x4176, 0xDDFC },
+ { 0x4177, 0x0000 },
+ { 0x4424, 0xDDFD },
+ { 0x554B, 0xEBF1 },
+ { 0x554C, 0xE8A3 },
+ { 0x5722, 0xECDE },
+ { 0x5723, 0xE976 },
+ { 0x5A28, 0xF0CB },
+ { 0x5A29, 0xEB5B },
+ { 0x5B3F, 0xEBF2 },
+ { 0x5C6A, 0xECDF },
+ { 0x5D74, 0xF056 },
+ { 0x5D75, 0xEDAA },
+ { 0x6039, 0xEEEC },
+ { 0x6243, 0xF057 },
+ { 0x6337, 0xF0CC },
+ { 0x642F, 0xEEEB },
+ { 0x6430, 0xF163 },
+ { 0x6438, 0xF16C },
+ { 0x6573, 0xF269 },
+ { 0x664D, 0xF4B5 },
+ { 0x664E, 0xF2C3 },
+ { 0x6761, 0xF16B },
+ { 0x6762, 0xF375 },
+ { 0x6934, 0xF268 },
+ { 0x6935, 0xF466 },
+ { 0x6962, 0xF4B6 },
+ { 0x6A4B, 0xF663 },
+ { 0x6A4C, 0xF4FD },
+ { 0x6C52, 0xF664 },
+ { 0x7166, 0xF9C4 },
+ { 0x7167, 0xF977 },
+ { 0x7234, 0xF9C5 },
+ { 0x7235, 0xF9C7 },
+ { 0x7240, 0xF9C6 },
+ { 0x7241, 0xF9D2 },
+ { 0x7245, 0x0000 }
+};
+
+/* Big Five Level 1 Correspondence to CNS 11643-1992 Plane 4 */
+static unsigned short b1c4[][2] = {
+ {0xC879, 0x2123},
+ {0xC87B, 0x2124},
+ {0xC87D, 0x212A},
+ {0xC8A2, 0x2152}
+};
+
+/* Big Five Level 2 Correspondence to CNS 11643-1992 Plane 3 */
+static unsigned short b2c3[][2] = {
+ {0xF9D6, 0x4337},
+ {0xF9D7, 0x4F50},
+ {0xF9D8, 0x444E},
+ {0xF9D9, 0x504A},
+ {0xF9DA, 0x2C5D},
+ {0xF9DB, 0x3D7E},
+ {0xF9DC, 0x4B5C}
+};
+
+static unsigned short BinarySearchRange
+(codes_t *array, int high, unsigned short code )
+{
+ int low, mid, distance, tmp;
+
+ low = 0;
+ mid = high >> 1;
+
+ for( ; low <= high ; mid = ( low + high ) >> 1 ){
+ if( ( array[ mid ].code <= code ) && ( array[ mid + 1 ].code > code ) ){
+ if( 0 == array[ mid ].peer )
+ return 0;
+ if( code >= 0xa140U ){
+ /* big5 to cns */
+ tmp = ( ( code & 0xff00 ) - ( array[ mid ].code & 0xff00 ) ) >> 8;
+ high = code & 0x00ff;
+ low = array[ mid ].code & 0x00ff;
+ /*
+ * NOTE: big5 high_byte: 0xa1-0xfe, low_byte: 0x40-0x7e, 0xa1-0xfe
+ * (radicals: 0x00-0x3e, 0x3f-0x9c)
+ * big5 radix is 0x9d. [region_low, region_high]
+ * We should remember big5 has two different regions (above).
+ * There is a bias for the distance between these regions.
+ * 0xa1 - 0x7e + bias = 1 (Distance between 0xa1 and 0x7e is 1.)
+ * bias = - 0x22.
+ */
+ distance = tmp * 0x9d + high - low +
+ ( high >= 0xa1 ? ( low >= 0xa1 ? 0 : - 0x22 )
+ : ( low >= 0xa1 ? + 0x22 : 0 ) );
+ /*
+ * NOTE: we have to convert the distance into a code point.
+ * The code point's low_byte is 0x21 plus mod_0x5e.
+ * In the first, we extract the mod_0x5e of the starting
+ * code point, subtracting 0x21, and add distance to it.
+ * Then we calculate again mod_0x5e of them, and restore
+ * the final codepoint, adding 0x21.
+ */
+ tmp = ( array[ mid ].peer & 0x00ff ) + distance - 0x21;
+ tmp = ( array[ mid ].peer & 0xff00 ) + ( ( tmp / 0x5e ) << 8 )
+ + 0x21 + tmp % 0x5e;
+ return tmp;
+ } else {
+ /* cns to big5 */
+ tmp = ( ( code & 0xff00 ) - ( array[ mid ].code & 0xff00 ) ) >> 8;
+ /*
+ * NOTE: ISO charsets ranges between 0x21-0xfe (94charset).
+ * Its radix is 0x5e. But there is no distance bias like big5.
+ */
+ distance = tmp * 0x5e
+ + ( (int)( code & 0x00ff ) - (int)( array[ mid ].code & 0x00ff ) );
+ /*
+ * NOTE: Similar to big5 to cns conversion, we extract mod_0x9d and
+ * restore mod_0x9d into a code point.
+ */
+ low = array[ mid ].peer & 0x00ff;
+ tmp = low + distance - ( low >= 0xa1 ? 0x62 : 0x40 );
+ low = tmp % 0x9d;
+ tmp = ( array[ mid ].peer & 0xff00 ) + ( ( tmp / 0x9d ) << 8 )
+ + ( low > 0x3e ? 0x62 : 0x40 ) + low;
+ return tmp;
+ }
+ } else if( array[ mid ].code > code ){
+ high = mid - 1;
+ } else {
+ low = mid + 1;
+ }
+ }
+
+ return 0;
+}
+
+
+unsigned short BIG5toCNS(unsigned short big5, unsigned char *lc )
+{
+ unsigned short cns = 0;
+ int i;
+
+ if( big5 < 0xc940U ){
+ /* level 1 */
+
+ for (i=0;i<sizeof(b1c4)/sizeof(unsigned short);i++) {
+ if (b1c4[i][0] == big5) {
+ *lc = LC_CNS11643_4;
+ return(b1c4[i][1] | 0x8080U);
+ }
+ }
+
+ if( 0 < (cns = BinarySearchRange( big5Level1ToCnsPlane1, 23, big5 )) )
+ *lc = LC_CNS11643_1;
+ } else if( big5 == 0xc94aU ){
+ /* level 2 */
+ *lc = LC_CNS11643_1;
+ cns = 0x4442;
+ } else {
+ /* level 2 */
+ for (i=0;i<sizeof(b2c3)/sizeof(unsigned short);i++) {
+ if (b2c3[i][0] == big5) {
+ *lc = LC_CNS11643_3;
+ return(b2c3[i][1]);
+ }
+ }
+
+ if( 0 < (cns = BinarySearchRange( big5Level2ToCnsPlane2, 46, big5 )) )
+ *lc = LC_CNS11643_2;
+ }
+
+ if( 0 == cns ){ /* no mapping Big5 to CNS 11643-1992 */
+ *lc = 0;
+ return (unsigned short)'?';
+ }
+
+ return cns | 0x8080;
+}
+
+unsigned short CNStoBIG5( unsigned short cns, unsigned char lc )
+{
+ int i;
+ unsigned int big5 = 0;
+
+ cns &= 0x7f7f;
+
+ switch( lc ){
+ case LC_CNS11643_1:
+ big5 = BinarySearchRange( cnsPlane1ToBig5Level1, 24, cns );
+ break;
+ case LC_CNS11643_2:
+ big5 = BinarySearchRange( cnsPlane2ToBig5Level2, 47, cns );
+ break;
+ case LC_CNS11643_3:
+ for (i=0;i<sizeof(b2c3)/sizeof(unsigned short);i++) {
+ if (b2c3[i][1] == cns) {
+ return(b2c3[i][0]);
+ }
+ }
+ break;
+ case LC_CNS11643_4:
+ for (i=0;i<sizeof(b1c4)/sizeof(unsigned short);i++) {
+ if (b1c4[i][1] == cns) {
+ return(b1c4[i][0]);
+ }
+ }
+ default:
+ break;
+ }
+ return big5;
+}
diff --git a/src/backend/utils/mb/conv.c b/src/backend/utils/mb/conv.c
index 1a4493c8e07dc395d729e63264e0335f9dc9b7db..9aac4e96a921458cb77eab39de07b4d616901091 100644
--- a/src/backend/utils/mb/conv.c
+++ b/src/backend/utils/mb/conv.c
@@ -2,7 +2,7 @@
* conversion between client encoding and server internal encoding
* (currently mule internal code (mic) is used)
* Tatsuo Ishii
- * $Id: conv.c,v 1.4 1998/12/14 04:59:58 momjian Exp $
+ * $Id: conv.c,v 1.5 1999/02/02 18:51:23 momjian Exp $
*/
#include <stdio.h>
#include <string.h>
@@ -369,6 +369,94 @@ mic2euc_tw(unsigned char *mic, unsigned char *p, int len)
*p = '\0';
}
+/*
+ * Big5 ---> MIC
+ */
+static void
+big52mic(unsigned char *big5, unsigned char *p, int len)
+{
+ unsigned short c1;
+ unsigned short big5buf, cnsBuf;
+ unsigned char lc;
+ char bogusBuf[2];
+ int i;
+
+ while (len > 0 && (c1 = *big5++))
+ {
+ if (c1 <= 0x007fU) { /* ASCII */
+ len--;
+ *p++ = c1;
+ } else {
+ len -= 2;
+ big5buf = c1 << 8;
+ c1 = *big5++;
+ big5buf |= c1;
+ cnsBuf = BIG5toCNS(big5buf, &lc);
+ if (lc != 0) {
+ if (lc == LC_CNS11643_3 || lc == LC_CNS11643_4) {
+ *p++ = 0x9d; /* LCPRV2 */
+ }
+ *p++ = lc; /* Plane No. */
+ *p++ = (cnsBuf >> 8) & 0x00ff;
+ *p++ = cnsBuf & 0x00ff;
+ } else { /* cannot convert */
+ big5 -= 2;
+ *p++ = '(';
+ for (i=0;i<2;i++) {
+ sprintf(bogusBuf,"%02x",*big5++);
+ *p++ = bogusBuf[0];
+ *p++ = bogusBuf[1];
+ }
+ *p++ = ')';
+ }
+ }
+ }
+ *p = '\0';
+}
+
+/*
+ * MIC ---> Big5
+ */
+static void
+mic2big5(unsigned char *mic, unsigned char *p, int len)
+{
+ int l;
+ unsigned short c1;
+ unsigned short big5buf, cnsBuf;
+
+ while (len > 0 && (c1 = *mic))
+ {
+ l = pg_mic_mblen(mic++);
+ len -= l;
+
+ /* 0x9d means LCPRV2 */
+ if (c1 == LC_CNS11643_1 || c1 == LC_CNS11643_2 || c1 == 0x9d)
+ {
+ if (c1 == 0x9d) {
+ c1 = *mic++; /* get plane no. */
+ }
+ cnsBuf = (*mic++)<<8;
+ cnsBuf |= (*mic++) & 0x00ff;
+ big5buf = CNStoBIG5(cnsBuf, c1);
+ if (big5buf == 0) { /* cannot convert to Big5! */
+ mic -= l;
+ printBogusChar(&mic, &p);
+ } else {
+ *p++ = (big5buf >> 8) & 0x00ff;
+ *p++ = big5buf & 0x00ff;
+ }
+ }
+ else if (c1 <= 0x7f) /* ASCII */
+ {
+ *p++ = c1;
+ } else { /* cannot convert to Big5! */
+ mic--;
+ printBogusChar(&mic, &p);
+ }
+ }
+ *p = '\0';
+}
+
/*
* LATINn ---> MIC
*/
@@ -514,5 +602,6 @@ pg_encoding_conv_tbl pg_conv_tbl[] = {
{LATIN4, "LATIN4", 0, latin42mic, mic2latin4}, /* ISO 8859 Latin 4 */
{LATIN5, "LATIN5", 0, latin52mic, mic2latin5}, /* ISO 8859 Latin 5 */
{SJIS, "SJIS", 1, sjis2mic, mic2sjis}, /* SJIS */
+ {BIG5, "BIG5", 1, big52mic, mic2big5}, /* Big5 */
{-1, "", 0, 0, 0} /* end mark */
};
diff --git a/src/backend/utils/mb/mbutils.c b/src/backend/utils/mb/mbutils.c
index e359a53df3d4350b2c673d52f39a00daf6dc7548..f354ccc96a24ff63135167ad62c9b552e6e23019 100644
--- a/src/backend/utils/mb/mbutils.c
+++ b/src/backend/utils/mb/mbutils.c
@@ -3,14 +3,14 @@
* client encoding and server internal encoding.
* (currently mule internal code (mic) is used)
* Tatsuo Ishii
- * $Id: mbutils.c,v 1.4 1998/09/25 01:46:23 momjian Exp $ */
+ * $Id: mbutils.c,v 1.5 1999/02/02 18:51:23 momjian Exp $ */
#include <stdio.h>
#include <string.h>
#include "mb/pg_wchar.h"
-static client_encoding = -1;
+static int client_encoding = -1;
static void (*client_to_mic) ();/* something to MIC */
static void (*client_from_mic) (); /* MIC to something */
static void (*server_to_mic) ();/* something to MIC */
diff --git a/src/backend/utils/mb/wchar.c b/src/backend/utils/mb/wchar.c
index edd786f29f80470f789a3a203b59c0c514097155..2a1141fbadbf776b58f4f0a7c15b2baf2f41bf68 100644
--- a/src/backend/utils/mb/wchar.c
+++ b/src/backend/utils/mb/wchar.c
@@ -1,7 +1,7 @@
/*
* conversion functions between pg_wchar and multi-byte streams.
* Tatsuo Ishii
- * $Id: wchar.c,v 1.4 1998/09/01 04:33:26 momjian Exp $
+ * $Id: wchar.c,v 1.5 1999/02/02 18:51:23 momjian Exp $
*/
#include "mb/pg_wchar.h"
@@ -396,6 +396,25 @@ pg_sjis_mblen(const unsigned char *s)
return (len);
}
+/*
+ * Big5
+ */
+static int
+pg_big5_mblen(const unsigned char *s)
+{
+ int len;
+
+ if (*s > 0x7f)
+ { /* kanji? */
+ len = 2;
+ }
+ else
+ { /* should be ASCII */
+ len = 1;
+ }
+ return (len);
+}
+
pg_wchar_tbl pg_wchar_table[] = {
{pg_ascii2wchar_with_len, pg_ascii_mblen},
{pg_eucjp2wchar_with_len, pg_eucjp_mblen},
@@ -429,7 +448,8 @@ pg_wchar_tbl pg_wchar_table[] = {
{0, 0},
{0, 0},
{0, 0},
- {0, pg_sjis_mblen}
+ {0, pg_sjis_mblen},
+ {0, pg_big5_mblen}
};
/* returns the byte length of a word for mule internal code */
diff --git a/src/bin/psql/psqlHelp.h b/src/bin/psql/psqlHelp.h
index ab0696d1ae44f82a4146842e715a6a8e0c33fff8..570f0226e344b1fc8a4230a7b79ae49ec3777d4c 100644
--- a/src/bin/psql/psqlHelp.h
+++ b/src/bin/psql/psqlHelp.h
@@ -5,7 +5,7 @@
*
* Copyright (c) 1994, Regents of the University of California
*
- * $Id: psqlHelp.h,v 1.57 1999/02/02 18:41:17 momjian Exp $
+ * $Id: psqlHelp.h,v 1.58 1999/02/02 18:51:24 momjian Exp $
*
*-------------------------------------------------------------------------
*/
@@ -330,7 +330,7 @@ static struct _helpStruct QL_HELP[] = {
set GEQO TO 'ON[=#]'|'OFF'\n\
set R_PLANS TO 'ON'|'OFF'\n\
set QUERY_LIMIT TO #\n\
-set CLIENT_ENCODING TO 'EUC_JP'|'SJIS'|'EUC_CN'|'EUC_KR'|'EUC_TW'|'MULE_INTERNAL'|'LATIN1'|'LATIN2'|'LATIN3'|'LATIN4'|'LATIN5'"},
+set CLIENT_ENCODING TO 'EUC_JP'|'SJIS'|'EUC_CN'|'EUC_KR'|'EUC_TW'|'BIG5'|'MULE_INTERNAL'|'LATIN1'|'LATIN2'|'LATIN3'|'LATIN4'|'LATIN5'"},
#else
"\
\tSET DateStyle TO 'ISO'|'SQL'|'Postgres'|'European'|'US'|'NonEuropean'\n\
diff --git a/src/include/mb/pg_wchar.h b/src/include/mb/pg_wchar.h
index 62bbe60cb1623b36311f09407200184fcc5c9335..889606361180fcc3442b3de2d18cecf37003a801 100644
--- a/src/include/mb/pg_wchar.h
+++ b/src/include/mb/pg_wchar.h
@@ -1,4 +1,4 @@
-/* $Id: pg_wchar.h,v 1.5 1998/09/25 01:46:25 momjian Exp $ */
+/* $Id: pg_wchar.h,v 1.6 1999/02/02 18:51:25 momjian Exp $ */
#ifndef PG_WCHAR_H
#define PG_WCHAR_H
@@ -11,7 +11,7 @@
#define EUC_JP 1 /* EUC for Japanese */
#define EUC_CN 2 /* EUC for Chinese */
#define EUC_KR 3 /* EUC for Korean */
-#define EUC_TW 3 /* EUC for Taiwan */
+#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 */
@@ -25,6 +25,7 @@
#define LATIN9 15 /* ISO-8859 Latin 9 */
/* followings are for client encoding only */
#define SJIS 32 /* Shift JIS */
+#define BIG5 33 /* Big5 */
#ifdef MULTIBYTE
typedef unsigned int pg_wchar;
@@ -122,6 +123,8 @@ extern int GetDatabaseEncoding(void);
extern void SetDatabaseEncoding(int);
extern void SetTemplateEncoding(int);
extern int GetTemplateEncoding(void);
+extern unsigned short BIG5toCNS(unsigned short, unsigned char *);
+extern unsigned short CNStoBIG5(unsigned short, unsigned char);
#endif /* MULTIBYTE */
diff --git a/src/interfaces/libpq/Makefile.in b/src/interfaces/libpq/Makefile.in
index 471fda1773df1258e5a8ebbc142bd5399e6a0754..b5ad798affcdbcb2fff131ca1025fcea3e31956f 100644
--- a/src/interfaces/libpq/Makefile.in
+++ b/src/interfaces/libpq/Makefile.in
@@ -6,7 +6,7 @@
# Copyright (c) 1994, Regents of the University of California
#
# IDENTIFICATION
-# $Header: /cvsroot/pgsql/src/interfaces/libpq/Attic/Makefile.in,v 1.42 1999/01/17 06:19:34 momjian Exp $
+# $Header: /cvsroot/pgsql/src/interfaces/libpq/Attic/Makefile.in,v 1.43 1999/02/02 18:51:29 momjian Exp $
#
#-------------------------------------------------------------------------
@@ -31,7 +31,7 @@ OBJS= fe-auth.o fe-connect.o fe-exec.o fe-misc.o fe-print.o fe-lobj.o \
dllist.o pqsignal.o
ifdef MULTIBYTE
-OBJS+= common.o wchar.o conv.o
+OBJS+= common.o wchar.o conv.o big5.o
endif
# If crypt is a separate library, rather than part of libc,
@@ -62,6 +62,9 @@ wchar.c: $(SRCDIR)/backend/utils/mb/wchar.c
conv.c: $(SRCDIR)/backend/utils/mb/conv.c
-$(LN_S) $(SRCDIR)/backend/utils/mb/conv.c .
+
+big5.c: $(SRCDIR)/backend/utils/mb/big5.c
+ -$(LN_S) $(SRCDIR)/backend/utils/mb/big5.c .
endif
# The following rules cause dependencies in the backend directory to
@@ -142,7 +145,7 @@ beforeinstall-headers:
.PHONY: clean
clean:
rm -f libpq.a $(shlib) $(OBJS)
- rm -f dllist.c common.c wchar.c conv.c
+ rm -f dllist.c common.c wchar.c conv.c big5.c
ifeq ($(PORTNAME), win)
rm -f pq.def
endif
diff --git a/src/test/mb/expected/big5.out b/src/test/mb/expected/big5.out
new file mode 100644
index 0000000000000000000000000000000000000000..0ca86167ece0fd667b31cd492b3fe7d3741df251
--- /dev/null
+++ b/src/test/mb/expected/big5.out
@@ -0,0 +1,84 @@
+QUERY: drop table �t�Ӹ��;
+QUERY: create table �t�Ӹ�� (��~�O text, ���q���Y varchar, �a�} varchar(16));
+QUERY: create index �t�Ӹ��index1 on �t�Ӹ�� using btree (��~�O);
+QUERY: create index �t�Ӹ��index2 on �t�Ӹ�� using hash (���q���Y);
+QUERY: insert into �t�Ӹ�� values ('�q���~', '�F�F���', '�_A01��');
+QUERY: insert into �t�Ӹ�� values ('�s�y�~', '�]���������q', '��B10��');
+QUERY: insert into �t�Ӹ�� values ('�\���~', '�����ѥ��������q', '��Z01�E');
+QUERY: vacuum �t�Ӹ��;
+QUERY: select * from �t�Ӹ��;
+��~�O|���q���Y |�a�}
+------+----------------+-------
+�q���~|�F�F��� |�_A01��
+�s�y�~|�]���������q |��B10��
+�\���~|�����ѥ��������q|��Z01�E
+(3 rows)
+
+QUERY: select * from �t�Ӹ�� where �a�} = '��Z01�E';
+��~�O|���q���Y |�a�}
+------+----------------+-------
+�\���~|�����ѥ��������q|��Z01�E
+(1 row)
+
+QUERY: select * from �t�Ӹ�� where �a�} ~* '��z01�E';
+��~�O|���q���Y |�a�}
+------+----------------+-------
+�\���~|�����ѥ��������q|��Z01�E
+(1 row)
+
+QUERY: select * from �t�Ӹ�� where �a�} like '_Z01_';
+��~�O|���q���Y |�a�}
+------+----------------+-------
+�\���~|�����ѥ��������q|��Z01�E
+(1 row)
+
+QUERY: select * from �t�Ӹ�� where �a�} like '_Z%';
+��~�O|���q���Y |�a�}
+------+----------------+-------
+�\���~|�����ѥ��������q|��Z01�E
+(1 row)
+
+QUERY: select * from �t�Ӹ�� where ���q���Y ~ '�F�F��[�H�O��]';
+��~�O|���q���Y|�a�}
+------+--------+-------
+�q���~|�F�F���|�_A01��
+(1 row)
+
+QUERY: select * from �t�Ӹ�� where ���q���Y ~* '�F�F��[�H�O��]';
+��~�O|���q���Y|�a�}
+------+--------+-------
+�q���~|�F�F���|�_A01��
+(1 row)
+
+QUERY: select *, character_length(��~�O) from �t�Ӹ��;
+��~�O|���q���Y |�a�} |length
+------+----------------+-------+------
+�q���~|�F�F��� |�_A01��| 3
+�s�y�~|�]���������q |��B10��| 3
+�\���~|�����ѥ��������q|��Z01�E| 3
+(3 rows)
+
+QUERY: select *, octet_length(��~�O) from �t�Ӹ��;
+��~�O|���q���Y |�a�} |octet_length
+------+----------------+-------+------------
+�q���~|�F�F��� |�_A01��| 6
+�s�y�~|�]���������q |��B10��| 6
+�\���~|�����ѥ��������q|��Z01�E| 6
+(3 rows)
+
+QUERY: select *, position('����' in ���q���Y) from �t�Ӹ��;
+��~�O|���q���Y |�a�} |strpos
+------+----------------+-------+------
+�q���~|�F�F��� |�_A01��| 0
+�s�y�~|�]���������q |��B10��| 3
+�\���~|�����ѥ��������q|��Z01�E| 5
+(3 rows)
+
+QUERY: select *, substring(���q���Y from 3 for 6 ) from �t�Ӹ��;
+��~�O|���q���Y |�a�} |substr
+------+----------------+-------+------------
+�q���~|�F�F��� |�_A01��|���
+�s�y�~|�]���������q |��B10��|�������q
+�\���~|�����ѥ��������q|��Z01�E|�ѥ��������q
+(3 rows)
+
diff --git a/src/test/mb/mbregress.sh b/src/test/mb/mbregress.sh
index 2ef2eb04124a1cbb6bac7f92c0b66dfd946526b0..9216d3c58be05a763d62ad64baa32ac7e67649e2 100644
--- a/src/test/mb/mbregress.sh
+++ b/src/test/mb/mbregress.sh
@@ -1,5 +1,5 @@
#! /bin/sh
-# $Header: /cvsroot/pgsql/src/test/mb/mbregress.sh,v 1.2 1998/07/26 04:31:38 scrappy Exp $
+# $Header: /cvsroot/pgsql/src/test/mb/mbregress.sh,v 1.3 1999/02/02 18:51:32 momjian Exp $
if echo '\c' | grep -s c >/dev/null 2>&1
then
@@ -15,7 +15,7 @@ if [ ! -d results ];then
fi
PSQL="psql -n -e -q"
-tests="euc_jp sjis euc_kr euc_cn unicode mule_internal"
+tests="euc_jp sjis euc_kr euc_cn euc_tw big5 unicode mule_internal"
unset PGCLIENTENCODING
for i in $tests
do
@@ -26,6 +26,11 @@ do
export PGCLIENTENCODING
$PSQL euc_jp < sql/sjis.sql > results/sjis.out 2>&1
unset PGCLIENTENCODING
+ elif [ $i = big5 ];then
+ PGCLIENTENCODING=BIG5
+ export PGCLIENTENCODING
+ $PSQL euc_tw < sql/big5.sql > results/big5.out 2>&1
+ unset PGCLIENTENCODING
else
destroydb $i >/dev/null 2>&1
createdb -E `echo $i|tr "[a-z]" "[A-Z]"` $i
diff --git a/src/test/mb/sql/big5.sql b/src/test/mb/sql/big5.sql
new file mode 100644
index 0000000000000000000000000000000000000000..4e2d10019962d0b902f1f6d97c51551afdd4a9cf
--- /dev/null
+++ b/src/test/mb/sql/big5.sql
@@ -0,0 +1,20 @@
+drop table �t�Ӹ��;
+create table �t�Ӹ�� (��~�O text, ���q���Y varchar, �a�} varchar(16));
+create index �t�Ӹ��index1 on �t�Ӹ�� using btree (��~�O);
+create index �t�Ӹ��index2 on �t�Ӹ�� using hash (���q���Y);
+insert into �t�Ӹ�� values ('�q���~', '�F�F���', '�_A01��');
+insert into �t�Ӹ�� values ('�s�y�~', '�]���������q', '��B10��');
+insert into �t�Ӹ�� values ('�\���~', '�����ѥ��������q', '��Z01�E');
+vacuum �t�Ӹ��;
+select * from �t�Ӹ��;
+select * from �t�Ӹ�� where �a�} = '��Z01�E';
+select * from �t�Ӹ�� where �a�} ~* '��z01�E';
+select * from �t�Ӹ�� where �a�} like '_Z01_';
+select * from �t�Ӹ�� where �a�} like '_Z%';
+select * from �t�Ӹ�� where ���q���Y ~ '�F�F��[�H�O��]';
+select * from �t�Ӹ�� where ���q���Y ~* '�F�F��[�H�O��]';
+
+select *, character_length(��~�O) from �t�Ӹ��;
+select *, octet_length(��~�O) from �t�Ӹ��;
+select *, position('����' in ���q���Y) from �t�Ӹ��;
+select *, substring(���q���Y from 3 for 6 ) from �t�Ӹ��;