Korean encoding patch for v0.59

1 files changed, 345 insertions, 0 deletions

diff --git a/net-im/gaim/files/gaim-0.59-korean.patch b/net-im/gaim/files/gaim-0.59-korean.patch
new file mode 100644
index 000000000000..2468e7386848
--- /dev/null
+++ b/net-im/gaim/files/gaim-0.59-korean.patch
@@ -0,0 +1,345 @@
+--- gaim-0.59.orig/src/aim.c	2002-05-31 14:13:02.000000000 +0900
++++ gaim-0.59/src/aim.c	2002-06-27 10:51:43.000000000 +0900
+@@ -644,6 +644,7 @@
+ 	}		
+ #endif
+ 
++	init_utf8_converter();
+ 
+ #ifdef USE_APPLET
+ 	init_applet_mgr(argc, argv);
+--- gaim-0.59.orig/src/gaim.h	2002-06-25 10:29:59.000000000 +0900
++++ gaim-0.59/src/gaim.h	2002-06-27 10:53:21.000000000 +0900
+@@ -432,8 +432,9 @@
+ extern gchar *strdup_withhtml(const gchar *);
+ extern void away_on_login(char *);
+ extern void system_log(enum log_event, struct gaim_connection *, struct buddy *, int);
+-extern unsigned char *utf8_to_str(unsigned char *);
+-extern char *str_to_utf8(unsigned char *);
++extern void init_utf8_converter();
++extern char *utf8_to_str(const char *);
++extern char *str_to_utf8(const char *);
+ extern char *add_cr(char *);
+ extern void strip_linefeed(char *);
+ extern time_t get_time(int, int, int, int, int, int);
+--- gaim-0.59.orig/src/util.c	2002-06-25 10:29:59.000000000 +0900
++++ gaim-0.59/src/util.c	2002-06-27 12:34:50.000000000 +0900
+@@ -33,6 +33,8 @@
+ #include <sys/wait.h>
+ #include <ctype.h>
+ #include <math.h>
++#include <langinfo.h>
++#include <iconv.h>
+ #include "gaim.h"
+ #include "prpl.h"
+ #include "gtkspell.h"
+@@ -1118,90 +1120,194 @@
+ 	fclose(fd);
+ }
+ 
+-unsigned char *utf8_to_str(unsigned char *in)
++/* Default system encoding */
++static char system_encoding[256] = "ASCII";
++
++/* Acquire and remember default character encoding for future conversion */
++void init_utf8_converter()
++{
++	char* tmp;
++	setlocale(LC_ALL, "");
++	tmp = nl_langinfo(CODESET);
++	if ( tmp != NULL ) {
++		strcpy(system_encoding, tmp);
++        }
++}
++
++char *utf8_to_str(const char *in)
+ {
+ 	int n = 0, i = 0;
+ 	int inlen;
+-	unsigned char *result;
++        char *result;
++        iconv_t cd1, cd2; /* cd1: UTF-8->UCS-4, cd2: UCS-4->system_encoding */
++	size_t outleft, ucs4_outleft;
++	size_t old_inlen, old_outleft;
++	size_t tmp;
++	char *result_ptr;
++	char *ucs4_result;
++	char *ucs4_result_ptr;
+ 
+ 	if (!in)
+ 		return NULL;
+ 
+-	inlen = strlen(in);
+-
+-	result = g_malloc(inlen + 1);
++	cd1 = iconv_open("UCS-4", "UTF-8");
++	cd2 = iconv_open(system_encoding, "UCS-4");
++	
++	if ( cd1 == (iconv_t)-1 || cd2 == (iconv_t)-1 ) {
++		if ( cd1 != (iconv_t)-1 ) {
++			iconv_close(cd1);
++		}
++		else if ( cd2 != (iconv_t)-1 ) {
++			iconv_close(cd2);
++		}
++		
++		/* failed to open converter. fall back to default conversion */
++		inlen = strlen(in);
++
++		result = g_malloc(inlen + 1);
++
++		while (n <= inlen - 1) {
++			long c = (long)in[n];
++			if (c < 0x80)
++				result[i++] = (char)c;
++			else {
++				if ((c & 0xC0) == 0xC0)
++					result[i++] =
++					(char)(((c & 0x03) << 6) | (((unsigned char)in[++n]) & 0x3F));
++				else if ((c & 0xE0) == 0xE0) {
++					if (n + 2 <= inlen) {
++						result[i] =
++						(char)(((c & 0xF) << 4) | (((unsigned char)in[++n]) & 0x3F));
++						result[i] =
++						(char)(((unsigned char)result[i]) |
++							(((unsigned char)in[++n])  & 0x3F));
++						i++;
++					} else
++						n += 2;
++				} else if ((c & 0xF0) == 0xF0)
++					n += 3;
++				else if ((c & 0xF8) == 0xF8)
++					n += 4;
++				else if ((c & 0xFC) == 0xFC)
++					n += 5;
++			}
++			n++;
+ 
+-	while (n <= inlen - 1) {
+-		long c = (long)in[n];
+-		if (c < 0x80)
+-			result[i++] = (char)c;
+-		else {
+-			if ((c & 0xC0) == 0xC0)
+-				result[i++] =
+-				    (char)(((c & 0x03) << 6) | (((unsigned char)in[++n]) & 0x3F));
+-			else if ((c & 0xE0) == 0xE0) {
+-				if (n + 2 <= inlen) {
+-					result[i] =
+-					    (char)(((c & 0xF) << 4) | (((unsigned char)in[++n]) & 0x3F));
+-					result[i] =
+-					    (char)(((unsigned char)result[i]) |
+-						   (((unsigned char)in[++n]) & 0x3F));
+-					i++;
+-				} else
+-					n += 2;
+-			} else if ((c & 0xF0) == 0xF0)
+-				n += 3;
+-			else if ((c & 0xF8) == 0xF8)
+-				n += 4;
+-			else if ((c & 0xFC) == 0xFC)
+-				n += 5;
+ 		}
+-		n++;
++		result[i] = '\0';
++		
++		return result;
+ 	}
+-	result[i] = '\0';
+ 
++	/* use iconv to convert UTF-8 to system encoding */
++	/* we will convert UTF-8 to UCS-4 and convert it again to system encoding
++	  one character by one because of the bug of iconv which doesn't convert
++	  all character when it encountered a character which doesn't exist
++	  in the target encoding charset. */
++
++	inlen = strlen(in);
++	if ( inlen <= 0 ) { /* no need to process an empty string. */
++		result = g_malloc(1);
++		result[0] = '\0';
++		return result;
++	}
++	outleft = inlen * 6; /* in maximum, one character will occupy 6 bytes in UTF-8 */
++	result = g_malloc(outleft + 1);
++	ucs4_outleft = inlen * 4;
++	ucs4_result = g_malloc(ucs4_outleft); /* ucs4 conversion buffer */
++	result_ptr = ucs4_result;
++	tmp = ucs4_outleft;
++	iconv(cd1, &in, &inlen, &result_ptr, &ucs4_outleft);
++	n = (tmp - ucs4_outleft) / 4; /* 'n' will indicate the number of chars */
++	debug_printf("ucs4_bytes: %d", (tmp-ucs4_outleft));
++
++	result_ptr = result;
++	ucs4_result_ptr = ucs4_result;
++	tmp = outleft; /* remember last buffer remainder */
++	for ( i = 0; i < n; i ++ ) {
++		inlen = 4;
++		iconv(cd2, &ucs4_result_ptr, &inlen, &result_ptr, &outleft);
++		if ( tmp == outleft ) {
++			/* if nothing has ben converted, add '?' symbol */
++			*(result_ptr++) = '?';
++			outleft --;
++			ucs4_result_ptr += 4;
++		}
++		
++		tmp = outleft; /* remember last buffer remainder */
++	}
++	iconv_close(cd2);
++	iconv_close(cd1);
++	
++	*result_ptr = '\0';
++	
++	g_free(ucs4_result);
++	
+ 	return result;
+ }
+ 
+-char *str_to_utf8(unsigned char *in)
++char *str_to_utf8(const char *in)
+ {
+ 	int n = 0, i = 0;
+ 	int inlen;
+ 	char *result = NULL;
++	iconv_t cd;
++	size_t outbytesleft;
++	char *result_ptr;
+ 
+ 	if (!in)
+ 		return NULL;
+ 
+-	inlen = strlen(in);
+-
+-	result = g_malloc(inlen * 2 + 1);
+-
+-	while (n < inlen) {
+-		long c = (long)in[n];
+-		if (c == 27) {
+-			n += 2;
+-			if (in[n] == 'x')
+-				n++;
+-			if (in[n] == '3')
++	cd = iconv_open("UTF-8", system_encoding);
++	if ( cd == (iconv_t)-1 ) {
++		/* failed to open converter. fall back to default conversion */
++		inlen = strlen(in);
++		result = g_malloc(inlen * 2 + 1);
++		
++		while (n < inlen) {
++			long c = (long)in[n];
++			if (c == 27) {
++				n += 2;
++				if (in[n] == 'x')
++					n++;
++				if (in[n] == '3')
++					n++;
++				n += 2;
++				continue;
++			}
++			/* why are we removing newlines and carriage returns?
++			if ((c == 0x0D) || (c == 0x0A)) {
+ 				n++;
+-			n += 2;
+-			continue;
+-		}
+-		/* why are we removing newlines and carriage returns?
+-		if ((c == 0x0D) || (c == 0x0A)) {
++				continue;
++			}
++			*/
++			if (c < 128)
++				result[i++] = (char)c;
++			else {
++				result[i++] = (char)((c >> 6) | 192);
++				result[i++] = (char)((c & 63) | 128);
++			}
+ 			n++;
+-			continue;
+-		}
+-		*/
+-		if (c < 128)
+-			result[i++] = (char)c;
+-		else {
+-			result[i++] = (char)((c >> 6) | 192);
+-			result[i++] = (char)((c & 63) | 128);
+ 		}
+-		n++;
++		result[i] = '\0';
++		
++		return result;
+ 	}
+-	result[i] = '\0';
++
++	/* use iconv to convert UTF-8 to system encoding */
++	inlen = strlen(in);
++	if ( inlen <= 0 ) { /* No need to process an empty string. */
++		result = g_malloc(1);
++		result[0] = '\0';
++		return result;
++	}
++	outbytesleft = inlen * 6; /* in maximum, one character will occupy 6 bytes in UTF-8 */
++	result = g_malloc(outbytesleft + 1);
++	result_ptr = result;
++	/* convert message */
++	iconv(cd, &in, &inlen, &result_ptr, &outbytesleft);
++	iconv_close(cd);
++	*result_ptr = '\0';
+ 
+ 	return result;
+ }
+--- gaim-0.59.orig/src/protocols/msn/msn.c	2002-06-25 10:29:59.000000000 +0900
++++ gaim-0.59/src/protocols/msn/msn.c	2002-06-27 12:46:41.000000000 +0900
+@@ -179,6 +179,7 @@
+ {
+ 	static char buf[MSN_BUF_LEN];
+ 	int i, j = 0;
++	char* result;
+ 
+ 	bzero(buf, sizeof(buf));
+ 	for (i = 0; i < strlen(msg); i++) {
+@@ -195,6 +196,10 @@
+ 	}
+ 	buf[j] = 0;
+ 
++	result = utf8_to_str(buf);
++	strcpy(buf, result);
++	g_free(result);
++
+ 	return buf;
+ }
+ 
+@@ -202,18 +207,23 @@
+ {
+ 	static char buf[MSN_BUF_LEN];
+ 	int i, j = 0;
++	char* tmp;
++
++	tmp = str_to_utf8(msg);
+ 
+ 	bzero(buf, sizeof(buf));
+-	for (i = 0; i < strlen(msg); i++) {
+-		if (isalnum(msg[i]))
+-			buf[j++] = msg[i];
++	for (i = 0; i < strlen(tmp); i++) {
++		if (isalnum(tmp[i]))
++			buf[j++] = tmp[i];
+ 		else {
+-			sprintf(buf + j, "%%%02x", (unsigned char)msg[i]);
++			sprintf(buf + j, "%%%02x", (unsigned char)tmp[i]);
+ 			j += 3;
+ 		}
+ 	}
+ 	buf[j] = 0;
+ 
++	g_free(tmp);
++
+ 	return buf;
+ }
+ 
+@@ -343,6 +353,7 @@
+ 			g_snprintf(msg, sizeof(msg), _("User unverified"));
+ 			break;
+ 		default:
++			return msg; /* Do not popup Unknown Error Code. */
+ 			g_snprintf(msg, sizeof(msg), _("Unknown Error Code"));
+ 			break;
+ 	}