MSN UTF-8 support correctly

Submitted by:	Cheng-Lung Sung <clsung@dragon2.net>

3 files changed, 218 insertions, 199 deletions

diff --git a/net-im/centericq/files/patch-msn_commands.cc b/net-im/centericq/files/patch-msn_commands.cc
deleted file mode 100644
index 00838c3..0000000
--- a/net-im/centericq/files/patch-msn_commands.cc
+++ /dev/null
@@ -1,199 +0,0 @@
---- libmsn-0.1/msn_commands.cc.orig	Wed Oct 16 18:20:59 2002
-+++ libmsn-0.1/msn_commands.cc	Wed Oct 30 16:44:55 2002
-@@ -30,7 +30,10 @@
- #include <sys/socket.h>
- #include <errno.h>
- #include <netdb.h>
--
-+#define HAVE_ICONV_H 1
-+#if HAVE_ICONV_H
-+#include <iconv.h>
-+#endif
- #include "libmsn.h"
- #include "msn_commands.h"
- #include "parse_utils.h"
-@@ -426,6 +429,128 @@
-     return 0;
- }
- 
-+#if HAVE_ICONV_H
-+/* 
-+** Name:    safe_iconv
-+** Purpose: 'Fault-tolerant' version if iconv. Replaces invalid seq with '?'
-+** Input:   see iconv manpage
-+*/
-+static int safe_iconv( iconv_t handle,
-+	     const char **inbuf, size_t *inbytesleft,
-+	     char **outbuf, size_t *outbytesleft)
-+{
-+    int ret;
-+    while (*inbytesleft) {
-+	 ret = iconv( handle, inbuf, inbytesleft,
-+		outbuf, outbytesleft);
-+	if (!*inbytesleft || !*outbytesleft)
-+	    return ret;
-+	/*got invalid seq - so replace it with '?' */
-+	**outbuf = '?'; (*outbuf)++; (*outbytesleft)--;
-+	(*inbuf)++; (*inbytesleft)--;
-+    }
-+    return ret;
-+}
-+
-+/* charset name cache buffer */
-+char loc_charset[32];
-+
-+#define DEFAULT_CHARSET "ISO-8859-1"
-+
-+/*
-+** Name:    guess_current_locale_charset
-+** Purpose: Try to guess default charset for the current locale
-+** Output:  charset name
-+** FIXME:   is there more right method for guessing charset
-+	    than scanning $LANG ?
-+*/
-+static char* guess_current_locale_charset()
-+{
-+    char *lang, *ch;
-+    /* Return previously learned charset */
-+    return "big5";
-+    if (loc_charset[0])
-+	return loc_charset;
-+
-+    lang = getenv("LANG");
-+    ch = strrchr( lang, '.' ) + 1;
-+    if (!ch)
-+	strcpy( loc_charset, DEFAULT_CHARSET );
-+    else {
-+	iconv_t pt;
-+	strncpy( loc_charset, ch, sizeof(loc_charset) );
-+	/* try to open iconv handle using guessed charset */
-+	if ( (pt = iconv_open( loc_charset, loc_charset )) == (iconv_t)(-1) )
-+	{
-+	    strcpy( loc_charset, DEFAULT_CHARSET );
-+	} else {
-+	    iconv_close(pt);
-+	};
-+	
-+    }
-+
-+    return loc_charset;
-+}
-+#endif /* HAVE_ICONV_H */
-+
-+/*
-+** Name:    Str2Utf8
-+** Purpose: convert a string in UTF-8 format
-+** Input:   inbuf     - the string to convert
-+** Output:  a new string in UTF-8 format
-+*/
-+static char *StrToUtf8( const char *inbuf )
-+{
-+#if HAVE_ICONV_H
-+    size_t length = strlen( inbuf );
-+    size_t outmaxlength = length * 4; /* FIXME: Is x4 multiplier enoght? */
-+    char *outbuf = (char*) malloc( outmaxlength + 1 );
-+    char *outbuf_save = outbuf;
-+    int ret;
-+
-+    iconv_t handle = iconv_open( "utf-8", guess_current_locale_charset() );
-+    ret = safe_iconv( handle, (const char **) &inbuf, &length, &outbuf, &outmaxlength );
-+    
-+    iconv_close( handle );
-+
-+    return outbuf_save;
-+#else
-+    char *outbuf = strdup( inbuf );
-+    char *outbuf_save = outbuf;
-+    /* Clear eight bit */
-+    for (; *outbuf; ++outbuf)
-+        *outbuf &= 0x7F;
-+    return outbuf_save;
-+#endif
-+}
-+ 
-+/*
-+** Name:    Utf8ToStr
-+** Purpose: revert UTF-8 string conversion
-+** Input:   inbuf  - the string to decode
-+** Output:  a new decoded string
-+*/
-+static char *Utf8ToStr( char *inbuf )
-+{
-+#if HAVE_ICONV_H
-+    size_t length = strlen( inbuf );
-+    size_t outmaxlength = length;
-+    char *outbuf = (char*) malloc( outmaxlength + 1 );
-+    char *outbuf_save = outbuf;
-+    int ret;
-+
-+    iconv_t handle = iconv_open( guess_current_locale_charset(), "utf-8" );
-+
-+    ret = safe_iconv( handle, (const char **) &inbuf, &length, &outbuf, &outmaxlength );
-+    
-+    iconv_close( handle );
-+
-+    return outbuf_save;
-+#else
-+    return strdup( inbuf );
-+#endif
-+}
-+
- /*
- ** Name:    HandleMessage
- ** Purpose: This function handles an instant message from either the server
-@@ -440,6 +565,7 @@
- {
-     MSN_InstantMessage newIm;
-     char *message;
-+    char *decodedIm;
-     char *mimeInfo, *im;
-     int  length, nread; 
-     struct timeval t;
-@@ -448,6 +574,8 @@
-     message = NULL;
-     mimeInfo = NULL;
-     im = NULL; 
-+    decodedIm = NULL;
-+
- 	
-     if (numOfArgs != 4)
- 	return -1;
-@@ -470,13 +598,15 @@
-  
-     if (mimeInfo != NULL) {
- 	if (strstr(mimeInfo, "text/plain") != NULL) {
-+
-+	    decodedIm = Utf8ToStr( im );
- 	    newIm.year = 0;
- 	    newIm.month = 0;
- 	    newIm.day = 0;
- 	    newIm.hour = 0;
- 	    newIm.minute = 0;
- 	    newIm.sec = 0;
--	    newIm.msg = im;
-+	    newIm.msg = decodedIm;
- 	    RemoveHotmail(args[1], &newIm.sender);
- 	    newIm.friendlyhandle = args[2];
- 	    newIm.fd = conn->fd;
-@@ -571,15 +701,18 @@
- int SendMessage(MSN_Conn *conn, const char *message)
- {
-     char   *commandLine;
-+    char   *UTFmessage;
-     int    length;
- 
-     if (!message)
- 	return -1;
-+    UTFmessage = NULL;
- 
--    commandLine = (char *)malloc(strlen(MIME_HEADER)+strlen(message)+25);
-+    UTFmessage = StrToUtf8( message );
-+    commandLine = (char *)malloc(strlen(MIME_HEADER)+strlen(UTFmessage)+25);
-     length = sprintf(commandLine, "%s %lu N %d\r\n%s%s", CommandString[MSG],
--		     TrID++, strlen(message)+strlen(MIME_HEADER), MIME_HEADER, 
--		     message);
-+		     TrID++, strlen(UTFmessage)+strlen(MIME_HEADER), MIME_HEADER, 
-+		     UTFmessage);
- 
-     write(conn->fd, commandLine, length);
-     free(commandLine);
diff --git a/net-im/centericq/files/patch-msnhook.cc b/net-im/centericq/files/patch-msnhook.cc
new file mode 100644
index 0000000..12ceb55
--- /dev/null
+++ b/net-im/centericq/files/patch-msnhook.cc
@@ -0,0 +1,159 @@
+--- src/hooks/msnhook.cc	Tue Nov 19 10:18:22 2002
++++ src/hooks/msnhook.cc	Tue Nov 19 10:18:35 2002
+@@ -30,6 +30,7 @@
+ #include "eventmanager.h"
+ #include "centericq.h"
+ #include "imlogger.h"
++#include "utf8conv.h"
+ 
+ msnhook mhook;
+ 
+@@ -159,7 +160,8 @@
+     }
+ 
+     icqcontact *c = clist.get(ev.getcontact());
+-    text = siconv(text, conf.getrussian() ? "koi8-u" : DEFAULT_CHARSET, "utf8");
++    text = StrToUtf8(text);
++//    text = siconv(text, conf.getrussian() ? "koi8-u" : DEFAULT_CHARSET, "utf8");
+ //    text = toutf8(rusconv("kw", text));
+ 
+     if(c)
+@@ -274,8 +276,8 @@
+ 	m.homepage = "http://members.msn.com/" + b.email;
+ 
+ 	if(!friendlynicks[ic.nickname].empty()) {
+-	    c->setdispnick(friendlynicks[ic.nickname]);
+-	    b.fname = friendlynicks[ic.nickname];
++	    c->setdispnick(Utf8ToStr(friendlynicks[ic.nickname]));
++	    b.fname = Utf8ToStr(friendlynicks[ic.nickname]);
+ 	}
+ 
+ 	c->setbasicinfo(b);
+@@ -371,7 +373,8 @@
+ 	}
+ 
+ //        text = rusconv("wk", fromutf8(d->msg));
+-	text = siconv(d->msg, "utf8", conf.getrussian() ? "koi8-u" : DEFAULT_CHARSET);
++//	text = siconv(d->msg, "utf8", conf.getrussian() ? "koi8-u" : DEFAULT_CHARSET);
++	text = Utf8ToStr(d->msg);
+ 	em.store(immessage(ic, imevent::incoming, text));
+ 
+ 	if(c)
+@@ -453,3 +456,117 @@
+ 	clist.get(contactroot)->playsound(imevent::email);
+     }
+ }
++#if HAVE_ICONV_H
++int safe_iconv( iconv_t handle, const char **inbuf, size_t *inbytesleft,
++	     char **outbuf, size_t *outbytesleft)
++{
++    int ret;
++    while (*inbytesleft) {
++	 ret = iconv( handle, inbuf, inbytesleft,
++		outbuf, outbytesleft);
++	if (!*inbytesleft || !*outbytesleft)
++	    return ret;
++	/*got invalid seq - so replace it with '?' */
++	**outbuf = '?'; (*outbuf)++; (*outbytesleft)--;
++	(*inbuf)++; (*inbytesleft)--;
++    }
++    return ret;
++}
++
++const char* guess_current_locale_charset()
++{
++    char *lang, *ch;
++    /* Return previously learned charset */
++    if (loc_charset[0])
++	return loc_charset;
++
++    lang = getenv("LANG");
++    ch = strrchr( lang, '.' ) + 1;
++    if (!ch)
++	strcpy( loc_charset, DEFAULT_CHARSET );
++    else {
++	iconv_t pt;
++	strncpy( loc_charset, ch, sizeof(loc_charset) );
++	/* try to open iconv handle using guessed charset */
++	if ( (pt = iconv_open( loc_charset, loc_charset )) == (iconv_t)(-1) )
++	{
++	    strcpy( loc_charset, DEFAULT_CHARSET );
++	} else {
++	    iconv_close(pt);
++	};
++	
++    }
++
++    return loc_charset;
++}
++
++char *StrToUtf8( const char *inbuf )
++{
++    size_t length = strlen( inbuf );
++    size_t outmaxlength = length * 4; /* FIXME: Is x4 multiplier enoght? */
++    char *outbuf = (char*) malloc( outmaxlength + 1 );
++    char *outbuf_save = outbuf;
++    int ret;
++
++    iconv_t handle = iconv_open( "utf-8", guess_current_locale_charset() );
++    ret = safe_iconv( handle, (const char **) &inbuf, &length, &outbuf, &outmaxlength );
++    
++    iconv_close( handle );
++
++    return outbuf_save;
++}
++ 
++std::string StrToUtf8( const std::string &instr )
++{
++    size_t length = instr.length();
++    size_t outmaxlength = length * 4; /* FIXME: Is x4 multiplier enoght? */
++    const char *inbuf = instr.c_str();
++    char *outbuf = (char*) malloc( outmaxlength + 1 );
++    char *outbuf_save = outbuf;
++    int ret;
++
++    iconv_t handle = iconv_open( "utf-8", guess_current_locale_charset() );
++    ret = safe_iconv( handle, (const char **) &inbuf, &length, &outbuf, &outmaxlength );
++    
++    iconv_close( handle );
++
++    std::string return_me = outbuf_save;
++    return return_me;
++}
++ 
++char *Utf8ToStr( const char *inbuf )
++{
++    size_t length = strlen( inbuf );
++    size_t outmaxlength = length;
++    char *outbuf = (char*) malloc( outmaxlength + 1 );
++    char *outbuf_save = outbuf;
++    int ret;
++
++    iconv_t handle = iconv_open( guess_current_locale_charset(), "utf-8" );
++
++    ret = safe_iconv( handle, (const char **) &inbuf, &length, &outbuf, &outmaxlength );
++    
++    iconv_close( handle );
++
++    return outbuf_save;
++}
++
++std::string Utf8ToStr( const std::string &instr )
++{
++    size_t length = instr.length();
++    size_t outmaxlength = length;
++    const char *inbuf = instr.c_str();
++    char *outbuf = (char*) malloc( outmaxlength + 1 );
++    char *outbuf_save = outbuf;
++    int ret;
++
++    iconv_t handle = iconv_open( guess_current_locale_charset(), "utf-8" );
++
++    ret = safe_iconv( handle, (const char **) &inbuf, &length, &outbuf, &outmaxlength );
++
++    iconv_close( handle );
++
++    std::string return_me = outbuf_save;
++    return return_me;
++}
++#endif /* HAVE_ICONV_H */
diff --git a/net-im/centericq/files/patch-utf8conv.h b/net-im/centericq/files/patch-utf8conv.h
new file mode 100644
index 0000000..ece334f
--- /dev/null
+++ b/net-im/centericq/files/patch-utf8conv.h
@@ -0,0 +1,59 @@
+--- src/hooks/utf8conv.h	Thu Jan  1 08:00:00 1970
++++ src/hooks/utf8conv.h	Tue Nov 19 15:19:44 2002
+@@ -0,0 +1,56 @@
++/*-------------------------------------------------------*/
++/* utf8conv.h     ( Utf8 Converter )                     */
++/*-------------------------------------------------------*/
++/* target : Converting Utf8 from/to string/char          */
++/* modifier : clsung@dragon2.net                         */
++/* create : unknown                                      */
++/* update : 02/11/19                                     */
++/*-------------------------------------------------------*/
++#ifndef _UTF8CONV_
++#define _UTF8CONV_
++
++#ifndef HAVE_ICONV_H
++#define HAVE_ICONV_H
++#endif
++
++#ifdef HAVE_ICONV_H
++#include <iconv.h>
++#endif
++#include <string>
++#define DEFAULT_CHARSET "ISO-8859-1"
++/* charset name cache buffer */
++static char loc_charset[32];
++/* 
++** Name:    safe_iconv
++** Purpose: 'Fault-tolerant' version if iconv. Replaces invalid seq with '?'
++** Input:   see iconv manpage
++*/
++int safe_iconv( iconv_t handle, const char **inbuf, size_t *inbytesleft,
++	     char **outbuf, size_t *outbytesleft);
++
++/*
++** Name:    guess_current_locale_charset
++** Purpose: Try to guess default charset for the current locale
++** Output:  charset name
++** FIXME:   is there more right method for guessing charset
++	    than scanning $LANG ?
++*/
++const char* guess_current_locale_charset();
++
++/*
++** Name:    Str2Utf8
++** Purpose: convert a string in UTF-8 format
++** Input:   inbuf     - the string to convert
++** Output:  a new string in UTF-8 format
++*/
++char *StrToUtf8( const char *inbuf );
++std::string StrToUtf8( const std::string &instr );
++/*
++** Name:    Utf8ToStr
++** Purpose: revert UTF-8 string conversion
++** Input:   inbuf  - the string to decode
++** Output:  a new decoded string
++*/
++char *Utf8ToStr( const char *inbuf );
++std::string Utf8ToStr( const std::string &instr );
++#endif