Add UTF-8 charset

NEWS

@@ -38,6 +38,8 @@
       was just wrong, so you might notice bad signature in some very
       big files.  It may be wise to keep an old copy of GnuPG around.
 
+    * New --charset=utf-8 to bypass all internal translations.
+
 
 Noteworthy changes in version 1.0.4 (2000-10-17)
 ------------------------------------------------

VERSION

@@ -1 +1 @@
-1.0.4h
+1.0.4i

doc/ChangeLog

@@ -1,3 +1,7 @@
+2001-04-20  Werner Koch  <wk@gnupg.org>
+
+	* gpg.sgml: Add new --charset UTF-8.
+
 2001-04-19  Werner Koch  <wk@gnupg.org>
 
 	* faq.raw: Add a note about dates displayed as ????-??-??.

doc/gpg.sgml

@@ -367,7 +367,8 @@ assigned owner trust and the second is the calculated
 trust value.  Letters are used for the values:</para>
     <variablelist>
       <varlistentry><term>-</term><listitem><para>No ownertrust assigned / not yet calculated.</para></listitem></varlistentry>
-      <varlistentry><term>e</term><listitem><para>Trust calculation has failed.</para></listitem></varlistentry>
+      <varlistentry><term>e</term><listitem><para>Trust
+calculation has failed; probably due to an expired key.</para></listitem></varlistentry>
       <varlistentry><term>q</term><listitem><para>Not enough information for calculation.</para></listitem></varlistentry>
       <varlistentry><term>n</term><listitem><para>Never trust this key.</para></listitem></varlistentry>
       <varlistentry><term>m</term><listitem><para>Marginally trusted.</para></listitem></varlistentry>
@@ -846,6 +847,10 @@ Valid values for &ParmName; are:</para>
 <varlistentry>
 <term>koi8-r</term><listitem><para>The usual Russian set (rfc1489).</para></listitem>
 </varlistentry>
+<varlistentry>
+<term>utf-8</term><listitem><para>Bypass all translations and assume
+that the OS uses native UTF-8 encoding.</para></listitem>
+</varlistentry>
 </variablelist>
 </listitem></varlistentry>
 

g10/ChangeLog

@@ -1,6 +1,10 @@
+2001-04-20  Werner Koch  <wk@gnupg.org>
+
+	* options.skel: Add some more comments.
+
 2001-04-19  Werner Koch  <wk@gnupg.org>
 
-	* keyid.c (mk_datestr): New.  Handles negative times.  we must do
+	* keyid.c (mk_datestr): New.  Handles negative times.  We must do
 	this because Windoze segvs on negative times passed to gmtime().
 	Changed all datestr_from function to use this one.
 

g10/options.skel

@@ -31,6 +31,18 @@ $Id$
 
 #default-key 621CC013
 
+# GnuPG ultimately trusts all keys in the secret keyring.  If you do
+# not have all your secret keys online available you should use this
+# option to tell GnuPG about ultimately trusted keys.
+# You have to give the long keyID here which can be obtained by using
+# the --list-key command along with the option --with-colons; you will
+# get a line similiar to this one:
+#    pub:u:1024:17:5DE249965B0358A2:1999-03-15:2006-02-04:59:f:
+# the 5th field is what you want.
+
+#trusted-key 12345678ABCDEF01
+
+
 # If you do not pass a recipient to gpg, it will ask for one.
 # Using this option you can encrypt to a default key.  key validation
 # will not be done in this case.
@@ -56,7 +68,7 @@ escape-from-lines
 # If you do not use the Latin-1 (ISO-8859-1) charset, you should
 # tell GnuPG which is the native character set.  Please check
 # the man page for supported character sets. 
-#charset koi8-r
+#charset utf-8
 
 
 # You may define aliases like this:
@@ -92,4 +104,3 @@ lock-once
 # this option is set.
 
 honor-http-proxy
-

util/ChangeLog

@@ -1,3 +1,10 @@
+2001-04-20  Werner Koch  <wk@gnupg.org>
+
+	* strgutil.c (set_native_charset): Allow utf-8 by introducing the
+	new no_translation variable.
+	(native_to_utf8): Handle no_translation.
+	(utf8_to_native): Ditto.
+
 2001-04-19  Werner Koch  <wk@gnupg.org>
 
 	* miscutil.c (asctimestamp): Handle negative times.  We must do

util/strgutil.c

@@ -67,7 +67,7 @@ static ushort latin2_unicode[128] = {
 
 static const char *active_charset_name = "iso-8859-1";
 static ushort *active_charset = NULL;
-
+static int no_translation = 0;
 
 void
 free_strlist( STRLIST sl )
@@ -327,16 +327,24 @@ set_native_charset( const char *newset )
 {
     if( !stricmp( newset, "iso-8859-1" ) ) {
 	active_charset_name = "iso-8859-1";
+        no_translation = 0;
 	active_charset = NULL;
     }
     else if( !stricmp( newset, "iso-8859-2" ) ) {
 	active_charset_name = "iso-8859-2";
+        no_translation = 0;
 	active_charset = latin2_unicode;
     }
     else if( !stricmp( newset, "koi8-r" ) ) {
 	active_charset_name = "koi8-r";
+        no_translation = 0;
 	active_charset = koi8_unicode;
     }
+    else if( !stricmp (newset, "utf8" ) || !stricmp(newset, "utf-8") ) {
+	active_charset_name = "utf-8";
+        no_translation = 1;
+	active_charset = NULL;
+    }
     else
 	return G10ERR_GENERAL;
     return 0;
@@ -360,7 +368,10 @@ native_to_utf8( const char *string )
     byte *p;
     size_t length=0;
 
-    if( active_charset ) {
+    if (no_translation) {
+        buffer = m_strdup (string);
+    }
+    else if( active_charset ) {
 	for(s=string; *s; s++ ) {
 	    length++;
 	    if( *s & 0x80 )
@@ -424,6 +435,13 @@ utf8_to_native( const char *string, size_t length )
     size_t slen;
     int resync = 0;
 
+    if (no_translation) {
+        buffer = m_alloc (length+1);
+        memcpy (buffer, string, length);
+        buffer[length] = 0; /* make sure it is a string */
+        return buffer;
+    }
+
     /* 1. pass (p==NULL): count the extended utf-8 characters */
     /* 2. pass (p!=NULL): create string */
     for( ;; ) {