It's probably well past time for this: change PuTTY's default

character set configuration to UTF-8, on both Windows and Unix, and
reorganise the dropdown lists in the Translation menu so that UTF-8
appears at the top (and Unix's odd "use font encoding" is relegated to
the bottom of the list like the special-purpose oddity it is).

git-svn-id: svn://svn.tartarus.org/sgt/putty@9843 cda61777-01e9-0310-a592-d414129be87e

diff --git a/charset/localenc.c b/charset/localenc.c
index 5eabeaf..a24126d 100644 (file)
--- a/charset/localenc.c
+++ b/charset/localenc.c
@@ -21,6 +21,7 @@ static const struct {
     int return_in_enum;   /* enumeration misses some charsets */
 } localencs[] = {
     { "<UNKNOWN>", CS_NONE, 0 },
+    { "UTF-8", CS_UTF8, 1 },
     { "ISO-8859-1", CS_ISO8859_1, 1 },
     { "ISO-8859-1 with X11 line drawing", CS_ISO8859_1_X11, 0 },
     { "ISO-8859-2", CS_ISO8859_2, 1 },
@@ -75,7 +76,6 @@ static const struct {
     { "VISCII", CS_VISCII, 1 },
     { "HP ROMAN8", CS_HP_ROMAN8, 1 },
     { "DEC MCS", CS_DEC_MCS, 1 },
-    { "UTF-8", CS_UTF8, 1 },
 };
 
 const char *charset_to_localenc(int charset)

diff --git a/unix/uxucs.c b/unix/uxucs.c
index 9d8242f..8cd300d 100644 (file)
--- a/unix/uxucs.c
+++ b/unix/uxucs.c
@@ -253,17 +253,19 @@ const char *cp_name(int codepage)
 const char *cp_enumerate(int index)
 {
     int charset;
-    if (index == 0)
-       return "Use font encoding";
-    charset = charset_localenc_nth(index-1);
-    if (charset == CS_NONE)
+    charset = charset_localenc_nth(index);
+    if (charset == CS_NONE) {
+        /* "Use font encoding" comes after all the named charsets */
+        if (charset_localenc_nth(index-1) != CS_NONE)
+            return "Use font encoding";
        return NULL;
+    }
     return charset_to_localenc(charset);
 }
 
 int decode_codepage(char *cp_name)
 {
     if (!*cp_name)
-       return CS_NONE;                /* use font encoding */
+       return CS_UTF8;
     return charset_from_localenc(cp_name);
 }

diff --git a/windows/winucs.c b/windows/winucs.c
index 2ce1693..cd50983 100644 (file)
--- a/windows/winucs.c
+++ b/windows/winucs.c
@@ -390,6 +390,8 @@ struct cp_list_item {
 };
 
 static const struct cp_list_item cp_list[] = {
+    {"UTF-8", CP_UTF8},
+
     {"ISO-8859-1:1998 (Latin-1, West Europe)", 0, 96, iso_8859_1},
     {"ISO-8859-2:1999 (Latin-2, East Europe)", 0, 96, iso_8859_2},
     {"ISO-8859-3:1999 (Latin-3, South Europe)", 0, 96, iso_8859_3},
@@ -406,8 +408,6 @@ static const struct cp_list_item cp_list[] = {
     {"ISO-8859-15:1999 (Latin-9, \"euro\")", 0, 96, iso_8859_15},
     {"ISO-8859-16:2001 (Latin-10, Balkan)", 0, 96, iso_8859_16},
 
-    {"UTF-8", CP_UTF8},
-
     {"KOI8-U", 0, 128, koi8_u},
     {"KOI8-R", 20866},
     {"HP-ROMAN8", 0, 96, roman8},
@@ -1016,48 +1016,8 @@ int decode_codepage(char *cp_name)
     int codepage = -1;
     CPINFO cpinfo;
 
-    if (!*cp_name) {
-       /*
-        * Here we select a plausible default code page based on
-        * the locale the user is in. We wish to select an ISO code
-        * page or appropriate local default _rather_ than go with
-        * the Win125* series, because it's more important to have
-        * CSI and friends enabled by default than the ghastly
-        * Windows extra quote characters, and because it's more
-        * likely the user is connecting to a remote server that
-        * does something Unixy or VMSy and hence standards-
-        * compliant than that they're connecting back to a Windows
-        * box using horrible nonstandard charsets.
-        * 
-        * Accordingly, Robert de Bath suggests a method for
-        * picking a default character set that runs as follows:
-        * first call GetACP to get the system's ANSI code page
-        * identifier, and translate as follows:
-        * 
-        * 1250 -> ISO 8859-2
-        * 1251 -> KOI8-U
-        * 1252 -> ISO 8859-1
-        * 1253 -> ISO 8859-7
-        * 1254 -> ISO 8859-9
-        * 1255 -> ISO 8859-8
-        * 1256 -> ISO 8859-6
-        * 1257 -> ISO 8859-13 (changed from 8859-4 on advice of a Lithuanian)
-        * 
-        * and for anything else, choose direct-to-font.
-        */
-       int cp = GetACP();
-       switch (cp) {
-         case 1250: cp_name = "ISO-8859-2"; break;
-         case 1251: cp_name = "KOI8-U"; break;
-         case 1252: cp_name = "ISO-8859-1"; break;
-         case 1253: cp_name = "ISO-8859-7"; break;
-         case 1254: cp_name = "ISO-8859-9"; break;
-         case 1255: cp_name = "ISO-8859-8"; break;
-         case 1256: cp_name = "ISO-8859-6"; break;
-         case 1257: cp_name = "ISO-8859-13"; break;
-           /* default: leave it blank, which will select -1, direct->font */
-       }
-    }
+    if (!*cp_name)
+        return CP_UTF8;                /* default */
 
     if (cp_name && *cp_name)
        for (cpi = cp_list; cpi->name; cpi++) {