String literals for UTF-8 language names using octal escapes only...

... Leaving absolutely no doubt about the portability.

New Perl script in locale emits the required C++ source when LanguageNames.txt
is updated.

locale/CeePlusPlusifyLanguageNames.pl

@@ -0,0 +1,23 @@
+#!/usr/bin/perl
+# Take LangugeNames.txt and write C++ string literals to standard out,
+# which describe the UTF-8 byte sequence using only the basic coding
+# character set.
+# Copy and paste that over the table of strings in src/Languages.cpp.
+
+open LANGUAGES, "LanguageNames.txt" or die "Can't open LanguageNames.txt: $!\n";
+while( <LANGUAGES> ) {
+    chop;
+    print "\"";
+    @codes = unpack 'C*';
+    @newCodes = ();
+    for $code (@codes) {
+	if ($code <= ord 'z') {
+	    push @newCodes, $code;
+	}
+	else {
+	    push @newCodes, unpack('C*', sprintf("\\%03o", $code));
+	}
+    }
+    print pack 'C*', @newCodes;
+    print "\",\n";
+}

locale/LanguageNames.txt

@@ -1,59 +1,56 @@
-// To edit this file, open as UTF-8 encoded Unicode
-// To use the updated file, open as 8-bit ASCII, copy the strings, and paste
-// over the table of names in src/Languages.cpp
-"af Afrikaans",
-"ar العربية",
-"be Беларуская",
-"bg Български",
-"bn বাংলা",
-"bs Bosanski",
-"ca Català",
-"ca_ES@valencia Valencià",
-"cs Čeština",
-"cy Cymraeg",
-"da Dansk",
-"de Deutsch",
-"el Ελληνικά",
-"en English",
-"es Español",
-"eu Euskara",
-"eu_ES Euskara (Espainiako)",
-"fa فارسی",
-"fi Suomi",
-"fr Français",
-"ga Gaeilge",
-"gl Galego",
-"he עברית",
-"hi हिन्दी",
-"hr Hrvatski",
-"hu Magyar",
-"hy Հայերեն",
-"id Bahasa Indonesia",
-"it Italiano",
-"ja 日本語",
-"ka ქართული",
-"km ខេមរភាសា",
-"ko 한국어",
-"lt Lietuvių",
-"mk Македонски",
-"my မြန်မာစာ",
-"nb Norsk",
-"nl Nederlands",
-"oc Occitan",
-"pl Polski",
-"pt Português",
-"pt_BR Português (Brasil)",
-"ro Română",
-"ru Русский",
-"sk Slovenčina",
-"sl Slovenščina",
-"sr_RS Српски",
-"sr_RS@latin Srpski",
-"sv Svenska",
-"ta தமிழ்",
-"tg Тоҷикӣ",
-"tr Türkçe",
-"uk Українська",
-"vi Tiếng Việt",
-"zh_CN 中文",
-"zh_TW 中文",
+af Afrikaans
+ar العربية
+be Беларуская
+bg Български
+bn বাংলা
+bs Bosanski
+ca Català
+ca_ES@valencia Valencià
+cs Čeština
+cy Cymraeg
+da Dansk
+de Deutsch
+el Ελληνικά
+en English
+es Español
+eu Euskara
+eu_ES Euskara (Espainiako)
+fa فارسی
+fi Suomi
+fr Français
+ga Gaeilge
+gl Galego
+he עברית
+hi हिन्दी
+hr Hrvatski
+hu Magyar
+hy Հայերեն
+id Bahasa Indonesia
+it Italiano
+ja 日本語
+ka ქართული
+km ខេមរភាសា
+ko 한국어
+lt Lietuvių
+mk Македонски
+my မြန်မာစာ
+nb Norsk
+nl Nederlands
+oc Occitan
+pl Polski
+pt Português
+pt_BR Português (Brasil)
+ro Română
+ru Русский
+sk Slovenčina
+sl Slovenščina
+sr_RS Српски
+sr_RS@latin Srpski
+sv Svenska
+ta தமிழ்
+tg Тоҷикӣ
+tr Türkçe
+uk Українська
+vi Tiếng Việt
+zh_CN 中文
+zh_TW 中文