Windows: Use UTF-8 locale when active code page is UTF-8

XZ Utils 5.6.3 set the active code page to UTF-8 to fix CVE-2024-47611.
This wasn't paired with UCRT-specific setlocale(LC_ALL, ".UTF8"), thus
non-ASCII characters from translations became mojibake.

Fixes: 46ee0061629fb075d61d83839e14dd193337af59
(cherry picked from commit 0d0b574cc45045d6150d397776340c068df59e2a)

diff --git a/src/common/tuklib_gettext.h b/src/common/tuklib_gettext.h
index 3ef5cb7292b59b57756d4abfc7dcec4fb808b005..4021c98fa9becf978c60190eae47067c19fe7040 100644 (file)
--- a/src/common/tuklib_gettext.h
+++ b/src/common/tuklib_gettext.h
@@ -15,6 +15,34 @@
 #include "tuklib_common.h"
 #include <locale.h>
 
+#if defined(_WIN32) && !defined(__CYGWIN__)
+#      define WIN32_LEAN_AND_MEAN
+#      include <windows.h>
+       // To use UTF-8 code page on Windows 10 version 1903 and later, the
+       // active code page has to be set to UTF-8 in the application manifest
+       // and UCRT-specific setlocale(LC_ALL, ".UTF8") must be called. The
+       // manifest makes argv[] use UTF-8 (which setlocale() cannot affect)
+       // and the special setlocale() call makes mbrtowc() and such functions
+       // use UTF-8. (It's weird why regular setlocale(LC_ALL, "") doesn't
+       // use the code page from the application manifest.)
+       //
+       // The two things have quite a bit of overlap though. For example,
+       // both affect the code page used in the file system APIs. Thus,
+       // if argv[] isn't in UTF-8, using setlocale() to set UTF-8 will
+       // break non-ASCII filenames that have been passed as command line
+       // arguments. Thus, it's best to set an UTF-8 locale only when
+       // the active code page is UTF-8.
+       //
+       // NOTE: Only UCRT supports the UTF-8 locale string, thus this
+       // will fail with MSVCRT if the active code page is UTF-8. That
+       // shouldn't be too bad because UTF-8 doesn't work properly with
+       // MSVCRT anyway.
+#      define tuklib_gettext_setlocale() \
+               setlocale(LC_ALL, GetACP() == CP_UTF8 ? ".UTF8" : "")
+#else
+#      define tuklib_gettext_setlocale() setlocale(LC_ALL, "")
+#endif
+
 #ifndef TUKLIB_GETTEXT
 #      ifdef ENABLE_NLS
 #              define TUKLIB_GETTEXT 1
@@ -27,14 +55,14 @@
 #      include <libintl.h>
 #      define tuklib_gettext_init(package, localedir) \
                do { \
-                       setlocale(LC_ALL, ""); \
+                       tuklib_gettext_setlocale(); \
                        bindtextdomain(package, localedir); \
                        textdomain(package); \
                } while (0)
 #      define _(msgid) gettext(msgid)
 #else
 #      define tuklib_gettext_init(package, localedir) \
-               setlocale(LC_ALL, "")
+               tuklib_gettext_setlocale()
 #      define _(msgid) (msgid)
 #      define ngettext(msgid1, msgid2, n) ((n) == 1 ? (msgid1) : (msgid2))
 #endif