* cygheap.cc (cygheap_init): Default locale.charset to "UTF-8".

* dcrt0.cc (dll_crt0_1): Call setlocale (LC_CTYPE, "") here, just
	before resetting application locale to "C".
	* environ.cc (environ_init): Drop code setting locale here.
	* syscalls.cc (check_codepage): Break out check for required codepage
	from internal_setlocale.
	(internal_setlocale): Just convert CWD and $PATH from old charset to
	new charset and set internal charset.
	(setlocale): Change accordingly.  Don't require LC_ALL or LC_CTYPE to
	store old locale value.  Always call internal_setlocale if charset
	changed due to locale environment changes.

winsup/cygwin/ChangeLog

@@ -1,3 +1,17 @@
+2009-09-28  Corinna Vinschen  <corinna@vinschen.de>
+
+	* cygheap.cc (cygheap_init): Default locale.charset to "UTF-8".
+	* dcrt0.cc (dll_crt0_1): Call setlocale (LC_CTYPE, "") here, just
+	before resetting application locale to "C".
+	* environ.cc (environ_init): Drop code setting locale here.
+	* syscalls.cc (check_codepage): Break out check for required codepage
+	from internal_setlocale.
+	(internal_setlocale): Just convert CWD and $PATH from old charset to
+	new charset and set internal charset.
+	(setlocale): Change accordingly.  Don't require LC_ALL or LC_CTYPE to
+	store old locale value.  Always call internal_setlocale if charset
+	changed due to locale environment changes.
+
 2009-09-26  Eric Blake  <ebb9@byu.net>
 
 	* exec.cc (fexecve): New function.

winsup/cygwin/cygheap.cc

@@ -164,7 +164,7 @@ cygheap_init ()
       /* Default locale settings. */
       cygheap->locale.mbtowc = __utf8_mbtowc;
       cygheap->locale.wctomb = __utf8_wctomb;
-      strcpy (cygheap->locale.charset, "ASCII");
+      strcpy (cygheap->locale.charset, "UTF-8");
       /* Set umask to a sane default. */
       cygheap->umask = 022;
     }

winsup/cygwin/dcrt0.cc

@@ -939,7 +939,9 @@ dll_crt0_1 (void *)
      do this for noncygwin case since the signal thread is blocked due to
      LoadLibrary serialization. */
   ld_preload ();
-  /* Reset current application locale to "C" per POSIX */
+  /* Set internal locale to the environment settings. */
+  setlocale (LC_CTYPE, "");
+  /* Reset application locale to "C" per POSIX */
   _setlocale_r (_REENT, LC_CTYPE, "C");
   if (user_data->main)
     cygwin_exit (user_data->main (__argc, __argv, *user_data->envptr));

winsup/cygwin/environ.cc

@@ -732,7 +732,6 @@ environ_init (char **envp, int envc)
   static char NO_COPY cygterm[] = "TERM=cygwin";
   myfault efault;
   tmp_pathbuf tp;
-  static const char *lc_arr[] = { "LC_ALL", "LC_CTYPE", "LANG", NULL };
 
   if (efault.faulted ())
     api_fatal ("internal error reading the windows environment - too many environment variables?");
@@ -777,22 +776,6 @@ environ_init (char **envp, int envc)
   /* Allocate space for environment + trailing NULL + CYGWIN env. */
   lastenviron = envp = (char **) malloc ((4 + (envc = 100)) * sizeof (char *));
 
-  /* We need the locale variables' content before we can loop through
-     the whole environment, so that the wide-char to multibyte conversion
-     can be done according to the $LC_ALL/$LC_CTYPE/$LANG/current_codepage
-     setting, as well as the uppercasing according to the "upcaseenv"
-     setting.  Note that we have to reset the LC_CTYPE setting to "C"
-     before calling main() for POSIX compatibility. */
-  for (int lc = 0; lc_arr[lc]; ++lc)
-    {
-      if ((i = GetEnvironmentVariableA (lc_arr[lc], NULL, 0)))
-      	{
-	  char *buf = (char *) alloca (i);
-	  GetEnvironmentVariableA (lc_arr[lc], buf, i);
-	  if (setlocale (LC_CTYPE, buf))
-	    break;
-	}
-    }
   /* We also need the CYGWIN variable early to know the value of the
      CYGWIN=upcaseenv setting for the below loop. */
   if ((i = GetEnvironmentVariableA ("CYGWIN", NULL, 0)))

winsup/cygwin/syscalls.cc

@@ -4151,24 +4151,7 @@ unlinkat (int dirfd, const char *pathname, int flags)
 }
 
 static char *
-internal_setlocale (char *ret)
+check_codepage (char *ret)
-{
-  tmp_pathbuf tp;
-
-  /* Each setlocale potentially changes the multibyte representation
-     of the CWD.  Therefore we have to reevaluate the CWD's posix path and
-     store in the new charset. */
-  /* FIXME: Other buffered paths might be affected as well. */
-  wchar_t *w_cwd = tp.w_get ();
-  cwdstuff::cwd_lock.acquire ();
-  sys_mbstowcs (w_cwd, 32768, cygheap->cwd.get_posix ());
-
-  if (*__locale_charset () == 'A')
-    {
-      cygheap->locale.mbtowc = __utf8_mbtowc;
-      cygheap->locale.wctomb = __utf8_wctomb;
-    }
-  else
 {
   if (!wincap.has_always_all_codepages ())
     {
@@ -4193,27 +4176,54 @@ internal_setlocale (char *ret)
 	  && GetLastError () == ERROR_INVALID_PARAMETER)
 	return NULL;
     }
+  return ret;
+}
+
+static void
+internal_setlocale ()
+{
+  /* Each setlocale from the environment potentially changes the
+     multibyte representation of the CWD.  Therefore we have to
+     reevaluate the CWD's posix path and store in the new charset.
+     Same for the PATH environment variable. */
+  /* FIXME: Other buffered paths might be affected as well. */
+  tmp_pathbuf tp;
+  wchar_t *w_path, *w_cwd;
+
+  debug_printf ("Cygwin charset changed from %s to %s",
+		cygheap->locale.charset, __locale_charset ());
+  /* Fetch CWD and PATH and convert to wchar_t in previous charset. */
+  w_path = tp.w_get ();
+  sys_mbstowcs (w_path, 32768, getenv ("PATH"));
+  w_cwd = tp.w_get ();
+  cwdstuff::cwd_lock.acquire ();
+  sys_mbstowcs (w_cwd, 32768, cygheap->cwd.get_posix ());
+  /* Set charset for internal conversion functions. */
   cygheap->locale.mbtowc = __mbtowc;
   cygheap->locale.wctomb = __wctomb;
-    }
   strcpy (cygheap->locale.charset, __locale_charset ());
-
+  /* Restore CWD and PATH in new charset. */
-  /* See above. */
   cygheap->cwd.reset_posix (w_cwd);
   cwdstuff::cwd_lock.release ();
-  return ret;
+  char *c_path = tp.c_get ();
+  sys_wcstombs (c_path, 32768, w_path);
+  setenv ("PATH", c_path, 1);
 }
 
 extern "C" char *
 setlocale (int category, const char *locale)
 {
   char old[(LC_MESSAGES + 1) * (ENCODING_LEN + 1/*"/"*/ + 1)];
-  if (locale && (category == LC_ALL || category == LC_CTYPE)
+  if (locale && !wincap.has_always_all_codepages ())
-      && !wincap.has_always_all_codepages ())
     stpcpy (old, _setlocale_r (_REENT, category, NULL));
   char *ret = _setlocale_r (_REENT, category, locale);
-  if (ret && locale && (category == LC_ALL || category == LC_CTYPE)
+  if (ret && locale)
-      && !(ret = internal_setlocale (ret)))
+    {
+      if (!(ret = check_codepage (ret)))
 	_setlocale_r (_REENT, category, old);
+      else if (!*locale && strcmp (cygheap->locale.charset,
+				   __locale_charset ()) != 0)
+	internal_setlocale ();
+    }
   return ret;
 }