* cygpath.cc (codepage): New variable.

(long_options): Add --codepage option.
	(options): Add -C option.
	(usage): Add -C/--codepage description.
	(my_wcstombs): New function.  Use througout instead of wcstombs.
	(do_options): Handle -C/--codepage option.
	* utils.sgml (cygpath): Add description for new -C/--codepage option.

winsup/utils/ChangeLog

@@ -1,3 +1,13 @@
+2009-05-17  Corinna Vinschen  <corinna@vinschen.de>
+
+	* cygpath.cc (codepage): New variable.
+	(long_options): Add --codepage option.
+	(options): Add -C option.
+	(usage): Add -C/--codepage description.
+	(my_wcstombs): New function.  Use througout instead of wcstombs.
+	(do_options): Handle -C/--codepage option.
+	* utils.sgml (cygpath): Add description for new -C/--codepage option.
+
 2009-05-16  Corinna Vinschen  <corinna@vinschen.de>
 
 	* Makefile.in (cygpath.exe): Link against cygwin.a before linking

winsup/utils/cygpath.cc

@@ -38,6 +38,7 @@ static int path_flag, unix_flag, windows_flag, absolute_flag;
 static int shortname_flag, longname_flag;
 static int ignore_flag, allusers_flag, output_flag;
 static int mixed_flag, options_from_file_flag, mode_flag;
+static UINT codepage;
 
 static const char *format_type_arg;
 
@@ -66,10 +67,11 @@ static struct option long_options[] = {
   {(char *) "sysdir", no_argument, NULL, 'S'},
   {(char *) "windir", no_argument, NULL, 'W'},
   {(char *) "folder", required_argument, NULL, 'F'},
+  {(char *) "codepage", required_argument, NULL, 'C'},
   {0, no_argument, 0, 0}
 };
 
-static char options[] = "ac:df:hilmMopst:uvwADHOPSWF:";
+static char options[] = "ac:df:hilmMopst:uvwAC:DHOPSWF:";
 
 static void
 usage (FILE * stream, int status)
@@ -94,6 +96,11 @@ Path conversion options:\n\
   -l, --long-name       print Windows long form of NAMEs (with -w, -m only)\n\
   -p, --path            NAME is a PATH list (i.e., '/bin:/usr/bin')\n\
   -s, --short-name      print DOS (short) form of NAMEs (with -w, -m only)\n\
+  -C, --codepage CP     print DOS, Windows, or mixed pathname in Windows\n\
+                        codepage CP.  CP can be a numeric codepage identifier,\n\
+                        or one of the reserved words ANSI, OEM, or UTF8.\n\
+                        If this option is missing, %s defaults to the\n\
+                        character set defined by the current locale.\n\
 System information:\n\
   -A, --allusers        use `All Users' instead of current user for -D, -O, -P\n\
   -D, --desktop         output `Desktop' directory and exit\n\
@@ -103,7 +110,7 @@ System information:\n\
   -S, --sysdir          output system directory and exit\n\
   -W, --windir          output `Windows' directory and exit\n\
   -F, --folder ID       output special folder with numeric ID and exit\n\
-", prog_name, prog_name, prog_name, prog_name);
+", prog_name, prog_name, prog_name, prog_name, prog_name);
   if (ignore_flag)
     /* nothing to do */;
   else if (stream != stdout)
@@ -132,6 +139,15 @@ RtlAllocateUnicodeString (PUNICODE_STRING uni, ULONG size)
   return uni->Buffer != NULL;
 }
 
+static size_t
+my_wcstombs (char *dest, const wchar_t *src, size_t n)
+{
+  if (codepage)
+    return WideCharToMultiByte (codepage, 0, src, -1, dest, n, NULL, NULL);
+  else
+    return wcstombs (dest, src, n);
+}
+
 static char *
 get_device_name (char *path)
 {
@@ -331,14 +347,14 @@ get_short_paths (char *path)
       ++ptr, ++sptr;
       acc -= len + 1;
     }
-  len = wcstombs (NULL, sbuf, 0) + 1;
+  len = my_wcstombs (NULL, sbuf, 0) + 1;
   ptr = (char *) malloc (len);
   if (ptr == NULL)
     {
       fprintf (stderr, "%s: out of memory\n", prog_name);
       exit (1);
     }
-  wcstombs (ptr, sbuf, len);
+  my_wcstombs (ptr, sbuf, len);
   return ptr;
 }
 
@@ -355,14 +371,14 @@ get_short_name (const char *filename)
 	       filename);
       exit (2);
     }
-  len = wcstombs (NULL, buf, 0) + 1;
+  len = my_wcstombs (NULL, buf, 0) + 1;
   sbuf = (char *) malloc (len);
   if (sbuf == NULL)
     {
       fprintf (stderr, "%s: out of memory\n", prog_name);
       exit (1);
     }
-  wcstombs (sbuf, buf, len);
+  my_wcstombs (sbuf, buf, len);
   return sbuf;
 }
 
@@ -452,14 +468,14 @@ get_long_name (const char *filename, DWORD& len)
 	  wcsncat (buf, wpath, 32767);
 	}
     }
-  len = wcstombs (NULL, buf, 0);
+  len = my_wcstombs (NULL, buf, 0);
   sbuf = (char *) malloc (len + 1);
   if (!sbuf)
     {
       fprintf (stderr, "%s: out of memory\n", prog_name);
       exit (1);
     }
-  wcstombs (sbuf, buf, len + 1);
+  my_wcstombs (sbuf, buf, len + 1);
   return sbuf;
 }
 
@@ -738,7 +754,7 @@ do_pathconv (char *filename)
 	}
       if (!unix_flag)
 	{
-	  wcstombs (buf, buf2, 32768);
+	  my_wcstombs (buf, buf2, 32768);
 	  buf = get_device_name (buf);
 	  if (shortname_flag)
 	    buf = get_short_name (buf);
@@ -802,6 +818,7 @@ do_options (int argc, char **argv, int from_file)
   allusers_flag = 0;
   output_flag = 0;
   mode_flag = 0;
+  codepage = 0;
   if (!from_file)
     options_from_file_flag = 0;
   optind = 0;
@@ -893,6 +910,25 @@ do_options (int argc, char **argv, int from_file)
 	  allusers_flag = 1;
 	  break;
 
+	case 'C':
+	  if (!optarg)
+	    usage (stderr, 1);
+	  if (!strcasecmp (optarg, "ANSI"))
+	    codepage = GetACP ();
+	  else if (!strcasecmp (optarg, "OEM"))
+	    codepage = GetOEMCP ();
+	  else if (!strcasecmp (optarg, "UTF8")
+		   || !strcasecmp (optarg, "UTF-8"))
+	    codepage = CP_UTF8;
+	  else
+	    {
+	      char *c;
+	      codepage = (UINT) strtoul (optarg, &c, 10);
+	      if (*c)
+		usage (stderr, 1);
+	    }
+	  break;
+
 	case 'D':
 	case 'H':
 	case 'O':

winsup/utils/utils.sgml

@@ -207,6 +207,11 @@ Path conversion options:
   -l, --long-name       print Windows long form of NAMEs (with -w, -m only)
   -p, --path            NAME is a PATH list (i.e., '/bin:/usr/bin')
   -s, --short-name      print DOS (short) form of NAMEs (with -w, -m only)
+  -C, --codepage CP     print DOS, Windows, or mixed pathname in Windows
+                        codepage CP.  CP can be a numeric codepage identifier,
+                        or one of the reserved words ANSI, OEM, or UTF8.
+                        If this option is missing, cygpath defaults to the
+                        character set defined by the current locale.
 System information:
   -A, --allusers        use `All Users' instead of current user for -D, -P
   -D, --desktop         output `Desktop' directory and exit
@@ -249,6 +254,30 @@ the <literal>-l</literal> and <literal>-s</literal> options to use normal
 identical to <literal>-w</literal> and <literal>-s</literal> together.
 </para>
 
+<para>The <literal>-C</literal> option allows to specify a Windows codepage
+to print DOS and Windows paths created with one of the <literal>-d</literal>,
+<literal>-m</literal>, or <literal>-w</literal> options.  The default is to
+use the character set of the current locale defined by one of the
+internationalization environment variables <envar>LC_ALL</envar>,
+<envar>LC_CTYPE</envar>, or <envar>LANG</envar>, see
+<xref linkend="setup-locale"></xref>.  This is sometimes not sufficent for
+interaction with native Windows tools, which might expect native, non-ASCII
+characters in a specific Windows codepage.  Console tools, for instance, might
+expect pathnames in the current OEM codepage, while graphical tools like
+Windows Explorer might expect pathnames in the current ANSI codepage.</para>
+
+<para>The <literal>-C</literal> option takes a single parameter:</para>
+<itemizedlist spacing="compact">
+<listitem><para><literal>ANSI</literal>, to specify the current ANSI codepage</para></listitem>
+<listitem><para><literal>OEM</literal>, to specify the current OEM (console) codepage</para></listitem>
+<listitem><para><literal>UTF8</literal>, to specify UTF-8.</para></listitem>
+<listitem><para>A numerical, decimal codepage number, for instance 936 for GBK,
+28593 for ISO-8859-3, etc.  A full list of supported codepages is listed on the
+Microsoft MSDN page
+<ulink url="http://msdn.microsoft.com/en-us/library/dd317756(VS.85).aspx">Code Page Identifiers</ulink>.  A codepage of 0 is the same as if the
+<literal>-C</literal> hasn't been specified at all.</para></listitem>
+</itemizedlist>
+
 <para>The <literal>-p</literal> option means that you want to convert
 a path-style string rather than a single filename.  For example, the
 PATH environment variable is semicolon-delimited in Windows, but
@@ -272,7 +301,7 @@ if [ "${1}" = "" ];
 	then
 		XPATH=".";
 	else
-		XPATH="$(cygpath -w "${1}")";
+		XPATH="$(cygpath -C ANSI -w "${1}")";
 fi
 explorer $XPATH &
 ]]>