kernel: Handle unicode in os:cmd

Allow unicode strings in os:cmd and try to convert the
result bytes to unicode list.

Also test it.

3 files changed, 69 insertions, 22 deletions

diff --git a/lib/kernel/src/os.erl b/lib/kernel/src/os.erl
index 86f2f94f1f..c92bb463c1 100644
--- a/lib/kernel/src/os.erl
+++ b/lib/kernel/src/os.erl
@@ -183,20 +183,25 @@ extensions() ->
       Command :: atom() | io_lib:chars().
 cmd(Cmd) ->
     validate(Cmd),
-    case type() of
-	{unix, _} ->
-	    unix_cmd(Cmd);
-	{win32, Wtype} ->
-	    Command0 = case {os:getenv("COMSPEC"),Wtype} of
-			  {false,windows} -> lists:concat(["command.com /c", Cmd]);
-			  {false,_} -> lists:concat(["cmd /c", Cmd]);
-			  {Cspec,_} -> lists:concat([Cspec," /c",Cmd])
-		      end,
-            %% open_port/2 awaits string() in Command, but io_lib:chars() can be
-            %% deep lists according to io_lib module description.
-            Command = lists:flatten(Command0),
-	    Port = open_port({spawn, Command}, [stream, in, eof, hide]),
-	    get_data(Port, [])
+    Bytes = case type() of
+		{unix, _} ->
+		    unix_cmd(Cmd);
+		{win32, Wtype} ->
+		    Command0 = case {os:getenv("COMSPEC"),Wtype} of
+				  {false,windows} -> lists:concat(["command.com /c", Cmd]);
+				  {false,_} -> lists:concat(["cmd /c", Cmd]);
+				  {Cspec,_} -> lists:concat([Cspec," /c",Cmd])
+			      end,
+		    %% open_port/2 awaits string() in Command, but io_lib:chars() can be
+		    %% deep lists according to io_lib module description.
+		    Command = lists:flatten(Command0),
+		    Port = open_port({spawn, Command}, [stream, in, eof, hide]),
+		    get_data(Port, [])
+	    end,
+    String = unicode:characters_to_list(list_to_binary(Bytes)),
+    if  %% Convert to unicode list if possible otherwise return bytes
+	is_list(String) -> String;
+	true -> Bytes
     end.
 
 unix_cmd(Cmd) ->
@@ -337,7 +342,7 @@ mk_cmd(Cmd) when is_atom(Cmd) ->		% backward comp.
 mk_cmd(Cmd) ->
     %% We insert a new line after the command, in case the command
     %% contains a comment character.
-    io_lib:format("(~ts\n) </dev/null; echo  \"\^D\"\n", [Cmd]).
+    [$(, unicode:characters_to_binary(Cmd), "\n) </dev/null; echo  \"\^D\"\n"].
 
 
 validate(Atom) when is_atom(Atom) ->
@@ -345,7 +350,7 @@ validate(Atom) when is_atom(Atom) ->
 validate(List) when is_list(List) ->
     validate1(List).
 
-validate1([C|Rest]) when is_integer(C), 0 =< C, C < 256 ->
+validate1([C|Rest]) when is_integer(C) ->
     validate1(Rest);
 validate1([List|Rest]) when is_list(List) ->
     validate1(List),
diff --git a/lib/kernel/test/os_SUITE.erl b/lib/kernel/test/os_SUITE.erl
index 73ed704ae3..9b474c4cdf 100644
--- a/lib/kernel/test/os_SUITE.erl
+++ b/lib/kernel/test/os_SUITE.erl
@@ -20,7 +20,7 @@
 
 -export([all/0, suite/0,groups/0,init_per_suite/1, end_per_suite/1,
 	 init_per_group/2,end_per_group/2]).
--export([space_in_cwd/1, quoting/1, space_in_name/1, bad_command/1,
+-export([space_in_cwd/1, quoting/1, cmd_unicode/1, space_in_name/1, bad_command/1,
 	 find_executable/1, unix_comment_in_command/1, deep_list_command/1, evil/1]).
 
 -include_lib("test_server/include/test_server.hrl").
@@ -28,9 +28,8 @@
 suite() -> [{ct_hooks,[ts_install_cth]}].
 
 all() ->
-    [space_in_cwd, quoting, space_in_name, bad_command,
-     find_executable, unix_comment_in_command, deep_list_command,
-     evil].
+    [space_in_cwd, quoting, cmd_unicode, space_in_name, bad_command,
+     find_executable, unix_comment_in_command, deep_list_command, evil].
 
 groups() ->
     [].
@@ -95,6 +94,21 @@ quoting(Config) when is_list(Config) ->
     ?line [] = receive_all(),
     ok.
 
+
+cmd_unicode(doc) -> "Test that unicode arguments work.";
+cmd_unicode(suite) -> [];
+cmd_unicode(Config) when is_list(Config) ->
+    ?line DataDir = ?config(data_dir, Config),
+    ?line Echo = filename:join(DataDir, "my_echo"),
+
+    ?line comp("one", os:cmd(Echo ++ " one")),
+    ?line comp("one::two", os:cmd(Echo ++ " one two")),
+    ?line comp("åäö::ϼΩ", os:cmd(Echo ++ " åäö " ++ [1020, 937])),
+    ?t:sleep(5),
+    ?line [] = receive_all(),
+    ok.
+
+
 space_in_name(doc) ->
     "Test that program with a space in its name can be executed.";
 space_in_name(suite) -> [];
@@ -302,8 +316,8 @@ comp(Expected, Got) ->
 	Expected ->
 	    ok;
 	Other ->
-	    ok = io:format("Expected: ~s\n", [Expected]),
-	    ok = io:format("Got:      ~s\n", [Other]),
+	    ok = io:format("Expected: ~ts\n", [Expected]),
+	    ok = io:format("Got:      ~ts\n", [Other]),
 	    test_server:fail()
     end.
 
diff --git a/lib/kernel/test/os_SUITE_data/my_echo.c b/lib/kernel/test/os_SUITE_data/my_echo.c
index 2127511dd1..712c828bb5 100644
--- a/lib/kernel/test/os_SUITE_data/my_echo.c
+++ b/lib/kernel/test/os_SUITE_data/my_echo.c
@@ -1,3 +1,30 @@
+#ifdef __WIN32__
+#include <windows.h>
+
+int wmain(int argc, wchar_t **argv)
+{
+    char* sep = "";
+    int len;
+
+    /*
+     * Echo all arguments separated with '::', so that we can check that
+     * quotes are interpreted correctly.
+     */
+
+    while (argc-- > 1) {
+	char *utf8;
+	len = WideCharToMultiByte(CP_UTF8, 0, argv[1], -1, NULL, 0, NULL, NULL);
+	utf8 = malloc(len*sizeof(char));
+	WideCharToMultiByte(CP_UTF8, 0, argv++[1], -1, utf8, len, NULL, NULL);
+	printf("%s%s", sep, utf8);
+	free(utf8);
+	sep = "::";
+    }
+    putchar('\n');
+    return 0;
+}
+#else
+
 #include <stdio.h>
 
 int
@@ -17,3 +44,4 @@ main(int argc, char** argv)
     putchar('\n');
     return 0;
 }
+#endif