Make prim_file skip invalid filenames in unicode mode

The fix affects list_dir and read_link. Raw filenames are now
never produced, just consumed even if +fnu or +fna is used on
Linux etc.

This also adds the options to get error return or error handler
warning messages with +fn{u|a}{i|w|e} as an option to erl.

This is still not documented and there needs to be other versions
of read_dir and read_link to facilitate reading of all types
of filenames and links.

A check that we will not change to an invalid directory is also needed.

9 files changed, 125 insertions, 13 deletions

diff --git a/erts/emulator/beam/atom.names b/erts/emulator/beam/atom.names
index f138324e1f..ce60bb9bbc 100644
--- a/erts/emulator/beam/atom.names
+++ b/erts/emulator/beam/atom.names
@@ -269,6 +269,7 @@ atom hipe_architecture
 atom http httph https http_response http_request http_header http_eoh http_error http_bin httph_bin
 atom id
 atom if_clause
+atom ignore
 atom imports
 atom in
 atom in_exiting
diff --git a/erts/emulator/beam/erl_init.c b/erts/emulator/beam/erl_init.c
index 223c9c4d7e..ec3e0d54cb 100644
--- a/erts/emulator/beam/erl_init.c
+++ b/erts/emulator/beam/erl_init.c
@@ -981,19 +981,64 @@ erl_start(int argc, char **argv)
 	    break;
 	case 'f':
 	    if (!strncmp(argv[i],"-fn",3)) {
+		int warning_type =  ERL_FILENAME_WARNING_WARNING;
 		arg = get_arg(argv[i]+3, argv[i+1], &i);
 		switch (*arg) {
 		case 'u':
-		    erts_set_user_requested_filename_encoding(ERL_FILENAME_UTF8);
+		    switch (*(argv[i]+4)) {
+		    case 'w':
+		    case 0:
+			break;
+		    case 'i':
+			warning_type =  ERL_FILENAME_WARNING_IGNORE;
+			break;
+		    case 'e':
+			warning_type =  ERL_FILENAME_WARNING_ERROR;
+			break;
+		    default:
+			erts_fprintf(stderr, "bad type of warnings for "
+				     "wrongly coded filename: %s\n", argv[i]+4);
+			erts_usage();
+		    }
+		    erts_set_user_requested_filename_encoding
+			(
+			 ERL_FILENAME_UTF8,
+			 warning_type
+			 );
 		    break;
 		case 'l':
-		    erts_set_user_requested_filename_encoding(ERL_FILENAME_LATIN1);
+		    erts_set_user_requested_filename_encoding
+			(
+			 ERL_FILENAME_LATIN1,
+			 warning_type
+			 );
 		    break;
 		case 'a':
-		    erts_set_user_requested_filename_encoding(ERL_FILENAME_UNKNOWN);
+		    switch (*(argv[i]+4)) {
+		    case 'w':
+		    case 0:
+			break;
+		    case 'i':
+			warning_type =  ERL_FILENAME_WARNING_IGNORE;
+			break;
+		    case 'e':
+			warning_type =  ERL_FILENAME_WARNING_ERROR;
+			break;
+		    default:
+			erts_fprintf(stderr, "bad type of warnings for "
+				     "wrongly coded filename: %s\n", argv[i]+4);
+			erts_usage();
+		    }
+		    erts_set_user_requested_filename_encoding
+			(
+			 ERL_FILENAME_UNKNOWN,
+			 warning_type
+			 );
 		    break;
 		default:
-		    erts_fprintf(stderr, "bad filename encoding %s, can be (l,u or a)\n", arg);
+		    erts_fprintf(stderr, "bad filename encoding %s, can be "
+				 "(l,u or a, optionally followed by w, "
+				 "i or e)\n", arg);
 		    erts_usage();
 		}
 		break;
diff --git a/erts/emulator/beam/erl_unicode.c b/erts/emulator/beam/erl_unicode.c
index 99108af937..8d7930e5fa 100644
--- a/erts/emulator/beam/erl_unicode.c
+++ b/erts/emulator/beam/erl_unicode.c
@@ -2573,8 +2573,20 @@ BIF_RETTYPE prim_file_internal_native2name_1(BIF_ALIST_1)
     case ERL_FILENAME_UTF8:
 	bytes = erts_get_aligned_binary_bytes(BIF_ARG_1, &temp_alloc);
 	if (erts_analyze_utf8(bytes,size,&err_pos,&num_chars,NULL) != ERTS_UTF8_OK) {
+	    Eterm *hp = HAlloc(BIF_P,3);
+	    Eterm warn_type = NIL;
 	    erts_free_aligned_binary_bytes(temp_alloc);
-	    goto noconvert;
+	    switch (erts_get_filename_warning_type()) {
+	    case ERL_FILENAME_WARNING_IGNORE:
+		warn_type = am_ignore;
+		break;
+	    case ERL_FILENAME_WARNING_ERROR:
+		warn_type = am_error;
+		break;
+	    default:
+		warn_type = am_warning;
+	    }
+	    BIF_RET(TUPLE2(hp,am_error,warn_type));
 	}
 	num_built = 0;
 	num_eaten = 0;
@@ -2607,9 +2619,8 @@ BIF_RETTYPE prim_file_internal_native2name_1(BIF_ALIST_1)
 	erts_free_aligned_binary_bytes(temp_alloc);
 	BIF_RET(ret);
     default:
-	goto noconvert;
+	break;
     }
- noconvert:
     BIF_RET(BIF_ARG_1);
 }
 
diff --git a/erts/emulator/beam/sys.h b/erts/emulator/beam/sys.h
index 249a9c05c2..9416a91480 100644
--- a/erts/emulator/beam/sys.h
+++ b/erts/emulator/beam/sys.h
@@ -1029,10 +1029,22 @@ char* win32_errorstr(int);
 #define ERL_FILENAME_UTF8_MAC  (3)
 #define ERL_FILENAME_WIN_WCHAR (4)
 
+/************************************************************************
+ * If a filename in for example list_dir is not in the right encoding, it
+ * will be skipped in the resulting list, but depending on a startup setting
+ * we will inform the user in different ways. These macros define the
+ * different reactions to wrongly coded filenames. In the error case an
+ * exception will be thrown by prim_file.
+ ************************************************************************/
+#define ERL_FILENAME_WARNING_WARNING (0)
+#define ERL_FILENAME_WARNING_IGNORE (1)
+#define ERL_FILENAME_WARNING_ERROR (2)
+
 int erts_get_native_filename_encoding(void);
 /* The set function is only to be used by erl_init! */
-void erts_set_user_requested_filename_encoding(int encoding); 
+void erts_set_user_requested_filename_encoding(int encoding, int warning);
 int erts_get_user_requested_filename_encoding(void);
+int erts_get_filename_warning_type(void);
 
 void erts_init_sys_common_misc(void);
 
diff --git a/erts/emulator/sys/common/erl_sys_common_misc.c b/erts/emulator/sys/common/erl_sys_common_misc.c
index 1bf5fa89f4..a0cd4cd10a 100644
--- a/erts/emulator/sys/common/erl_sys_common_misc.c
+++ b/erts/emulator/sys/common/erl_sys_common_misc.c
@@ -47,14 +47,16 @@
 /* Written once and only once */
 
 static int filename_encoding = ERL_FILENAME_UNKNOWN;
+static int filename_warning = ERL_FILENAME_WARNING_WARNING;
 #if defined(__WIN32__) || defined(__DARWIN__)
 static int user_filename_encoding = ERL_FILENAME_UTF8; /* Default unicode on windows */
 #else
 static int user_filename_encoding = ERL_FILENAME_LATIN1;
 #endif
-void erts_set_user_requested_filename_encoding(int encoding)
+void erts_set_user_requested_filename_encoding(int encoding, int warning)
 {
     user_filename_encoding = encoding;
+    filename_warning = warning;
 }
 
 int erts_get_user_requested_filename_encoding(void)
@@ -62,6 +64,11 @@ int erts_get_user_requested_filename_encoding(void)
     return user_filename_encoding;
 }
 
+int erts_get_filename_warning_type(void)
+{
+    return filename_warning;
+}
+
 void erts_init_sys_common_misc(void)
 {
 #if defined(__WIN32__)
diff --git a/erts/preloaded/ebin/erl_prim_loader.beam b/erts/preloaded/ebin/erl_prim_loader.beam
index f8c2df3a5a..4a3af265c1 100644
--- a/erts/preloaded/ebin/erl_prim_loader.beam
+++ b/erts/preloaded/ebin/erl_prim_loader.beam
diff --git a/erts/preloaded/ebin/prim_file.beam b/erts/preloaded/ebin/prim_file.beam
index ca93edbe25..b1b54ca050 100644
--- a/erts/preloaded/ebin/prim_file.beam
+++ b/erts/preloaded/ebin/prim_file.beam
diff --git a/erts/preloaded/src/erl_prim_loader.erl b/erts/preloaded/src/erl_prim_loader.erl
index c2fac8c0fc..7490954f2d 100644
--- a/erts/preloaded/src/erl_prim_loader.erl
+++ b/erts/preloaded/src/erl_prim_loader.erl
@@ -150,7 +150,17 @@ start_it("inet", Id, Pid, Hosts) ->
 start_it("efile", Id, Pid, _Hosts) ->
     process_flag(trap_exit, true),
     {ok, Port} = prim_file:start(),
-    init_ack(Pid),
+    %% Check that we started in a valid directory.
+    case prim_file:get_cwd(Port) of
+	{error, _} ->
+	    %% At this point in the startup, we have no error_logger at all.
+	    Report = "Invalid current directory or invalid filename "
+		"mode: loader cannot read current directory\n",
+	    erlang:display(Report),
+	    exit({error, invalid_current_directory});
+	_ ->
+	    init_ack(Pid)
+    end,
     MultiGet = case erlang:system_info(thread_pool_size) of
                    0 -> false;
                    _ -> true
diff --git a/erts/preloaded/src/prim_file.erl b/erts/preloaded/src/prim_file.erl
index af1a6127c9..27e7640b7e 100644
--- a/erts/preloaded/src/prim_file.erl
+++ b/erts/preloaded/src/prim_file.erl
@@ -155,13 +155,15 @@
          internal_normalize_utf8/1]).
 
 -type prim_file_name() :: string() | unicode:unicode_binary().
+-type prim_file_name_error() :: 'error' | 'ignore' | 'warning'.
 
 -spec internal_name2native(prim_file_name()) -> binary().
 
 internal_name2native(_) ->
     erlang:nif_error(undefined).
 
--spec internal_native2name(binary()) -> prim_file_name().
+-spec internal_native2name(binary()) ->
+        prim_file_name() | {'error',prim_file_name_error()}.
 
 internal_native2name(_) ->
     erlang:nif_error(undefined).
@@ -924,7 +926,21 @@ list_dir_response(Port, Acc0) ->
     end.
 
 list_dir_convert([Name|Names]) ->
-    [prim_file:internal_native2name(Name)|list_dir_convert(Names)];
+    %% If the filename cannot be converted, return error or ignore
+    %% with optional error logger warning, depending on +fn{u|a}{i|e|w}
+    %% emulator switches.
+    case prim_file:internal_native2name(Name) of
+	{error, warning} ->
+	    error_logger:warning_msg("Non-unicode filename ~p ignored\n",
+				     [Name]),
+	    list_dir_convert(Names);
+	{error, ignore} ->
+	    list_dir_convert(Names);
+	{error, error} ->
+	    {error, {no_translation, Name}};
+	Converted when is_list(Converted) ->
+	    [Converted|list_dir_convert(Names)]
+    end;
 list_dir_convert([]) -> [].
 
 %%%-----------------------------------------------------------------
@@ -933,7 +949,17 @@ list_dir_convert([]) -> [].
 handle_fname_response(Port) ->
     case drv_get_response(Port) of
 	{fname, Name} ->
-	    {ok, prim_file:internal_native2name(Name)};
+	    case prim_file:internal_native2name(Name) of
+		{error, warning} ->
+		    error_logger:warning_msg("Non-unicode filename ~p "
+					     "ignored when reading link\n",
+					     [Name]),
+		    {error, einval};
+		{error, _} ->
+		    {error, einval};
+		Converted when is_list(Converted) ->
+		    {ok, Converted}
+	    end;
 	Error ->
 	    Error
     end.