stdlib: Add Unicode modifier t to control sequences w and W

As of the introduction of Unicode characters in atoms, the control
sequences 'w' and 'W' can return non-Latin-1 characters, unless some
measure is taken.

This commit makes sure that '~w' and '~W' always return Latin-1
characters, or bytes, which can be output to ports or written to raw
files.

The Unicode translation modifier 't' is needed to return non-Latin-1
characters.

6 files changed, 83 insertions, 40 deletions

diff --git a/lib/stdlib/doc/src/io.xml b/lib/stdlib/doc/src/io.xml
index 11a64c7f8a..74b57457ea 100644
--- a/lib/stdlib/doc/src/io.xml
+++ b/lib/stdlib/doc/src/io.xml
@@ -4,7 +4,7 @@
 <erlref>
   <header>
     <copyright>
-      <year>1996</year><year>2016</year>
+      <year>1996</year><year>2017</year>
       <holder>Ericsson AB. All Rights Reserved.</holder>
     </copyright>
     <legalnotice>
@@ -265,6 +265,8 @@ ok
             <p>Writes data with the standard syntax. This is used to
               output Erlang terms. Atoms are printed within quotes if
               they contain embedded non-printable characters.
+	      Atom characters &gt; 255 are escaped unless the
+	      Unicode translation modifier (<c>t</c>) is used.
               Floats are printed accurately as the shortest, correctly
               rounded string.</p>
           </item>
diff --git a/lib/stdlib/doc/src/io_lib.xml b/lib/stdlib/doc/src/io_lib.xml
index 5ae400da62..bc1d77ac83 100644
--- a/lib/stdlib/doc/src/io_lib.xml
+++ b/lib/stdlib/doc/src/io_lib.xml
@@ -356,7 +356,8 @@
 
     <func>
       <name name="write" arity="1"/>
-      <name name="write" arity="2"/>
+      <name name="write" arity="2" clause_i="1"/>
+      <name name="write" arity="2" clause_i="2"/>
       <fsummary>Write a term.</fsummary>
       <desc>
         <p>Returns a character list that represents <c><anno>Term</anno></c>.
diff --git a/lib/stdlib/src/erl_lint.erl b/lib/stdlib/src/erl_lint.erl
index 78b7a0e751..7c40058dd8 100644
--- a/lib/stdlib/src/erl_lint.erl
+++ b/lib/stdlib/src/erl_lint.erl
@@ -3883,6 +3883,10 @@ extract_sequence(4, [$t, $p | Fmt], Need) ->
     extract_sequence(5, [$p|Fmt], Need);
 extract_sequence(4, [$t, $P | Fmt], Need) ->
     extract_sequence(5, [$P|Fmt], Need);
+extract_sequence(4, [$t, $w | Fmt], Need) ->
+    extract_sequence(5, [$w|Fmt], Need);
+extract_sequence(4, [$t, $W | Fmt], Need) ->
+    extract_sequence(5, [$W|Fmt], Need);
 extract_sequence(4, [$t, C | _Fmt], _Need) ->
     {error,"invalid control ~t" ++ [C]};
 extract_sequence(4, [$l, $p | Fmt], Need) ->
diff --git a/lib/stdlib/src/io_lib.erl b/lib/stdlib/src/io_lib.erl
index 28e5007e5a..5ed2f4d888 100644
--- a/lib/stdlib/src/io_lib.erl
+++ b/lib/stdlib/src/io_lib.erl
@@ -268,47 +268,61 @@ write(Term, D, false) ->
 
 -spec write(Term, Depth) -> chars() when
       Term :: term(),
+      Depth :: depth();
+           (Term, Options) -> chars() when
+      Term :: term(),
+      Options :: [Option],
+      Option :: {'depth', Depth}
+              | {'encoding', 'latin1' | 'utf8' | 'unicode'},
       Depth :: depth().
 
-write(_Term, 0) -> "...";
-write(Term, _D) when is_integer(Term) -> integer_to_list(Term);
-write(Term, _D) when is_float(Term) -> io_lib_format:fwrite_g(Term);
-write(Atom, _D) when is_atom(Atom) -> write_atom(Atom);
-write(Term, _D) when is_port(Term) -> write_port(Term);
-write(Term, _D) when is_pid(Term) -> pid_to_list(Term);
-write(Term, _D) when is_reference(Term) -> write_ref(Term);
-write(<<_/bitstring>>=Term, D) -> write_binary(Term, D);
-write([], _D) -> "[]";
-write({}, _D) -> "{}";
-write([H|T], D) ->
+write(Term, Options) when is_list(Options) ->
+    Depth = get_option(depth, Options, -1),
+    Encoding = get_option(encoding, Options, epp:default_encoding()),
+    write1(Term, Depth, Encoding);
+write(Term, Depth) ->
+    write1(Term, Depth, latin1).
+
+write1(_Term, 0, _E) -> "...";
+write1(Term, _D, _E) when is_integer(Term) -> integer_to_list(Term);
+write1(Term, _D, _E) when is_float(Term) -> io_lib_format:fwrite_g(Term);
+write1(Atom, _D, latin1) when is_atom(Atom) -> write_atom_as_latin1(Atom);
+write1(Atom, _D, _E) when is_atom(Atom) -> write_atom(Atom);
+write1(Term, _D, _E) when is_port(Term) -> write_port(Term);
+write1(Term, _D, _E) when is_pid(Term) -> pid_to_list(Term);
+write1(Term, _D, _E) when is_reference(Term) -> write_ref(Term);
+write1(<<_/bitstring>>=Term, D, _E) -> write_binary(Term, D);
+write1([], _D, _E) -> "[]";
+write1({}, _D, _E) -> "{}";
+write1([H|T], D, E) ->
     if
 	D =:= 1 -> "[...]";
 	true ->
-	    [$[,[write(H, D-1)|write_tail(T, D-1, $|)],$]]
+	    [$[,[write1(H, D-1, E)|write_tail(T, D-1, E, $|)],$]]
     end;
-write(F, _D) when is_function(F) ->
+write1(F, _D, _E) when is_function(F) ->
     erlang:fun_to_list(F);
-write(Term, D) when is_map(Term) ->
-    write_map(Term, D);
-write(T, D) when is_tuple(T) ->
+write1(Term, D, E) when is_map(Term) ->
+    write_map(Term, D, E);
+write1(T, D, E) when is_tuple(T) ->
     if
 	D =:= 1 -> "{...}";
 	true ->
 	    [${,
-	     [write(element(1, T), D-1)|
-              write_tail(tl(tuple_to_list(T)), D-1, $,)],
+	     [write1(element(1, T), D-1, E)|
+              write_tail(tl(tuple_to_list(T)), D-1, E, $,)],
 	     $}]
     end.
 
 %% write_tail(List, Depth, CharacterBeforeDots)
 %%  Test the terminating case first as this looks better with depth.
 
-write_tail([], _D, _S) -> "";
-write_tail(_, 1, S) -> [S | "..."];
-write_tail([H|T], D, S) ->
-    [$,,write(H, D-1)|write_tail(T, D-1, S)];
-write_tail(Other, D, S) ->
-    [S,write(Other, D-1)].
+write_tail([], _D, _E, _S) -> "";
+write_tail(_, 1, _E, S) -> [S | "..."];
+write_tail([H|T], D, E, S) ->
+    [$,,write1(H, D-1, E)|write_tail(T, D-1, E, S)];
+write_tail(Other, D, E, S) ->
+    [S,write1(Other, D-1, E)].
 
 write_port(Port) ->
     erlang:port_to_list(Port).
@@ -316,17 +330,17 @@ write_port(Port) ->
 write_ref(Ref) ->
     erlang:ref_to_list(Ref).
 
-write_map(Map, D) when is_integer(D) ->
-    [$#,${,write_map_body(maps:to_list(Map), D),$}].
+write_map(Map, D, E) when is_integer(D) ->
+    [$#,${,write_map_body(maps:to_list(Map), D, E),$}].
 
-write_map_body(_, 0) -> "...";
-write_map_body([],_) -> [];
-write_map_body([{K,V}],D) -> write_map_assoc(K,V,D);
-write_map_body([{K,V}|KVs], D) ->
-    [write_map_assoc(K,V,D),$, | write_map_body(KVs,D-1)].
+write_map_body(_, 0, _E) -> "...";
+write_map_body([], _, _E) -> [];
+write_map_body([{K,V}], D, E) -> write_map_assoc(K, V, D, E);
+write_map_body([{K,V}|KVs], D, E) ->
+    [write_map_assoc(K, V, D, E),$, | write_map_body(KVs, D-1, E)].
 
-write_map_assoc(K,V,D) ->
-    [write(K,D - 1),"=>",write(V,D-1)].
+write_map_assoc(K, V, D, E) ->
+    [write1(K, D - 1, E),"=>",write1(V, D-1, E)].
 
 write_binary(B, D) when is_integer(D) ->
     [$<,$<,write_binary_body(B, D),$>,$>].
@@ -344,6 +358,13 @@ write_binary_body(B, _D) ->
     <<X:L>> = B,
     [integer_to_list(X),$:,integer_to_list(L)].
 
+get_option(Key, TupleList, Default) ->
+    case lists:keyfind(Key, 1, TupleList) of
+	false -> Default;
+	{Key, Value} -> Value;
+	_ -> Default
+    end.
+
 %%% There are two functions to write Unicode atoms:
 %%% - they both escape control characters < 160;
 %%% - write_atom() never escapes characters >= 160;
diff --git a/lib/stdlib/src/io_lib_format.erl b/lib/stdlib/src/io_lib_format.erl
index 3113767614..14d925bacf 100644
--- a/lib/stdlib/src/io_lib_format.erl
+++ b/lib/stdlib/src/io_lib_format.erl
@@ -1,7 +1,7 @@
 %%
 %% %CopyrightBegin%
 %% 
-%% Copyright Ericsson AB 1996-2016. All Rights Reserved.
+%% Copyright Ericsson AB 1996-2017. All Rights Reserved.
 %% 
 %% Licensed under the Apache License, Version 2.0 (the "License");
 %% you may not use this file except in compliance with the License.
@@ -257,12 +257,12 @@ indentation([], I) -> I.
 %%  This is the main dispatch function for the various formatting commands.
 %%  Field widths and precisions have already been calculated.
 
-control($w, [A], F, Adj, P, Pad, _Enc, _Str, _I) ->
-    term(io_lib:write(A, -1), F, Adj, P, Pad);
+control($w, [A], F, Adj, P, Pad, Enc, _Str, _I) ->
+    term(io_lib:write(A, [{depth,-1}, {encoding, Enc}]), F, Adj, P, Pad);
 control($p, [A], F, Adj, P, Pad, Enc, Str, I) ->
     print(A, -1, F, Adj, P, Pad, Enc, Str, I);
-control($W, [A,Depth], F, Adj, P, Pad, _Enc, _Str, _I) when is_integer(Depth) ->
-    term(io_lib:write(A, Depth), F, Adj, P, Pad);
+control($W, [A,Depth], F, Adj, P, Pad, Enc, _Str, _I) when is_integer(Depth) ->
+    term(io_lib:write(A, [{depth,Depth}, {encoding, Enc}]), F, Adj, P, Pad);
 control($P, [A,Depth], F, Adj, P, Pad, Enc, Str, I) when is_integer(Depth) ->
     print(A, Depth, F, Adj, P, Pad, Enc, Str, I);
 control($s, [A], F, Adj, P, Pad, latin1, _Str, _I) when is_atom(A) ->
diff --git a/lib/stdlib/test/io_SUITE.erl b/lib/stdlib/test/io_SUITE.erl
index b2754e47ba..fb62e3aa3c 100644
--- a/lib/stdlib/test/io_SUITE.erl
+++ b/lib/stdlib/test/io_SUITE.erl
@@ -2348,4 +2348,19 @@ otp_14285(_Config) ->
     L1 = [S || C <- Chars, S <- io_lib:write_atom(list_to_atom([C])),
                not is_latin1(S)],
     L1 = lists:seq(256, 512),
+
+    latin1_fmt("~w", ['кирилли́ческий атом']),
+    latin1_fmt("~w", ['\x{10FFFF}']),
+    "'кирилли́ческий атом'" = fmt("~tw", ['кирилли́ческий атом']),
+    [$',16#10FFFF,$'] = fmt("~tw", ['\x{10FFFF}']),
+
+    latin1_fmt("~W", ['кирилли́ческий атом', 13]),
+    latin1_fmt("~W", ['\x{10FFFF}', 13]),
+    "'кирилли́ческий атом'" = fmt("~tW", ['кирилли́ческий атом', 13]),
+    [$',16#10FFFF,$'] = fmt("~tW", ['\x{10FFFF}', 13]),
+
     ok.
+
+latin1_fmt(Fmt, Args) ->
+    L = fmt(Fmt, Args),
+    true = lists:all(fun is_latin1/1, L).