Merge branch 'hb/stdlib/escript_unicode/OTP-10951' into maint

* hb/stdlib/escript_unicode/OTP-10951:
  Let escript recognize an encoding comment on the second line

6 files changed, 48 insertions, 2 deletions

diff --git a/lib/stdlib/src/escript.erl b/lib/stdlib/src/escript.erl
index 32742e419b..fea718541d 100644
--- a/lib/stdlib/src/escript.erl
+++ b/lib/stdlib/src/escript.erl
@@ -602,9 +602,15 @@ parse_beam(S, File, HeaderSz, CheckOnly) ->
 parse_source(S, File, Fd, StartLine, HeaderSz, CheckOnly) ->
     {PreDefMacros, Module} = pre_def_macros(File),
     IncludePath = [],
-    {ok, _} = file:position(Fd, {bof, HeaderSz}),
+    %% Read the encoding on the second line, if there is any:
+    {ok, _} = file:position(Fd, 0),
+    _ = io:get_line(Fd, ''),
+    Encoding = epp:set_encoding(Fd),
+    {ok, _} = file:position(Fd, HeaderSz),
     case epp:open(File, Fd, StartLine, IncludePath, PreDefMacros) of
         {ok, Epp} ->
+            _ = [io:setopts(Fd, [{encoding,Encoding}]) ||
+                    Encoding =/= none],
             {ok, FileForm} = epp:parse_erl_form(Epp),
             OptModRes = epp:parse_erl_form(Epp),
             S2 = S#state{source = text, module = Module},
diff --git a/lib/stdlib/test/escript_SUITE.erl b/lib/stdlib/test/escript_SUITE.erl
index cf5fb12686..b6cdd0a9c7 100644
--- a/lib/stdlib/test/escript_SUITE.erl
+++ b/lib/stdlib/test/escript_SUITE.erl
@@ -919,6 +919,9 @@ unicode(Config) when is_list(Config) ->
            " an arithmetic expression\n  in operator  '/'/2\n     "
            "called as <<\"\xaa\">> / <<\"\xaa\">>\nExitCode:127">>]),
     run(Dir, "unicode3", [<<"ExitCode:0">>]),
+    run(Dir, "unicode4", [<<"ExitCode:0">>]),
+    run(Dir, "unicode5", [<<"ExitCode:0">>]),
+    run(Dir, "unicode6", [<<"ExitCode:0">>]),
     ok.
 
 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
diff --git a/lib/stdlib/test/escript_SUITE_data/unicode3 b/lib/stdlib/test/escript_SUITE_data/unicode3
index 944487dcae..0702ecf309 100755
--- a/lib/stdlib/test/escript_SUITE_data/unicode3
+++ b/lib/stdlib/test/escript_SUITE_data/unicode3
@@ -1,5 +1,5 @@
 #!/usr/bin/env escript
-%% -*- erlang; coding: utf-8 -*-
+%% -*- erlang; coding: latin-1 -*-
 
 -export([main/1]).
 
diff --git a/lib/stdlib/test/escript_SUITE_data/unicode4 b/lib/stdlib/test/escript_SUITE_data/unicode4
new file mode 100755
index 0000000000..a7563a613a
--- /dev/null
+++ b/lib/stdlib/test/escript_SUITE_data/unicode4
@@ -0,0 +1,12 @@
+#!/usr/bin/env escript
+%% -*- erlang; encoding:utf-8 -*-
+
+-export([main/1]).
+
+main(_) ->
+    ok = io:setopts([{encoding,unicode}]),
+    Bin1 = <<"örn_Ѐ שלום-שלום+של 日本語">>,
+    L = [246,114,110,95,1024,32,1513,1500,1493,1501,45,1513,1500,1493,
+         1501,43,1513,1500,32,26085,26412,35486],
+    L = unicode:characters_to_list(Bin1, utf8),
+    ok.
diff --git a/lib/stdlib/test/escript_SUITE_data/unicode5 b/lib/stdlib/test/escript_SUITE_data/unicode5
new file mode 100755
index 0000000000..e95da3361d
--- /dev/null
+++ b/lib/stdlib/test/escript_SUITE_data/unicode5
@@ -0,0 +1,12 @@
+#!/usr/bin/env escript
+%% -*- erlang -*-
+-export([main/1]).
+%% -*- encoding:latin-1 -*-
+
+main(_) ->
+    ok = io:setopts([{encoding,unicode}]),
+    Bin1 = <<"örn_Ѐ שלום-שלום+של 日本語">>,
+    L = [246,114,110,95,1024,32,1513,1500,1493,1501,45,1513,1500,1493,
+         1501,43,1513,1500,32,26085,26412,35486],
+    L = unicode:characters_to_list(Bin1, utf8),
+    ok.
diff --git a/lib/stdlib/test/escript_SUITE_data/unicode6 b/lib/stdlib/test/escript_SUITE_data/unicode6
new file mode 100755
index 0000000000..8027a2a08c
--- /dev/null
+++ b/lib/stdlib/test/escript_SUITE_data/unicode6
@@ -0,0 +1,13 @@
+#!/usr/bin/env escript
+%% -*- erlang -*-
+%%! +pc unicode
+-export([main/1]).
+%% -*- encoding:utf-8 -*-
+
+main(_) ->
+    ok = io:setopts([{encoding,unicode}]),
+    Bin1 = <<"örn_Ѐ שלום-שלום+של 日本語">>,
+    L = [246,114,110,95,1024,32,1513,1500,1493,1501,45,1513,1500,1493,
+         1501,43,1513,1500,32,26085,26412,35486],
+    L = unicode:characters_to_list(Bin1, utf8),
+    ok.