Let escript recognize an encoding comment on the second line

The manual says that an Emacs directive can be placed on the second
line. With this patch that directive is also recognized when selecting
encoding of the script.

7 files changed, 57 insertions, 4 deletions

diff --git a/erts/doc/src/escript.xml b/erts/doc/src/escript.xml
index 66e904f64f..9e2a87dde6 100644
--- a/erts/doc/src/escript.xml
+++ b/erts/doc/src/escript.xml
@@ -4,7 +4,7 @@
 <comref>
   <header>
     <copyright>
-      <year>2007</year><year>2011</year>
+      <year>2007</year><year>2013</year>
       <holder>Ericsson AB. All Rights Reserved.</holder>
     </copyright>
     <legalnotice>
@@ -85,6 +85,11 @@ $ <input>escript factorial 5</input>        </pre>
           enter the major mode for editing Erlang source files. If the
           directive is present it must be located on the second
           line.</p>
+
+        <p>If there is a comment selecting the <seealso
+          marker="stdlib:epp#encoding">encoding</seealso> it can be
+          located on the second line.</p>
+
 	<p>On the third line (or second line depending on the presence
           of the Emacs directive), it is possible to give arguments to
           the emulator, such as </p>
@@ -133,7 +138,9 @@ halt(1).</pre>
         <pre>
 -include_lib("kernel/include/file.hrl").</pre>
         <p>to include the record definitions for the records used by the
-          <c>file:read_link_info/1</c> function.</p>
+          <c>file:read_link_info/1</c> function. You can also select
+          encoding here, but if there is a valid encoding comment on
+          the second line it takes precedence.</p>
 
         <p>The script will be checked for syntactic and semantic
           correctness before being run. If there are warnings (such as
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.