load.c: search in OS path encoding

* load.c (rb_load_internal): use rb_load_file_str() to keep path
  encoding.
* load.c (rb_require_safe): search in OS path encoding for Windows.
* ruby.c (rb_load_file_str): load file with keeping path encoding.
* win32/file.c (rb_file_load_ok): use WCHAR type API assuming incoming
  path is encoded in UTF-8.  [ruby-core:56136] [Bug #8676]

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@42183 b2dd03c8-39d4-4d8f-98ff-823fe69b080e

6 files changed, 42 insertions, 10 deletions

diff --git a/ChangeLog b/ChangeLog
index e5285235c5..8f8862805c 100644
--- a/ChangeLog
+++ b/ChangeLog
@@ -1,4 +1,14 @@
-Fri Jul 26 13:01:57 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+Fri Jul 26 13:04:15 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* load.c (rb_load_internal): use rb_load_file_str() to keep path
+	  encoding.
+
+	* load.c (rb_require_safe): search in OS path encoding for Windows.
+
+	* ruby.c (rb_load_file_str): load file with keeping path encoding.
+
+	* win32/file.c (rb_file_load_ok): use WCHAR type API assuming incoming
+	  path is encoded in UTF-8.  [ruby-core:56136] [Bug #8676]
 
 	* file.c (rb_str_encode_ospath): simplify using rb_str_conv_enc().
 
diff --git a/include/ruby/intern.h b/include/ruby/intern.h
index 21adcca752..f30a94d6b5 100644
--- a/include/ruby/intern.h
+++ b/include/ruby/intern.h
@@ -644,6 +644,7 @@ int rb_reg_options(VALUE);
 RUBY_EXTERN VALUE rb_argv0;
 VALUE rb_get_argv(void);
 void *rb_load_file(const char*);
+void *rb_load_file_str(VALUE);
 /* signal.c */
 VALUE rb_f_kill(int, VALUE*);
 #ifdef POSIX_SIGNAL
diff --git a/load.c b/load.c
index 9d95c95fff..03ce062f14 100644
--- a/load.c
+++ b/load.c
@@ -591,7 +591,7 @@ rb_load_internal(VALUE fname, int wrap)
 	VALUE iseq;
 
 	th->mild_compile_error++;
-	node = (NODE *)rb_load_file(RSTRING_PTR(fname));
+	node = (NODE *)rb_load_file_str(fname);
 	loaded = TRUE;
 	iseq = rb_iseq_new_top(node, rb_str_new2("<top (required)>"), fname, rb_realpath_internal(Qnil, fname, 1), Qfalse);
 	th->mild_compile_error--;
@@ -941,7 +941,8 @@ rb_require_safe(VALUE fname, int safe)
 					   rb_sourceline());
 	}
 
-	found = search_required(fname, &path, safe);
+	path = rb_str_encode_ospath(fname);
+	found = search_required(path, &path, safe);
 
 	if (RUBY_DTRACE_FIND_REQUIRE_RETURN_ENABLED()) {
 	    RUBY_DTRACE_FIND_REQUIRE_RETURN(StringValuePtr(fname),
diff --git a/ruby.c b/ruby.c
index 2b2e548de8..df3f089267 100644
--- a/ruby.c
+++ b/ruby.c
@@ -1768,8 +1768,14 @@ load_file(VALUE parser, VALUE fname, int script, struct cmdline_options *opt)
 void *
 rb_load_file(const char *fname)
 {
-    struct cmdline_options opt;
     VALUE fname_v = rb_str_new_cstr(fname);
+    return rb_load_file_str(fname_v);
+}
+
+void *
+rb_load_file_str(VALUE fname_v)
+{
+    struct cmdline_options opt;
 
     return load_file(rb_parser_new(), fname_v, 0, cmdline_options_init(&opt));
 }
diff --git a/test/ruby/test_require.rb b/test/ruby/test_require.rb
index 8d40f3dd80..41cbff5c02 100644
--- a/test/ruby/test_require.rb
+++ b/test/ruby/test_require.rb
@@ -61,10 +61,17 @@ class TestRequire < Test::Unit::TestCase
 
   def test_require_nonascii_path
     bug8165 = '[ruby-core:53733] [Bug #8165]'
-    encoding = /mswin|mingw/ =~ RUBY_PLATFORM ? 'filesystem' : 'UTF-8'
+    encoding = 'filesystem'
     assert_require_nonascii_path(encoding, bug8165)
   end
 
+  def test_require_nonascii_path_utf8
+    bug8676 = '[ruby-core:56136] [Bug #8676]'
+    encoding = Encoding::UTF_8
+    return if Encoding.find('filesystem') == encoding
+    assert_require_nonascii_path(encoding, bug8676)
+  end
+
   def assert_require_nonascii_path(encoding, bug)
     Dir.mktmpdir {|tmp|
       dir = "\u3042" * 5
diff --git a/win32/file.c b/win32/file.c
index f7b419530f..d43db14c50 100644
--- a/win32/file.c
+++ b/win32/file.c
@@ -681,16 +681,22 @@ rb_file_expand_path_internal(VALUE fname, VALUE dname, int abs_mode, int long_na
 int
 rb_file_load_ok(const char *path)
 {
+    DWORD attr;
     int ret = 1;
-    DWORD attr = GetFileAttributes(path);
+    size_t len;
+    wchar_t* wpath;
+
+    convert_mb_to_wchar(path, &wpath, &len, CP_UTF8);
+
+    attr = GetFileAttributesW(wpath);
     if (attr == INVALID_FILE_ATTRIBUTES ||
-	attr & FILE_ATTRIBUTE_DIRECTORY) {
+	(attr & FILE_ATTRIBUTE_DIRECTORY)) {
 	ret = 0;
     }
     else {
-	HANDLE h = CreateFile(path, GENERIC_READ,
-			      FILE_SHARE_READ | FILE_SHARE_WRITE,
-			      NULL, OPEN_EXISTING, FILE_ATTRIBUTE_NORMAL, NULL);
+	HANDLE h = CreateFileW(wpath, GENERIC_READ,
+			       FILE_SHARE_READ | FILE_SHARE_WRITE,
+			       NULL, OPEN_EXISTING, FILE_ATTRIBUTE_NORMAL, NULL);
 	if (h != INVALID_HANDLE_VALUE) {
 	    CloseHandle(h);
 	}
@@ -698,6 +704,7 @@ rb_file_load_ok(const char *path)
 	    ret = 0;
 	}
     }
+    xfree(wpath);
     return ret;
 }