* win32/win32.c, include/ruby/win32.h (rb_w32_ugetenv): new API to

  accept and to return UTF-8 strings.

* win32/win32.c (rb_w32_getenv): follow above change.

* win32/win32.c (rb_w32_get_environ): returns UTF-8 environment area.

* hash.c (env_str_new, rb_f_getenv, env_fetch): follow above changes.
  [Bug #5570] [ruby-core:40737]


git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@35030 b2dd03c8-39d4-4d8f-98ff-823fe69b080e

4 files changed, 86 insertions, 8 deletions

diff --git a/ChangeLog b/ChangeLog
index 76bbe927b1..4247852ec6 100644
--- a/ChangeLog
+++ b/ChangeLog
@@ -1,3 +1,15 @@
+Thu Mar 15 12:44:50 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c, include/ruby/win32.h (rb_w32_ugetenv): new API to
+	  accept and to return UTF-8 strings.
+
+	* win32/win32.c (rb_w32_getenv): follow above change.
+
+	* win32/win32.c (rb_w32_get_environ): returns UTF-8 environment area.
+
+	* hash.c (env_str_new, rb_f_getenv, env_fetch): follow above changes.
+	  [Bug #5570] [ruby-core:40737]
+
 Thu Mar 15 10:57:27 2012  Shugo Maeda  <shugo@ruby-lang.org>
 
 	* enumerator.c (lazy_cycle): add Enumerable::Lazy#cycle.
diff --git a/hash.c b/hash.c
index 84a80bd086..ee703f59d4 100644
--- a/hash.c
+++ b/hash.c
@@ -2008,6 +2008,8 @@ static char **origenviron;
 static char **my_environ;
 #undef environ
 #define environ my_environ
+#undef getenv
+#define getenv(n) rb_w32_ugetenv(n)
 #elif defined(__APPLE__)
 #undef environ
 #define environ (*_NSGetEnviron())
@@ -2029,7 +2031,11 @@ extern char **environ;
 static VALUE
 env_str_new(const char *ptr, long len)
 {
+#ifdef _WIN32
+    VALUE str = rb_str_encode(rb_enc_str_new(ptr, len, rb_utf8_encoding()), rb_enc_from_encoding(rb_locale_encoding()), 0, Qnil);
+#else
     VALUE str = rb_locale_str_new(ptr, len);
+#endif
 
     rb_obj_freeze(str);
     return str;
@@ -2108,7 +2114,11 @@ rb_f_getenv(VALUE obj, VALUE name)
     env = getenv(nam);
     if (env) {
 	if (ENVMATCH(nam, PATH_ENV) && !env_path_tainted(env)) {
+#ifdef _WIN32
+	    VALUE str = rb_str_encode(rb_enc_str_new(env, strlen(env), rb_utf8_encoding()), rb_enc_from_encoding(rb_filesystem_encoding()), 0, Qnil);
+#else
 	    VALUE str = rb_filesystem_str_new_cstr(env);
+#endif
 
 	    rb_obj_freeze(str);
 	    return str;
@@ -2159,7 +2169,11 @@ env_fetch(int argc, VALUE *argv)
 	return if_none;
     }
     if (ENVMATCH(nam, PATH_ENV) && !env_path_tainted(env))
+#ifdef _WIN32
+	return rb_str_encode(rb_enc_str_new(env, strlen(env), rb_utf8_encoding()), rb_enc_from_encoding(rb_filesystem_encoding()), 0, Qnil);
+#else
 	return rb_filesystem_str_new_cstr(env);
+#endif
     return env_str_new2(env);
 }
 
diff --git a/include/ruby/win32.h b/include/ruby/win32.h
index d49a9b4f71..807d388ed9 100644
--- a/include/ruby/win32.h
+++ b/include/ruby/win32.h
@@ -268,6 +268,7 @@ extern struct servent  *WSAAPI rb_w32_getservbyname(const char *, const char *);
 extern struct servent  *WSAAPI rb_w32_getservbyport(int, const char *);
 extern int    rb_w32_socketpair(int, int, int, int *);
 extern char * rb_w32_getcwd(char *, int);
+extern char * rb_w32_ugetenv(const char *);
 extern char * rb_w32_getenv(const char *);
 extern int    rb_w32_rename(const char *, const char *);
 extern int    rb_w32_urename(const char *, const char *);
diff --git a/win32/win32.c b/win32/win32.c
index 744c61489e..37b3943036 100644
--- a/win32/win32.c
+++ b/win32/win32.c
@@ -630,6 +630,7 @@ static CRITICAL_SECTION select_mutex;
 static int NtSocketsInitialized = 0;
 static st_table *socklist = NULL;
 static char *envarea;
+static char *uenvarea;
 
 /* License: Ruby's */
 static void
@@ -648,6 +649,10 @@ exit_handler(void)
 	FreeEnvironmentStrings(envarea);
 	envarea = NULL;
     }
+    if (uenvarea) {
+	free(uenvarea);
+	uenvarea = NULL;
+    }
 }
 
 /* License: Artistic or GPL */
@@ -4236,13 +4241,58 @@ wait(int *status)
 
 /* License: Ruby's */
 char *
+rb_w32_ugetenv(const char *name)
+{
+    WCHAR *wenvarea, *wenv;
+    int len = strlen(name);
+    char *env;
+    int wlen;
+
+    if (len == 0) return NULL;
+
+    if (uenvarea) {
+	free(uenvarea);
+	uenvarea = NULL;
+    }
+    if (envarea) {
+	FreeEnvironmentStrings(envarea);
+	envarea = NULL;
+    }
+    wenvarea = GetEnvironmentStringsW();
+    if (!wenvarea) {
+	map_errno(GetLastError());
+	return NULL;
+    }
+    for (wenv = wenvarea, wlen = 1; *wenv; wenv += lstrlenW(wenv) + 1)
+	wlen += lstrlenW(wenv) + 1;
+    uenvarea = wstr_to_mbstr(CP_UTF8, wenvarea, wlen, NULL);
+    FreeEnvironmentStringsW(wenvarea);
+    if (!uenvarea)
+	return NULL;
+
+    for (env = uenvarea; *env; env += strlen(env) + 1)
+	if (strncasecmp(env, name, len) == 0 && *(env + len) == '=')
+	    return env + len + 1;
+
+    return NULL;
+}
+
+/* License: Ruby's */
+char *
 rb_w32_getenv(const char *name)
 {
     int len = strlen(name);
     char *env;
 
     if (len == 0) return NULL;
-    if (envarea) FreeEnvironmentStrings(envarea);
+    if (uenvarea) {
+	free(uenvarea);
+	uenvarea = NULL;
+    }
+    if (envarea) {
+	FreeEnvironmentStrings(envarea);
+	envarea = NULL;
+    }
     envarea = GetEnvironmentStrings();
     if (!envarea) {
 	map_errno(GetLastError());
@@ -5011,7 +5061,7 @@ rb_w32_asynchronize(asynchronous_func_t func, uintptr_t self,
 char **
 rb_w32_get_environ(void)
 {
-    char *envtop, *env;
+    WCHAR *envtop, *env;
     char **myenvtop, **myenv;
     int num;
 
@@ -5022,23 +5072,24 @@ rb_w32_get_environ(void)
      * CygWin deals these values by changing first `=' to '!'. But we don't
      * use such trick and follow cmd.exe's way that just doesn't show these
      * values.
-     * (U.N. 2001-11-15)
+     *
+     * This function returns UTF-8 strings.
      */
-    envtop = GetEnvironmentStrings();
-    for (env = envtop, num = 0; *env; env += strlen(env) + 1)
+    envtop = GetEnvironmentStringsW();
+    for (env = envtop, num = 0; *env; env += lstrlenW(env) + 1)
 	if (*env != '=') num++;
 
     myenvtop = (char **)malloc(sizeof(char *) * (num + 1));
-    for (env = envtop, myenv = myenvtop; *env; env += strlen(env) + 1) {
+    for (env = envtop, myenv = myenvtop; *env; env += lstrlenW(env) + 1) {
 	if (*env != '=') {
-	    if (!(*myenv = strdup(env))) {
+	    if (!(*myenv = wstr_to_utf8(env, NULL))) {
 		break;
 	    }
 	    myenv++;
 	}
     }
     *myenv = NULL;
-    FreeEnvironmentStrings(envtop);
+    FreeEnvironmentStringsW(envtop);
 
     return myenvtop;
 }