* win32/{win32.c,dir.h} (rb_w32_uopendir): new API to pass UTF-8 path.

* win32/win32.c (opendir_internal, rb_w32_opendir): extract and merge
  common part of rb_w32_opendir() and rb_w32_uopendir().

* dir.c (do_opendir, glob_helper): encoding.

* dir.c (dir_initialize, do_opendir): convert path to UTF-8 and call
  rb_w32_uopendir() instead of rb_w32_opendir() on Windows.
  fixes #4491, reported by Joey Zhou.


git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@31372 b2dd03c8-39d4-4d8f-98ff-823fe69b080e

ChangeLog

@@ -1,3 +1,16 @@
+Thu Apr 28 16:15:49 2011  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/{win32.c,dir.h} (rb_w32_uopendir): new API to pass UTF-8 path.
+
+	* win32/win32.c (opendir_internal, rb_w32_opendir): extract and merge
+	  common part of rb_w32_opendir() and rb_w32_uopendir().
+
+	* dir.c (do_opendir, glob_helper): encoding.
+
+	* dir.c (dir_initialize, do_opendir): convert path to UTF-8 and call
+	  rb_w32_uopendir() instead of rb_w32_opendir() on Windows.
+	  fixes #4491, reported by Joey Zhou.
+
 Thu Apr 28 15:32:53 2011  NAKAMURA Usaku  <usa@ruby-lang.org>
 
 	* test/dl/test_base.rb (DL::LIBC_SO): its always msvc*.dll on

dir.c

@@ -75,6 +75,8 @@ char *strchr(char*,char);
 #define mkdir(p, m) rb_w32_umkdir((p), (m))
 #undef rmdir
 #define rmdir(p) rb_w32_urmdir(p)
+#undef opendir
+#define opendir(p) rb_w32_uopendir(p)
 #endif
 
 #define FNM_NOESCAPE	0x01
@@ -402,6 +404,7 @@ dir_initialize(int argc, VALUE *argv, VALUE dir)
     }
 
     GlobPathValue(dirname, FALSE);
+    dirname = rb_str_encode_ospath(dirname);
 
     TypedData_Get_Struct(dir, struct dir_data, &dir_data_type, dp);
     if (dp->dir) closedir(dp->dir);
@@ -1033,9 +1036,20 @@ do_lstat(const char *path, struct stat *pst, int flags)
 }
 
 static DIR *
-do_opendir(const char *path, int flags)
+do_opendir(const char *path, int flags, rb_encoding *enc)
 {
-    DIR *dirp = opendir(path);
+    DIR *dirp;
+#ifdef _WIN32
+    volatile VALUE tmp;
+    if (enc != rb_usascii_encoding() &&
+	enc != rb_ascii8bit_encoding() &&
+	enc != rb_utf8_encoding()) {
+	tmp = rb_enc_str_new(path, strlen(path), enc);
+	tmp = rb_str_encode_ospath(tmp);
+	path = RSTRING_PTR(tmp);
+    }
+#endif
+    dirp = opendir(path);
     if (dirp == NULL && !to_be_ignored(errno))
 	sys_warning(path);
 
@@ -1354,7 +1368,7 @@ glob_helper(
 	struct dirent *dp;
 	DIR *dirp;
 	IF_HAVE_READDIR_R(DEFINE_STRUCT_DIRENT entry);
-	dirp = do_opendir(*path ? path : ".", flags);
+	dirp = do_opendir(*path ? path : ".", flags, enc);
 	if (dirp == NULL) return 0;
 
 	while (READDIR(dirp, enc, &STRUCT_DIRENT(entry), dp)) {

win32/dir.h

@@ -28,6 +28,7 @@ typedef struct {
 
 
 DIR*           rb_w32_opendir(const char*);
+DIR*           rb_w32_uopendir(const char*);
 struct direct* rb_w32_readdir(DIR *);
 struct direct* rb_w32_readdir_with_enc(DIR *, rb_encoding *);
 long           rb_w32_telldir(DIR *);

win32/win32.c

@@ -1690,14 +1690,30 @@ open_dir_handle(const WCHAR *filename, WIN32_FIND_DATAW *fd)
 }
 
 static DIR *
-opendir_internal(HANDLE fh, WIN32_FIND_DATAW *fd)
+opendir_internal(WCHAR *wpath, const char *filename)
 {
+    struct stati64 sbuf;
+    WIN32_FIND_DATAW fd;
+    HANDLE fh;
     DIR *p;
     long len;
     long idx;
     WCHAR *tmpW;
     char *tmp;
 
+    //
+    // check to see if we've got a directory
+    //
+    if (wstati64(wpath, &sbuf) < 0) {
+	return NULL;
+    }
+    if (!(sbuf.st_mode & S_IFDIR) &&
+	(!ISALPHA(filename[0]) || filename[1] != ':' || filename[2] != '\0' ||
+	 ((1 << ((filename[0] & 0x5f) - 'A')) & GetLogicalDrives()) == 0)) {
+	errno = ENOTDIR;
+	return NULL;
+    }
+    fh = open_dir_handle(wpath, &fd);
     if (fh == INVALID_HANDLE_VALUE) {
 	return NULL;
     }
@@ -1718,7 +1734,7 @@ opendir_internal(HANDLE fh, WIN32_FIND_DATAW *fd)
     // of the previous string found.
     //
     do {
-	len = lstrlenW(fd->cFileName) + 1;
+	len = lstrlenW(fd.cFileName) + 1;
 
 	//
 	// bump the string table size by enough for the
@@ -1734,7 +1750,7 @@ opendir_internal(HANDLE fh, WIN32_FIND_DATAW *fd)
 	}
 
 	p->start = tmpW;
-	memcpy(&p->start[idx], fd->cFileName, len * sizeof(WCHAR));
+	memcpy(&p->start[idx], fd.cFileName, len * sizeof(WCHAR));
 
 	if (p->nfiles % DIRENT_PER_CHAR == 0) {
 	    tmp = realloc(p->bits, p->nfiles / DIRENT_PER_CHAR + 1);
@@ -1743,14 +1759,14 @@ opendir_internal(HANDLE fh, WIN32_FIND_DATAW *fd)
 	    p->bits = tmp;
 	    p->bits[p->nfiles / DIRENT_PER_CHAR] = 0;
 	}
-	if (fd->dwFileAttributes & FILE_ATTRIBUTE_DIRECTORY)
+	if (fd.dwFileAttributes & FILE_ATTRIBUTE_DIRECTORY)
 	    SetBit(p->bits, BitOfIsDir(p->nfiles));
-	if (fd->dwFileAttributes & FILE_ATTRIBUTE_REPARSE_POINT)
+	if (fd.dwFileAttributes & FILE_ATTRIBUTE_REPARSE_POINT)
 	    SetBit(p->bits, BitOfIsRep(p->nfiles));
 
 	p->nfiles++;
 	idx += len;
-    } while (FindNextFileW(fh, fd));
+    } while (FindNextFileW(fh, &fd));
     FindClose(fh);
     p->size = idx;
     p->curr = p->start;
@@ -1789,31 +1805,25 @@ mbstr_to_wstr(UINT cp, const char *str, int clen, long *plen)
 DIR *
 rb_w32_opendir(const char *filename)
 {
-    struct stati64 sbuf;
+    DIR *ret;
-    WIN32_FIND_DATAW fd;
+    WCHAR *wpath = filecp_to_wstr(filename, NULL);
-    HANDLE fh;
+    if (!wpath)
-    WCHAR *wpath;
+	return NULL;
+    ret = opendir_internal(wpath, filename);
+    free(wpath);
+    return ret;
+}
 
-    if (!(wpath = filecp_to_wstr(filename, NULL)))
+DIR *
+rb_w32_uopendir(const char *filename)
+{
+    DIR *ret;
+    WCHAR *wpath = utf8_to_wstr(filename, NULL);
+    if (!wpath)
 	return NULL;
-
+    ret = opendir_internal(wpath, filename);
-    //
-    // check to see if we've got a directory
-    //
-    if (wstati64(wpath, &sbuf) < 0) {
     free(wpath);
-	return NULL;
+    return ret;
-    }
-    if (!(sbuf.st_mode & S_IFDIR) &&
-	(!ISALPHA(filename[0]) || filename[1] != ':' || filename[2] != '\0' ||
-	 ((1 << ((filename[0] & 0x5f) - 'A')) & GetLogicalDrives()) == 0)) {
-	free(wpath);
-	errno = ENOTDIR;
-	return NULL;
-    }
-    fh = open_dir_handle(wpath, &fd);
-    free(wpath);
-    return opendir_internal(fh, &fd);
 }
 
 //