Dep: Add efsw (entropia file system watcher)

* Uses system API's which reduces the overhead
  instead of checking periodically for changes.
* Will be used in the hotswap system to reload
  shared libraries on changes.

(cherry picked from commit 46daaf768911f91a9a65bebc7e3fa62400c4ccad)

79 files changed, 10579 insertions, 1 deletions

diff --git a/dep/CMakeLists.txt b/dep/CMakeLists.txt
index aeffd5b8469..0932c7d4058 100644
--- a/dep/CMakeLists.txt
+++ b/dep/CMakeLists.txt
@@ -34,6 +34,7 @@ if(SERVERS)
   add_subdirectory(readline)
   add_subdirectory(gsoap)
   add_subdirectory(process)
+  add_subdirectory(efsw)
 endif()
 
 if(TOOLS)
diff --git a/dep/PackageList.txt b/dep/PackageList.txt
index f855e8d493d..66421127162 100644
--- a/dep/PackageList.txt
+++ b/dep/PackageList.txt
@@ -16,6 +16,10 @@ cppformat (type safe format library)
   https://github.com/cppformat/cppformat
   Version: 5174b8ca281426af604b85fdf53be8a748b33f56
 
+efws (Entropia File System Watcher - crossplatform file system watcher)
+  https://bitbucket.org/SpartanJ/efsw
+  ff0b69daeca1edf7785a8a580518e462be5a6f3d
+
 G3D (a commercial-grade C++ 3D engine available as Open Source (BSD License)
   http://g3d.sourceforge.net/
   Version: 9.0-Release r4036
diff --git a/dep/efsw/.hg_archival.txt b/dep/efsw/.hg_archival.txt
new file mode 100644
index 00000000000..19565afb668
--- /dev/null
+++ b/dep/efsw/.hg_archival.txt
@@ -0,0 +1,5 @@
+repo: 78c2ea8c48b213ee0078d6326a1dd719d0844764
+node: ff0b69daeca1edf7785a8a580518e462be5a6f3d
+branch: default
+latesttag: null
+latesttagdistance: 144
diff --git a/dep/efsw/CMakeLists.txt b/dep/efsw/CMakeLists.txt
new file mode 100644
index 00000000000..81a1d20b204
--- /dev/null
+++ b/dep/efsw/CMakeLists.txt
@@ -0,0 +1,85 @@
+if (WITH_DYNAMIC_LINKING)
+  set(SRCS
+    src/efsw/DirectorySnapshot.cpp
+    src/efsw/DirectorySnapshotDiff.cpp
+    src/efsw/DirWatcherGeneric.cpp
+    src/efsw/FileInfo.cpp
+    src/efsw/FileSystem.cpp
+    src/efsw/FileWatcher.cpp
+    src/efsw/FileWatcherCWrapper.cpp
+    src/efsw/FileWatcherGeneric.cpp
+    src/efsw/FileWatcherImpl.cpp
+    src/efsw/Log.cpp
+    src/efsw/Mutex.cpp
+    src/efsw/String.cpp
+    src/efsw/System.cpp
+    src/efsw/Thread.cpp
+    src/efsw/Watcher.cpp
+    src/efsw/WatcherGeneric.cpp)
+
+  if(WIN32)
+    list(APPEND SRCS
+      src/efsw/platform/win/FileSystemImpl.cpp
+      src/efsw/platform/win/MutexImpl.cpp
+      src/efsw/platform/win/SystemImpl.cpp
+      src/efsw/platform/win/ThreadImpl.cpp)
+  else()
+    list(APPEND SRCS
+      src/efsw/platform/posix/FileSystemImpl.cpp
+      src/efsw/platform/posix/MutexImpl.cpp
+      src/efsw/platform/posix/SystemImpl.cpp
+      src/efsw/platform/posix/ThreadImpl.cpp)
+  endif()
+
+  if (WIN32)
+    list(APPEND SRCS
+      src/efsw/WatcherWin32.cpp
+      src/efsw/FileWatcherWin32.cpp)
+  elseif (${CMAKE_SYSTEM_NAME} MATCHES "Linux")
+    list(APPEND SRCS
+      src/efsw/FileWatcherInotify.cpp
+      src/efsw/WatcherInotify.cpp)
+
+    if (NOT EXISTS "/usr/include/sys/inotify.h" AND NOT EXISTS "/usr/local/include/sys/inotify.h")
+      add_definitions(-DEFSW_INOTIFY_NOSYS)
+    endif()
+  elseif (${CMAKE_SYSTEM_NAME} MATCHES "FreeBSD" OR APPLE)
+    list(APPEND SRCS
+      src/efsw/FileWatcherKqueue.cpp
+      src/efsw/WatcherKqueue.cpp)
+
+    if (APPLE)
+      list(APPEND SRCS
+        src/efsw/FileWatcherFSEvents.cpp
+        src/efsw/WatcherFSEvents.cpp)
+
+      exec_program(uname ARGS -v  OUTPUT_VARIABLE OSX_VERSION)
+      string(REGEX MATCH "[0-9]+" OSX_VERSION ${OSX_VERSION})
+      if (NOT OSX_VERSION GREATER 9)
+        add_definitions(-DEFSW_FSEVENTS_NOT_SUPPORTED)
+      endif()
+
+      set(OPTIONAL_MAC_LINK_LIBRARIES "-framework CoreFoundation" "-framework CoreServices")
+    endif()
+  endif()
+
+  add_library(efsw STATIC ${SRCS})
+
+  target_include_directories(efsw
+    PUBLIC
+      ${CMAKE_CURRENT_SOURCE_DIR}/include
+    PRIVATE
+      ${CMAKE_CURRENT_SOURCE_DIR}/src)
+
+  target_link_libraries(efsw
+    PUBLIC
+      threads
+      ${OPTIONAL_MAC_LINK_LIBRARIES})
+
+  set_target_properties(efsw
+    PROPERTIES
+      FOLDER
+        "dep")
+else()
+  add_library(efsw INTERFACE)
+endif()
diff --git a/dep/efsw/LICENSE b/dep/efsw/LICENSE
new file mode 100644
index 00000000000..ac8ac28988d
--- /dev/null
+++ b/dep/efsw/LICENSE
@@ -0,0 +1,22 @@
+Copyright (c) 2012 Martín Lucas Golini
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in
+all copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
+THE SOFTWARE.
+
+This software is a fork of the "simplefilewatcher" by James Wynn (james@jameswynn.com)
+http://code.google.com/p/simplefilewatcher/ also MIT licensed.
\ No newline at end of file
diff --git a/dep/efsw/README.md b/dep/efsw/README.md
new file mode 100644
index 00000000000..fda0bb599aa
--- /dev/null
+++ b/dep/efsw/README.md
@@ -0,0 +1,138 @@
+Entropia File System Watcher
+============================
+**efsw** is a C++ cross-platform file system watcher and notifier.
+
+**efsw** monitors the file system asynchronously for changes to files and directories by watching a list of specified paths, and raises events when a directory or file change.
+
+**efsw** supports recursive directories watch, tracking the entire sub directory tree.
+
+**efsw** currently supports the following platforms:
+
+* Linux via [inotify](http://en.wikipedia.org/wiki/Inotify)
+
+* Windows via [I/O Completion Ports](http://en.wikipedia.org/wiki/IOCP)
+
+* Mac OS X via [FSEvents](http://en.wikipedia.org/wiki/FSEvents) or [kqueue](http://en.wikipedia.org/wiki/Kqueue)
+
+* FreeBSD/BSD via [kqueue](http://en.wikipedia.org/wiki/Kqueue)
+
+* OS-independent generic watcher
+(polling the disk for directory snapshots and comparing them periodically)
+
+If any of the backend fails to start by any reason, it will fallback to the OS-independent implementation.
+This should never happen, except for the Kqueue implementation, see `Platform limitations and clarifications`.
+
+**Code License**
+--------------
+[MIT License](http://www.opensource.org/licenses/mit-license.php)
+
+**Some example code:**
+--------------------
+
+    :::c++
+	// Inherits from the abstract listener class, and implements the the file action handler
+	class UpdateListener : public efsw::FileWatchListener
+	{
+	public:
+		UpdateListener() {}
+
+		void handleFileAction( efsw::WatchID watchid, const std::string& dir, const std::string& filename, efsw::Action action, std::string oldFilename = "" )
+		{
+			switch( action )
+			{
+			case efsw::Actions::Add:
+				std::cout << "DIR (" << dir << ") FILE (" << filename << ") has event Added" << std::endl;
+				break;
+			case efsw::Actions::Delete:
+				std::cout << "DIR (" << dir << ") FILE (" << filename << ") has event Delete" << std::endl;
+				break;
+			case efsw::Actions::Modified:
+				std::cout << "DIR (" << dir << ") FILE (" << filename << ") has event Modified" << std::endl;
+				break;
+			case efsw::Actions::Moved:
+					std::cout << "DIR (" << dir << ") FILE (" << filename << ") has event Moved from (" << oldFilename << ")" << std::endl;
+				break;
+			default:
+				std::cout << "Should never happen!" << std::endl;
+			}
+		}
+	};
+
+	// Create the file system watcher instance
+    // efsw::FileWatcher allow a first boolean parameter that indicates if it should start with the generic file watcher instead of the platform specific backend
+	efsw::FileWatcher * fileWatcher = new efsw::FileWatcher();
+
+	// Create the instance of your efsw::FileWatcherListener implementation
+	UpdateListener * listener = new UpdateListener();
+
+	// Add a folder to watch, and get the efsw::WatchID
+	// It will watch the /tmp folder recursively ( the third parameter indicates that is recursive )
+	// Reporting the files and directories changes to the instance of the listener
+	efsw::WatchID watchID = fileWatcher->addWatch( "/tmp", listener, true );
+
+    // Adds another directory to watch. This time as non-recursive.
+    efsw::WatchID watchID2 = fileWatcher->addWatch( "/usr", listener, false );
+
+	// Start watching asynchronously the directories
+	fileWatcher.watch();
+
+	// Remove the second watcher added
+    // You can also call removeWatch by passing the watch path ( it must end with an slash or backslash in windows, since that's how internally it's saved )
+	fileWatcher->removeWatch( watchID2 );
+	
+**Dependencies**
+--------------
+None :)
+
+**Compiling**
+------------
+To generate project files you will need to [download and install](http://industriousone.com/premake/download) [Premake](http://industriousone.com/what-premake)
+
+Then you can generate the project for your platform just going to the project directory where the premake4.lua file is located and then execute:
+
+`premake4 gmake` to generate project Makefiles, then `cd make/*YOURPLATFORM*/`, and finally `make` or `make config=release` ( it will generate the static lib, the shared lib and the test application ).
+
+or 
+
+`premake4 vs2010` to generate Visual Studio 2010 project.
+
+or
+
+`premake4 xcode4` to generate Xcode 4 project.
+
+There is also a cmake file that i don't oficially support but it works just fine, provided by [Mohammed Nafees](https://bitbucket.org/binaryking).
+
+**Platform limitations and clarifications**
+-------------------------------------------
+
+Directory paths are expected to be encoded as UTF-8 strings in all platforms.
+
+handleFileAction returns UTF-8 strings in all platforms.
+
+Windows and FSEvents Mac OS X implementation can't follow symlinks ( it will ignore followSymlinks() and allowOutOfScopeLinks() ).
+
+Kqueue implementation is limited by the maximun number of file descriptors allowed per process by the OS, in the case of reaching the file descriptors limit ( in BSD around 18000 and in OS X around 10240 ) it will fallback to the generic file watcher.
+
+OS X will only use Kqueue if OS X version is below to 10.5, and this implementation needs to be compiled separately from the OS X >= 10.5 implementation. Since there's no way to compile FSEvents backend in OS X below 10.5.
+
+FSEvents for OS X Lion and beyond in some cases will generate more actions that in reality ocurred, since fine-grained implementation of FSEvents doesn't give the order of the actions retrieved, in some cases i need to guess/aproximate the order of them.
+
+Generic watcher relies on the inode information to detect file and directories renames/move. Since Windows has no concept of inodes as Unix platforms do, there is no current reliable way of determining file/directory movement on Windows without help from the Windows API ( this is replaced with Add/Delete events ).
+
+Linux versions below 2.6.13 are not supported, since inotify wasn't implemented yet. I'm not interested in support older kernels, since i don't see the point. If someone needs this open an issue in the issue tracker and i may consider implenent a dnotify backend.
+
+OS-independent watcher, Kqueue and FSEvents for OS X below 10.5 keep cache of the directories structures, to be able to detect changes in the directories. This means that there's a memory overhead for this backends.
+
+**Useful information**
+--------------------
+The project also comes with a C API wrapper, contributed by [Sepul Sepehr Taghdisian](https://bitbucket.org/sepul).
+
+There's a string manipulation class not exposed in the efsw header ( efsw::String ) that can be used to make string encoding conversion.
+
+
+**Clarifications**
+----------------
+
+This software started as a fork of the [simplefilewatcher](http://code.google.com/p/simplefilewatcher/) by James Wynn (james[at]jameswynn.com), [MIT licensed](http://www.opensource.org/licenses/mit-license.html).
+
+The icon used for the project is part of the [Haiku®'s Icons](http://www.haiku-inc.org/haiku-icons.html), [MIT licensed](http://www.opensource.org/licenses/mit-license.html).
diff --git a/dep/efsw/include/efsw/efsw.h b/dep/efsw/include/efsw/efsw.h
new file mode 100644
index 00000000000..28e63e2139e
--- /dev/null
+++ b/dep/efsw/include/efsw/efsw.h
@@ -0,0 +1,151 @@
+/**
+	@author Sepul Sepehr Taghdisian
+
+	Copyright (c) 2013 Martin Lucas Golini
+
+	Permission is hereby granted, free of charge, to any person obtaining a copy
+	of this software and associated documentation files (the "Software"), to deal
+	in the Software without restriction, including without limitation the rights
+	to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+	copies of the Software, and to permit persons to whom the Software is
+	furnished to do so, subject to the following conditions:
+
+	The above copyright notice and this permission notice shall be included in
+	all copies or substantial portions of the Software.
+
+	THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+	IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+	FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+	AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+	LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+	OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
+	THE SOFTWARE.
+
+	This software is a fork of the "simplefilewatcher" by James Wynn (james@jameswynn.com)
+	http://code.google.com/p/simplefilewatcher/ also MIT licensed.
+*/
+/** This is the C API wrapper of EFSW */
+#ifndef ESFW_H
+#define ESFW_H
+
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+#if defined(_WIN32)
+	#ifdef EFSW_DYNAMIC
+		// Windows platforms
+		#ifdef EFSW_EXPORTS
+			// From DLL side, we must export
+			#define EFSW_API __declspec(dllexport)
+		#else
+			// From client application side, we must import
+			#define EFSW_API __declspec(dllimport)
+		#endif
+	#else
+		// No specific directive needed for static build
+		#ifndef EFSW_API
+		#define EFSW_API
+		#endif
+	#endif
+#else
+	#if ( __GNUC__ >= 4 ) && defined( EFSW_EXPORTS )
+		#define EFSW_API __attribute__ ((visibility("default")))
+	#endif
+
+	// Other platforms don't need to define anything
+	#ifndef EFSW_API
+	#define EFSW_API
+	#endif
+#endif
+
+/// Type for a watch id
+typedef long efsw_watchid;
+
+/// Type for watcher
+typedef void* efsw_watcher;
+
+enum efsw_action
+{
+	EFSW_ADD = 1,		/// Sent when a file is created or renamed
+	EFSW_DELETE = 2,	/// Sent when a file is deleted or renamed
+	EFSW_MODIFIED = 3,	/// Sent when a file is modified
+	EFSW_MOVED = 4		/// Sent when a file is moved
+};
+
+enum efsw_error
+{
+	EFSW_NOTFOUND		= -1,
+	EFSW_REPEATED		= -2,
+	EFSW_OUTOFSCOPE		= -3,
+	EFSW_NOTREADABLE	= -4,
+	EFSW_REMOTE			= -5,
+	EFSW_UNSPECIFIED	= -6
+};
+
+/// Basic interface for listening for file events.
+typedef void (*efsw_pfn_fileaction_callback) (
+		efsw_watcher watcher,
+		efsw_watchid watchid,
+		const char* dir,
+		const char* filename,
+		enum efsw_action action,
+		const char* old_filename,
+		void* param
+);
+
+/**
+ * Creates a new file-watcher
+ * @param generic_mode Force the use of the Generic file watcher
+ */
+efsw_watcher EFSW_API efsw_create(int generic_mode);
+
+/// Release the file-watcher and unwatch any directories
+void EFSW_API efsw_release(efsw_watcher watcher);
+
+/// Retreive last error occured by file-watcher
+EFSW_API const char* efsw_getlasterror();
+
+/// Add a directory watch. Same as the other addWatch, but doesn't have recursive option.
+/// For backwards compatibility.
+/// On error returns WatchID with Error type.
+efsw_watchid EFSW_API efsw_addwatch(efsw_watcher watcher, const char* directory, 
+	efsw_pfn_fileaction_callback callback_fn, int recursive, void* param);
+
+/// Remove a directory watch. This is a brute force search O(nlogn).
+void EFSW_API efsw_removewatch(efsw_watcher watcher, const char* directory);
+
+/// Remove a directory watch. This is a map lookup O(logn).
+void EFSW_API efsw_removewatch_byid(efsw_watcher watcher, efsw_watchid watchid);
+
+/// Starts watching ( in other thread )
+void EFSW_API efsw_watch(efsw_watcher watcher);
+
+/**
+ * Allow recursive watchers to follow symbolic links to other directories
+ * followSymlinks is disabled by default
+ */
+void EFSW_API efsw_follow_symlinks(efsw_watcher watcher, int enable);
+
+/** @return If can follow symbolic links to directorioes */
+int EFSW_API efsw_follow_symlinks_isenabled(efsw_watcher watcher);
+
+/**
+ * When enable this it will allow symlinks to watch recursively out of the pointed directory.
+ * follorSymlinks must be enabled to this work.
+ * For example, added symlink to /home/folder, and the symlink points to /, this by default is not allowed,
+ * it's only allowed to symlink anything from /home/ and deeper. This is to avoid great levels of recursion.
+ * Enabling this could lead in infinite recursion, and crash the watcher ( it will try not to avoid this ).
+ * Buy enabling out of scope links, it will allow this behavior.
+ * allowOutOfScopeLinks are disabled by default.
+ */
+void EFSW_API efsw_allow_outofscopelinks(efsw_watcher watcher, int allow);
+
+/// @return Returns if out of scope links are allowed
+int EFSW_API efsw_outofscopelinks_isallowed(efsw_watcher watcher);
+
+#ifdef __cplusplus
+}
+#endif
+
+#endif
diff --git a/dep/efsw/include/efsw/efsw.hpp b/dep/efsw/include/efsw/efsw.hpp
new file mode 100644
index 00000000000..0693bb296f0
--- /dev/null
+++ b/dep/efsw/include/efsw/efsw.hpp
@@ -0,0 +1,197 @@
+/**
+	@author Martín Lucas Golini
+
+	Copyright (c) 2013 Martín Lucas Golini
+
+	Permission is hereby granted, free of charge, to any person obtaining a copy
+	of this software and associated documentation files (the "Software"), to deal
+	in the Software without restriction, including without limitation the rights
+	to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+	copies of the Software, and to permit persons to whom the Software is
+	furnished to do so, subject to the following conditions:
+
+	The above copyright notice and this permission notice shall be included in
+	all copies or substantial portions of the Software.
+
+	THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+	IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+	FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+	AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+	LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+	OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
+	THE SOFTWARE.
+
+	This software is a fork of the "simplefilewatcher" by James Wynn (james@jameswynn.com)
+	http://code.google.com/p/simplefilewatcher/ also MIT licensed.
+*/
+
+#ifndef ESFW_HPP
+#define ESFW_HPP
+
+#include <string>
+#include <list>
+
+#if defined(_WIN32)
+	#ifdef EFSW_DYNAMIC
+		// Windows platforms
+		#ifdef EFSW_EXPORTS
+			// From DLL side, we must export
+			#define EFSW_API __declspec(dllexport)
+		#else
+			// From client application side, we must import
+			#define EFSW_API __declspec(dllimport)
+		#endif
+	#else
+		// No specific directive needed for static build
+		#ifndef EFSW_API
+		#define EFSW_API
+		#endif
+	#endif
+#else
+	#if ( __GNUC__ >= 4 ) && defined( EFSW_EXPORTS )
+		#define EFSW_API __attribute__ ((visibility("default")))
+	#endif
+
+	// Other platforms don't need to define anything
+	#ifndef EFSW_API
+	#define EFSW_API
+	#endif
+#endif
+
+namespace efsw {
+
+/// Type for a watch id
+typedef long WatchID;
+
+// forward declarations
+class FileWatcherImpl;
+class FileWatchListener;
+
+/// Actions to listen for. Rename will send two events, one for
+/// the deletion of the old file, and one for the creation of the
+/// new file.
+namespace Actions {
+	enum Action
+	{
+		/// Sent when a file is created or renamed
+		Add = 1,
+		/// Sent when a file is deleted or renamed
+		Delete = 2,
+		/// Sent when a file is modified
+		Modified = 3,
+		/// Sent when a file is moved
+		Moved = 4
+	};
+}
+typedef Actions::Action Action;
+
+/// Errors log namespace
+namespace Errors {
+
+enum Error
+{
+	FileNotFound	= -1,
+	FileRepeated	= -2,
+	FileOutOfScope	= -3,
+	FileNotReadable	= -4,
+	FileRemote		= -5, /** Directory in remote file system ( create a generic FileWatcher instance to watch this directory ). */
+	Unspecified		= -6
+};
+
+class EFSW_API Log
+{
+	public:
+		/// @return The last error logged
+		static std::string getLastErrorLog();
+
+		/// Creates an error of the type specified
+		static Error createLastError( Error err, std::string log );
+};
+
+}
+typedef Errors::Error Error;
+
+/// Listens to files and directories and dispatches events
+/// to notify the listener of files and directories changes.
+/// @class FileWatcher
+class EFSW_API FileWatcher
+{
+	public:
+		/// Default constructor, will use the default platform file watcher
+		FileWatcher();
+
+		/// Constructor that lets you force the use of the Generic File Watcher
+		FileWatcher( bool useGenericFileWatcher );
+
+		virtual ~FileWatcher();
+
+		/// Add a directory watch. Same as the other addWatch, but doesn't have recursive option.
+		/// For backwards compatibility.
+		/// On error returns WatchID with Error type.
+		WatchID addWatch(const std::string& directory, FileWatchListener* watcher);
+
+		/// Add a directory watch
+		/// On error returns WatchID with Error type.
+		WatchID addWatch(const std::string& directory, FileWatchListener* watcher, bool recursive);
+
+		/// Remove a directory watch. This is a brute force search O(nlogn).
+		void removeWatch(const std::string& directory);
+
+		/// Remove a directory watch. This is a map lookup O(logn).
+		void removeWatch(WatchID watchid);
+
+		/// Starts watching ( in other thread )
+		void watch();
+
+		/// @return Returns a list of the directories that are being watched
+		std::list<std::string> directories();
+
+		/** Allow recursive watchers to follow symbolic links to other directories
+		* followSymlinks is disabled by default
+		*/
+		void followSymlinks( bool follow );
+
+		/** @return If can follow symbolic links to directorioes */
+		const bool& followSymlinks() const;
+
+		/** When enable this it will allow symlinks to watch recursively out of the pointed directory.
+		* follorSymlinks must be enabled to this work.
+		* For example, added symlink to /home/folder, and the symlink points to /, this by default is not allowed,
+		* it's only allowed to symlink anything from /home/ and deeper. This is to avoid great levels of recursion.
+		* Enabling this could lead in infinite recursion, and crash the watcher ( it will try not to avoid this ).
+		* Buy enabling out of scope links, it will allow this behavior.
+		* allowOutOfScopeLinks are disabled by default.
+		*/
+		void allowOutOfScopeLinks( bool allow );
+
+		/// @return Returns if out of scope links are allowed
+		const bool& allowOutOfScopeLinks() const;
+	private:
+		/// The implementation
+		FileWatcherImpl	*	mImpl;
+		bool				mFollowSymlinks;
+		bool				mOutOfScopeLinks;
+};
+
+/// Basic interface for listening for file events.
+/// @class FileWatchListener
+class FileWatchListener
+{
+	public:
+		FileWatchListener() {}
+
+		virtual ~FileWatchListener() {}
+
+		/// Handles the action file action
+		/// @param watchid The watch id for the directory
+		/// @param dir The directory
+		/// @param filename The filename that was accessed (not full path)
+		/// @param action Action that was performed
+		/// @param oldFilename The name of the file or directory moved
+		virtual void handleFileAction(WatchID watchid, const std::string& dir, const std::string& filename, Action action, std::string oldFilename = "" ) = 0;
+
+};
+
+}
+
+#endif
diff --git a/dep/efsw/src/efsw/Debug.cpp b/dep/efsw/src/efsw/Debug.cpp
new file mode 100644
index 00000000000..9c4ee02e5b6
--- /dev/null
+++ b/dep/efsw/src/efsw/Debug.cpp
@@ -0,0 +1,85 @@
+#include <efsw/Debug.hpp>
+#include <iostream>
+
+#ifdef EFSW_COMPILER_MSVC
+#define WIN32_LEAN_AND_MEAN
+#include <windows.h>
+#include <crtdbg.h>
+#endif
+
+#include <cassert>
+#include <cstdio>
+#include <cstdarg>
+
+namespace efsw {
+
+#ifdef DEBUG
+
+void efREPORT_ASSERT( const char * File, int Line, const char * Exp )
+{
+	#ifdef EFSW_COMPILER_MSVC
+		_CrtDbgReport( _CRT_ASSERT, File, Line, "", Exp);
+
+		DebugBreak();
+	#else
+		std::cout << "ASSERT: " << Exp << " file: " << File << " line: " << Line << std::endl;
+
+		#if defined(EFSW_COMPILER_GCC) && defined(EFSW_32BIT) && !defined(EFSW_ARM)
+			asm("int3");
+		#else
+			assert( false );
+		#endif
+	#endif
+}
+
+void efPRINT( const char * format, ... )
+{
+	char		buf[2048];
+	va_list		args;
+
+	va_start( args, format );
+
+	#ifdef EFSW_COMPILER_MSVC
+		_vsnprintf_s( buf, sizeof( buf ), sizeof( buf ) / sizeof( buf[0]), format, args );
+	#else
+		vsnprintf( buf, sizeof( buf ) / sizeof( buf[0]), format, args );
+	#endif
+
+	va_end( args );
+
+	#ifdef EFSW_COMPILER_MSVC
+		OutputDebugStringA( buf );
+	#else
+		std::cout << buf;
+	#endif
+}
+
+void efPRINTC( unsigned int cond, const char * format, ...)
+{
+	if ( 0 == cond )
+		return;
+
+	char		buf[2048];
+	va_list		args;
+
+	va_start( args, format );
+
+	#ifdef EFSW_COMPILER_MSVC
+		_vsnprintf_s( buf, efARRAY_SIZE( buf ), efARRAY_SIZE( buf ), format, args );
+	#else
+		vsnprintf( buf, sizeof( buf ) / sizeof( buf[0]), format, args );
+	#endif
+
+	va_end( args );
+
+	#ifdef EFSW_COMPILER_MSVC
+		OutputDebugStringA( buf );
+	#else
+		std::cout << buf;
+	#endif
+}
+
+#endif
+
+}
+
diff --git a/dep/efsw/src/efsw/Debug.hpp b/dep/efsw/src/efsw/Debug.hpp
new file mode 100644
index 00000000000..75d6dce0466
--- /dev/null
+++ b/dep/efsw/src/efsw/Debug.hpp
@@ -0,0 +1,50 @@
+#ifndef EFSW_DEBUG_HPP
+#define EFSW_DEBUG_HPP
+
+#include <efsw/base.hpp>
+
+namespace efsw {
+
+#ifdef DEBUG
+
+void efREPORT_ASSERT( const char * File, const int Line, const char * Exp );
+
+#define efASSERT( expr )		if ( !(expr) ) { efREPORT_ASSERT( __FILE__, __LINE__, #expr	);	}
+#define efASSERTM( expr, msg )	if ( !(expr) ) { efREPORT_ASSERT( __FILE__, __LINE__, #msg	);	}
+
+void efPRINT 	( const char * format, ... );
+void efPRINTC	( unsigned int cond, const char * format, ... );
+
+#else
+
+#define efASSERT( expr )
+#define efASSERTM( expr, msg )
+
+#ifndef EFSW_COMPILER_MSVC
+	#define efPRINT( format, args... ) {}
+	#define efPRINTC( cond, format, args... ) {}
+#else
+	#define efPRINT
+	#define efPRINTC
+#endif
+
+#endif
+
+#ifdef EFSW_VERBOSE
+	#define efDEBUG efPRINT
+	#define efDEBUGC efPRINTC
+#else
+
+	#ifndef EFSW_COMPILER_MSVC
+		#define efDEBUG( format, args... ) {}
+		#define efDEBUGC( cond, format, args... ) {}
+	#else
+		#define efDEBUG
+		#define efDEBUGC
+	#endif
+
+#endif
+
+}
+
+#endif
diff --git a/dep/efsw/src/efsw/DirWatcherGeneric.cpp b/dep/efsw/src/efsw/DirWatcherGeneric.cpp
new file mode 100644
index 00000000000..b80c14d24ce
--- /dev/null
+++ b/dep/efsw/src/efsw/DirWatcherGeneric.cpp
@@ -0,0 +1,451 @@
+#include <efsw/DirWatcherGeneric.hpp>
+#include <efsw/FileSystem.hpp>
+#include <efsw/Debug.hpp>
+#include <efsw/String.hpp>
+
+namespace efsw {
+
+DirWatcherGeneric::DirWatcherGeneric( DirWatcherGeneric * parent, WatcherGeneric * ws, const std::string& directory, bool recursive, bool reportNewFiles ) :
+	Parent( parent ),
+	Watch( ws ),
+	Recursive( recursive ),
+	Deleted( false )
+{
+	resetDirectory( directory );
+
+	if ( !reportNewFiles )
+	{
+		DirSnap.scan();
+	}
+	else
+	{
+		DirectorySnapshotDiff Diff = DirSnap.scan();
+
+		if ( Diff.changed() )
+		{
+			FileInfoList::iterator it;
+
+			DiffIterator( FilesCreated )
+			{
+				handleAction( ( *it ).Filepath, Actions::Add );
+			}
+		}
+	}
+}
+
+DirWatcherGeneric::~DirWatcherGeneric()
+{
+	/// If the directory was deleted mark the files as deleted
+	if ( Deleted )
+	{
+		DirectorySnapshotDiff Diff = DirSnap.scan();
+
+		if ( !DirSnap.exists() )
+		{
+			FileInfoList::iterator it;
+
+			DiffIterator( FilesDeleted )
+			{
+				handleAction( (*it).Filepath, Actions::Delete );
+			}
+
+			DiffIterator( DirsDeleted )
+			{
+				handleAction( (*it).Filepath, Actions::Delete );
+			}
+		}
+	}
+
+	DirWatchMap::iterator it = Directories.begin();
+
+	for ( ; it != Directories.end(); it++ )
+	{
+		if ( Deleted )
+		{
+			/// If the directory was deleted, mark the flag for file deletion
+			it->second->Deleted = true;
+		}
+
+		efSAFE_DELETE( it->second );
+	}
+}
+
+void DirWatcherGeneric::resetDirectory( std::string directory )
+{
+	std::string dir( directory );
+
+	/// Is this a recursive watch?
+	if ( Watch->Directory != directory )
+	{
+		if ( !( directory.size() && ( directory.at(0) == FileSystem::getOSSlash() || directory.at( directory.size() - 1 ) == FileSystem::getOSSlash() ) ) )
+		{
+			/// Get the real directory
+			if ( NULL != Parent )
+			{
+				FileSystem::dirAddSlashAtEnd(directory);
+
+				dir = Parent->DirSnap.DirectoryInfo.Filepath + directory;
+			}
+			else
+			{
+				efDEBUG( "resetDirectory(): Parent is NULL. Fatal error." );
+			}
+		}
+	}
+
+	DirSnap.setDirectoryInfo( dir );
+}
+
+void DirWatcherGeneric::handleAction( const std::string &filename, unsigned long action, std::string oldFilename)
+{
+	Watch->Listener->handleFileAction( Watch->ID, DirSnap.DirectoryInfo.Filepath, FileSystem::fileNameFromPath( filename ), (Action)action, oldFilename );
+}
+
+void DirWatcherGeneric::addChilds( bool reportNewFiles )
+{
+	if ( Recursive )
+	{
+		/// Create the subdirectories watchers
+		std::string dir;
+
+		for ( FileInfoMap::iterator it = DirSnap.Files.begin(); it != DirSnap.Files.end(); it++ )
+		{
+			if ( it->second.isDirectory() && it->second.isReadable() && !FileSystem::isRemoteFS( it->second.Filepath ) )
+			{
+				/// Check if the directory is a symbolic link
+				std::string curPath;
+				std::string link( FileSystem::getLinkRealPath( it->second.Filepath, curPath ) );
+
+				dir = it->first;
+
+				if ( "" != link )
+				{
+					/// Avoid adding symlinks directories if it's now enabled
+					if ( !Watch->WatcherImpl->mFileWatcher->followSymlinks() )
+					{
+						continue;
+					}
+
+					/// If it's a symlink check if the realpath exists as a watcher, or
+					/// if the path is outside the current dir
+					if ( Watch->WatcherImpl->pathInWatches( link ) || Watch->pathInWatches( link ) || !Watch->WatcherImpl->linkAllowed( curPath, link ) )
+					{
+						continue;
+					}
+					else
+					{
+						dir = link;
+					}
+				}
+				else
+				{
+					if ( Watch->pathInWatches( dir ) || Watch->WatcherImpl->pathInWatches( dir ) )
+					{
+						continue;
+					}
+				}
+
+				if ( reportNewFiles )
+				{
+					handleAction( dir, Actions::Add );
+				}
+
+				Directories[dir] = new DirWatcherGeneric( this, Watch, dir, Recursive, reportNewFiles );
+
+				Directories[dir]->addChilds( reportNewFiles  );
+			}
+		}
+	}
+}
+
+void DirWatcherGeneric::watch( bool reportOwnChange )
+{
+	DirectorySnapshotDiff Diff = DirSnap.scan();
+
+	if ( reportOwnChange && Diff.DirChanged && NULL != Parent )
+	{
+		Watch->Listener->handleFileAction( Watch->ID, FileSystem::pathRemoveFileName( DirSnap.DirectoryInfo.Filepath ), FileSystem::fileNameFromPath( DirSnap.DirectoryInfo.Filepath ), Actions::Modified );
+	}
+
+	if ( Diff.changed() )
+	{
+		FileInfoList::iterator it;
+		MovedList::iterator mit;
+
+		/// Files
+		DiffIterator( FilesCreated )
+		{
+			handleAction( (*it).Filepath, Actions::Add );
+		}
+
+		DiffIterator( FilesModified )
+		{
+			handleAction( (*it).Filepath, Actions::Modified );
+		}
+
+		DiffIterator( FilesDeleted )
+		{
+			handleAction( (*it).Filepath, Actions::Delete );
+		}
+
+		DiffMovedIterator( FilesMoved )
+		{
+			handleAction( (*mit).second.Filepath, Actions::Moved, (*mit).first );
+		}
+
+		/// Directories
+		DiffIterator( DirsCreated )
+		{
+			createDirectory( (*it).Filepath );
+		}
+
+		DiffIterator( DirsModified )
+		{
+			handleAction( (*it).Filepath, Actions::Modified );
+		}
+
+		DiffIterator( DirsDeleted )
+		{
+			handleAction( (*it).Filepath, Actions::Delete );
+			removeDirectory( (*it).Filepath );
+		}
+
+		DiffMovedIterator( DirsMoved )
+		{
+			handleAction( (*mit).second.Filepath, Actions::Moved, (*mit).first );
+			moveDirectory( (*mit).first, (*mit).second.Filepath );
+		}
+	}
+
+	/// Process the subdirectories looking for changes
+	for ( DirWatchMap::iterator dit = Directories.begin(); dit != Directories.end(); dit++ )
+	{
+		/// Just watch
+		dit->second->watch();
+	}
+}
+
+void DirWatcherGeneric::watchDir( std::string &dir )
+{
+	DirWatcherGeneric * watcher = Watch->WatcherImpl->mFileWatcher->allowOutOfScopeLinks() ?
+									findDirWatcher( dir ) :
+									findDirWatcherFast( dir );
+
+	if ( NULL != watcher )
+	{
+		watcher->watch( true );
+	}
+}
+
+DirWatcherGeneric * DirWatcherGeneric::findDirWatcherFast( std::string dir )
+{
+	// remove the common base ( dir should always start with the same base as the watcher )
+	efASSERT( !dir.empty() );
+	efASSERT( dir.size() >= DirSnap.DirectoryInfo.Filepath.size() );
+	efASSERT( DirSnap.DirectoryInfo.Filepath == dir.substr( 0, DirSnap.DirectoryInfo.Filepath.size() ) );
+
+	if ( dir.size() >= DirSnap.DirectoryInfo.Filepath.size() )
+	{
+		dir = dir.substr( DirSnap.DirectoryInfo.Filepath.size() - 1 );
+	}
+
+	if ( dir.size() == 1 )
+	{
+		efASSERT( dir[0] == FileSystem::getOSSlash() );
+		return this;
+	}
+
+	size_t level = 0;
+	std::vector<std::string> dirv = String::split( dir, FileSystem::getOSSlash(), false );
+
+	DirWatcherGeneric * watcher = this;
+
+	while ( level < dirv.size() )
+	{
+		// search the dir level in the current watcher
+		DirWatchMap::iterator it = watcher->Directories.find( dirv[ level ] );
+
+		// found? continue with the next level
+		if ( it != watcher->Directories.end() )
+		{
+			watcher = it->second;
+
+			level++;
+		}
+		else
+		{
+			// couldn't found the folder level?
+			// directory not watched
+			return NULL;
+		}
+	}
+
+	return watcher;
+}
+
+DirWatcherGeneric * DirWatcherGeneric::findDirWatcher( std::string dir )
+{
+	if ( DirSnap.DirectoryInfo.Filepath == dir )
+	{
+		return this;
+	}
+	else
+	{
+		DirWatcherGeneric * watcher = NULL;
+
+		for ( DirWatchMap::iterator it = Directories.begin(); it != Directories.end(); it++ )
+		{
+			watcher = it->second->findDirWatcher( dir );
+
+			if ( NULL != watcher )
+			{
+				return watcher;
+			}
+		}
+	}
+
+	return NULL;
+}
+
+DirWatcherGeneric * DirWatcherGeneric::createDirectory( std::string newdir )
+{
+	FileSystem::dirRemoveSlashAtEnd( newdir );
+	newdir = FileSystem::fileNameFromPath( newdir );
+
+	DirWatcherGeneric * dw = NULL;
+
+	/// Check if the directory is a symbolic link
+	std::string dir( DirSnap.DirectoryInfo.Filepath + newdir );
+
+	FileSystem::dirAddSlashAtEnd( dir );
+
+	FileInfo fi( dir );
+
+	if ( !fi.isDirectory() || !fi.isReadable() || FileSystem::isRemoteFS( dir ) )
+	{
+		return NULL;
+	}
+
+	std::string curPath;
+	std::string link( FileSystem::getLinkRealPath( dir, curPath ) );
+	bool skip = false;
+
+	if ( "" != link )
+	{
+		/// Avoid adding symlinks directories if it's now enabled
+		if ( !Watch->WatcherImpl->mFileWatcher->followSymlinks() )
+		{
+			skip = true;
+		}
+
+		/// If it's a symlink check if the realpath exists as a watcher, or
+		/// if the path is outside the current dir
+		if ( Watch->WatcherImpl->pathInWatches( link ) || Watch->pathInWatches( link ) || !Watch->WatcherImpl->linkAllowed( curPath, link ) )
+		{
+			skip = true;
+		}
+		else
+		{
+			dir = link;
+		}
+	}
+	else
+	{
+		if ( Watch->pathInWatches( dir ) || Watch->WatcherImpl->pathInWatches( dir ) )
+		{
+			skip = true;
+		}
+	}
+
+	if ( !skip )
+	{
+		handleAction( newdir, Actions::Add );
+
+		/// Creates the new directory watcher of the subfolder and check for new files
+		dw = new DirWatcherGeneric( this, Watch, dir, Recursive );
+		
+		dw->addChilds();
+
+		dw->watch();
+
+		/// Add it to the list of directories
+		Directories[ newdir ] = dw;
+	}
+
+	return dw;
+}
+
+void DirWatcherGeneric::removeDirectory( std::string dir )
+{
+	FileSystem::dirRemoveSlashAtEnd( dir );
+	dir = FileSystem::fileNameFromPath( dir );
+
+	DirWatcherGeneric * dw			= NULL;
+	DirWatchMap::iterator dit;
+
+	/// Folder deleted
+
+	/// Search the folder, it should exists
+	dit = Directories.find( dir );
+
+	if ( dit != Directories.end() )
+	{
+		dw = dit->second;
+
+		/// Flag it as deleted so it fire the event for every file inside deleted
+		dw->Deleted = true;
+
+		/// Delete the DirWatcherGeneric
+		efSAFE_DELETE( dw );
+
+		/// Remove the directory from the map
+		Directories.erase( dit->first );
+	}
+}
+
+void DirWatcherGeneric::moveDirectory( std::string oldDir, std::string newDir )
+{
+	FileSystem::dirRemoveSlashAtEnd( oldDir );
+	oldDir = FileSystem::fileNameFromPath( oldDir );
+
+	FileSystem::dirRemoveSlashAtEnd( newDir );
+	newDir = FileSystem::fileNameFromPath( newDir );
+
+	DirWatcherGeneric * dw			= NULL;
+	DirWatchMap::iterator dit;
+
+	/// Directory existed?
+	dit = Directories.find( oldDir );
+
+	if ( dit != Directories.end() )
+	{
+		dw = dit->second;
+
+		/// Remove the directory from the map
+		Directories.erase( dit->first );
+
+		Directories[ newDir ] = dw;
+
+		dw->resetDirectory( newDir );
+	}
+}
+
+bool DirWatcherGeneric::pathInWatches( std::string path )
+{
+	if ( DirSnap.DirectoryInfo.Filepath == path )
+	{
+		return true;
+	}
+
+	for ( DirWatchMap::iterator it = Directories.begin(); it != Directories.end(); it++ )
+	{
+		 if ( it->second->pathInWatches( path ) )
+		 {
+			 return true;
+		 }
+	}
+
+	return false;
+}
+
+} 
diff --git a/dep/efsw/src/efsw/DirWatcherGeneric.hpp b/dep/efsw/src/efsw/DirWatcherGeneric.hpp
new file mode 100644
index 00000000000..a7581904422
--- /dev/null
+++ b/dep/efsw/src/efsw/DirWatcherGeneric.hpp
@@ -0,0 +1,55 @@
+#ifndef EFSW_DIRWATCHERGENERIC_HPP
+#define EFSW_DIRWATCHERGENERIC_HPP
+
+#include <efsw/WatcherGeneric.hpp>
+#include <efsw/FileInfo.hpp>
+#include <efsw/DirectorySnapshot.hpp>
+#include <map>
+
+namespace efsw {
+
+class DirWatcherGeneric
+{
+	public:
+		typedef std::map<std::string, DirWatcherGeneric*> DirWatchMap;
+
+		DirWatcherGeneric *	Parent;
+		WatcherGeneric *	Watch;
+		DirectorySnapshot	DirSnap;
+		DirWatchMap			Directories;
+		bool				Recursive;
+
+		DirWatcherGeneric( DirWatcherGeneric * parent, WatcherGeneric * ws, const std::string& directory, bool recursive, bool reportNewFiles = false );
+
+		~DirWatcherGeneric();
+
+		void watch( bool reportOwnChange = false );
+
+		void watchDir( std::string& dir );
+
+		static bool isDir( const std::string& directory );
+
+		bool pathInWatches( std::string path );
+
+		void addChilds( bool reportNewFiles = true );
+
+		DirWatcherGeneric * findDirWatcher( std::string dir );
+
+		DirWatcherGeneric * findDirWatcherFast( std::string dir );
+	protected:
+		bool				Deleted;
+
+		DirWatcherGeneric * createDirectory( std::string newdir );
+
+		void removeDirectory( std::string dir );
+
+		void moveDirectory( std::string oldDir, std::string newDir );
+
+		void resetDirectory( std::string directory );
+
+		void handleAction( const std::string& filename, unsigned long action, std::string oldFilename = "");
+};
+
+}
+
+#endif
diff --git a/dep/efsw/src/efsw/DirectorySnapshot.cpp b/dep/efsw/src/efsw/DirectorySnapshot.cpp
new file mode 100644
index 00000000000..c0ef747548a
--- /dev/null
+++ b/dep/efsw/src/efsw/DirectorySnapshot.cpp
@@ -0,0 +1,261 @@
+#include <efsw/DirectorySnapshot.hpp>
+#include <efsw/FileSystem.hpp>
+
+namespace efsw {
+
+DirectorySnapshot::DirectorySnapshot()
+{
+}
+
+DirectorySnapshot::DirectorySnapshot( std::string directory )
+{
+	init( directory );
+}
+
+DirectorySnapshot::~DirectorySnapshot()
+{
+}
+
+void DirectorySnapshot::init( std::string directory )
+{
+	setDirectoryInfo( directory );
+	initFiles();
+}
+
+bool DirectorySnapshot::exists()
+{
+	return DirectoryInfo.exists();
+}
+
+void DirectorySnapshot::deleteAll( DirectorySnapshotDiff& Diff )
+{
+	FileInfo fi;
+
+	for ( FileInfoMap::iterator it = Files.begin(); it != Files.end(); it++ )
+	{
+		fi = it->second;
+
+		if ( fi.isDirectory() )
+		{
+			Diff.DirsDeleted.push_back( fi );
+		}
+		else
+		{
+			Diff.FilesDeleted.push_back( fi );
+		}
+	}
+}
+
+void DirectorySnapshot::setDirectoryInfo( std::string directory )
+{
+	DirectoryInfo = FileInfo( directory );
+}
+
+void DirectorySnapshot::initFiles()
+{
+	Files = FileSystem::filesInfoFromPath( DirectoryInfo.Filepath );
+
+	FileInfoMap::iterator it = Files.begin();
+	std::list<std::string> eraseFiles;
+
+	/// Remove all non regular files and non directories
+	for ( ; it != Files.end(); it++ )
+	{
+		if ( !it->second.isRegularFile() && !it->second.isDirectory() )
+		{
+			eraseFiles.push_back( it->first );
+		}
+	}
+
+	for ( std::list<std::string>::iterator eit = eraseFiles.begin(); eit != eraseFiles.end(); eit++ )
+	{
+		Files.erase( *eit );
+	}
+}
+
+DirectorySnapshotDiff DirectorySnapshot::scan()
+{
+	DirectorySnapshotDiff	Diff;
+
+	Diff.clear();
+
+	FileInfo curFI( DirectoryInfo.Filepath );
+
+	Diff.DirChanged	= DirectoryInfo != curFI;
+
+	if ( Diff.DirChanged )
+	{
+		DirectoryInfo = curFI;
+	}
+
+	/// If the directory was erased, create the events for files and directories deletion
+	if ( !curFI.exists() )
+	{
+		deleteAll( Diff );
+
+		return Diff;
+	}
+
+	FileInfoMap files = FileSystem::filesInfoFromPath( DirectoryInfo.Filepath );
+
+	if ( files.empty() && Files.empty() )
+	{
+		return Diff;
+	}
+
+	FileInfo fi;
+	FileInfoMap FilesCpy;
+	FileInfoMap::iterator it;
+	FileInfoMap::iterator fiIt;
+
+	if ( Diff.DirChanged )
+	{
+		FilesCpy = Files;
+	}
+
+	for ( it = files.begin(); it != files.end(); it++ )
+	{
+		fi	= it->second;
+
+		/// File existed before?
+		fiIt = Files.find( it->first );
+
+		if ( fiIt != Files.end() )
+		{
+			/// Erase from the file list copy
+			FilesCpy.erase( it->first );
+
+			/// File changed?
+			if ( (*fiIt).second != fi )
+			{
+				/// Update the new file info
+				Files[ it->first ] = fi;
+
+				/// handle modified event
+				if ( fi.isDirectory() )
+				{
+					Diff.DirsModified.push_back( fi );
+				}
+				else
+				{
+					Diff.FilesModified.push_back( fi );
+				}
+			}
+		}
+		/// Only add regular files or directories
+		else if ( fi.isRegularFile() || fi.isDirectory() )
+		{
+			/// New file found
+			Files[ it->first ] = fi;
+
+			FileInfoMap::iterator fit;
+			std::string oldFile = "";
+
+			/// Check if the same inode already existed
+			if ( ( fit = nodeInFiles( fi ) ) != Files.end() )
+			{
+				oldFile = fit->first;
+
+				/// Avoid firing a Delete event
+				FilesCpy.erase( fit->first );
+
+				/// Delete the old file name
+				Files.erase( fit->first );
+
+				if ( fi.isDirectory() )
+				{
+					Diff.DirsMoved.push_back( std::make_pair( oldFile, fi ) );
+				}
+				else
+				{
+					Diff.FilesMoved.push_back( std::make_pair( oldFile, fi ) );
+				}
+			}
+			else
+			{
+				if ( fi.isDirectory() )
+				{
+					Diff.DirsCreated.push_back( fi );
+				}
+				else
+				{
+					Diff.FilesCreated.push_back( fi );
+				}
+			}
+		}
+	}
+
+	if ( !Diff.DirChanged )
+	{
+		return Diff;
+	}
+
+	/// The files or directories that remains were deleted
+	for ( it = FilesCpy.begin(); it != FilesCpy.end(); it++ )
+	{
+		fi = it->second;
+
+		if ( fi.isDirectory() )
+		{
+			Diff.DirsDeleted.push_back( fi );
+		}
+		else
+		{
+			Diff.FilesDeleted.push_back( fi );
+		}
+
+		/// Remove the file or directory from the list of files
+		Files.erase( it->first );
+	}
+
+	return Diff;
+}
+
+FileInfoMap::iterator DirectorySnapshot::nodeInFiles( FileInfo& fi )
+{
+	FileInfoMap::iterator it;
+
+	if ( FileInfo::inodeSupported() )
+	{
+		for ( it = Files.begin(); it != Files.end(); it++ )
+		{
+			if ( it->second.sameInode( fi ) && it->second.Filepath != fi.Filepath )
+			{
+				return it;
+			}
+		}
+	}
+
+	return Files.end();
+}
+
+void DirectorySnapshot::addFile( std::string path )
+{
+	std::string name( FileSystem::fileNameFromPath( path ) );
+	Files[ name ] = FileInfo( path );
+}
+
+void DirectorySnapshot::removeFile( std::string path )
+{
+	std::string name( FileSystem::fileNameFromPath( path ) );
+
+	FileInfoMap::iterator it = Files.find( name );
+
+	if ( Files.end() != it )
+	{
+		Files.erase( it );
+	}
+}
+
+void DirectorySnapshot::moveFile( std::string oldPath, std::string newPath )
+{
+	removeFile( oldPath );
+	addFile( newPath );
+}
+
+void DirectorySnapshot::updateFile(std::string path)
+{
+	addFile( path );
+}
+
+}
diff --git a/dep/efsw/src/efsw/DirectorySnapshot.hpp b/dep/efsw/src/efsw/DirectorySnapshot.hpp
new file mode 100644
index 00000000000..1ada66fe980
--- /dev/null
+++ b/dep/efsw/src/efsw/DirectorySnapshot.hpp
@@ -0,0 +1,46 @@
+#ifndef EFSW_DIRECTORYSNAPSHOT_HPP
+#define EFSW_DIRECTORYSNAPSHOT_HPP
+
+#include <efsw/DirectorySnapshotDiff.hpp>
+
+namespace efsw {
+
+class DirectorySnapshot
+{
+	public:
+		FileInfo		DirectoryInfo;
+		FileInfoMap		Files;
+
+		void setDirectoryInfo( std::string directory );
+
+		DirectorySnapshot();
+
+		DirectorySnapshot( std::string directory );
+
+		~DirectorySnapshot();
+
+		void init( std::string directory );
+
+		bool exists();
+
+		DirectorySnapshotDiff scan();
+
+		FileInfoMap::iterator nodeInFiles( FileInfo& fi );
+
+		void addFile( std::string path );
+
+		void removeFile( std::string path );
+
+		void moveFile( std::string oldPath, std::string newPath );
+
+		void updateFile( std::string path );
+	protected:
+		void initFiles();
+
+		void deleteAll( DirectorySnapshotDiff &Diff );
+};
+
+}
+
+#endif
+
diff --git a/dep/efsw/src/efsw/DirectorySnapshotDiff.cpp b/dep/efsw/src/efsw/DirectorySnapshotDiff.cpp
new file mode 100644
index 00000000000..eabc6fdbda1
--- /dev/null
+++ b/dep/efsw/src/efsw/DirectorySnapshotDiff.cpp
@@ -0,0 +1,29 @@
+#include <efsw/DirectorySnapshotDiff.hpp>
+
+namespace efsw {
+
+void DirectorySnapshotDiff::clear()
+{
+	FilesCreated.clear();
+	FilesModified.clear();
+	FilesMoved.clear();
+	FilesDeleted.clear();
+	DirsCreated.clear();
+	DirsModified.clear();
+	DirsMoved.clear();
+	DirsDeleted.clear();
+}
+
+bool DirectorySnapshotDiff::changed()
+{
+	return	!FilesCreated.empty()	||
+			!FilesModified.empty()	||
+			!FilesMoved.empty()		||
+			!FilesDeleted.empty()	||
+			!DirsCreated.empty()	||
+			!DirsModified.empty()	||
+			!DirsMoved.empty()		||
+			!DirsDeleted.empty();
+}
+
+}
diff --git a/dep/efsw/src/efsw/DirectorySnapshotDiff.hpp b/dep/efsw/src/efsw/DirectorySnapshotDiff.hpp
new file mode 100644
index 00000000000..042de9ce02c
--- /dev/null
+++ b/dep/efsw/src/efsw/DirectorySnapshotDiff.hpp
@@ -0,0 +1,35 @@
+#ifndef EFSW_DIRECTORYSNAPSHOTDIFF_HPP
+#define EFSW_DIRECTORYSNAPSHOTDIFF_HPP
+
+#include <efsw/FileInfo.hpp>
+
+namespace efsw {
+
+class DirectorySnapshotDiff
+{
+	public:
+		FileInfoList	FilesDeleted;
+		FileInfoList	FilesCreated;
+		FileInfoList	FilesModified;
+		MovedList		FilesMoved;
+		FileInfoList	DirsDeleted;
+		FileInfoList	DirsCreated;
+		FileInfoList	DirsModified;
+		MovedList		DirsMoved;
+		bool			DirChanged;
+
+		void clear();
+
+		bool changed();
+};
+
+#define DiffIterator( FileInfoListName )	it = Diff.FileInfoListName.begin(); \
+											for ( ; it != Diff.FileInfoListName.end(); it++ )
+
+#define DiffMovedIterator( MovedListName )	mit = Diff.MovedListName.begin(); \
+											for ( ; mit != Diff.MovedListName.end(); mit++ )
+
+}
+
+#endif
+
diff --git a/dep/efsw/src/efsw/FileInfo.cpp b/dep/efsw/src/efsw/FileInfo.cpp
new file mode 100644
index 00000000000..7003afc2a15
--- /dev/null
+++ b/dep/efsw/src/efsw/FileInfo.cpp
@@ -0,0 +1,274 @@
+#include <efsw/FileInfo.hpp>
+#include <efsw/FileSystem.hpp>
+#include <efsw/String.hpp>
+
+#ifndef _DARWIN_FEATURE_64_BIT_INODE
+#define _DARWIN_FEATURE_64_BIT_INODE
+#endif
+
+#ifndef _FILE_OFFSET_BITS
+#define _FILE_OFFSET_BITS 64
+#endif
+
+#include <sys/stat.h>
+
+#include <limits.h>
+#include <stdlib.h>
+
+#ifdef EFSW_COMPILER_MSVC
+	#ifndef S_ISDIR
+	#define S_ISDIR(f) ((f)&_S_IFDIR)
+	#endif
+
+	#ifndef S_ISREG
+	#define S_ISREG(f) ((f)&_S_IFREG)
+	#endif
+
+	#ifndef S_ISRDBL
+	#define S_ISRDBL(f) ((f)&_S_IREAD)
+	#endif
+#else
+	#include <unistd.h>
+
+	#ifndef S_ISRDBL
+	#define S_ISRDBL(f) ((f)&S_IRUSR)
+	#endif
+#endif
+
+namespace efsw {
+
+bool FileInfo::exists( const std::string& filePath )
+{
+	FileInfo fi( filePath );
+	return fi.exists();
+}
+
+bool FileInfo::isLink( const std::string& filePath )
+{
+	FileInfo fi( filePath, true );
+	return fi.isLink();
+}
+
+bool FileInfo::inodeSupported()
+{
+	#if EFSW_PLATFORM != EFSW_PLATFORM_WIN32
+	return true;
+	#else
+	return false;
+	#endif
+}
+
+FileInfo::FileInfo() :
+	ModificationTime(0),
+	OwnerId(0),
+	GroupId(0),
+	Permissions(0),
+	Inode(0)
+{}
+
+FileInfo::FileInfo( const std::string& filepath ) :
+	Filepath( filepath ),
+	ModificationTime(0),
+	OwnerId(0),
+	GroupId(0),
+	Permissions(0),
+	Inode(0)
+{
+	getInfo();
+}
+
+FileInfo::FileInfo( const std::string& filepath, bool linkInfo ) :
+	Filepath( filepath ),
+	ModificationTime(0),
+	OwnerId(0),
+	GroupId(0),
+	Permissions(0),
+	Inode(0)
+{
+	if ( linkInfo )
+	{
+		getRealInfo();
+	}
+	else
+	{
+		getInfo();
+	}
+}
+
+void FileInfo::getInfo()
+{
+	#if EFSW_PLATFORM == EFSW_PLATFORM_WIN32
+	if ( Filepath.size() == 3 && Filepath[1] == ':' && Filepath[2] == FileSystem::getOSSlash() )
+	{
+		Filepath += FileSystem::getOSSlash();
+	}
+	#endif
+
+	/// Why i'm doing this? stat in mingw32 doesn't work for directories if the dir path ends with a path slash
+	bool slashAtEnd = FileSystem::slashAtEnd( Filepath );
+
+	if ( slashAtEnd )
+	{
+		FileSystem::dirRemoveSlashAtEnd( Filepath );
+	}
+
+	#if EFSW_PLATFORM != EFSW_PLATFORM_WIN32
+	struct stat st;
+	int res = stat( Filepath.c_str(), &st );
+	#else
+	struct _stat st;
+	int res = _wstat( String::fromUtf8( Filepath ).toWideString().c_str(), &st );
+	#endif
+
+	if ( 0 == res )
+	{
+		ModificationTime	= st.st_mtime;
+		Size				= st.st_size;
+		OwnerId				= st.st_uid;
+		GroupId				= st.st_gid;
+		Permissions			= st.st_mode;
+		Inode				= st.st_ino;
+	}
+
+	if ( slashAtEnd )
+	{
+		FileSystem::dirAddSlashAtEnd( Filepath );
+	}
+}
+
+void FileInfo::getRealInfo()
+{
+	bool slashAtEnd = FileSystem::slashAtEnd( Filepath );
+
+	if ( slashAtEnd )
+	{
+		FileSystem::dirRemoveSlashAtEnd( Filepath );
+	}
+
+	#if EFSW_PLATFORM != EFSW_PLATFORM_WIN32
+	struct stat st;
+	int res = lstat( Filepath.c_str(), &st );
+	#else
+	struct _stat st;
+	int res = _wstat( String::fromUtf8( Filepath ).toWideString().c_str(), &st );
+	#endif
+
+	if ( 0 == res )
+	{
+		ModificationTime	= st.st_mtime;
+		Size				= st.st_size;
+		OwnerId				= st.st_uid;
+		GroupId				= st.st_gid;
+		Permissions			= st.st_mode;
+		Inode				= st.st_ino;
+	}
+
+	if ( slashAtEnd )
+	{
+		FileSystem::dirAddSlashAtEnd( Filepath );
+	}
+}
+
+bool FileInfo::operator==( const FileInfo& Other ) const
+{
+	return (	ModificationTime	== Other.ModificationTime &&
+				Size				== Other.Size &&
+				OwnerId				== Other.OwnerId &&
+				GroupId				== Other.GroupId &&
+				Permissions			== Other.Permissions &&
+				Inode				== Other.Inode
+	);
+}
+
+bool FileInfo::isDirectory()
+{
+	return 0 != S_ISDIR(Permissions);
+}
+
+bool FileInfo::isRegularFile()
+{
+	return 0 != S_ISREG(Permissions);
+}
+
+bool FileInfo::isReadable()
+{
+	return 0 != S_ISRDBL(Permissions);
+}
+
+bool FileInfo::isLink()
+{
+#if EFSW_PLATFORM != EFSW_PLATFORM_WIN32
+	return S_ISLNK(Permissions);
+#else
+	return false;
+#endif
+}
+
+std::string FileInfo::linksTo()
+{
+#if EFSW_PLATFORM != EFSW_PLATFORM_WIN32
+	if ( isLink() )
+	{
+		char * ch = realpath( Filepath.c_str(), NULL);
+
+		if ( NULL != ch )
+		{
+			std::string tstr( ch );
+
+			free( ch );
+
+			return tstr;
+		}
+	}
+#endif
+	return std::string("");
+}
+
+bool FileInfo::exists()
+{
+	bool slashAtEnd = FileSystem::slashAtEnd( Filepath );
+
+	if ( slashAtEnd )
+	{
+		FileSystem::dirRemoveSlashAtEnd(Filepath);
+	}
+
+#if EFSW_PLATFORM != EFSW_PLATFORM_WIN32
+	struct stat st;
+	int res = stat( Filepath.c_str(), &st );
+#else
+	struct _stat st;
+	int res = _wstat( String::fromUtf8( Filepath ).toWideString().c_str(), &st );
+#endif
+
+	if (slashAtEnd)
+	{
+		FileSystem::dirAddSlashAtEnd(Filepath);
+	}
+
+	return 0 == res;
+}
+
+FileInfo& FileInfo::operator=( const FileInfo& Other )
+{
+	this->Filepath			= Other.Filepath;
+	this->Size				= Other.Size;
+	this->ModificationTime	= Other.ModificationTime;
+	this->GroupId			= Other.GroupId;
+	this->OwnerId			= Other.OwnerId;
+	this->Permissions		= Other.Permissions;
+	this->Inode				= Other.Inode;
+	return *this;
+}
+
+bool FileInfo::sameInode( const FileInfo& Other ) const
+{
+	return inodeSupported() && Inode == Other.Inode;
+}
+
+bool FileInfo::operator!=( const FileInfo& Other ) const
+{
+	return !(*this == Other);
+}
+
+} 
diff --git a/dep/efsw/src/efsw/FileInfo.hpp b/dep/efsw/src/efsw/FileInfo.hpp
new file mode 100644
index 00000000000..45cca6a7953
--- /dev/null
+++ b/dep/efsw/src/efsw/FileInfo.hpp
@@ -0,0 +1,66 @@
+#ifndef EFSW_FILEINFO_HPP
+#define EFSW_FILEINFO_HPP
+
+#include <efsw/base.hpp>
+#include <string>
+#include <map>
+#include <list>
+
+namespace efsw {
+
+class FileInfo
+{
+	public:
+		static bool exists( const std::string& filePath );
+
+		static bool isLink( const std::string& filePath );
+
+		static bool inodeSupported();
+
+		FileInfo();
+
+		FileInfo( const std::string& filepath );
+
+		FileInfo( const std::string& filepath, bool linkInfo );
+
+		bool operator==( const FileInfo& Other ) const;
+
+		bool operator!=( const FileInfo& Other ) const;
+
+		FileInfo& operator=( const FileInfo& Other );
+
+		bool isDirectory();
+
+		bool isRegularFile();
+
+		bool isReadable();
+
+		bool sameInode( const FileInfo& Other ) const;
+
+		bool isLink();
+
+		std::string linksTo();
+
+		bool exists();
+
+		void getInfo();
+
+		void getRealInfo();
+
+		std::string		Filepath;
+		Uint64			ModificationTime;
+		Uint64			Size;
+		Uint32			OwnerId;
+		Uint32			GroupId;
+		Uint32			Permissions;
+		Uint64			Inode;
+};
+
+typedef std::map<std::string, FileInfo>					FileInfoMap;
+typedef std::list<FileInfo>								FileInfoList;
+typedef std::list< std::pair< std::string, FileInfo> >	MovedList;
+
+}
+
+#endif
+
diff --git a/dep/efsw/src/efsw/FileSystem.cpp b/dep/efsw/src/efsw/FileSystem.cpp
new file mode 100644
index 00000000000..e3afa0b4046
--- /dev/null
+++ b/dep/efsw/src/efsw/FileSystem.cpp
@@ -0,0 +1,124 @@
+#include <efsw/FileSystem.hpp>
+#include <efsw/platform/platformimpl.hpp>
+
+#if EFSW_OS == EFSW_OS_MACOSX
+#include <CoreFoundation/CoreFoundation.h>
+#endif
+
+namespace efsw {
+
+bool FileSystem::isDirectory( const std::string& path )
+{
+	return Platform::FileSystem::isDirectory( path );
+}
+
+FileInfoMap FileSystem::filesInfoFromPath( std::string path ) {
+	dirAddSlashAtEnd( path );
+
+	return Platform::FileSystem::filesInfoFromPath( path );
+}
+
+char FileSystem::getOSSlash()
+{
+	return Platform::FileSystem::getOSSlash();
+}
+
+bool FileSystem::slashAtEnd( std::string &dir )
+{
+	return ( dir.size() && dir[ dir.size() - 1 ] == getOSSlash() );
+}
+
+void FileSystem::dirAddSlashAtEnd( std::string& dir )
+{
+	if ( dir.size() > 1 && dir[ dir.size() - 1 ] != getOSSlash() )
+	{
+		dir.push_back( getOSSlash() );
+	}
+}
+
+void FileSystem::dirRemoveSlashAtEnd( std::string& dir )
+{
+	if ( dir.size() > 1 && dir[ dir.size() - 1 ] == getOSSlash() )
+	{
+		dir.erase( dir.size() - 1 );
+	}
+}
+
+std::string FileSystem::fileNameFromPath( std::string filepath )
+{
+	dirRemoveSlashAtEnd( filepath );
+
+	size_t pos = filepath.find_last_of( getOSSlash() );
+
+	if ( pos != std::string::npos )
+	{
+		return filepath.substr( pos + 1 );
+	}
+
+	return filepath;
+}
+
+std::string FileSystem::pathRemoveFileName( std::string filepath )
+{
+	dirRemoveSlashAtEnd( filepath );
+
+	size_t pos = filepath.find_last_of( getOSSlash() );
+
+	if ( pos != std::string::npos )
+	{
+		return filepath.substr( 0, pos + 1 );
+	}
+
+	return filepath;
+}
+
+std::string FileSystem::getLinkRealPath( std::string dir, std::string& curPath )
+{
+	FileSystem::dirRemoveSlashAtEnd( dir );
+	FileInfo fi( dir, true );
+
+	/// Check with lstat and see if it's a link
+	if ( fi.isLink() )
+	{
+		/// get the real path of the link
+		std::string link( fi.linksTo() );
+
+		/// get the current path of the directory without the link dir path
+		curPath = FileSystem::pathRemoveFileName( dir );
+
+		/// ensure that ends with the os directory slash
+		FileSystem::dirAddSlashAtEnd( link );
+
+		return link;
+	}
+
+	/// if it's not a link return nothing
+	return "";
+}
+
+std::string FileSystem::precomposeFileName( const std::string& name )
+{
+#if EFSW_OS == EFSW_OS_MACOSX
+	CFStringRef cfStringRef = CFStringCreateWithCString(kCFAllocatorDefault, name.c_str(), kCFStringEncodingUTF8);
+	CFMutableStringRef cfMutable = CFStringCreateMutableCopy(NULL, 0, cfStringRef);
+
+	CFStringNormalize(cfMutable,kCFStringNormalizationFormC);
+
+	char c_str[255 + 1];
+	CFStringGetCString(cfMutable, c_str, sizeof(c_str)-1, kCFStringEncodingUTF8);
+
+	CFRelease(cfStringRef);
+	CFRelease(cfMutable);
+
+	return std::string(c_str);
+#else
+	return name;
+#endif
+}
+
+bool FileSystem::isRemoteFS( const std::string& directory )
+{
+	return Platform::FileSystem::isRemoteFS( directory );
+}
+
+}
diff --git a/dep/efsw/src/efsw/FileSystem.hpp b/dep/efsw/src/efsw/FileSystem.hpp
new file mode 100644
index 00000000000..4e2e1aeb7cd
--- /dev/null
+++ b/dep/efsw/src/efsw/FileSystem.hpp
@@ -0,0 +1,40 @@
+#ifndef EFSW_FILESYSTEM_HPP
+#define EFSW_FILESYSTEM_HPP
+
+#include <efsw/base.hpp>
+#include <efsw/FileInfo.hpp>
+#include <map>
+
+namespace efsw {
+
+class FileSystem
+{
+	public:
+		static bool isDirectory( const std::string& path );
+
+		static FileInfoMap filesInfoFromPath( std::string path );
+
+		static char getOSSlash();
+
+		static bool slashAtEnd( std::string& dir );
+
+		static void dirAddSlashAtEnd( std::string& dir );
+
+		static void dirRemoveSlashAtEnd( std::string& dir );
+
+		static std::string fileNameFromPath( std::string filepath );
+
+		static std::string pathRemoveFileName( std::string filepath );
+
+		static void realPath( std::string curdir, std::string& path );
+
+		static std::string getLinkRealPath( std::string dir, std::string& curPath );
+
+		static std::string precomposeFileName(const std::string& name);
+
+		static bool isRemoteFS( const std::string& directory );
+};
+
+}
+
+#endif
diff --git a/dep/efsw/src/efsw/FileWatcher.cpp b/dep/efsw/src/efsw/FileWatcher.cpp
new file mode 100644
index 00000000000..e33d5ec46fb
--- /dev/null
+++ b/dep/efsw/src/efsw/FileWatcher.cpp
@@ -0,0 +1,145 @@
+#include <efsw/efsw.hpp>
+#include <efsw/FileWatcherImpl.hpp>
+#include <efsw/FileWatcherGeneric.hpp>
+#include <efsw/FileSystem.hpp>
+
+#if EFSW_PLATFORM == EFSW_PLATFORM_WIN32
+#	include <efsw/FileWatcherWin32.hpp>
+#	define FILEWATCHER_IMPL FileWatcherWin32
+#	define BACKEND_NAME "Win32"
+#elif EFSW_PLATFORM == EFSW_PLATFORM_INOTIFY
+#	include <efsw/FileWatcherInotify.hpp>
+#	define FILEWATCHER_IMPL FileWatcherInotify
+#	define BACKEND_NAME "Inotify"
+#elif EFSW_PLATFORM == EFSW_PLATFORM_KQUEUE
+#	include <efsw/FileWatcherKqueue.hpp>
+#	define FILEWATCHER_IMPL FileWatcherKqueue
+#	define BACKEND_NAME "Kqueue"
+#elif EFSW_PLATFORM == EFSW_PLATFORM_FSEVENTS
+#	include <efsw/FileWatcherFSEvents.hpp>
+#	define FILEWATCHER_IMPL FileWatcherFSEvents
+#	define BACKEND_NAME "FSEvents"
+#else
+#	define FILEWATCHER_IMPL FileWatcherGeneric
+#	define BACKEND_NAME "Generic"
+#endif
+
+#include <efsw/Debug.hpp>
+
+namespace efsw {
+
+FileWatcher::FileWatcher() :
+	mFollowSymlinks(false),
+	mOutOfScopeLinks(false)
+{
+	efDEBUG( "Using backend: %s\n", BACKEND_NAME );
+
+	mImpl = new FILEWATCHER_IMPL( this );
+
+	if ( !mImpl->initOK() )
+	{
+		efSAFE_DELETE( mImpl );
+
+		efDEBUG( "Falled back to backend: %s\n", BACKEND_NAME );
+
+		mImpl = new FileWatcherGeneric( this );
+	}
+}
+
+FileWatcher::FileWatcher( bool useGenericFileWatcher ) :
+	mFollowSymlinks(false),
+	mOutOfScopeLinks(false)
+{
+	if ( useGenericFileWatcher )
+	{
+		efDEBUG( "Using backend: Generic\n" );
+
+		mImpl = new FileWatcherGeneric( this );
+	}
+	else
+	{
+		efDEBUG( "Using backend: %s\n", BACKEND_NAME );
+
+		mImpl = new FILEWATCHER_IMPL( this );
+
+		if ( !mImpl->initOK() )
+		{
+			efSAFE_DELETE( mImpl );
+
+			efDEBUG( "Falled back to backend: %s\n", BACKEND_NAME );
+
+			mImpl = new FileWatcherGeneric( this );
+		}
+	}
+}
+
+FileWatcher::~FileWatcher()
+{
+	efSAFE_DELETE( mImpl );
+}
+
+WatchID FileWatcher::addWatch(const std::string& directory, FileWatchListener* watcher)
+{
+	if ( mImpl->mIsGeneric || !FileSystem::isRemoteFS( directory ) )
+	{
+		return mImpl->addWatch(directory, watcher, false);
+	}
+	else
+	{
+		return Errors::Log::createLastError( Errors::FileRemote, directory );
+	}
+}
+
+WatchID FileWatcher::addWatch(const std::string& directory, FileWatchListener* watcher, bool recursive)
+{
+	if ( mImpl->mIsGeneric || !FileSystem::isRemoteFS( directory ) )
+	{
+		return mImpl->addWatch(directory, watcher, recursive);
+	}
+	else
+	{
+		return Errors::Log::createLastError( Errors::FileRemote, directory );
+	}
+}
+
+void FileWatcher::removeWatch(const std::string& directory)
+{
+	mImpl->removeWatch(directory);
+}
+
+void FileWatcher::removeWatch(WatchID watchid)
+{
+	mImpl->removeWatch(watchid);
+}
+
+void FileWatcher::watch()
+{
+	mImpl->watch();
+}
+
+std::list<std::string> FileWatcher::directories()
+{
+	return mImpl->directories();
+}
+
+void FileWatcher::followSymlinks( bool follow )
+{
+	mFollowSymlinks = follow;
+}
+
+const bool& FileWatcher::followSymlinks() const
+{
+	return mFollowSymlinks;
+}
+
+void FileWatcher::allowOutOfScopeLinks( bool allow )
+{
+	mOutOfScopeLinks = allow;
+}
+
+const bool& FileWatcher::allowOutOfScopeLinks() const
+{
+	return mOutOfScopeLinks;
+}
+
+}
diff --git a/dep/efsw/src/efsw/FileWatcherCWrapper.cpp b/dep/efsw/src/efsw/FileWatcherCWrapper.cpp
new file mode 100644
index 00000000000..2739e756bb2
--- /dev/null
+++ b/dep/efsw/src/efsw/FileWatcherCWrapper.cpp
@@ -0,0 +1,132 @@
+#include <efsw/efsw.h>
+#include <efsw/efsw.hpp>
+#include <vector>
+
+#define TOBOOL(i) ((i) == 0 ? false : true)
+
+/*************************************************************************************************/
+class Watcher_CAPI : public efsw::FileWatchListener
+{
+public:
+	efsw_watcher mWatcher;
+	efsw_pfn_fileaction_callback mFn;
+	void* mParam;
+public:
+	Watcher_CAPI(efsw_watcher watcher, efsw_pfn_fileaction_callback fn, void* param)
+	{
+		mWatcher = watcher;
+		mFn = fn;
+		mParam = param;
+	}
+
+	void handleFileAction(efsw::WatchID watchid, const std::string& dir, const std::string& filename,
+		efsw::Action action, std::string oldFilename = "")
+	{
+		mFn(mWatcher, watchid, dir.c_str(), filename.c_str(), (enum efsw_action)action,
+			oldFilename.c_str(), mParam );
+	}
+};
+
+/*************************************************************************************************
+ * globals
+ */
+static std::vector<Watcher_CAPI*> g_callbacks;
+
+Watcher_CAPI* find_callback(efsw_watcher watcher, efsw_pfn_fileaction_callback fn)
+{
+	for (std::vector<Watcher_CAPI*>::iterator i = g_callbacks.begin(); i != g_callbacks.end(); i++ )
+	{
+		Watcher_CAPI* callback = *i;
+
+		if (callback->mFn == fn && callback->mWatcher == watcher)
+			return *i;
+	}
+
+	return NULL;
+}
+
+Watcher_CAPI* remove_callback(efsw_watcher watcher)
+{
+	std::vector<Watcher_CAPI*>::iterator i = g_callbacks.begin();
+
+	while (i != g_callbacks.end())  {
+		Watcher_CAPI* callback = *i;
+
+		if (callback->mWatcher == watcher)
+			i = g_callbacks.erase(i);
+		else
+			i++;
+	}
+
+	return NULL;
+}
+
+
+/*************************************************************************************************/
+efsw_watcher efsw_create(int generic_mode)
+{
+	return (efsw_watcher)new efsw::FileWatcher(TOBOOL(generic_mode));
+}
+
+void efsw_release(efsw_watcher watcher)
+{
+	remove_callback(watcher);
+	delete (efsw::FileWatcher*)watcher;
+}
+
+const char* efsw_getlasterror()
+{
+	static std::string log_str;
+	log_str = efsw::Errors::Log::getLastErrorLog();
+	return log_str.c_str();
+}
+
+efsw_watchid efsw_addwatch(efsw_watcher watcher, const char* directory,
+	efsw_pfn_fileaction_callback callback_fn, int recursive, void * param)
+{
+	Watcher_CAPI* callback = find_callback(watcher, callback_fn);
+
+	if (callback == NULL)   {
+		callback = new Watcher_CAPI(watcher, callback_fn, param);
+		g_callbacks.push_back(callback);
+	}
+
+	return ((efsw::FileWatcher*)watcher)->addWatch(std::string(directory), callback,
+		TOBOOL(recursive));
+}
+
+void efsw_removewatch(efsw_watcher watcher, const char* directory)
+{
+	((efsw::FileWatcher*)watcher)->removeWatch(std::string(directory));
+}
+
+void efsw_removewatch_byid(efsw_watcher watcher, efsw_watchid watchid)
+{
+	((efsw::FileWatcher*)watcher)->removeWatch(watchid);
+}
+
+void efsw_watch(efsw_watcher watcher)
+{
+	((efsw::FileWatcher*)watcher)->watch();
+}
+
+void efsw_follow_symlinks(efsw_watcher watcher, int enable)
+{
+	((efsw::FileWatcher*)watcher)->followSymlinks(TOBOOL(enable));
+}
+
+int efsw_follow_symlinks_isenabled(efsw_watcher watcher)
+{
+	return (int)((efsw::FileWatcher*)watcher)->followSymlinks();
+}
+
+void efsw_allow_outofscopelinks(efsw_watcher watcher, int allow)
+{
+	((efsw::FileWatcher*)watcher)->allowOutOfScopeLinks(TOBOOL(allow));
+}
+
+int efsw_outofscopelinks_isallowed(efsw_watcher watcher)
+{
+	return (int)((efsw::FileWatcher*)watcher)->allowOutOfScopeLinks();
+}
+
diff --git a/dep/efsw/src/efsw/FileWatcherFSEvents.cpp b/dep/efsw/src/efsw/FileWatcherFSEvents.cpp
new file mode 100644
index 00000000000..40156674132
--- /dev/null
+++ b/dep/efsw/src/efsw/FileWatcherFSEvents.cpp
@@ -0,0 +1,278 @@
+#include <efsw/FileWatcherFSEvents.hpp>
+#include <efsw/FileSystem.hpp>
+#include <efsw/System.hpp>
+#include <efsw/Debug.hpp>
+#include <efsw/String.hpp>
+
+#if EFSW_PLATFORM == EFSW_PLATFORM_FSEVENTS
+
+#include <sys/utsname.h>
+
+namespace efsw
+{
+
+int getOSXReleaseNumber()
+{
+	static int osxR = -1;
+
+	if ( -1 == osxR )
+	{
+		struct utsname os;
+
+		if ( -1 != uname( &os ) ) {
+			std::string release( os.release );
+			
+			size_t pos = release.find_first_of( '.' );
+			
+			if ( pos != std::string::npos )
+			{
+				release = release.substr( 0, pos );
+			}
+			
+			int rel = 0;
+			
+			if ( String::fromString<int>( rel, release ) )
+			{
+				osxR = rel;
+			}
+		}
+	}
+	
+	return osxR;
+}
+
+bool FileWatcherFSEvents::isGranular()
+{
+	return getOSXReleaseNumber() >= 11;
+}
+
+void FileWatcherFSEvents::FSEventCallback(	ConstFSEventStreamRef streamRef,
+												void *userData, 
+												size_t numEvents, 
+												void *eventPaths, 
+												const FSEventStreamEventFlags eventFlags[], 
+												const FSEventStreamEventId eventIds[] )
+{
+	WatcherFSEvents * watcher = static_cast<WatcherFSEvents*>( userData );
+
+	std::vector<FSEvent> events;
+	events.reserve( numEvents );
+
+	for ( size_t i = 0; i < numEvents; i++ )
+	{
+		events.push_back( FSEvent( std::string( ((char**)eventPaths)[i] ), (long)eventFlags[i], (Uint64)eventIds[i] ) );
+	}
+
+	watcher->handleActions( events );
+
+	watcher->process();
+
+	efDEBUG( "\n" );
+}
+
+FileWatcherFSEvents::FileWatcherFSEvents( FileWatcher * parent ) :
+	FileWatcherImpl( parent ),
+	mRunLoopRef( NULL ),
+	mLastWatchID(0),
+	mThread( NULL )
+{
+	mInitOK = true;
+	
+	watch();
+}
+
+FileWatcherFSEvents::~FileWatcherFSEvents()
+{
+	WatchMap::iterator iter = mWatches.begin();
+
+	for( ; iter != mWatches.end(); ++iter )
+	{
+		WatcherFSEvents * watch = iter->second;
+		
+		efSAFE_DELETE( watch );
+	}
+
+	mWatches.clear();
+
+	mInitOK = false;
+	
+	if ( NULL != mRunLoopRef )
+	{
+		CFRunLoopStop( mRunLoopRef );
+	}
+
+	efSAFE_DELETE( mThread );
+}
+
+WatchID FileWatcherFSEvents::addWatch( const std::string& directory, FileWatchListener* watcher, bool recursive )
+{
+	/// Wait to the RunLoopRef to be ready
+	while ( NULL == mRunLoopRef )
+	{
+		System::sleep( 1 );
+	}
+	
+	std::string dir( directory );
+
+	FileInfo fi( dir );
+
+	if ( !fi.isDirectory() )
+	{
+		return Errors::Log::createLastError( Errors::FileNotFound, dir );
+	}
+	else if ( !fi.isReadable() )
+	{
+		return Errors::Log::createLastError( Errors::FileNotReadable, dir );
+	}
+
+	FileSystem::dirAddSlashAtEnd( dir );
+
+	if ( pathInWatches( dir ) )
+	{
+		return Errors::Log::createLastError( Errors::FileRepeated, directory );
+	}
+
+	/// Check if the directory is a symbolic link
+	std::string curPath;
+	std::string link( FileSystem::getLinkRealPath( dir, curPath ) );
+
+	if ( "" != link )
+	{
+		/// If it's a symlink check if the realpath exists as a watcher, or
+		/// if the path is outside the current dir
+		if ( pathInWatches( link ) )
+		{
+			return Errors::Log::createLastError( Errors::FileRepeated, directory );
+		}
+		else if ( !linkAllowed( curPath, link ) )
+		{
+			return Errors::Log::createLastError( Errors::FileOutOfScope, dir );
+		}
+		else
+		{
+			dir = link;
+		}
+	}
+	
+	mLastWatchID++;
+
+	WatcherFSEvents * pWatch	= new WatcherFSEvents();
+	pWatch->Listener			= watcher;
+	pWatch->ID					= mLastWatchID;
+	pWatch->Directory			= dir;
+	pWatch->Recursive			= recursive;
+	pWatch->FWatcher			= this;
+	
+	pWatch->init();
+
+	mWatchesLock.lock();
+	mWatches.insert(std::make_pair(mLastWatchID, pWatch));
+	mWatchesLock.unlock();
+
+	return pWatch->ID;
+}
+
+void FileWatcherFSEvents::removeWatch(const std::string& directory)
+{
+	mWatchesLock.lock();
+
+	WatchMap::iterator iter = mWatches.begin();
+
+	for(; iter != mWatches.end(); ++iter)
+	{
+		if( directory == iter->second->Directory )
+		{
+			removeWatch( iter->second->ID );
+			return;
+		}
+	}
+
+	mWatchesLock.unlock();
+}
+
+void FileWatcherFSEvents::removeWatch(WatchID watchid)
+{
+	mWatchesLock.lock();
+
+	WatchMap::iterator iter = mWatches.find( watchid );
+
+	if( iter == mWatches.end() )
+		return;
+
+	WatcherFSEvents * watch = iter->second;
+
+	mWatches.erase( iter );
+
+	efDEBUG( "Removed watch %s\n", watch->Directory.c_str() );
+
+	efSAFE_DELETE( watch );
+
+	mWatchesLock.unlock();
+}
+
+void FileWatcherFSEvents::watch()
+{
+	if ( NULL == mThread )
+	{
+		mThread = new Thread( &FileWatcherFSEvents::run, this );
+		mThread->launch();
+	}
+}
+
+void FileWatcherFSEvents::run()
+{
+	mRunLoopRef = CFRunLoopGetCurrent();
+	
+	while ( mInitOK )
+	{
+		if ( !mNeedInit.empty() )
+		{
+			for ( std::list<WatcherFSEvents*>::iterator it = mNeedInit.begin(); it != mNeedInit.end(); it++ )
+			{
+				(*it)->initAsync();
+			}
+
+			mNeedInit.clear();
+		}
+
+		CFRunLoopRunInMode( kCFRunLoopDefaultMode, 0.5, kCFRunLoopRunTimedOut );
+	}
+}
+
+void FileWatcherFSEvents::handleAction(Watcher* watch, const std::string& filename, unsigned long action, std::string oldFilename)
+{
+	/// Not used
+}
+
+std::list<std::string> FileWatcherFSEvents::directories()
+{
+	std::list<std::string> dirs;
+
+	mWatchesLock.lock();
+
+	for ( WatchMap::iterator it = mWatches.begin(); it != mWatches.end(); it++ )
+	{
+		dirs.push_back( std::string( it->second->Directory ) );
+	}
+
+	mWatchesLock.unlock();
+
+	return dirs;
+}
+
+bool FileWatcherFSEvents::pathInWatches( const std::string& path )
+{
+	for ( WatchMap::iterator it = mWatches.begin(); it != mWatches.end(); it++ )
+	{
+		if ( it->second->Directory == path )
+		{
+			return true;
+		}
+	}
+
+	return false;
+}
+
+}
+
+#endif
diff --git a/dep/efsw/src/efsw/FileWatcherFSEvents.hpp b/dep/efsw/src/efsw/FileWatcherFSEvents.hpp
new file mode 100644
index 00000000000..6aafbc0b5ea
--- /dev/null
+++ b/dep/efsw/src/efsw/FileWatcherFSEvents.hpp
@@ -0,0 +1,107 @@
+#ifndef EFSW_FILEWATCHERFSEVENTS_HPP
+#define EFSW_FILEWATCHERFSEVENTS_HPP
+
+#include <efsw/FileWatcherImpl.hpp>
+
+#if EFSW_PLATFORM == EFSW_PLATFORM_FSEVENTS
+
+#include <CoreFoundation/CoreFoundation.h>
+#include <CoreServices/CoreServices.h>
+#include <efsw/WatcherFSEvents.hpp>
+#include <map>
+#include <list>
+#include <vector>
+
+namespace efsw
+{
+
+/* OSX < 10.7 has no file events */
+/* So i declare the events constants */
+enum FSEventEvents
+{
+	efswFSEventStreamCreateFlagFileEvents			= 0x00000010,
+	efswFSEventStreamEventFlagItemCreated			= 0x00000100,
+	efswFSEventStreamEventFlagItemRemoved			= 0x00000200,
+	efswFSEventStreamEventFlagItemInodeMetaMod		= 0x00000400,
+	efswFSEventStreamEventFlagItemRenamed			= 0x00000800,
+	efswFSEventStreamEventFlagItemModified			= 0x00001000,
+	efswFSEventStreamEventFlagItemFinderInfoMod		= 0x00002000,
+	efswFSEventStreamEventFlagItemChangeOwner		= 0x00004000,
+	efswFSEventStreamEventFlagItemXattrMod			= 0x00008000,
+	efswFSEventStreamEventFlagItemIsFile			= 0x00010000,
+	efswFSEventStreamEventFlagItemIsDir				= 0x00020000,
+	efswFSEventStreamEventFlagItemIsSymlink			= 0x00040000,
+	efswFSEventsModified							= efswFSEventStreamEventFlagItemFinderInfoMod	|
+													  efswFSEventStreamEventFlagItemModified		|
+													  efswFSEventStreamEventFlagItemInodeMetaMod	|
+													  efswFSEventStreamEventFlagItemChangeOwner		|
+													  efswFSEventStreamEventFlagItemXattrMod
+};
+
+/// Implementation for Win32 based on ReadDirectoryChangesW.
+/// @class FileWatcherFSEvents
+class FileWatcherFSEvents : public FileWatcherImpl
+{
+	friend class WatcherFSEvents;
+	public:
+		/// @return If FSEvents supports file-level notifications ( true if OS X >= 10.7 )
+		static bool isGranular();
+		
+		/// type for a map from WatchID to WatcherWin32 pointer
+		typedef std::map<WatchID, WatcherFSEvents*> WatchMap;
+
+		FileWatcherFSEvents( FileWatcher * parent );
+
+		virtual ~FileWatcherFSEvents();
+
+		/// Add a directory watch
+		/// On error returns WatchID with Error type.
+		WatchID addWatch(const std::string& directory, FileWatchListener* watcher, bool recursive);
+
+		/// Remove a directory watch. This is a brute force lazy search O(nlogn).
+		void removeWatch(const std::string& directory);
+
+		/// Remove a directory watch. This is a map lookup O(logn).
+		void removeWatch(WatchID watchid);
+
+		/// Updates the watcher. Must be called often.
+		void watch();
+
+		/// Handles the action
+		void handleAction(Watcher* watch, const std::string& filename, unsigned long action, std::string oldFilename = "");
+
+		/// @return Returns a list of the directories that are being watched
+		std::list<std::string> directories();
+	protected:
+		static void FSEventCallback(	ConstFSEventStreamRef streamRef,
+										void *userData, 
+										size_t numEvents, 
+										void *eventPaths, 
+										const FSEventStreamEventFlags eventFlags[], 
+										const FSEventStreamEventId eventIds[]
+		);
+		
+		CFRunLoopRef mRunLoopRef;
+		
+		/// Vector of WatcherWin32 pointers
+		WatchMap mWatches;
+		
+		/// The last watchid
+		WatchID mLastWatchID;
+
+		Thread * mThread;
+
+		Mutex mWatchesLock;
+
+		bool pathInWatches( const std::string& path );
+
+		std::list<WatcherFSEvents*> mNeedInit;
+	private:
+		void run();
+};
+
+}
+
+#endif
+
+#endif
diff --git a/dep/efsw/src/efsw/FileWatcherGeneric.cpp b/dep/efsw/src/efsw/FileWatcherGeneric.cpp
new file mode 100644
index 00000000000..1534b6a9279
--- /dev/null
+++ b/dep/efsw/src/efsw/FileWatcherGeneric.cpp
@@ -0,0 +1,197 @@
+#include <efsw/FileWatcherGeneric.hpp>
+#include <efsw/FileSystem.hpp>
+#include <efsw/System.hpp>
+
+namespace efsw
+{
+
+FileWatcherGeneric::FileWatcherGeneric( FileWatcher * parent ) :
+	FileWatcherImpl( parent ),
+	mThread( NULL ),
+	mLastWatchID( 0 )
+{
+	mInitOK = true;
+	mIsGeneric = true;
+}
+
+FileWatcherGeneric::~FileWatcherGeneric()
+{
+	mInitOK = false;
+
+	mThread->wait();
+
+	efSAFE_DELETE( mThread );
+
+	/// Delete the watches
+	WatchList::iterator it = mWatches.begin();
+
+	for ( ; it != mWatches.end(); it++ )
+	{
+		efSAFE_DELETE( (*it) );
+	}
+}
+
+WatchID FileWatcherGeneric::addWatch(const std::string& directory, FileWatchListener* watcher, bool recursive)
+{
+	std::string dir( directory );
+
+	FileSystem::dirAddSlashAtEnd( dir );
+
+	FileInfo fi( dir );
+
+	if ( !fi.isDirectory() )
+	{
+		return Errors::Log::createLastError( Errors::FileNotFound, dir );
+	}
+	else if ( !fi.isReadable() )
+	{
+		return Errors::Log::createLastError( Errors::FileNotReadable, dir );
+	}
+	else if ( pathInWatches( dir ) )
+	{
+		return Errors::Log::createLastError( Errors::FileRepeated, dir );
+	}
+
+	std::string curPath;
+	std::string link( FileSystem::getLinkRealPath( dir, curPath ) );
+
+	if ( "" != link )
+	{
+		if ( pathInWatches( link ) )
+		{
+			return Errors::Log::createLastError( Errors::FileRepeated, dir );
+		}
+		else if ( !linkAllowed( curPath, link ) )
+		{
+			return Errors::Log::createLastError( Errors::FileOutOfScope, dir );
+		}
+		else
+		{
+			dir = link;
+		}
+	}
+
+	mLastWatchID++;
+
+	WatcherGeneric * pWatch		= new WatcherGeneric( mLastWatchID, dir, watcher, this, recursive );
+
+	mWatchesLock.lock();
+	mWatches.push_back(pWatch);
+	mWatchesLock.unlock();
+
+	return pWatch->ID;
+}
+
+void FileWatcherGeneric::removeWatch( const std::string& directory )
+{
+	WatchList::iterator it = mWatches.begin();
+
+	for ( ; it != mWatches.end(); it++ )
+	{
+		if ( (*it)->Directory == directory )
+		{
+			WatcherGeneric * watch = (*it);
+
+			mWatchesLock.lock();
+
+			mWatches.erase( it );
+
+			efSAFE_DELETE( watch ) ;
+
+			mWatchesLock.unlock();
+
+			return;
+		}
+	}
+}
+
+void FileWatcherGeneric::removeWatch(WatchID watchid)
+{
+	WatchList::iterator it = mWatches.begin();
+
+	for ( ; it != mWatches.end(); it++ )
+	{
+		if ( (*it)->ID == watchid )
+		{
+			WatcherGeneric * watch = (*it);
+
+			mWatchesLock.lock();
+
+			mWatches.erase( it );
+
+			efSAFE_DELETE( watch ) ;
+
+			mWatchesLock.unlock();
+
+			return;
+		}
+	}
+}
+
+void FileWatcherGeneric::watch()
+{
+	if ( NULL == mThread )
+	{
+		mThread = new Thread( &FileWatcherGeneric::run, this );
+		mThread->launch();
+	}
+}
+
+void FileWatcherGeneric::run()
+{
+	do
+	{
+		mWatchesLock.lock();
+
+		WatchList::iterator it = mWatches.begin();
+
+		for ( ; it != mWatches.end(); it++ )
+		{
+			(*it)->watch();
+		}
+
+		mWatchesLock.unlock();
+
+		if ( mInitOK ) System::sleep( 1000 );
+	} while ( mInitOK );
+}
+
+void FileWatcherGeneric::handleAction(Watcher * watch, const std::string& filename, unsigned long action, std::string oldFilename)
+{
+	/// Not used
+}
+
+std::list<std::string> FileWatcherGeneric::directories()
+{
+	std::list<std::string> dirs;
+
+	mWatchesLock.lock();
+
+	WatchList::iterator it = mWatches.begin();
+
+	for ( ; it != mWatches.end(); it++ )
+	{
+		dirs.push_back( (*it)->Directory );
+	}
+
+	mWatchesLock.unlock();
+
+	return dirs;
+}
+
+bool FileWatcherGeneric::pathInWatches( const std::string& path )
+{
+	WatchList::iterator it = mWatches.begin();
+
+	for ( ; it != mWatches.end(); it++ )
+	{
+		if ( (*it)->Directory == path || (*it)->pathInWatches( path ) )
+		{
+			return true;
+		}
+	}
+
+	return false;
+}
+
+}
diff --git a/dep/efsw/src/efsw/FileWatcherGeneric.hpp b/dep/efsw/src/efsw/FileWatcherGeneric.hpp
new file mode 100644
index 00000000000..fc9826580ab
--- /dev/null
+++ b/dep/efsw/src/efsw/FileWatcherGeneric.hpp
@@ -0,0 +1,59 @@
+#ifndef EFSW_FILEWATCHERGENERIC_HPP
+#define EFSW_FILEWATCHERGENERIC_HPP
+
+#include <efsw/FileWatcherImpl.hpp>
+#include <efsw/WatcherGeneric.hpp>
+#include <efsw/DirWatcherGeneric.hpp>
+#include <list>
+
+namespace efsw
+{
+
+/// Implementation for Generic File Watcher.
+/// @class FileWatcherGeneric
+class FileWatcherGeneric : public FileWatcherImpl
+{
+	public:
+		typedef std::list<WatcherGeneric*> WatchList;
+
+		FileWatcherGeneric( FileWatcher * parent );
+
+		virtual ~FileWatcherGeneric();
+
+		/// Add a directory watch
+		/// On error returns WatchID with Error type.
+		WatchID addWatch(const std::string& directory, FileWatchListener* watcher, bool recursive);
+
+		/// Remove a directory watch. This is a brute force lazy search O(nlogn).
+		void removeWatch(const std::string& directory);
+
+		/// Remove a directory watch. This is a map lookup O(logn).
+		void removeWatch(WatchID watchid);
+
+		/// Updates the watcher. Must be called often.
+		void watch();
+
+		/// Handles the action
+		void handleAction(Watcher * watch, const std::string& filename, unsigned long action, std::string oldFilename = "");
+
+		/// @return Returns a list of the directories that are being watched
+		std::list<std::string> directories();
+	protected:
+		Thread * mThread;
+
+		/// The last watchid
+		WatchID mLastWatchID;
+
+		/// Map of WatchID to WatchStruct pointers
+		WatchList mWatches;
+
+		Mutex mWatchesLock;
+
+		bool pathInWatches( const std::string& path );
+	private:
+		void run();
+};
+
+}
+
+#endif
diff --git a/dep/efsw/src/efsw/FileWatcherImpl.cpp b/dep/efsw/src/efsw/FileWatcherImpl.cpp
new file mode 100644
index 00000000000..e6e0fc72a13
--- /dev/null
+++ b/dep/efsw/src/efsw/FileWatcherImpl.cpp
@@ -0,0 +1,29 @@
+#include <efsw/FileWatcherImpl.hpp>
+#include <efsw/String.hpp>
+#include <efsw/System.hpp>
+
+namespace efsw {
+
+FileWatcherImpl::FileWatcherImpl( FileWatcher * parent ) :
+	mFileWatcher( parent ),
+	mInitOK( false ),
+	mIsGeneric( false )
+{
+	System::maxFD();
+}
+
+FileWatcherImpl::~FileWatcherImpl()
+{
+}
+
+bool FileWatcherImpl::initOK()
+{
+	return mInitOK;
+}
+
+bool FileWatcherImpl::linkAllowed( const std::string& curPath, const std::string& link )
+{
+	return ( mFileWatcher->followSymlinks() && mFileWatcher->allowOutOfScopeLinks() ) || -1 != String::strStartsWith( curPath, link );
+}
+
+}
diff --git a/dep/efsw/src/efsw/FileWatcherImpl.hpp b/dep/efsw/src/efsw/FileWatcherImpl.hpp
new file mode 100644
index 00000000000..8f472bf56c5
--- /dev/null
+++ b/dep/efsw/src/efsw/FileWatcherImpl.hpp
@@ -0,0 +1,54 @@
+#ifndef EFSW_FILEWATCHERIMPL_HPP
+#define EFSW_FILEWATCHERIMPL_HPP
+
+#include <efsw/base.hpp>
+#include <efsw/efsw.hpp>
+#include <efsw/Watcher.hpp>
+#include <efsw/Thread.hpp>
+#include <efsw/Mutex.hpp>
+
+namespace efsw {
+
+class FileWatcherImpl
+{
+	public:
+		FileWatcherImpl( FileWatcher * parent );
+
+		virtual ~FileWatcherImpl();
+
+		/// Add a directory watch
+		/// On error returns WatchID with Error type.
+		virtual WatchID addWatch(const std::string& directory, FileWatchListener* watcher, bool recursive) = 0;
+
+		/// Remove a directory watch. This is a brute force lazy search O(nlogn).
+		virtual void removeWatch(const std::string& directory) = 0;
+
+		/// Remove a directory watch. This is a map lookup O(logn).
+		virtual void removeWatch(WatchID watchid) = 0;
+
+		/// Updates the watcher. Must be called often.
+		virtual void watch() = 0;
+
+		/// Handles the action
+		virtual void handleAction(Watcher * watch, const std::string& filename, unsigned long action, std::string oldFilename = "") = 0;
+
+		/// @return Returns a list of the directories that are being watched
+		virtual std::list<std::string> directories() = 0;
+
+		/// @return true if the backend init successfully
+		virtual bool initOK();
+
+		/// @return If the link is allowed according to the current path and the state of out scope links
+		virtual bool linkAllowed( const std::string& curPath, const std::string& link );
+
+		/// Search if a directory already exists in the watches
+		virtual bool pathInWatches( const std::string& path ) = 0;
+
+		FileWatcher *	mFileWatcher;
+		bool			mInitOK;
+		bool			mIsGeneric;
+};
+
+}
+
+#endif
diff --git a/dep/efsw/src/efsw/FileWatcherInotify.cpp b/dep/efsw/src/efsw/FileWatcherInotify.cpp
new file mode 100644
index 00000000000..19c20761663
--- /dev/null
+++ b/dep/efsw/src/efsw/FileWatcherInotify.cpp
@@ -0,0 +1,531 @@
+#include <efsw/FileWatcherInotify.hpp>
+
+#if EFSW_PLATFORM == EFSW_PLATFORM_INOTIFY
+
+#include <unistd.h>
+#include <sys/stat.h>
+#include <fcntl.h>
+#include <string.h>
+#include <stdio.h>
+#include <errno.h>
+
+#ifdef EFSW_INOTIFY_NOSYS
+#include <efsw/inotify-nosys.h>
+#else
+#include <sys/inotify.h>
+#endif
+
+#include <efsw/FileSystem.hpp>
+#include <efsw/System.hpp>
+#include <efsw/Debug.hpp>
+
+#define BUFF_SIZE ((sizeof(struct inotify_event)+FILENAME_MAX)*1024)
+
+namespace efsw
+{
+
+FileWatcherInotify::FileWatcherInotify( FileWatcher * parent ) :
+	FileWatcherImpl( parent ),
+	mFD(-1),
+	mThread(NULL)
+{
+	mFD = inotify_init();
+
+	if (mFD < 0)
+	{
+		efDEBUG( "Error: %s\n", strerror(errno) );
+	}
+	else
+	{
+		mInitOK = true;
+	}
+}
+
+FileWatcherInotify::~FileWatcherInotify()
+{
+	mInitOK = false;
+
+	efSAFE_DELETE( mThread );
+	
+	WatchMap::iterator iter = mWatches.begin();
+	WatchMap::iterator end = mWatches.end();
+
+	for(; iter != end; ++iter)
+	{
+		efSAFE_DELETE( iter->second );
+	}
+
+	mWatches.clear();
+
+	if ( mFD != -1 )
+	{
+		close(mFD);
+		mFD = -1;
+	}
+}
+
+WatchID FileWatcherInotify::addWatch( const std::string& directory, FileWatchListener* watcher, bool recursive )
+{
+	return addWatch( directory, watcher, recursive, NULL );
+}
+
+WatchID FileWatcherInotify::addWatch( const std::string& directory, FileWatchListener* watcher, bool recursive, WatcherInotify * parent )
+{
+	std::string dir( directory );
+
+	FileSystem::dirAddSlashAtEnd( dir );
+
+	FileInfo fi( dir );
+
+	if ( !fi.isDirectory() )
+	{
+		return Errors::Log::createLastError( Errors::FileNotFound, dir );
+	}
+	else if ( !fi.isReadable() )
+	{
+		return Errors::Log::createLastError( Errors::FileNotReadable, dir );
+	}
+	else if ( pathInWatches( dir ) )
+	{
+		return Errors::Log::createLastError( Errors::FileRepeated, directory );
+	}
+	else if ( NULL != parent && FileSystem::isRemoteFS( dir ) )
+	{
+		return Errors::Log::createLastError( Errors::FileRemote, dir );
+	}
+
+	/// Check if the directory is a symbolic link
+	std::string curPath;
+	std::string link( FileSystem::getLinkRealPath( dir, curPath ) );
+
+	if ( "" != link )
+	{
+		/// Avoid adding symlinks directories if it's now enabled
+		if ( NULL != parent && !mFileWatcher->followSymlinks() )
+		{
+			return Errors::Log::createLastError( Errors::FileOutOfScope, dir );
+		}
+
+		/// If it's a symlink check if the realpath exists as a watcher, or
+		/// if the path is outside the current dir
+		if ( pathInWatches( link ) )
+		{
+			return Errors::Log::createLastError( Errors::FileRepeated, directory );
+		}
+		else if ( !linkAllowed( curPath, link ) )
+		{
+			return Errors::Log::createLastError( Errors::FileOutOfScope, dir );
+		}
+		else
+		{
+			dir = link;
+		}
+	}
+
+	int wd = inotify_add_watch (mFD, dir.c_str(), IN_CLOSE_WRITE | IN_MOVED_TO | IN_CREATE | IN_MOVED_FROM | IN_DELETE | IN_MODIFY);
+
+	if ( wd < 0 )
+	{
+		if( errno == ENOENT )
+		{
+			return Errors::Log::createLastError( Errors::FileNotFound, dir );
+		}
+		else
+		{
+			return Errors::Log::createLastError( Errors::Unspecified, std::string(strerror(errno)) );
+		}
+	}
+
+	efDEBUG( "Added watch %s with id: %d\n", dir.c_str(), wd );
+
+	WatcherInotify * pWatch	= new WatcherInotify();
+	pWatch->Listener	= watcher;
+	pWatch->ID			= wd;
+	pWatch->Directory	= dir;
+	pWatch->Recursive	= recursive;
+	pWatch->Parent		= parent;
+
+	mWatchesLock.lock();
+	mWatches.insert(std::make_pair(wd, pWatch));
+	mWatchesLock.unlock();
+
+	if ( NULL == pWatch->Parent )
+	{
+		mRealWatches[ pWatch->ID ] = pWatch;
+	}
+
+	if ( pWatch->Recursive )
+	{
+		std::map<std::string, FileInfo> files = FileSystem::filesInfoFromPath( pWatch->Directory );
+		std::map<std::string, FileInfo>::iterator it = files.begin();
+
+		for ( ; it != files.end(); it++ )
+		{
+			FileInfo fi = it->second;
+
+			if ( fi.isDirectory() && fi.isReadable() )
+			{
+				addWatch( fi.Filepath, watcher, recursive, pWatch );
+			}
+		}
+	}
+
+	return wd;
+}
+
+void FileWatcherInotify::removeWatchLocked(WatchID watchid)
+{
+	WatchMap::iterator iter = mWatches.find( watchid );
+
+	WatcherInotify * watch = iter->second;
+
+	if ( watch->Recursive )
+	{
+		WatchMap::iterator it = mWatches.begin();
+		std::list<WatchID> eraseWatches;
+
+		for(; it != mWatches.end(); ++it)
+		{
+			if ( it->second != watch &&
+				 it->second->inParentTree( watch )
+			)
+			{
+				eraseWatches.push_back( it->second->ID );
+			}
+		}
+
+		for ( std::list<WatchID>::iterator eit = eraseWatches.begin(); eit != eraseWatches.end(); eit++ )
+		{
+			removeWatch( *eit );
+		}
+	}
+
+	mWatches.erase( iter );
+
+	if ( NULL == watch->Parent )
+	{
+		WatchMap::iterator eraseit = mRealWatches.find( watch->ID );
+
+		if ( eraseit != mRealWatches.end() )
+		{
+			mRealWatches.erase( eraseit );
+		}
+	}
+
+	int err = inotify_rm_watch(mFD, watchid);
+
+	if ( err < 0 )
+	{
+		efDEBUG( "Error removing watch %d: %s\n", watchid, strerror(errno) );
+	}
+	else
+	{
+		efDEBUG( "Removed watch %s with id: %d\n", watch->Directory.c_str(), watchid );
+	}
+
+	efSAFE_DELETE( watch );
+}
+
+void FileWatcherInotify::removeWatch(const std::string& directory)
+{
+	mWatchesLock.lock();
+
+	WatchMap::iterator iter = mWatches.begin();
+
+	for(; iter != mWatches.end(); ++iter)
+	{
+		if( directory == iter->second->Directory )
+		{
+			WatcherInotify * watch = iter->second;
+
+			if ( watch->Recursive )
+			{
+				WatchMap::iterator it = mWatches.begin();
+				std::list<WatchID> eraseWatches;
+
+				for(; it != mWatches.end(); ++it)
+				{
+					if ( it->second->inParentTree( watch ) )
+					{
+						eraseWatches.push_back( it->second->ID );
+					}
+				}
+
+				for ( std::list<WatchID>::iterator eit = eraseWatches.begin(); eit != eraseWatches.end(); eit++ )
+				{
+					removeWatchLocked( *eit );
+				}
+			}
+
+			mWatches.erase( iter );
+
+			if ( NULL == watch->Parent )
+			{
+				WatchMap::iterator eraseit = mRealWatches.find( watch->ID );
+
+				if ( eraseit != mRealWatches.end() )
+				{
+					mRealWatches.erase( eraseit );
+				}
+			}
+
+			int err = inotify_rm_watch(mFD, watch->ID);
+
+			if ( err < 0 )
+			{
+				efDEBUG( "Error removing watch %d: %s\n", watch->ID, strerror(errno) );
+			}
+			else
+			{
+				efDEBUG( "Removed watch %s with id: %d\n", watch->Directory.c_str(), watch->ID );
+			}
+
+			efSAFE_DELETE( watch );
+
+			break;
+		}
+	}
+
+	mWatchesLock.unlock();
+}
+
+void FileWatcherInotify::removeWatch( WatchID watchid )
+{
+	mWatchesLock.lock();
+
+	WatchMap::iterator iter = mWatches.find( watchid );
+
+	if( iter == mWatches.end() )
+	{
+		mWatchesLock.unlock();
+
+		return;
+	}
+
+	removeWatchLocked( watchid );
+
+	mWatchesLock.unlock();
+}
+
+void FileWatcherInotify::watch()
+{
+	if ( NULL == mThread )
+	{
+		mThread = new Thread( &FileWatcherInotify::run, this );
+		mThread->launch();
+	}
+}
+
+void FileWatcherInotify::run()
+{
+	static char buff[BUFF_SIZE] = {0};
+	WatchMap::iterator wit;
+	std::list<WatcherInotify*> movedOutsideWatches;
+
+	do
+	{
+		fd_set rfds;
+		FD_ZERO (&rfds);
+		FD_SET (mFD, &rfds);
+		timeval timeout;
+		timeout.tv_sec=0;
+		timeout.tv_usec=100000;
+
+		if( select (FD_SETSIZE, &rfds, NULL, NULL, &timeout) > 0 )
+		{
+			ssize_t len, i = 0;
+
+			len = read (mFD, buff, BUFF_SIZE);
+
+			if (len != -1)
+			{
+				while (i < len)
+				{
+					struct inotify_event *pevent = (struct inotify_event *)&buff[i];
+
+					mWatchesLock.lock();
+
+					wit = mWatches.find( pevent->wd );
+
+					if ( wit != mWatches.end() )
+					{
+						handleAction(wit->second, pevent->name, pevent->mask);
+
+						/// Keep track of the IN_MOVED_FROM events to known if the IN_MOVED_TO event is also fired
+						if ( !wit->second->OldFileName.empty() )
+						{
+							movedOutsideWatches.push_back( wit->second );
+						}
+					}
+
+					mWatchesLock.unlock();
+
+					i += sizeof(struct inotify_event) + pevent->len;
+				}
+
+				if ( !movedOutsideWatches.empty() )
+				{
+					/// In case that the IN_MOVED_TO is never fired means that the file was moved to other folder
+					for ( std::list<WatcherInotify*>::iterator it = movedOutsideWatches.begin(); it != movedOutsideWatches.end(); it++ )
+					{
+						if ( !(*it)->OldFileName.empty() )
+						{
+							/// So we send a IN_DELETE event for files that where moved outside of our scope
+							handleAction( *it, (*it)->OldFileName, IN_DELETE );
+
+							/// Remove the OldFileName
+							(*it)->OldFileName = "";
+						}
+					}
+
+					movedOutsideWatches.clear();
+				}
+			}
+		}
+	} while( mInitOK );
+}
+
+void FileWatcherInotify::checkForNewWatcher( Watcher* watch, std::string fpath )
+{
+	FileSystem::dirAddSlashAtEnd( fpath );
+
+	/// If the watcher is recursive, checks if the new file is a folder, and creates a watcher
+	if ( watch->Recursive && FileSystem::isDirectory( fpath ) )
+	{
+		bool found = false;
+
+		/// First check if exists
+		for ( WatchMap::iterator it = mWatches.begin(); it != mWatches.end(); it++ )
+		{
+			if ( it->second->Directory == fpath )
+			{
+				found = true;
+				break;
+			}
+		}
+
+		if ( !found )
+		{
+			addWatch( fpath, watch->Listener, watch->Recursive, static_cast<WatcherInotify*>( watch ) );
+		}
+	}
+}
+
+void FileWatcherInotify::handleAction( Watcher* watch, const std::string& filename, unsigned long action, std::string oldFilename )
+{
+	if ( !watch || !watch->Listener )
+	{
+		return;
+	}
+
+	std::string fpath( watch->Directory + filename );
+
+	if ( ( IN_CLOSE_WRITE & action ) || ( IN_MODIFY & action ) )
+	{
+		watch->Listener->handleFileAction( watch->ID, watch->Directory, filename, Actions::Modified );
+	}
+	else if( IN_MOVED_TO & action )
+	{
+		/// If OldFileName doesn't exist means that the file has been moved from other folder, so we just send the Add event
+		if ( watch->OldFileName.empty() )
+		{
+			watch->Listener->handleFileAction( watch->ID, watch->Directory, filename, Actions::Add );
+
+			watch->Listener->handleFileAction( watch->ID, watch->Directory, filename, Actions::Modified );
+
+			checkForNewWatcher( watch, fpath );
+		}
+		else
+		{
+			watch->Listener->handleFileAction( watch->ID, watch->Directory, filename, Actions::Moved, watch->OldFileName );
+		}
+
+		if ( watch->Recursive && FileSystem::isDirectory( fpath ) )
+		{
+			/// Update the new directory path
+			std::string opath( watch->Directory + watch->OldFileName );
+			FileSystem::dirAddSlashAtEnd( opath );
+			FileSystem::dirAddSlashAtEnd( fpath );
+
+			for ( WatchMap::iterator it = mWatches.begin(); it != mWatches.end(); it++ )
+			{
+				if ( it->second->Directory == opath && it->second->DirInfo.Inode == FileInfo( opath ).Inode )
+				{
+					it->second->Directory	= fpath;
+					it->second->DirInfo		= FileInfo( fpath );
+
+					break;
+				}
+			}
+		}
+
+		watch->OldFileName = "";
+	}
+	else if( IN_CREATE & action )
+	{
+		watch->Listener->handleFileAction( watch->ID, watch->Directory, filename, Actions::Add );
+
+		checkForNewWatcher( watch, fpath );
+	}
+	else if ( IN_MOVED_FROM & action )
+	{
+		watch->OldFileName = filename;
+	}
+	else if( IN_DELETE & action )
+	{
+		watch->Listener->handleFileAction( watch->ID, watch->Directory, filename, Actions::Delete );
+
+		FileSystem::dirAddSlashAtEnd( fpath );
+
+		/// If the file erased is a directory and recursive is enabled, removes the directory erased
+		if ( watch->Recursive )
+		{
+			for ( WatchMap::iterator it = mWatches.begin(); it != mWatches.end(); it++ )
+			{
+				if ( it->second->Directory == fpath )
+				{
+					removeWatch( it->second->ID );
+					break;
+				}
+			}
+		}
+	}
+}
+
+std::list<std::string> FileWatcherInotify::directories()
+{
+	std::list<std::string> dirs;
+
+	mWatchesLock.lock();
+
+	WatchMap::iterator it = mRealWatches.begin();
+
+	for ( ; it != mRealWatches.end(); it++ )
+	{
+		dirs.push_back( it->second->Directory );
+	}
+
+	mWatchesLock.unlock();
+
+	return dirs;
+}
+
+bool FileWatcherInotify::pathInWatches( const std::string& path )
+{
+	/// Search in the real watches, since it must allow adding a watch already watched as a subdir
+	WatchMap::iterator it = mRealWatches.begin();
+
+	for ( ; it != mRealWatches.end(); it++ )
+	{
+		if ( it->second->Directory == path )
+		{
+			return true;
+		}
+	}
+
+	return false;
+}
+
+}
+
+#endif
diff --git a/dep/efsw/src/efsw/FileWatcherInotify.hpp b/dep/efsw/src/efsw/FileWatcherInotify.hpp
new file mode 100644
index 00000000000..43ee9ca6afc
--- /dev/null
+++ b/dep/efsw/src/efsw/FileWatcherInotify.hpp
@@ -0,0 +1,73 @@
+#ifndef EFSW_FILEWATCHERLINUX_HPP
+#define EFSW_FILEWATCHERLINUX_HPP
+
+#include <efsw/FileWatcherImpl.hpp>
+
+#if EFSW_PLATFORM == EFSW_PLATFORM_INOTIFY
+
+#include <efsw/WatcherInotify.hpp>
+#include <map>
+
+namespace efsw
+{
+
+/// Implementation for Linux based on inotify.
+/// @class FileWatcherInotify
+class FileWatcherInotify : public FileWatcherImpl
+{
+	public:
+		/// type for a map from WatchID to WatchStruct pointer
+		typedef std::map<WatchID, WatcherInotify*> WatchMap;
+
+		FileWatcherInotify( FileWatcher * parent );
+
+		virtual ~FileWatcherInotify();
+
+		/// Add a directory watch
+		/// On error returns WatchID with Error type.
+		WatchID addWatch(const std::string& directory, FileWatchListener* watcher, bool recursive);
+
+		/// Remove a directory watch. This is a brute force lazy search O(nlogn).
+		void removeWatch(const std::string& directory);
+
+		/// Remove a directory watch. This is a map lookup O(logn).
+		void removeWatch(WatchID watchid);
+
+		/// Updates the watcher. Must be called often.
+		void watch();
+
+		/// Handles the action
+		void handleAction(Watcher * watch, const std::string& filename, unsigned long action, std::string oldFilename = "");
+
+		/// @return Returns a list of the directories that are being watched
+		std::list<std::string> directories();
+	protected:
+		/// Map of WatchID to WatchStruct pointers
+		WatchMap mWatches;
+
+		/// User added watches
+		WatchMap mRealWatches;
+
+		/// inotify file descriptor
+		int mFD;
+
+		Thread * mThread;
+
+		Mutex mWatchesLock;
+
+		WatchID addWatch(const std::string& directory, FileWatchListener* watcher, bool recursive, WatcherInotify * parent = NULL );
+
+		bool pathInWatches( const std::string& path );
+	private:
+		void run();
+
+		void removeWatchLocked(WatchID watchid);
+
+		void checkForNewWatcher( Watcher* watch, std::string fpath );
+};
+
+}
+
+#endif
+
+#endif
diff --git a/dep/efsw/src/efsw/FileWatcherKqueue.cpp b/dep/efsw/src/efsw/FileWatcherKqueue.cpp
new file mode 100644
index 00000000000..600fd085b35
--- /dev/null
+++ b/dep/efsw/src/efsw/FileWatcherKqueue.cpp
@@ -0,0 +1,274 @@
+#include <efsw/FileWatcherKqueue.hpp>
+
+#if EFSW_PLATFORM == EFSW_PLATFORM_KQUEUE || EFSW_PLATFORM == EFSW_PLATFORM_FSEVENTS
+
+#include <sys/time.h>
+#include <sys/stat.h>
+#include <unistd.h>
+#include <fcntl.h>
+#include <errno.h>
+#include <stdio.h>
+#include <stdlib.h>
+#include <dirent.h>
+#include <string.h>
+#include <efsw/FileSystem.hpp>
+#include <efsw/System.hpp>
+#include <efsw/Debug.hpp>
+#include <efsw/WatcherGeneric.hpp>
+
+namespace efsw
+{
+
+FileWatcherKqueue::FileWatcherKqueue( FileWatcher * parent ) :
+	FileWatcherImpl( parent ),
+	mLastWatchID(0),
+	mThread( NULL ),
+	mFileDescriptorCount( 1 ),
+	mAddingWatcher( false )
+{
+	mTimeOut.tv_sec		= 0;
+	mTimeOut.tv_nsec	= 0;
+	mInitOK				= true;
+}
+
+FileWatcherKqueue::~FileWatcherKqueue()
+{
+	WatchMap::iterator iter = mWatches.begin();
+
+	for(; iter != mWatches.end(); ++iter)
+	{
+		efSAFE_DELETE( iter->second );
+	}
+
+	mWatches.clear();
+
+	mInitOK = false;
+
+	mThread->wait();
+
+	efSAFE_DELETE( mThread );
+}
+
+WatchID FileWatcherKqueue::addWatch(const std::string& directory, FileWatchListener* watcher, bool recursive)
+{
+	static bool s_ug = false;
+
+	std::string dir( directory );
+
+	FileSystem::dirAddSlashAtEnd( dir );
+
+	FileInfo fi( dir );
+
+	if ( !fi.isDirectory() )
+	{
+		return Errors::Log::createLastError( Errors::FileNotFound, dir );
+	}
+	else if ( !fi.isReadable() )
+	{
+		return Errors::Log::createLastError( Errors::FileNotReadable, dir );
+	}
+	else if ( pathInWatches( dir ) )
+	{
+		return Errors::Log::createLastError( Errors::FileRepeated, directory );
+	}
+
+	std::string curPath;
+	std::string link( FileSystem::getLinkRealPath( dir, curPath ) );
+
+	if ( "" != link )
+	{
+		if ( pathInWatches( link ) )
+		{
+			return Errors::Log::createLastError( Errors::FileRepeated, directory );
+		}
+		else if ( !linkAllowed( curPath, link ) )
+		{
+			return Errors::Log::createLastError( Errors::FileOutOfScope, dir );
+		}
+		else
+		{
+			dir = link;
+		}
+	}
+
+	/// Check first if are enough file descriptors available to create another kqueue watcher, otherwise it creates a generic watcher
+	if ( availablesFD() )
+	{
+		mAddingWatcher = true;
+
+		WatcherKqueue * watch = new WatcherKqueue( ++mLastWatchID, dir, watcher, recursive, this );
+
+		mWatchesLock.lock();
+		mWatches.insert(std::make_pair(mLastWatchID, watch));
+		mWatchesLock.unlock();
+
+		watch->addAll();
+
+		// if failed to open the directory... erase the watcher
+		if ( !watch->initOK() )
+		{
+			int le = watch->lastErrno();
+
+			mWatches.erase( watch->ID );
+
+			efSAFE_DELETE( watch );
+
+			mLastWatchID--;
+
+			// Probably the folder has too many files, create a generic watcher
+			if ( EACCES != le )
+			{
+				WatcherGeneric * watch = new WatcherGeneric( ++mLastWatchID, dir, watcher, this, recursive );
+
+				mWatchesLock.lock();
+				mWatches.insert(std::make_pair(mLastWatchID, watch));
+				mWatchesLock.unlock();
+			}
+			else
+			{
+				return Errors::Log::createLastError( Errors::Unspecified, link );
+			}
+		}
+
+		mAddingWatcher = false;
+	}
+	else
+	{
+		if ( !s_ug )
+		{
+			efDEBUG( "Started using generic watcher, file descriptor limit reached: %ld\n", mFileDescriptorCount );
+			s_ug = true;
+		}
+
+		WatcherGeneric * watch = new WatcherGeneric( ++mLastWatchID, dir, watcher, this, recursive );
+
+		mWatchesLock.lock();
+		mWatches.insert(std::make_pair(mLastWatchID, watch));
+		mWatchesLock.unlock();
+	}
+
+	return mLastWatchID;
+}
+
+void FileWatcherKqueue::removeWatch(const std::string& directory)
+{
+	mWatchesLock.lock();
+
+	WatchMap::iterator iter = mWatches.begin();
+
+	for(; iter != mWatches.end(); ++iter)
+	{
+		if(directory == iter->second->Directory)
+		{
+			removeWatch(iter->first);
+			return;
+		}
+	}
+
+	mWatchesLock.unlock();
+}
+
+void FileWatcherKqueue::removeWatch(WatchID watchid)
+{
+	mWatchesLock.lock();
+
+	WatchMap::iterator iter = mWatches.find(watchid);
+
+	if(iter == mWatches.end())
+		return;
+
+	Watcher* watch = iter->second;
+
+	mWatches.erase(iter);
+
+	efSAFE_DELETE( watch );
+
+	mWatchesLock.unlock();
+}
+
+bool FileWatcherKqueue::isAddingWatcher() const
+{
+	return mAddingWatcher;
+}
+
+void FileWatcherKqueue::watch()
+{
+	if ( NULL == mThread )
+	{
+		mThread = new Thread( &FileWatcherKqueue::run, this );
+		mThread->launch();
+	}
+}
+
+void FileWatcherKqueue::run()
+{
+	do
+	{
+		mWatchesLock.lock();
+
+		for ( WatchMap::iterator it = mWatches.begin(); it != mWatches.end(); ++it )
+		{
+			it->second->watch();
+		}
+
+		mWatchesLock.unlock();
+
+		System::sleep( 500 );
+	} while( mInitOK );
+}
+
+void FileWatcherKqueue::handleAction(Watcher* watch, const std::string& filename, unsigned long action, std::string oldFilename)
+{
+}
+
+std::list<std::string> FileWatcherKqueue::directories()
+{
+	std::list<std::string> dirs;
+
+	mWatchesLock.lock();
+
+	WatchMap::iterator it = mWatches.begin();
+
+	for ( ; it != mWatches.end(); it++ )
+	{
+		dirs.push_back( it->second->Directory );
+	}
+
+	mWatchesLock.unlock();
+
+	return dirs;
+}
+
+bool FileWatcherKqueue::pathInWatches( const std::string& path )
+{
+	WatchMap::iterator it = mWatches.begin();
+
+	for ( ; it != mWatches.end(); it++ )
+	{
+		if ( it->second->Directory == path )
+		{
+			return true;
+		}
+	}
+
+	return false;
+}
+
+void FileWatcherKqueue::addFD()
+{
+	mFileDescriptorCount++;
+}
+
+void FileWatcherKqueue::removeFD()
+{
+	mFileDescriptorCount--;
+}
+
+bool FileWatcherKqueue::availablesFD()
+{
+	return mFileDescriptorCount <= (Int64)System::getMaxFD() - 500;
+}
+
+}
+
+#endif
diff --git a/dep/efsw/src/efsw/FileWatcherKqueue.hpp b/dep/efsw/src/efsw/FileWatcherKqueue.hpp
new file mode 100644
index 00000000000..0a2431e3777
--- /dev/null
+++ b/dep/efsw/src/efsw/FileWatcherKqueue.hpp
@@ -0,0 +1,78 @@
+#ifndef EFSW_FILEWATCHEROSX_HPP
+#define EFSW_FILEWATCHEROSX_HPP
+
+#include <efsw/FileWatcherImpl.hpp>
+
+#if EFSW_PLATFORM == EFSW_PLATFORM_KQUEUE || EFSW_PLATFORM == EFSW_PLATFORM_FSEVENTS
+
+#include <efsw/WatcherKqueue.hpp>
+
+namespace efsw
+{
+
+/// Implementation for OSX based on kqueue.
+/// @class FileWatcherKqueue
+class FileWatcherKqueue : public FileWatcherImpl
+{
+	friend class WatcherKqueue;
+	public:
+		FileWatcherKqueue( FileWatcher * parent );
+
+		virtual ~FileWatcherKqueue();
+
+		/// Add a directory watch
+		/// On error returns WatchID with Error type.
+		WatchID addWatch(const std::string& directory, FileWatchListener* watcher, bool recursive);
+
+		/// Remove a directory watch. This is a brute force lazy search O(nlogn).
+		void removeWatch(const std::string& directory);
+
+		/// Remove a directory watch. This is a map lookup O(logn).
+		void removeWatch(WatchID watchid);
+
+		/// Updates the watcher. Must be called often.
+		void watch();
+
+		/// Handles the action
+		void handleAction(Watcher* watch, const std::string& filename, unsigned long action, std::string oldFilename = "");
+
+		/// @return Returns a list of the directories that are being watched
+		std::list<std::string> directories();
+	protected:
+		/// Map of WatchID to WatchStruct pointers
+		WatchMap mWatches;
+
+		/// time out data
+		struct timespec mTimeOut;
+
+		/// WatchID allocator
+		int mLastWatchID;
+
+		Thread * mThread;
+
+		Mutex mWatchesLock;
+
+		std::list<WatchID> mRemoveList;
+
+		long mFileDescriptorCount;
+
+		bool mAddingWatcher;
+
+		bool isAddingWatcher() const;
+
+		bool pathInWatches( const std::string& path );
+
+		void addFD();
+
+		void removeFD();
+
+		bool availablesFD();
+	private:
+		void run();
+};
+
+}
+
+#endif
+
+#endif
diff --git a/dep/efsw/src/efsw/FileWatcherWin32.cpp b/dep/efsw/src/efsw/FileWatcherWin32.cpp
new file mode 100644
index 00000000000..fe78dd122cf
--- /dev/null
+++ b/dep/efsw/src/efsw/FileWatcherWin32.cpp
@@ -0,0 +1,291 @@
+#include <efsw/FileWatcherWin32.hpp>
+#include <efsw/FileSystem.hpp>
+#include <efsw/System.hpp>
+#include <efsw/String.hpp>
+
+#if EFSW_PLATFORM == EFSW_PLATFORM_WIN32
+
+namespace efsw
+{
+
+FileWatcherWin32::FileWatcherWin32( FileWatcher * parent ) :
+	FileWatcherImpl( parent ),
+	mLastWatchID(0),
+	mThread( NULL )
+{
+	mInitOK = true;
+}
+
+FileWatcherWin32::~FileWatcherWin32()
+{
+	WatchVector::iterator iter = mWatches.begin();
+
+	mWatchesLock.lock();
+
+	for(; iter != mWatches.end(); ++iter)
+	{
+		DestroyWatch((*iter));
+	}
+
+	mHandles.clear();
+	mWatches.clear();
+
+	mInitOK = false;
+
+	mWatchesLock.unlock();
+
+	efSAFE_DELETE( mThread );
+}
+
+WatchID FileWatcherWin32::addWatch(const std::string& directory, FileWatchListener* watcher, bool recursive)
+{
+	std::string dir( directory );
+
+	FileInfo fi( dir );
+
+	if ( !fi.isDirectory() )
+	{
+		return Errors::Log::createLastError( Errors::FileNotFound, dir );
+	}
+	else if ( !fi.isReadable() )
+	{
+		return Errors::Log::createLastError( Errors::FileNotReadable, dir );
+	}
+
+	FileSystem::dirAddSlashAtEnd( dir );
+
+	WatchID watchid = ++mLastWatchID;
+
+	mWatchesLock.lock();
+
+	WatcherStructWin32 * watch = CreateWatch( String::fromUtf8( dir ).toWideString().c_str(), recursive,		FILE_NOTIFY_CHANGE_CREATION |
+																			FILE_NOTIFY_CHANGE_LAST_WRITE |
+																			FILE_NOTIFY_CHANGE_FILE_NAME |
+																			FILE_NOTIFY_CHANGE_DIR_NAME |
+																			FILE_NOTIFY_CHANGE_SIZE
+	);
+
+	if( NULL == watch )
+	{
+		return Errors::Log::createLastError( Errors::FileNotFound, dir );
+	}
+
+	if ( pathInWatches( dir ) )
+	{
+		return Errors::Log::createLastError( Errors::FileRepeated, dir );
+	}
+
+	// Add the handle to the handles vector
+	watch->Watch->ID = watchid;
+	watch->Watch->Watch = this;
+	watch->Watch->Listener = watcher;
+	watch->Watch->DirName = new char[dir.length()+1];
+	strcpy(watch->Watch->DirName, dir.c_str());
+
+	mHandles.push_back( watch->Watch->DirHandle );
+	mWatches.push_back( watch );
+	mWatchesLock.unlock();
+
+	return watchid;
+}
+
+void FileWatcherWin32::removeWatch(const std::string& directory)
+{
+	mWatchesLock.lock();
+
+	WatchVector::iterator iter = mWatches.begin();
+
+	for(; iter != mWatches.end(); ++iter)
+	{
+		if(directory == (*iter)->Watch->DirName)
+		{
+			removeWatch((*iter)->Watch->ID);
+			return;
+		}
+	}
+
+	mWatchesLock.unlock();
+}
+
+void FileWatcherWin32::removeWatch(WatchID watchid)
+{
+	mWatchesLock.lock();
+
+	WatchVector::iterator iter = mWatches.begin();
+
+	WatcherStructWin32* watch = NULL;
+
+	for(; iter != mWatches.end(); ++iter)
+	{
+		// Find the watch ID
+		if ( (*iter)->Watch->ID == watchid )
+		{
+			watch	= (*iter);
+
+			mWatches.erase( iter );
+
+			// Remove handle from the handle vector
+			HandleVector::iterator it = mHandles.begin();
+
+			for ( ; it != mHandles.end(); it++ )
+			{
+				if ( watch->Watch->DirHandle == (*it) )
+				{
+					mHandles.erase( it );
+					break;
+				}
+			}
+
+			DestroyWatch(watch);
+
+			break;
+		}
+	}
+
+	mWatchesLock.unlock();
+}
+
+void FileWatcherWin32::watch()
+{
+	if ( NULL == mThread )
+	{
+		mThread = new Thread( &FileWatcherWin32::run, this );
+		mThread->launch();
+	}
+}
+
+void FileWatcherWin32::run()
+{
+	if ( mHandles.empty() )
+	{
+		return;
+	}
+
+	do
+	{
+		if ( !mHandles.empty() )
+		{
+			mWatchesLock.lock();
+
+			for ( std::size_t i = 0; i < mWatches.size(); i++ )
+			{
+				WatcherStructWin32 * watch = mWatches[ i ];
+
+				// If the overlapped struct was cancelled ( because the creator thread doesn't exists anymore ),
+				// we recreate the overlapped in the current thread and refresh the watch
+				if ( /*STATUS_CANCELED*/0xC0000120 == watch->Overlapped.Internal )
+				{
+					watch->Overlapped = OVERLAPPED();
+					RefreshWatch(watch);
+				}
+
+				// First ensure that the handle is the same, this means that the watch was not removed.
+				if ( HasOverlappedIoCompleted( &watch->Overlapped ) && mHandles[ i ] == watch->Watch->DirHandle )
+				{
+					DWORD bytes;
+
+					if ( GetOverlappedResult( watch->Watch->DirHandle, &watch->Overlapped, &bytes, FALSE ) )
+					{
+						WatchCallback( ERROR_SUCCESS, bytes, &watch->Overlapped );
+					}
+				}
+			}
+
+			mWatchesLock.unlock();
+
+			if ( mInitOK )
+			{
+				System::sleep( 10 );
+			}
+		}
+		else
+		{
+			// Wait for a new handle to be added
+			System::sleep( 10 );
+		}
+	} while ( mInitOK );
+}
+
+void FileWatcherWin32::handleAction(Watcher* watch, const std::string& filename, unsigned long action, std::string oldFilename)
+{
+	Action fwAction;
+
+	switch(action)
+	{
+	case FILE_ACTION_RENAMED_OLD_NAME:
+		watch->OldFileName = filename;
+		return;
+	case FILE_ACTION_ADDED:
+		fwAction = Actions::Add;
+		break;
+	case FILE_ACTION_RENAMED_NEW_NAME:
+	{
+		fwAction = Actions::Moved;
+
+		std::string fpath( watch->Directory + filename );
+
+		// Update the directory path
+		if ( watch->Recursive && FileSystem::isDirectory( fpath ) )
+		{
+			// Update the new directory path
+			std::string opath( watch->Directory + watch->OldFileName );
+			FileSystem::dirAddSlashAtEnd( opath );
+			FileSystem::dirAddSlashAtEnd( fpath );
+
+			for ( WatchVector::iterator it = mWatches.begin(); it != mWatches.end(); it++ )
+			{
+				if ( (*it)->Watch->Directory == opath )
+				{
+					(*it)->Watch->Directory = fpath;
+
+					break;
+				}
+			}
+		}
+
+		watch->Listener->handleFileAction(watch->ID, static_cast<WatcherWin32*>( watch )->DirName, filename, fwAction, watch->OldFileName);
+		return;
+	}
+	case FILE_ACTION_REMOVED:
+		fwAction = Actions::Delete;
+		break;
+	case FILE_ACTION_MODIFIED:
+		fwAction = Actions::Modified;
+		break;
+	};
+
+	watch->Listener->handleFileAction(watch->ID, static_cast<WatcherWin32*>( watch )->DirName, filename, fwAction);
+}
+
+std::list<std::string> FileWatcherWin32::directories()
+{
+	std::list<std::string> dirs;
+
+	mWatchesLock.lock();
+
+	for ( WatchVector::iterator it = mWatches.begin(); it != mWatches.end(); it++ )
+	{
+		dirs.push_back( std::string( (*it)->Watch->DirName ) );
+	}
+
+	mWatchesLock.unlock();
+
+	return dirs;
+}
+
+bool FileWatcherWin32::pathInWatches( const std::string& path )
+{
+	for ( WatchVector::iterator it = mWatches.begin(); it != mWatches.end(); it++ )
+	{
+		if ( (*it)->Watch->DirName == path )
+		{
+			return true;
+		}
+	}
+
+	return false;
+}
+
+}
+
+#endif
diff --git a/dep/efsw/src/efsw/FileWatcherWin32.hpp b/dep/efsw/src/efsw/FileWatcherWin32.hpp
new file mode 100644
index 00000000000..505cd772b82
--- /dev/null
+++ b/dep/efsw/src/efsw/FileWatcherWin32.hpp
@@ -0,0 +1,69 @@
+#ifndef EFSW_FILEWATCHERWIN32_HPP
+#define EFSW_FILEWATCHERWIN32_HPP
+
+#include <efsw/base.hpp>
+
+#if EFSW_PLATFORM == EFSW_PLATFORM_WIN32
+
+#include <efsw/WatcherWin32.hpp>
+#include <vector>
+#include <map>
+
+namespace efsw
+{
+
+/// Implementation for Win32 based on ReadDirectoryChangesW.
+/// @class FileWatcherWin32
+class FileWatcherWin32 : public FileWatcherImpl
+{
+	public:
+		/// type for a map from WatchID to WatcherWin32 pointer
+		typedef std::vector<WatcherStructWin32*> WatchVector;
+		typedef std::vector<HANDLE>	HandleVector;
+
+		FileWatcherWin32( FileWatcher * parent );
+
+		virtual ~FileWatcherWin32();
+
+		/// Add a directory watch
+		/// On error returns WatchID with Error type.
+		WatchID addWatch(const std::string& directory, FileWatchListener* watcher, bool recursive);
+
+		/// Remove a directory watch. This is a brute force lazy search O(nlogn).
+		void removeWatch(const std::string& directory);
+
+		/// Remove a directory watch. This is a map lookup O(logn).
+		void removeWatch(WatchID watchid);
+
+		/// Updates the watcher. Must be called often.
+		void watch();
+
+		/// Handles the action
+		void handleAction(Watcher* watch, const std::string& filename, unsigned long action, std::string oldFilename = "");
+
+		/// @return Returns a list of the directories that are being watched
+		std::list<std::string> directories();
+	protected:
+		/// Vector of WatcherWin32 pointers
+		WatchVector mWatches;
+
+		/// Keeps an updated handles vector
+		HandleVector mHandles;
+
+		/// The last watchid
+		WatchID mLastWatchID;
+
+		Thread * mThread;
+
+		Mutex mWatchesLock;
+
+		bool pathInWatches( const std::string& path );
+	private:
+		void run();
+};
+
+}
+
+#endif
+
+#endif
diff --git a/dep/efsw/src/efsw/Log.cpp b/dep/efsw/src/efsw/Log.cpp
new file mode 100644
index 00000000000..8e2860ac099
--- /dev/null
+++ b/dep/efsw/src/efsw/Log.cpp
@@ -0,0 +1,27 @@
+#include <efsw/efsw.hpp>
+
+namespace efsw { namespace Errors {
+
+static std::string LastError;
+
+std::string Log::getLastErrorLog()
+{
+	return LastError;
+}
+
+Error Log::createLastError( Error err, std::string log )
+{
+	switch ( err )
+	{
+		case FileNotFound:		LastError = "File not found ( " + log + " )";					break;
+		case FileRepeated:		LastError = "File reapeated in watches ( " + log + " )";		break;
+		case FileOutOfScope:	LastError = "Symlink file out of scope ( " + log + " )";		break;
+		case FileRemote:		LastError = "File is located in a remote file system, use a generic watcher. ( " + log + " )"; break;
+		case Unspecified:
+		default:				LastError = log;
+	}
+
+	return err;
+}
+
+}}
diff --git a/dep/efsw/src/efsw/Mutex.cpp b/dep/efsw/src/efsw/Mutex.cpp
new file mode 100644
index 00000000000..b34ba066ee9
--- /dev/null
+++ b/dep/efsw/src/efsw/Mutex.cpp
@@ -0,0 +1,26 @@
+#include <efsw/Mutex.hpp>
+#include <efsw/platform/platformimpl.hpp>
+
+namespace efsw {
+
+Mutex::Mutex() :
+	mMutexImpl( new Platform::MutexImpl() )
+{
+}
+
+Mutex::~Mutex()
+{
+	efSAFE_DELETE( mMutexImpl );
+}
+
+void Mutex::lock()
+{
+	mMutexImpl->lock();
+}
+
+void Mutex::unlock()
+{
+	mMutexImpl->unlock();
+}
+
+}
diff --git a/dep/efsw/src/efsw/Mutex.hpp b/dep/efsw/src/efsw/Mutex.hpp
new file mode 100644
index 00000000000..e6e89def175
--- /dev/null
+++ b/dep/efsw/src/efsw/Mutex.hpp
@@ -0,0 +1,28 @@
+#ifndef EFSW_MUTEX_HPP
+#define EFSW_MUTEX_HPP
+
+#include <efsw/base.hpp>
+
+namespace efsw {
+
+namespace Platform { class MutexImpl; }
+
+/** Simple mutex class */
+class Mutex {
+	public:
+		Mutex();
+
+		~Mutex();
+
+		/** Lock the mutex */
+		void lock();
+
+		/** Unlock the mutex */
+		void unlock();
+	private:
+		Platform::MutexImpl *		mMutexImpl;
+};
+
+}
+
+#endif
diff --git a/dep/efsw/src/efsw/String.cpp b/dep/efsw/src/efsw/String.cpp
new file mode 100644
index 00000000000..f703d25cb35
--- /dev/null
+++ b/dep/efsw/src/efsw/String.cpp
@@ -0,0 +1,813 @@
+#include <iterator>
+#include <efsw/String.hpp>
+#include <efsw/Utf.hpp>
+
+namespace efsw {
+
+const std::size_t String::InvalidPos = StringType::npos;
+
+std::vector < std::string > String::split ( const std::string& str, const char& splitchar, const bool& pushEmptyString )
+{
+	std::vector < std::string > tmp;
+	std::string tmpstr;
+
+	for ( size_t i = 0; i < str.size(); i++ )
+	{
+		if ( str[i] == splitchar )
+		{
+			if ( pushEmptyString || tmpstr.size() )
+			{
+				tmp.push_back(tmpstr);
+				tmpstr = "";
+			}
+		}
+		else
+		{
+			tmpstr += str[i];
+		}
+	}
+
+	if ( tmpstr.size() )
+	{
+		tmp.push_back( tmpstr );
+	}
+
+	return tmp;
+}
+
+std::vector < String > String::split ( const String& str, const Uint32& splitchar, const bool& pushEmptyString )
+{
+	std::vector < String > tmp;
+	String tmpstr;
+
+	for ( size_t i = 0; i < str.size(); i++ )
+	{
+		if ( str[i] == splitchar )
+		{
+			if ( pushEmptyString || tmpstr.size() )
+			{
+				tmp.push_back(tmpstr);
+				tmpstr = "";
+			}
+		}
+		else
+		{
+			tmpstr += str[i];
+		}
+	}
+
+	if ( tmpstr.size() )
+	{
+		tmp.push_back( tmpstr );
+	}
+
+	return tmp;
+}
+
+int String::strStartsWith( const std::string& start, const std::string& str )
+{
+	int pos		= -1;
+	size_t size	= start.size();
+
+	if ( str.size() >= size )
+	{
+		for ( std::size_t i = 0; i < size; i++ )
+		{
+			if ( start[i] == str[i] )
+			{
+				pos = (int)i;
+			}
+			else
+			{
+				pos = -1;
+				break;
+			}
+		}
+	}
+
+	return pos;
+}
+
+int String::strStartsWith( const String& start, const String& str )
+{
+	int pos		= -1;
+	size_t size	= start.size();
+
+	if ( str.size() >= size )
+	{
+		for ( std::size_t i = 0; i < size; i++ )
+		{
+			if ( start[i] == str[i] )
+			{
+				pos = (int)i;
+			}
+			else
+			{
+				pos = -1;
+				break;
+			}
+		}
+	}
+
+	return pos;
+}
+
+String::String()
+{
+}
+
+String::String(char ansiChar, const std::locale& locale)
+{
+	mString += Utf32::DecodeAnsi(ansiChar, locale);
+}
+
+#ifndef EFSW_NO_WIDECHAR
+String::String(wchar_t wideChar)
+{
+	mString += Utf32::DecodeWide(wideChar);
+}
+#endif
+
+String::String(StringBaseType utf32Char)
+{
+	mString += utf32Char;
+}
+
+String::String( const char* uf8String ) {
+	if (uf8String)
+	{
+		std::size_t length = strlen(uf8String);
+
+		if (length > 0)
+		{
+			mString.reserve(length + 1);
+
+			Utf8::ToUtf32(uf8String, uf8String + length, std::back_inserter(mString));
+		}
+	}
+}
+
+String::String( const std::string& utf8String ) {
+	mString.reserve( utf8String.length() + 1 );
+
+	Utf8::ToUtf32( utf8String.begin(), utf8String.end(), std::back_inserter( mString ) );
+}
+
+String::String(const char* ansiString, const std::locale& locale)
+{
+	if (ansiString)
+	{
+		std::size_t length = strlen(ansiString);
+		if (length > 0)
+		{
+			mString.reserve(length + 1);
+			Utf32::FromAnsi(ansiString, ansiString + length, std::back_inserter(mString), locale);
+		}
+	}
+}
+
+String::String(const std::string& ansiString, const std::locale& locale)
+{
+	mString.reserve(ansiString.length() + 1);
+	Utf32::FromAnsi(ansiString.begin(), ansiString.end(), std::back_inserter(mString), locale);
+}
+
+#ifndef EFSW_NO_WIDECHAR
+String::String(const wchar_t* wideString)
+{
+	if (wideString)
+	{
+		std::size_t length = std::wcslen(wideString);
+		if (length > 0)
+		{
+			mString.reserve(length + 1);
+			Utf32::FromWide(wideString, wideString + length, std::back_inserter(mString));
+		}
+	}
+}
+
+String::String(const std::wstring& wideString)
+{
+	mString.reserve(wideString.length() + 1);
+	Utf32::FromWide(wideString.begin(), wideString.end(), std::back_inserter(mString));
+}
+#endif
+
+String::String(const StringBaseType* utf32String)
+{
+	if (utf32String)
+		mString = utf32String;
+}
+
+String::String(const StringType& utf32String) :
+mString(utf32String)
+{
+}
+
+String::String(const String& str) :
+mString(str.mString)
+{
+}
+
+String String::fromUtf8( const std::string& utf8String )
+{
+	String::StringType utf32;
+
+	utf32.reserve( utf8String.length() + 1 );
+
+	Utf8::ToUtf32( utf8String.begin(), utf8String.end(), std::back_inserter( utf32 ) );
+
+	return String( utf32 );
+}
+
+String::operator std::string() const
+{
+	return toAnsiString();
+}
+
+std::string String::toAnsiString(const std::locale& locale) const
+{
+	// Prepare the output string
+	std::string output;
+	output.reserve(mString.length() + 1);
+
+	// Convert
+	Utf32::ToAnsi(mString.begin(), mString.end(), std::back_inserter(output), 0, locale);
+
+	return output;
+}
+
+#ifndef EFSW_NO_WIDECHAR
+std::wstring String::toWideString() const
+{
+	// Prepare the output string
+	std::wstring output;
+	output.reserve(mString.length() + 1);
+
+	// Convert
+	Utf32::ToWide(mString.begin(), mString.end(), std::back_inserter(output), 0);
+
+	return output;
+}
+#endif
+
+std::string String::toUtf8() const {
+	// Prepare the output string
+	std::string output;
+	output.reserve(mString.length() + 1);
+
+	// Convert
+	Utf32::toUtf8(mString.begin(), mString.end(), std::back_inserter(output) );
+
+	return output;
+}
+
+String& String::operator =(const String& right)
+{
+	mString = right.mString;
+	return *this;
+}
+
+String& String::operator =( const StringBaseType& right )
+{
+	mString = right;
+	return *this;
+}
+
+String& String::operator +=(const String& right)
+{
+	mString += right.mString;
+	return *this;
+}
+
+String& String::operator +=( const StringBaseType& right )
+{
+	mString += right;
+	return *this;
+}
+
+
+String::StringBaseType String::operator [](std::size_t index) const
+{
+	return mString[index];
+}
+
+String::StringBaseType& String::operator [](std::size_t index)
+{
+	return mString[index];
+}
+
+String::StringBaseType String::at( std::size_t index ) const
+{
+	return mString.at( index );
+}
+
+void String::push_back( StringBaseType c )
+{
+	mString.push_back( c );
+}
+
+void String::swap ( String& str )
+{
+	mString.swap( str.mString );
+}
+
+void String::clear()
+{
+	mString.clear();
+}
+
+std::size_t String::size() const
+{
+	return mString.size();
+}
+
+std::size_t String::length() const
+{
+	return mString.length();
+}
+
+bool String::empty() const
+{
+	return mString.empty();
+}
+
+void String::erase(std::size_t position, std::size_t count)
+{
+	mString.erase(position, count);
+}
+
+String& String::insert(std::size_t position, const String& str)
+{
+	mString.insert(position, str.mString);
+	return *this;
+}
+
+String& String::insert( std::size_t pos1, const String& str, std::size_t pos2, std::size_t n )
+{
+	mString.insert( pos1, str.mString, pos2, n );
+	return *this;
+}
+
+String& String::insert ( size_t pos1, const char* s, size_t n )
+{
+	String tmp( s );
+
+	mString.insert( pos1, tmp.data(), n );
+
+	return *this;
+}
+
+String& String::insert ( size_t pos1, size_t n, char c )
+{
+	mString.insert( pos1, n, c );
+	return *this;
+}
+
+String& String::insert ( size_t pos1, const char* s )
+{
+	String tmp( s );
+
+	mString.insert( pos1, tmp.data() );
+
+	return *this;
+}
+
+String::Iterator String::insert ( Iterator p, char c )
+{
+	return mString.insert( p, c );
+}
+
+void String::insert ( Iterator p, size_t n, char c )
+{
+	mString.insert( p, n, c );
+}
+
+const String::StringBaseType* String::c_str() const
+{
+	return mString.c_str();
+}
+
+const String::StringBaseType* String::data() const
+{
+	return mString.data();
+}
+
+String::Iterator String::begin()
+{
+	return mString.begin();
+}
+
+String::ConstIterator String::begin() const
+{
+	return mString.begin();
+}
+
+String::Iterator String::end()
+{
+	return mString.end();
+}
+
+String::ConstIterator String::end() const
+{
+	return mString.end();
+}
+
+String::ReverseIterator String::rbegin()
+{
+	return mString.rbegin();
+}
+
+String::ConstReverseIterator String::rbegin() const
+{
+	return mString.rbegin();
+}
+
+String::ReverseIterator String::rend()
+{
+	return mString.rend();
+}
+
+String::ConstReverseIterator String::rend() const
+{
+	return mString.rend();
+}
+
+void String::resize( std::size_t n, StringBaseType c )
+{
+	mString.resize( n, c );
+}
+
+void String::resize( std::size_t n )
+{
+	mString.resize( n );
+}
+
+std::size_t String::max_size() const
+{
+	return mString.max_size();
+}
+
+void String::reserve( size_t res_arg )
+{
+	mString.reserve( res_arg );
+}
+
+std::size_t String::capacity() const
+{
+	return mString.capacity();
+}
+
+String& String::assign ( const String& str )
+{
+	mString.assign( str.mString );
+	return *this;
+}
+
+String& String::assign ( const String& str, size_t pos, size_t n )
+{
+	mString.assign( str.mString, pos, n );
+	return *this;
+}
+
+String& String::assign ( const char* s, size_t n )
+{
+	String tmp( s );
+
+	mString.assign( tmp.mString );
+
+	return *this;
+}
+
+String& String::assign ( const char* s )
+{
+	String tmp( s );
+
+	mString.assign( tmp.mString );
+
+	return *this;
+}
+
+String& String::assign ( size_t n, char c )
+{
+	mString.assign( n, c );
+
+	return *this;
+}
+
+String& String::append ( const String& str )
+{
+	mString.append( str.mString );
+
+	return *this;
+}
+
+String& String::append ( const String& str, size_t pos, size_t n )
+{
+	mString.append( str.mString, pos, n );
+
+	return *this;
+}
+
+String& String::append ( const char* s, size_t n )
+{
+	String tmp( s );
+
+	mString.append( tmp.mString );
+
+	return *this;
+}
+
+String& String::append ( const char* s )
+{
+	String tmp( s );
+
+	mString.append( tmp.mString );
+
+	return *this;
+}
+
+String& String::append ( size_t n, char c )
+{
+	mString.append( n, c );
+
+	return *this;
+}
+
+String& String::append ( std::size_t n, StringBaseType c )
+{
+	mString.append( n, c );
+
+	return *this;
+}
+
+String& String::replace ( size_t pos1, size_t n1, const String& str )
+{
+	mString.replace( pos1, n1, str.mString );
+
+	return *this;
+}
+
+String& String::replace ( Iterator i1, Iterator i2, const String& str )
+{
+	mString.replace( i1, i2, str.mString );
+
+	return *this;
+}
+
+String& String::replace ( size_t pos1, size_t n1, const String& str, size_t pos2, size_t n2 )
+{
+	mString.replace( pos1, n1, str.mString, pos2, n2 );
+
+	return *this;
+}
+
+String& String::replace ( size_t pos1, size_t n1, const char* s, size_t n2 )
+{
+	String tmp( s );
+
+	mString.replace( pos1, n1, tmp.data(), n2 );
+
+	return *this;
+}
+
+String& String::replace ( Iterator i1, Iterator i2, const char* s, size_t n2 )
+{
+	String tmp( s );
+
+	mString.replace( i1, i2, tmp.data(), n2 );
+
+	return *this;
+}
+
+String& String::replace ( size_t pos1, size_t n1,   const char* s )
+{
+	String tmp( s );
+
+	mString.replace( pos1, n1, tmp.mString );
+
+	return *this;
+}
+
+String& String::replace ( Iterator i1, Iterator i2, const char* s )
+{
+	String tmp( s );
+
+	mString.replace( i1, i2, tmp.mString );
+
+	return *this;
+}
+
+String& String::replace ( size_t pos1, size_t n1,   size_t n2, char c )
+{
+	mString.replace( pos1, n1, n2, (StringBaseType)c );
+
+	return *this;
+}
+
+String& String::replace ( Iterator i1, Iterator i2, size_t n2, char c )
+{
+	mString.replace( i1, i2, n2, (StringBaseType)c );
+
+	return *this;
+}
+
+std::size_t String::find( const String& str, std::size_t start ) const
+{
+	return mString.find( str.mString, start );
+}
+
+std::size_t String::find ( const char* s, std::size_t pos, std::size_t n ) const
+{
+	return find( String( s ), pos );
+}
+
+std::size_t String::find ( const char* s, std::size_t pos ) const
+{
+	return find( String( s ), pos );
+}
+
+size_t String::find ( char c, std::size_t pos ) const
+{
+	return mString.find( (StringBaseType)c, pos );
+}
+
+std::size_t String::rfind ( const String& str, std::size_t pos ) const
+{
+	return mString.rfind( str.mString, pos );
+}
+
+std::size_t String::rfind ( const char* s, std::size_t pos, std::size_t n ) const
+{
+	return rfind( String( s ), pos );
+}
+
+std::size_t String::rfind ( const char* s, std::size_t pos ) const
+{
+	return rfind( String( s ), pos );
+}
+
+std::size_t String::rfind ( char c, std::size_t pos ) const
+{
+	return mString.rfind( c, pos );
+}
+
+std::size_t String::copy ( StringBaseType* s, std::size_t n, std::size_t pos ) const
+{
+	return mString.copy( s, n, pos );
+}
+
+String String::substr ( std::size_t pos, std::size_t n ) const
+{
+	return String( mString.substr( pos, n ) );
+}
+
+int String::compare ( const String& str ) const
+{
+	return mString.compare( str.mString );
+}
+
+int String::compare ( const char* s ) const
+{
+	return compare( String( s ) );
+}
+
+int String::compare ( std::size_t pos1, std::size_t n1, const String& str ) const
+{
+	return mString.compare( pos1, n1, str.mString );
+}
+
+int String::compare ( std::size_t pos1, std::size_t n1, const char* s) const
+{
+	return compare( pos1, n1, String( s ) );
+}
+
+int String::compare ( std::size_t pos1, std::size_t n1, const String& str, std::size_t pos2, std::size_t n2 ) const
+{
+	return mString.compare( pos1, n1, str.mString, pos2, n2 );
+}
+
+int String::compare ( std::size_t pos1, std::size_t n1, const char* s, std::size_t n2) const
+{
+	return compare( pos1, n1, String( s ), 0, n2 );
+}
+
+std::size_t String::find_first_of ( const String& str, std::size_t pos ) const
+{
+	return mString.find_first_of( str.mString, pos );
+}
+
+std::size_t String::find_first_of ( const char* s, std::size_t pos, std::size_t n ) const
+{
+	return find_first_of( String( s ), pos );
+}
+
+std::size_t String::find_first_of ( const char* s, std::size_t pos ) const
+{
+	return find_first_of( String( s ), pos );
+}
+
+std::size_t String::find_first_of ( StringBaseType c, std::size_t pos ) const
+{
+	return mString.find_first_of( c, pos );
+}
+
+std::size_t String::find_last_of ( const String& str, std::size_t pos ) const
+{
+	return mString.find_last_of( str.mString, pos );
+}
+
+std::size_t String::find_last_of ( const char* s, std::size_t pos, std::size_t n ) const
+{
+	return find_last_of( String( s ), pos );
+}
+
+std::size_t String::find_last_of ( const char* s, std::size_t pos ) const
+{
+	return find_last_of( String( s ), pos );
+}
+
+std::size_t String::find_last_of ( StringBaseType c, std::size_t pos) const
+{
+	return mString.find_last_of( c, pos );
+}
+
+std::size_t String::find_first_not_of ( const String& str, std::size_t pos ) const
+{
+	return mString.find_first_not_of( str.mString, pos );
+}
+
+std::size_t String::find_first_not_of ( const char* s, std::size_t pos, std::size_t n ) const
+{
+	return find_first_not_of( String( s ), pos );
+}
+
+std::size_t String::find_first_not_of ( const char* s, std::size_t pos ) const
+{
+	return find_first_not_of( String( s ), pos );
+}
+
+std::size_t String::find_first_not_of ( StringBaseType c, std::size_t pos ) const
+{
+	return mString.find_first_not_of( c, pos );
+}
+
+std::size_t String::find_last_not_of ( const String& str, std::size_t pos ) const
+{
+	return mString.find_last_not_of( str.mString, pos );
+}
+
+std::size_t String::find_last_not_of ( const char* s, std::size_t pos, std::size_t n ) const
+{
+	return find_last_not_of( String( s ), pos );
+}
+
+std::size_t String::find_last_not_of ( const char* s, std::size_t pos ) const
+{
+	return find_last_not_of( String( s ), pos );
+}
+
+std::size_t String::find_last_not_of ( StringBaseType c, std::size_t pos ) const
+{
+	return mString.find_last_not_of( c, pos );
+}
+
+bool operator ==(const String& left, const String& right)
+{
+	return left.mString == right.mString;
+}
+
+bool operator !=(const String& left, const String& right)
+{
+	return !(left == right);
+}
+
+bool operator <(const String& left, const String& right)
+{
+	return left.mString < right.mString;
+}
+
+bool operator >(const String& left, const String& right)
+{
+	return right < left;
+}
+
+bool operator <=(const String& left, const String& right)
+{
+	return !(right < left);
+}
+
+bool operator >=(const String& left, const String& right)
+{
+	return !(left < right);
+}
+
+String operator +(const String& left, const String& right)
+{
+	String string = left;
+	string += right;
+
+	return string;
+}
+
+}
diff --git a/dep/efsw/src/efsw/String.hpp b/dep/efsw/src/efsw/String.hpp
new file mode 100644
index 00000000000..ce7e3b75f89
--- /dev/null
+++ b/dep/efsw/src/efsw/String.hpp
@@ -0,0 +1,629 @@
+/** NOTE:
+*	This code is based on the Utf implementation from SFML2. License zlib/png ( http://www.sfml-dev.org/license.php )
+*	The class was modified to fit efsw own needs. This is not the original implementation from SFML2.
+*	Functions and methods are the same that in std::string to facilitate portability.
+**/
+
+#ifndef EFSW_STRING_HPP
+#define EFSW_STRING_HPP
+
+#include <efsw/base.hpp>
+#include <locale>
+#include <string>
+#include <cstring>
+#include <cstdlib>
+#include <iostream>
+#include <fstream>
+#include <sstream>
+#include <vector>
+
+namespace efsw {
+
+/** @brief Utility string class that automatically handles conversions between types and encodings **/
+class String
+{
+	public :
+	typedef Uint32								StringBaseType;
+	typedef std::basic_string<StringBaseType>	StringType;
+	typedef StringType::iterator				Iterator;				//! Iterator type
+	typedef StringType::const_iterator			ConstIterator;			//! Constant iterator type
+	typedef StringType::reverse_iterator		ReverseIterator;		//! Reverse Iterator type
+	typedef StringType::const_reverse_iterator	ConstReverseIterator;	//! Constant iterator type
+
+	static const std::size_t InvalidPos; ///< Represents an invalid position in the string
+
+	template <class T>
+	static std::string toStr(const T& i) {
+		std::ostringstream ss;
+		ss << i;
+		return ss.str();
+	}
+
+	/** Converts from a string to type */
+	template <class T>
+	static bool fromString(T& t, const std::string& s, std::ios_base& (*f)(std::ios_base&) = std::dec  ) {
+		std::istringstream iss(s);
+		return !(iss >> f >> t).fail();
+	}
+
+	/** Converts from a String to type */
+	template <class T>
+	static bool fromString(T& t, const String& s, std::ios_base& (*f)(std::ios_base&) = std::dec ) {
+		std::istringstream iss( s.toUtf8() );
+		return !(iss >> f >> t).fail();
+	}
+
+	/** Split a string and hold it on a vector */
+	static std::vector < std::string > split( const std::string& str, const char& splitchar, const bool& pushEmptyString = false );
+
+	/** Split a string and hold it on a vector */
+	static std::vector < String > split( const String& str, const Uint32& splitchar, const bool& pushEmptyString = false );
+
+	/** Determine if a string starts with the string passed
+	** @param start The substring expected to start
+	** @param str The string to compare
+	** @return -1 if the substring is no in str, otherwise the size of the substring
+	*/
+	static int strStartsWith( const std::string& start, const std::string& str );
+
+	static int strStartsWith( const String& start, const String& str );
+
+	/** @brief Construct from an UTF-8 string to UTF-32 according
+	** @param uf8String UTF-8 string to convert
+	**/
+	static String fromUtf8( const std::string& utf8String );
+
+	/** @brief Default constructor
+	** This constructor creates an empty string.
+	**/
+	String();
+
+	/** @brief Construct from a single ANSI character and a locale
+	** The source character is converted to UTF-32 according
+	** to the given locale. If you want to use the current global
+	** locale, rather use the other constructor.
+	** @param ansiChar ANSI character to convert
+	** @param locale   Locale to use for conversion
+	**/
+	String( char ansiChar, const std::locale& locale = std::locale() );
+
+#ifndef EFSW_NO_WIDECHAR
+	/** @brief Construct from single wide character
+	** @param wideChar Wide character to convert
+	**/
+	String( wchar_t wideChar );
+#endif
+
+	/** @brief Construct from single UTF-32 character
+	** @param utf32Char UTF-32 character to convert
+	**/
+	String( StringBaseType utf32Char );
+
+	/** @brief Construct from an from a null-terminated C-style UTF-8 string to UTF-32
+	** @param uf8String UTF-8 string to convert
+	**/
+	String( const char* uf8String );
+
+	/** @brief Construct from an UTF-8 string to UTF-32 according
+	** @param uf8String UTF-8 string to convert
+	**/
+	String( const std::string& utf8String );
+
+	/** @brief Construct from a null-terminated C-style ANSI string and a locale
+	** The source string is converted to UTF-32 according
+	** to the given locale. If you want to use the current global
+	** locale, rather use the other constructor.
+	** @param ansiString ANSI string to convert
+	** @param locale     Locale to use for conversion
+	**/
+	String( const char* ansiString, const std::locale& locale );
+
+	/** @brief Construct from an ANSI string and a locale
+	** The source string is converted to UTF-32 according
+	** to the given locale. If you want to use the current global
+	** locale, rather use the other constructor.
+	** @param ansiString ANSI string to convert
+	** @param locale     Locale to use for conversion
+	**/
+	String( const std::string& ansiString, const std::locale& locale );
+
+#ifndef EFSW_NO_WIDECHAR
+	/** @brief Construct from null-terminated C-style wide string
+	** @param wideString Wide string to convert
+	**/
+	String( const wchar_t* wideString );
+
+	/** @brief Construct from a wide string
+	** @param wideString Wide string to convert
+	**/
+	String( const std::wstring& wideString );
+#endif
+
+	/** @brief Construct from a null-terminated C-style UTF-32 string
+	** @param utf32String UTF-32 string to assign
+	**/
+	String( const StringBaseType* utf32String );
+
+	/** @brief Construct from an UTF-32 string
+	** @param utf32String UTF-32 string to assign
+	**/
+	String( const StringType& utf32String );
+
+
+	/** @brief Copy constructor
+	** @param str Instance to copy
+	**/
+	String( const String& str );
+
+	/** @brief Implicit cast operator to std::string (ANSI string)
+	** The current global locale is used for conversion. If you
+	** want to explicitely specify a locale, see toAnsiString.
+	** Characters that do not fit in the target encoding are
+	** discarded from the returned string.
+	** This operator is defined for convenience, and is equivalent
+	** to calling toAnsiString().
+	** @return Converted ANSI string
+	** @see toAnsiString, operator String
+	**/
+	operator std::string() const;
+
+	/** @brief Convert the unicode string to an ANSI string
+	** The UTF-32 string is converted to an ANSI string in
+	** the encoding defined by \a locale. If you want to use
+	** the current global locale, see the other overload
+	** of toAnsiString.
+	** Characters that do not fit in the target encoding are
+	** discarded from the returned string.
+	** @param locale Locale to use for conversion
+	** @return Converted ANSI string
+	** @see toWideString, operator std::string
+	**/
+	std::string toAnsiString( const std::locale& locale = std::locale() ) const;
+
+#ifndef EFSW_NO_WIDECHAR
+	/** @brief Convert the unicode string to a wide string
+	** Characters that do not fit in the target encoding are
+	** discarded from the returned string.
+	** @return Converted wide string
+	** @see toAnsiString, operator String
+	**/
+	std::wstring toWideString() const;
+#endif
+
+	std::string toUtf8() const;
+
+	/** @brief Overload of assignment operator
+	** @param right Instance to assign
+	** @return Reference to self
+	**/
+	String& operator =(const String& right);
+
+	String& operator =( const StringBaseType& right );
+
+	/** @brief Overload of += operator to append an UTF-32 string
+	** @param right String to append
+	** @return Reference to self
+	**/
+	String& operator +=(const String& right);
+
+	String& operator +=( const StringBaseType& right );
+
+	/** @brief Overload of [] operator to access a character by its position
+	** This function provides read-only access to characters.
+	** Note: this function doesn't throw if \a index is out of range.
+	** @param index Index of the character to get
+	** @return Character at position \a index
+	**/
+	StringBaseType operator [](std::size_t index) const;
+
+	/** @brief Overload of [] operator to access a character by its position
+	** This function provides read and write access to characters.
+	** Note: this function doesn't throw if \a index is out of range.
+	** @param index Index of the character to get
+	** @return Reference to the character at position \a index
+	**/
+
+	StringBaseType& operator [](std::size_t index);
+
+	/** @brief Get character in string
+	** Performs a range check, throwing an exception of type out_of_range in case that pos is not an actual position in the string.
+	** @return The character at position pos in the string.
+	*/
+	StringBaseType at( std::size_t index ) const;
+	
+	/** @brief clear the string
+	** This function removes all the characters from the string.
+	** @see empty, erase
+	**/
+	void clear();
+
+	/** @brief Get the size of the string
+	** @return Number of characters in the string
+	** @see empty
+	**/
+	std::size_t size() const;
+	
+	/** @see size() */
+	std::size_t length() const;
+
+	/** @brief Check whether the string is empty or not
+	** @return True if the string is empty (i.e. contains no character)
+	** @see clear, size
+	**/
+	bool empty() const;
+
+	/** @brief Erase one or more characters from the string
+	** This function removes a sequence of \a count characters
+	** starting from \a position.
+	** @param position Position of the first character to erase
+	** @param count    Number of characters to erase
+	**/
+	void erase(std::size_t position, std::size_t count = 1);
+
+
+	/** @brief Insert one or more characters into the string
+	** This function inserts the characters of \a str
+	** into the string, starting from \a position.
+	** @param position Position of insertion
+	** @param str      Characters to insert
+	**/
+	String& insert(std::size_t position, const String& str);
+
+	String& insert( std::size_t pos1, const String& str, std::size_t pos2, std::size_t n );
+
+	String& insert ( std::size_t pos1, const char* s, std::size_t n );
+
+	String& insert ( std::size_t pos1, const char* s );
+
+	String& insert ( std::size_t pos1, size_t n, char c );
+
+	Iterator insert ( Iterator p, char c );
+
+	void insert ( Iterator p, std::size_t n, char c );
+
+	template<class InputIterator>
+	void insert ( Iterator p, InputIterator first, InputIterator last )
+	{
+		mString.insert( p, first, last );
+	}
+
+	/** @brief Find a sequence of one or more characters in the string
+	** This function searches for the characters of \a str
+	** into the string, starting from \a start.
+	** @param str   Characters to find
+	** @param start Where to begin searching
+	** @return Position of \a str in the string, or String::InvalidPos if not found
+	**/
+	std::size_t find( const String& str, std::size_t start = 0 ) const;
+
+	std::size_t find ( const char* s, std::size_t pos, std::size_t n ) const;
+
+	std::size_t find ( const char* s, std::size_t pos = 0 ) const;
+
+	std::size_t find ( char c, std::size_t pos = 0 ) const;
+
+	/** @brief Get a pointer to the C-style array of characters
+	** This functions provides a read-only access to a
+	** null-terminated C-style representation of the string.
+	** The returned pointer is temporary and is meant only for
+	** immediate use, thus it is not recommended to store it.
+	** @return Read-only pointer to the array of characters
+	**/
+	const StringBaseType* c_str() const;
+	
+	/** @brief Get string data
+	** Notice that no terminating null character is appended (see member c_str for such a functionality).
+	** The returned array points to an internal location which should not be modified directly in the program.
+	** Its contents are guaranteed to remain unchanged only until the next call to a non-constant member function of the string object.
+	** @return Pointer to an internal array containing the same content as the string.
+	**/
+	const StringBaseType* data() const;
+
+	/** @brief Return an iterator to the beginning of the string
+	** @return Read-write iterator to the beginning of the string characters
+	** @see end
+	**/
+	Iterator begin();
+
+	/** @brief Return an iterator to the beginning of the string
+	** @return Read-only iterator to the beginning of the string characters
+	** @see end
+	**/
+	ConstIterator begin() const;
+
+	/** @brief Return an iterator to the beginning of the string
+	** The end iterator refers to 1 position past the last character;
+	** thus it represents an invalid character and should never be
+	** accessed.
+	** @return Read-write iterator to the end of the string characters
+	** @see begin
+	**/
+	Iterator end();
+
+	/** @brief Return an iterator to the beginning of the string
+	** The end iterator refers to 1 position past the last character;
+	** thus it represents an invalid character and should never be
+	** accessed.
+	** @return Read-only iterator to the end of the string characters
+	** @see begin
+	**/
+	ConstIterator end() const;
+	
+	/** @brief Return an reverse iterator to the beginning of the string
+	** @return Read-write reverse iterator to the beginning of the string characters
+	** @see end
+	**/
+	ReverseIterator rbegin();
+
+	/** @brief Return an reverse iterator to the beginning of the string
+	** @return Read-only reverse iterator to the beginning of the string characters
+	** @see end
+	**/
+	ConstReverseIterator rbegin() const;
+
+	/** @brief Return an reverse iterator to the beginning of the string
+	** The end reverse iterator refers to 1 position past the last character;
+	** thus it represents an invalid character and should never be
+	** accessed.
+	** @return Read-write reverse iterator to the end of the string characters
+	** @see begin
+	**/
+	ReverseIterator rend();
+
+
+	/** @brief Return an reverse iterator to the beginning of the string
+	** The end reverse iterator refers to 1 position past the last character;
+	** thus it represents an invalid character and should never be
+	** accessed.
+	** @return Read-only reverse iterator to the end of the string characters
+	** @see begin
+	**/
+	ConstReverseIterator rend() const;
+	
+	/** @brief Resize String */
+	void resize ( std::size_t n, StringBaseType c );
+	
+	/** @brief Resize String */
+	void resize ( std::size_t n );
+	
+	/** @return Maximum size of string */
+	std::size_t max_size() const;
+	
+	/** @brief Request a change in capacity */
+	void reserve ( size_t res_arg=0 );
+	
+	/** @return Size of allocated storage */
+	std::size_t capacity() const;
+
+	/** @brief Append character to string */
+	void push_back( StringBaseType c );
+	
+	/** @brief Swap contents with another string */
+	void swap ( String& str );
+	
+	String& assign ( const String& str );
+
+	String& assign ( const String& str, std::size_t pos, std::size_t n );
+
+	String& assign ( const char* s, std::size_t n );
+
+	String& assign ( const char* s );
+
+	String& assign ( std::size_t n, char c );
+
+	template <class InputIterator>
+	String& assign ( InputIterator first, InputIterator last )
+	{
+		mString.assign( first, last );
+		return *this;
+	}
+
+	String& append ( const String& str );
+
+	String& append ( const String& str, std::size_t pos, std::size_t n );
+
+	String& append ( const char* s, std::size_t n );
+
+	String& append ( const char* s );
+
+	String& append ( std::size_t n, char c );
+
+	String& append ( std::size_t n, StringBaseType c );
+
+	template <class InputIterator>
+	String& append ( InputIterator first, InputIterator last )
+	{
+		mString.append( first, last );
+		return *this;
+	}
+
+	String& replace ( std::size_t pos1, std::size_t n1,   const String& str );
+
+	String& replace ( Iterator i1, Iterator i2, const String& str );
+
+	String& replace ( std::size_t pos1, std::size_t n1, const String& str, std::size_t pos2, std::size_t n2 );
+
+	String& replace ( std::size_t pos1, std::size_t n1, const char* s, std::size_t n2 );
+
+	String& replace ( Iterator i1, Iterator i2, const char* s, std::size_t n2 );
+
+	String& replace ( std::size_t pos1, std::size_t n1,   const char* s );
+
+	String& replace ( Iterator i1, Iterator i2, const char* s );
+
+	String& replace ( std::size_t pos1, std::size_t n1, std::size_t n2, char c );
+
+	String& replace ( Iterator i1, Iterator i2, std::size_t n2, char c );
+
+	template<class InputIterator>
+	String& replace ( Iterator i1, Iterator i2, InputIterator j1, InputIterator j2 )
+	{
+		mString.replace( i1, i2, j1, j2 );
+		return *this;
+	}
+
+	std::size_t rfind ( const String& str, std::size_t pos = StringType::npos ) const;
+
+	std::size_t rfind ( const char* s, std::size_t pos, std::size_t n ) const;
+
+	std::size_t rfind ( const char* s, std::size_t pos = StringType::npos ) const;
+
+	std::size_t rfind ( char c, std::size_t pos = StringType::npos ) const;
+
+	String substr ( std::size_t pos = 0, std::size_t n = StringType::npos ) const;
+
+	std::size_t copy ( StringBaseType* s, std::size_t n, std::size_t pos = 0 ) const;
+
+	int compare ( const String& str ) const;
+
+	int compare ( const char* s ) const;
+
+	int compare ( std::size_t pos1, std::size_t n1, const String& str ) const;
+
+	int compare ( std::size_t pos1, std::size_t n1, const char* s) const;
+
+	int compare ( std::size_t pos1, std::size_t n1, const String& str, std::size_t pos2, std::size_t n2 ) const;
+
+	int compare ( std::size_t pos1, std::size_t n1, const char* s, std::size_t n2) const;
+
+	std::size_t find_first_of ( const String& str, std::size_t pos = 0 ) const;
+
+	std::size_t find_first_of ( const char* s, std::size_t pos, std::size_t n ) const;
+
+	std::size_t find_first_of ( const char* s, std::size_t pos = 0 ) const;
+
+	std::size_t find_first_of ( StringBaseType c, std::size_t pos = 0 ) const;
+
+	std::size_t find_last_of ( const String& str, std::size_t pos = StringType::npos ) const;
+
+	std::size_t find_last_of ( const char* s, std::size_t pos, std::size_t n ) const;
+
+	std::size_t find_last_of ( const char* s, std::size_t pos = StringType::npos ) const;
+
+	std::size_t find_last_of ( StringBaseType c, std::size_t pos = StringType::npos ) const;
+
+	std::size_t find_first_not_of ( const String& str, std::size_t pos = 0 ) const;
+
+	std::size_t find_first_not_of ( const char* s, std::size_t pos, std::size_t n ) const;
+
+	std::size_t find_first_not_of ( const char* s, std::size_t pos = 0 ) const;
+
+	std::size_t find_first_not_of ( StringBaseType c, std::size_t pos = 0 ) const;
+
+	std::size_t find_last_not_of ( const String& str, std::size_t pos = StringType::npos ) const;
+
+	std::size_t find_last_not_of ( const char* s, std::size_t pos, std::size_t n ) const;
+
+	std::size_t find_last_not_of ( const char* s, std::size_t pos = StringType::npos ) const;
+
+	std::size_t find_last_not_of ( StringBaseType c, std::size_t pos = StringType::npos ) const;
+private :
+	friend  bool operator ==(const String& left, const String& right);
+	friend  bool operator <(const String& left, const String& right);
+
+	StringType mString; ///< Internal string of UTF-32 characters
+};
+
+/** @relates String
+** @brief Overload of == operator to compare two UTF-32 strings
+** @param left  Left operand (a string)
+** @param right Right operand (a string)
+** @return True if both strings are equal
+**/
+ bool operator ==(const String& left, const String& right);
+
+/** @relates String
+** @brief Overload of != operator to compare two UTF-32 strings
+** @param left  Left operand (a string)
+** @param right Right operand (a string)
+** @return True if both strings are different
+**/
+ bool operator !=(const String& left, const String& right);
+
+/** @relates String
+** @brief Overload of < operator to compare two UTF-32 strings
+** @param left  Left operand (a string)
+** @param right Right operand (a string)
+** @return True if \a left is alphabetically lesser than \a right
+**/
+ bool operator <(const String& left, const String& right);
+
+/** @relates String
+** @brief Overload of > operator to compare two UTF-32 strings
+** @param left  Left operand (a string)
+** @param right Right operand (a string)
+** @return True if \a left is alphabetically greater than \a right
+**/
+ bool operator >(const String& left, const String& right);
+
+/** @relates String
+** @brief Overload of <= operator to compare two UTF-32 strings
+** @param left  Left operand (a string)
+** @param right Right operand (a string)
+** @return True if \a left is alphabetically lesser or equal than \a right
+**/
+ bool operator <=(const String& left, const String& right);
+
+/** @relates String
+** @brief Overload of >= operator to compare two UTF-32 strings
+** @param left  Left operand (a string)
+** @param right Right operand (a string)
+** @return True if \a left is alphabetically greater or equal than \a right
+**/
+ bool operator >=(const String& left, const String& right);
+
+/** @relates String
+** @brief Overload of binary + operator to concatenate two strings
+** @param left  Left operand (a string)
+** @param right Right operand (a string)
+** @return Concatenated string
+**/
+ String operator +( const String& left, const String& right );
+
+}
+
+#endif
+
+/** @class efsw::String
+** @ingroup system
+** efsw::String is a utility string class defined mainly for
+** convenience. It is a Unicode string (implemented using
+** UTF-32), thus it can store any character in the world
+** (european, chinese, arabic, hebrew, etc.).
+** It automatically handles conversions from/to ANSI and
+** wide strings, so that you can work with standard string
+** classes and still be compatible with functions taking a
+** efsw::String.
+** @code
+** efsw::String s;
+** std::string s1 = s;  // automatically converted to ANSI string
+** String s2 = s; // automatically converted to wide string
+** s = "hello";         // automatically converted from ANSI string
+** s = L"hello";        // automatically converted from wide string
+** s += 'a';            // automatically converted from ANSI string
+** s += L'a';           // automatically converted from wide string
+** @endcode
+** Conversions involving ANSI strings use the default user locale. However
+** it is possible to use a custom locale if necessary:
+** @code
+** std::locale locale;
+** efsw::String s;
+** ...
+** std::string s1 = s.toAnsiString(locale);
+** s = efsw::String("hello", locale);
+** @endcode
+**
+** efsw::String defines the most important functions of the
+** standard std::string class: removing, random access, iterating,
+** appending, comparing, etc. However it is a simple class
+** provided for convenience, and you may have to consider using
+** a more optimized class if your program requires complex string
+** handling. The automatic conversion functions will then take
+** care of converting your string to efsw::String whenever EE
+** requires it.
+**
+** Please note that EE also defines a low-level, generic
+** interface for Unicode handling, see the efsw::Utf classes.
+**
+** All credits to Laurent Gomila, i just modified and expanded a little bit the implementation.
+**/
diff --git a/dep/efsw/src/efsw/System.cpp b/dep/efsw/src/efsw/System.cpp
new file mode 100644
index 00000000000..f8767cbc58f
--- /dev/null
+++ b/dep/efsw/src/efsw/System.cpp
@@ -0,0 +1,26 @@
+#include <efsw/System.hpp>
+#include <efsw/platform/platformimpl.hpp>
+
+namespace efsw {
+
+void System::sleep( const unsigned long& ms )
+{
+	Platform::System::sleep( ms );
+}
+
+std::string System::getProcessPath()
+{
+	return Platform::System::getProcessPath();
+}
+
+void System::maxFD()
+{
+	Platform::System::maxFD();
+}
+
+Uint64 System::getMaxFD()
+{
+	return Platform::System::getMaxFD();
+}
+
+} 
diff --git a/dep/efsw/src/efsw/System.hpp b/dep/efsw/src/efsw/System.hpp
new file mode 100644
index 00000000000..1d6f4a70f91
--- /dev/null
+++ b/dep/efsw/src/efsw/System.hpp
@@ -0,0 +1,26 @@
+#ifndef EFSW_SYSTEM_HPP
+#define EFSW_SYSTEM_HPP
+
+#include <efsw/base.hpp>
+
+namespace efsw {
+
+class System
+{
+	public:
+		/// Sleep for x milliseconds
+		static void sleep( const unsigned long& ms );
+
+		/// @return The process binary path
+		static std::string getProcessPath();
+
+		/// Maximize the number of file descriptors allowed per process in the current OS
+		static void maxFD();
+
+		/// @return The number of supported file descriptors for the process
+		static Uint64 getMaxFD();
+};
+	
+}
+
+#endif
diff --git a/dep/efsw/src/efsw/Thread.cpp b/dep/efsw/src/efsw/Thread.cpp
new file mode 100644
index 00000000000..fff41517dc1
--- /dev/null
+++ b/dep/efsw/src/efsw/Thread.cpp
@@ -0,0 +1,51 @@
+#include <efsw/Thread.hpp>
+#include <efsw/platform/platformimpl.hpp>
+
+namespace efsw {
+
+Thread::Thread() :
+	mThreadImpl(NULL),
+	mEntryPoint(NULL)
+{
+}
+
+Thread::~Thread()
+{
+	wait();
+
+	efSAFE_DELETE( mEntryPoint );
+}
+
+void Thread::launch()
+{
+	wait();
+
+	mThreadImpl = new Platform::ThreadImpl( this );
+}
+
+void Thread::wait()
+{
+	if ( mThreadImpl )
+	{
+		mThreadImpl->wait();
+
+		efSAFE_DELETE( mThreadImpl );
+	}
+}
+
+void Thread::terminate()
+{
+	if ( mThreadImpl )
+	{
+		mThreadImpl->terminate();
+
+		efSAFE_DELETE( mThreadImpl );
+	}
+}
+
+void Thread::run()
+{
+	mEntryPoint->run();
+}
+
+}
diff --git a/dep/efsw/src/efsw/Thread.hpp b/dep/efsw/src/efsw/Thread.hpp
new file mode 100644
index 00000000000..c7da9e8b33c
--- /dev/null
+++ b/dep/efsw/src/efsw/Thread.hpp
@@ -0,0 +1,111 @@
+#ifndef EFSW_THREAD_HPP
+#define EFSW_THREAD_HPP
+
+#include <efsw/base.hpp>
+
+namespace efsw {
+
+namespace Platform { class ThreadImpl; }
+namespace Private { struct ThreadFunc; }
+
+/** @brief Thread manager class */
+class Thread {
+	public:
+		typedef void (*FuncType)(void*);
+
+		template <typename F>
+		Thread( F function );
+
+		template <typename F, typename A>
+		Thread( F function, A argument );
+
+		template <typename C>
+		Thread( void(C::*function)(), C* object );
+
+		virtual ~Thread();
+
+		/** Launch the thread */
+		virtual void	launch();
+
+		/** Wait the thread until end */
+		void			wait();
+
+		/** Terminate the thread */
+		void			terminate();
+	protected:
+		Thread();
+	private:
+		friend class Platform::ThreadImpl;
+
+		/** The virtual function to run in the thread */
+		virtual void	run();
+
+		Platform::ThreadImpl *		mThreadImpl;       ///< OS-specific implementation of the thread
+		Private::ThreadFunc *		mEntryPoint; ///< Abstraction of the function to run
+};
+
+//! NOTE: Taken from SFML2 threads
+namespace Private {
+
+// Base class for abstract thread functions
+struct ThreadFunc
+{
+	virtual ~ThreadFunc() {}
+	virtual void run() = 0;
+};
+
+// Specialization using a functor (including free functions) with no argument
+template <typename T>
+struct ThreadFunctor : ThreadFunc
+{
+	ThreadFunctor(T functor) : m_functor(functor) {}
+	virtual void run() {m_functor();}
+	T m_functor;
+};
+
+// Specialization using a functor (including free functions) with one argument
+template <typename F, typename A>
+struct ThreadFunctorWithArg : ThreadFunc
+{
+	ThreadFunctorWithArg(F function, A arg) : m_function(function), m_arg(arg) {}
+	virtual void run() {m_function(m_arg);}
+	F m_function;
+	A m_arg;
+};
+
+// Specialization using a member function
+template <typename C>
+struct ThreadMemberFunc : ThreadFunc
+{
+	ThreadMemberFunc(void(C::*function)(), C* object) : m_function(function), m_object(object) {}
+	virtual void run() {(m_object->*m_function)();}
+	void(C::*m_function)();
+	C* m_object;
+};
+
+}
+
+template <typename F>
+Thread::Thread(F functor) :
+	mThreadImpl      (NULL),
+	mEntryPoint( new Private::ThreadFunctor<F>(functor) )
+{
+}
+
+template <typename F, typename A>
+Thread::Thread(F function, A argument) :
+	mThreadImpl(NULL),
+	mEntryPoint( new Private::ThreadFunctorWithArg<F efCOMMA A>(function, argument) )
+{
+}
+
+template <typename C>
+Thread::Thread(void(C::*function)(), C* object) :
+	mThreadImpl(NULL),
+	mEntryPoint( new Private::ThreadMemberFunc<C>(function, object) )
+{
+}
+
+}
+
+#endif
diff --git a/dep/efsw/src/efsw/Utf.hpp b/dep/efsw/src/efsw/Utf.hpp
new file mode 100644
index 00000000000..925a7cc149b
--- /dev/null
+++ b/dep/efsw/src/efsw/Utf.hpp
@@ -0,0 +1,748 @@
+/** NOTE:
+*	This code is based on the Utf implementation from SFML2. License zlib/png ( http://www.sfml-dev.org/license.php )
+*	The class was modified to fit efsw own needs. This is not the original implementation from SFML2.
+* */
+
+#ifndef EFSW_UTF_HPP
+#define EFSW_UTF_HPP
+
+////////////////////////////////////////////////////////////
+// Headers
+////////////////////////////////////////////////////////////
+#include <locale>
+#include <string>
+#include <cstdlib>
+#include <efsw/base.hpp>
+
+namespace efsw {
+
+template <unsigned int N>
+class Utf;
+
+////////////////////////////////////////////////////////////
+/// \brief Specialization of the Utf template for UTF-8
+///
+////////////////////////////////////////////////////////////
+template <>
+class Utf<8> {
+public :
+    ////////////////////////////////////////////////////////////
+    /// \brief Decode a single UTF-8 character
+    ///
+    /// Decoding a character means finding its unique 32-bits
+    /// code (called the codepoint) in the Unicode standard.
+    ///
+    /// \param begin       Iterator pointing to the beginning of the input sequence
+    /// \param end         Iterator pointing to the end of the input sequence
+    /// \param output      Codepoint of the decoded UTF-8 character
+    /// \param replacement Replacement character to use in case the UTF-8 sequence is invalid
+    ///
+    /// \return Iterator pointing to one past the last read element of the input sequence
+    ///
+    ////////////////////////////////////////////////////////////
+    template <typename In>
+    static In Decode(In begin, In end, Uint32& output, Uint32 replacement = 0);
+
+    ////////////////////////////////////////////////////////////
+    /// \brief Encode a single UTF-8 character
+    ///
+    /// Encoding a character means converting a unique 32-bits
+    /// code (called the codepoint) in the target encoding, UTF-8.
+    ///
+    /// \param input       Codepoint to encode as UTF-8
+    /// \param output      Iterator pointing to the beginning of the output sequence
+    /// \param replacement Replacement for characters not convertible to UTF-8 (use 0 to skip them)
+    ///
+    /// \return Iterator to the end of the output sequence which has been written
+    ///
+    ////////////////////////////////////////////////////////////
+    template <typename Out>
+    static Out Encode(Uint32 input, Out output, Uint8 replacement = 0);
+
+    ////////////////////////////////////////////////////////////
+    /// \brief Advance to the next UTF-8 character
+    ///
+    /// This function is necessary for multi-elements encodings, as
+    /// a single character may use more than 1 storage element.
+    ///
+    /// \param begin Iterator pointing to the beginning of the input sequence
+    /// \param end   Iterator pointing to the end of the input sequence
+    ///
+    /// \return Iterator pointing to one past the last read element of the input sequence
+    ///
+    ////////////////////////////////////////////////////////////
+    template <typename In>
+    static In Next(In begin, In end);
+
+    ////////////////////////////////////////////////////////////
+    /// \brief Count the number of characters of a UTF-8 sequence
+    ///
+    /// This function is necessary for multi-elements encodings, as
+    /// a single character may use more than 1 storage element, thus the
+    /// total size can be different from (begin - end).
+    ///
+    /// \param begin Iterator pointing to the beginning of the input sequence
+    /// \param end   Iterator pointing to the end of the input sequence
+    ///
+    /// \return Iterator pointing to one past the last read element of the input sequence
+    ///
+    ////////////////////////////////////////////////////////////
+    template <typename In>
+    static std::size_t Count(In begin, In end);
+
+    ////////////////////////////////////////////////////////////
+    /// \brief Convert an ANSI characters range to UTF-8
+    ///
+    /// The current global locale will be used by default, unless you
+    /// pass a custom one in the \a locale parameter.
+    ///
+    /// \param begin  Iterator pointing to the beginning of the input sequence
+    /// \param end    Iterator pointing to the end of the input sequence
+    /// \param output Iterator pointing to the beginning of the output sequence
+    /// \param locale Locale to use for conversion
+    ///
+    /// \return Iterator to the end of the output sequence which has been written
+    ///
+    ////////////////////////////////////////////////////////////
+    template <typename In, typename Out>
+    static Out FromAnsi(In begin, In end, Out output, const std::locale& locale = std::locale());
+
+    ////////////////////////////////////////////////////////////
+    /// \brief Convert a wide characters range to UTF-8
+    ///
+    /// \param begin  Iterator pointing to the beginning of the input sequence
+    /// \param end    Iterator pointing to the end of the input sequence
+    /// \param output Iterator pointing to the beginning of the output sequence
+    ///
+    /// \return Iterator to the end of the output sequence which has been written
+    ///
+    ////////////////////////////////////////////////////////////
+    template <typename In, typename Out>
+    static Out FromWide(In begin, In end, Out output);
+
+    ////////////////////////////////////////////////////////////
+    /// \brief Convert a latin-1 (ISO-5589-1) characters range to UTF-8
+    ///
+    /// \param begin  Iterator pointing to the beginning of the input sequence
+    /// \param end    Iterator pointing to the end of the input sequence
+    /// \param output Iterator pointing to the beginning of the output sequence
+    /// \param locale Locale to use for conversion
+    ///
+    /// \return Iterator to the end of the output sequence which has been written
+    ///
+    ////////////////////////////////////////////////////////////
+    template <typename In, typename Out>
+    static Out FromLatin1(In begin, In end, Out output);
+
+    ////////////////////////////////////////////////////////////
+    /// \brief Convert an UTF-8 characters range to ANSI characters
+    ///
+    /// The current global locale will be used by default, unless you
+    /// pass a custom one in the \a locale parameter.
+    ///
+    /// \param begin       Iterator pointing to the beginning of the input sequence
+    /// \param end         Iterator pointing to the end of the input sequence
+    /// \param output      Iterator pointing to the beginning of the output sequence
+    /// \param replacement Replacement for characters not convertible to ANSI (use 0 to skip them)
+    /// \param locale      Locale to use for conversion
+    ///
+    /// \return Iterator to the end of the output sequence which has been written
+    ///
+    ////////////////////////////////////////////////////////////
+    template <typename In, typename Out>
+    static Out ToAnsi(In begin, In end, Out output, char replacement = 0, const std::locale& locale = std::locale());
+
+#ifndef EFSW_NO_WIDECHAR
+    ////////////////////////////////////////////////////////////
+    /// \brief Convert an UTF-8 characters range to wide characters
+    ///
+    /// \param begin       Iterator pointing to the beginning of the input sequence
+    /// \param end         Iterator pointing to the end of the input sequence
+    /// \param output      Iterator pointing to the beginning of the output sequence
+    /// \param replacement Replacement for characters not convertible to wide (use 0 to skip them)
+    ///
+    /// \return Iterator to the end of the output sequence which has been written
+    ///
+    ////////////////////////////////////////////////////////////
+    template <typename In, typename Out>
+    static Out ToWide(In begin, In end, Out output, wchar_t replacement = 0);
+#endif
+
+    ////////////////////////////////////////////////////////////
+    /// \brief Convert an UTF-8 characters range to latin-1 (ISO-5589-1) characters
+    ///
+    /// \param begin       Iterator pointing to the beginning of the input sequence
+    /// \param end         Iterator pointing to the end of the input sequence
+    /// \param output      Iterator pointing to the beginning of the output sequence
+    /// \param replacement Replacement for characters not convertible to wide (use 0 to skip them)
+    ///
+    /// \return Iterator to the end of the output sequence which has been written
+    ///
+    ////////////////////////////////////////////////////////////
+    template <typename In, typename Out>
+    static Out ToLatin1(In begin, In end, Out output, char replacement = 0);
+
+    ////////////////////////////////////////////////////////////
+    /// \brief Convert a UTF-8 characters range to UTF-8
+    ///
+    /// This functions does nothing more than a direct copy;
+    /// it is defined only to provide the same interface as other
+	/// specializations of the efsw::Utf<> template, and allow
+    /// generic code to be written on top of it.
+    ///
+    /// \param begin  Iterator pointing to the beginning of the input sequence
+    /// \param end    Iterator pointing to the end of the input sequence
+    /// \param output Iterator pointing to the beginning of the output sequence
+    ///
+    /// \return Iterator to the end of the output sequence which has been written
+    ///
+    ////////////////////////////////////////////////////////////
+    template <typename In, typename Out>
+	static Out toUtf8(In begin, In end, Out output);
+
+    ////////////////////////////////////////////////////////////
+    /// \brief Convert a UTF-8 characters range to UTF-16
+    ///
+    /// \param begin  Iterator pointing to the beginning of the input sequence
+    /// \param end    Iterator pointing to the end of the input sequence
+    /// \param output Iterator pointing to the beginning of the output sequence
+    ///
+    /// \return Iterator to the end of the output sequence which has been written
+    ///
+    ////////////////////////////////////////////////////////////
+    template <typename In, typename Out>
+    static Out ToUtf16(In begin, In end, Out output);
+
+    ////////////////////////////////////////////////////////////
+    /// \brief Convert a UTF-8 characters range to UTF-32
+    ///
+    /// \param begin  Iterator pointing to the beginning of the input sequence
+    /// \param end    Iterator pointing to the end of the input sequence
+    /// \param output Iterator pointing to the beginning of the output sequence
+    ///
+    /// \return Iterator to the end of the output sequence which has been written
+    ///
+    ////////////////////////////////////////////////////////////
+    template <typename In, typename Out>
+    static Out ToUtf32(In begin, In end, Out output);
+};
+
+////////////////////////////////////////////////////////////
+/// \brief Specialization of the Utf template for UTF-16
+///
+////////////////////////////////////////////////////////////
+template <>
+class Utf<16>
+{
+public :
+
+    ////////////////////////////////////////////////////////////
+    /// \brief Decode a single UTF-16 character
+    ///
+    /// Decoding a character means finding its unique 32-bits
+    /// code (called the codepoint) in the Unicode standard.
+    ///
+    /// \param begin       Iterator pointing to the beginning of the input sequence
+    /// \param end         Iterator pointing to the end of the input sequence
+    /// \param output      Codepoint of the decoded UTF-16 character
+    /// \param replacement Replacement character to use in case the UTF-8 sequence is invalid
+    ///
+    /// \return Iterator pointing to one past the last read element of the input sequence
+    ///
+    ////////////////////////////////////////////////////////////
+    template <typename In>
+    static In Decode(In begin, In end, Uint32& output, Uint32 replacement = 0);
+
+    ////////////////////////////////////////////////////////////
+    /// \brief Encode a single UTF-16 character
+    ///
+    /// Encoding a character means converting a unique 32-bits
+    /// code (called the codepoint) in the target encoding, UTF-16.
+    ///
+    /// \param input       Codepoint to encode as UTF-16
+    /// \param output      Iterator pointing to the beginning of the output sequence
+    /// \param replacement Replacement for characters not convertible to UTF-16 (use 0 to skip them)
+    ///
+    /// \return Iterator to the end of the output sequence which has been written
+    ///
+    ////////////////////////////////////////////////////////////
+    template <typename Out>
+    static Out Encode(Uint32 input, Out output, Uint16 replacement = 0);
+
+    ////////////////////////////////////////////////////////////
+    /// \brief Advance to the next UTF-16 character
+    ///
+    /// This function is necessary for multi-elements encodings, as
+    /// a single character may use more than 1 storage element.
+    ///
+    /// \param begin Iterator pointing to the beginning of the input sequence
+    /// \param end   Iterator pointing to the end of the input sequence
+    ///
+    /// \return Iterator pointing to one past the last read element of the input sequence
+    ///
+    ////////////////////////////////////////////////////////////
+    template <typename In>
+    static In Next(In begin, In end);
+
+    ////////////////////////////////////////////////////////////
+    /// \brief Count the number of characters of a UTF-16 sequence
+    ///
+    /// This function is necessary for multi-elements encodings, as
+    /// a single character may use more than 1 storage element, thus the
+    /// total size can be different from (begin - end).
+    ///
+    /// \param begin Iterator pointing to the beginning of the input sequence
+    /// \param end   Iterator pointing to the end of the input sequence
+    ///
+    /// \return Iterator pointing to one past the last read element of the input sequence
+    ///
+    ////////////////////////////////////////////////////////////
+    template <typename In>
+    static std::size_t Count(In begin, In end);
+
+    ////////////////////////////////////////////////////////////
+    /// \brief Convert an ANSI characters range to UTF-16
+    ///
+    /// The current global locale will be used by default, unless you
+    /// pass a custom one in the \a locale parameter.
+    ///
+    /// \param begin  Iterator pointing to the beginning of the input sequence
+    /// \param end    Iterator pointing to the end of the input sequence
+    /// \param output Iterator pointing to the beginning of the output sequence
+    /// \param locale Locale to use for conversion
+    ///
+    /// \return Iterator to the end of the output sequence which has been written
+    ///
+    ////////////////////////////////////////////////////////////
+    template <typename In, typename Out>
+    static Out FromAnsi(In begin, In end, Out output, const std::locale& locale = std::locale());
+
+    ////////////////////////////////////////////////////////////
+    /// \brief Convert a wide characters range to UTF-16
+    ///
+    /// \param begin  Iterator pointing to the beginning of the input sequence
+    /// \param end    Iterator pointing to the end of the input sequence
+    /// \param output Iterator pointing to the beginning of the output sequence
+    ///
+    /// \return Iterator to the end of the output sequence which has been written
+    ///
+    ////////////////////////////////////////////////////////////
+    template <typename In, typename Out>
+    static Out FromWide(In begin, In end, Out output);
+
+    ////////////////////////////////////////////////////////////
+    /// \brief Convert a latin-1 (ISO-5589-1) characters range to UTF-16
+    ///
+    /// \param begin  Iterator pointing to the beginning of the input sequence
+    /// \param end    Iterator pointing to the end of the input sequence
+    /// \param output Iterator pointing to the beginning of the output sequence
+    /// \param locale Locale to use for conversion
+    ///
+    /// \return Iterator to the end of the output sequence which has been written
+    ///
+    ////////////////////////////////////////////////////////////
+    template <typename In, typename Out>
+    static Out FromLatin1(In begin, In end, Out output);
+
+    ////////////////////////////////////////////////////////////
+    /// \brief Convert an UTF-16 characters range to ANSI characters
+    ///
+    /// The current global locale will be used by default, unless you
+    /// pass a custom one in the \a locale parameter.
+    ///
+    /// \param begin       Iterator pointing to the beginning of the input sequence
+    /// \param end         Iterator pointing to the end of the input sequence
+    /// \param output      Iterator pointing to the beginning of the output sequence
+    /// \param replacement Replacement for characters not convertible to ANSI (use 0 to skip them)
+    /// \param locale      Locale to use for conversion
+    ///
+    /// \return Iterator to the end of the output sequence which has been written
+    ///
+    ////////////////////////////////////////////////////////////
+    template <typename In, typename Out>
+    static Out ToAnsi(In begin, In end, Out output, char replacement = 0, const std::locale& locale = std::locale());
+
+#ifndef EFSW_NO_WIDECHAR
+    ////////////////////////////////////////////////////////////
+    /// \brief Convert an UTF-16 characters range to wide characters
+    ///
+    /// \param begin       Iterator pointing to the beginning of the input sequence
+    /// \param end         Iterator pointing to the end of the input sequence
+    /// \param output      Iterator pointing to the beginning of the output sequence
+    /// \param replacement Replacement for characters not convertible to wide (use 0 to skip them)
+    ///
+    /// \return Iterator to the end of the output sequence which has been written
+    ///
+    ////////////////////////////////////////////////////////////
+    template <typename In, typename Out>
+    static Out ToWide(In begin, In end, Out output, wchar_t replacement = 0);
+#endif
+
+    ////////////////////////////////////////////////////////////
+    /// \brief Convert an UTF-16 characters range to latin-1 (ISO-5589-1) characters
+    ///
+    /// \param begin       Iterator pointing to the beginning of the input sequence
+    /// \param end         Iterator pointing to the end of the input sequence
+    /// \param output      Iterator pointing to the beginning of the output sequence
+    /// \param replacement Replacement for characters not convertible to wide (use 0 to skip them)
+    ///
+    /// \return Iterator to the end of the output sequence which has been written
+    ///
+    ////////////////////////////////////////////////////////////
+    template <typename In, typename Out>
+    static Out ToLatin1(In begin, In end, Out output, char replacement = 0);
+
+    ////////////////////////////////////////////////////////////
+    /// \brief Convert a UTF-16 characters range to UTF-8
+    ///
+    /// \param begin  Iterator pointing to the beginning of the input sequence
+    /// \param end    Iterator pointing to the end of the input sequence
+    /// \param output Iterator pointing to the beginning of the output sequence
+    ///
+    /// \return Iterator to the end of the output sequence which has been written
+    ///
+    ////////////////////////////////////////////////////////////
+    template <typename In, typename Out>
+	static Out toUtf8(In begin, In end, Out output);
+
+    ////////////////////////////////////////////////////////////
+    /// \brief Convert a UTF-16 characters range to UTF-16
+    ///
+    /// This functions does nothing more than a direct copy;
+    /// it is defined only to provide the same interface as other
+	/// specializations of the efsw::Utf<> template, and allow
+    /// generic code to be written on top of it.
+    ///
+    /// \param begin  Iterator pointing to the beginning of the input sequence
+    /// \param end    Iterator pointing to the end of the input sequence
+    /// \param output Iterator pointing to the beginning of the output sequence
+    ///
+    /// \return Iterator to the end of the output sequence which has been written
+    ///
+    ////////////////////////////////////////////////////////////
+    template <typename In, typename Out>
+    static Out ToUtf16(In begin, In end, Out output);
+
+    ////////////////////////////////////////////////////////////
+    /// \brief Convert a UTF-16 characters range to UTF-32
+    ///
+    /// \param begin  Iterator pointing to the beginning of the input sequence
+    /// \param end    Iterator pointing to the end of the input sequence
+    /// \param output Iterator pointing to the beginning of the output sequence
+    ///
+    /// \return Iterator to the end of the output sequence which has been written
+    ///
+    ////////////////////////////////////////////////////////////
+    template <typename In, typename Out>
+    static Out ToUtf32(In begin, In end, Out output);
+};
+
+////////////////////////////////////////////////////////////
+/// \brief Specialization of the Utf template for UTF-32
+///
+////////////////////////////////////////////////////////////
+template <>
+class Utf<32>
+{
+public :
+
+    ////////////////////////////////////////////////////////////
+    /// \brief Decode a single UTF-32 character
+    ///
+    /// Decoding a character means finding its unique 32-bits
+    /// code (called the codepoint) in the Unicode standard.
+    /// For UTF-32, the character value is the same as the codepoint.
+    ///
+    /// \param begin       Iterator pointing to the beginning of the input sequence
+    /// \param end         Iterator pointing to the end of the input sequence
+    /// \param output      Codepoint of the decoded UTF-32 character
+    /// \param replacement Replacement character to use in case the UTF-8 sequence is invalid
+    ///
+    /// \return Iterator pointing to one past the last read element of the input sequence
+    ///
+    ////////////////////////////////////////////////////////////
+    template <typename In>
+    static In Decode(In begin, In end, Uint32& output, Uint32 replacement = 0);
+
+    ////////////////////////////////////////////////////////////
+    /// \brief Encode a single UTF-32 character
+    ///
+    /// Encoding a character means converting a unique 32-bits
+    /// code (called the codepoint) in the target encoding, UTF-32.
+    /// For UTF-32, the codepoint is the same as the character value.
+    ///
+    /// \param input       Codepoint to encode as UTF-32
+    /// \param output      Iterator pointing to the beginning of the output sequence
+    /// \param replacement Replacement for characters not convertible to UTF-32 (use 0 to skip them)
+    ///
+    /// \return Iterator to the end of the output sequence which has been written
+    ///
+    ////////////////////////////////////////////////////////////
+    template <typename Out>
+    static Out Encode(Uint32 input, Out output, Uint32 replacement = 0);
+
+    ////////////////////////////////////////////////////////////
+    /// \brief Advance to the next UTF-32 character
+    ///
+    /// This function is trivial for UTF-32, which can store
+    /// every character in a single storage element.
+    ///
+    /// \param begin Iterator pointing to the beginning of the input sequence
+    /// \param end   Iterator pointing to the end of the input sequence
+    ///
+    /// \return Iterator pointing to one past the last read element of the input sequence
+    ///
+    ////////////////////////////////////////////////////////////
+    template <typename In>
+    static In Next(In begin, In end);
+
+    ////////////////////////////////////////////////////////////
+    /// \brief Count the number of characters of a UTF-32 sequence
+    ///
+    /// This function is trivial for UTF-32, which can store
+    /// every character in a single storage element.
+    ///
+    /// \param begin Iterator pointing to the beginning of the input sequence
+    /// \param end   Iterator pointing to the end of the input sequence
+    ///
+    /// \return Iterator pointing to one past the last read element of the input sequence
+    ///
+    ////////////////////////////////////////////////////////////
+    template <typename In>
+    static std::size_t Count(In begin, In end);
+
+    ////////////////////////////////////////////////////////////
+    /// \brief Convert an ANSI characters range to UTF-32
+    ///
+    /// The current global locale will be used by default, unless you
+    /// pass a custom one in the \a locale parameter.
+    ///
+    /// \param begin  Iterator pointing to the beginning of the input sequence
+    /// \param end    Iterator pointing to the end of the input sequence
+    /// \param output Iterator pointing to the beginning of the output sequence
+    /// \param locale Locale to use for conversion
+    ///
+    /// \return Iterator to the end of the output sequence which has been written
+    ///
+    ////////////////////////////////////////////////////////////
+    template <typename In, typename Out>
+    static Out FromAnsi(In begin, In end, Out output, const std::locale& locale = std::locale());
+
+    ////////////////////////////////////////////////////////////
+    /// \brief Convert a wide characters range to UTF-32
+    ///
+    /// \param begin  Iterator pointing to the beginning of the input sequence
+    /// \param end    Iterator pointing to the end of the input sequence
+    /// \param output Iterator pointing to the beginning of the output sequence
+    ///
+    /// \return Iterator to the end of the output sequence which has been written
+    ///
+    ////////////////////////////////////////////////////////////
+    template <typename In, typename Out>
+    static Out FromWide(In begin, In end, Out output);
+
+    ////////////////////////////////////////////////////////////
+    /// \brief Convert a latin-1 (ISO-5589-1) characters range to UTF-32
+    ///
+    /// \param begin  Iterator pointing to the beginning of the input sequence
+    /// \param end    Iterator pointing to the end of the input sequence
+    /// \param output Iterator pointing to the beginning of the output sequence
+    /// \param locale Locale to use for conversion
+    ///
+    /// \return Iterator to the end of the output sequence which has been written
+    ///
+    ////////////////////////////////////////////////////////////
+    template <typename In, typename Out>
+    static Out FromLatin1(In begin, In end, Out output);
+
+    ////////////////////////////////////////////////////////////
+    /// \brief Convert an UTF-32 characters range to ANSI characters
+    ///
+    /// The current global locale will be used by default, unless you
+    /// pass a custom one in the \a locale parameter.
+    ///
+    /// \param begin       Iterator pointing to the beginning of the input sequence
+    /// \param end         Iterator pointing to the end of the input sequence
+    /// \param output      Iterator pointing to the beginning of the output sequence
+    /// \param replacement Replacement for characters not convertible to ANSI (use 0 to skip them)
+    /// \param locale      Locale to use for conversion
+    ///
+    /// \return Iterator to the end of the output sequence which has been written
+    ///
+    ////////////////////////////////////////////////////////////
+    template <typename In, typename Out>
+    static Out ToAnsi(In begin, In end, Out output, char replacement = 0, const std::locale& locale = std::locale());
+
+#ifndef EFSW_NO_WIDECHAR
+    ////////////////////////////////////////////////////////////
+    /// \brief Convert an UTF-32 characters range to wide characters
+    ///
+    /// \param begin       Iterator pointing to the beginning of the input sequence
+    /// \param end         Iterator pointing to the end of the input sequence
+    /// \param output      Iterator pointing to the beginning of the output sequence
+    /// \param replacement Replacement for characters not convertible to wide (use 0 to skip them)
+    ///
+    /// \return Iterator to the end of the output sequence which has been written
+    ///
+    ////////////////////////////////////////////////////////////
+    template <typename In, typename Out>
+    static Out ToWide(In begin, In end, Out output, wchar_t replacement = 0);
+#endif
+
+    ////////////////////////////////////////////////////////////
+    /// \brief Convert an UTF-16 characters range to latin-1 (ISO-5589-1) characters
+    ///
+    /// \param begin       Iterator pointing to the beginning of the input sequence
+    /// \param end         Iterator pointing to the end of the input sequence
+    /// \param output      Iterator pointing to the beginning of the output sequence
+    /// \param replacement Replacement for characters not convertible to wide (use 0 to skip them)
+    ///
+    /// \return Iterator to the end of the output sequence which has been written
+    ///
+    ////////////////////////////////////////////////////////////
+    template <typename In, typename Out>
+    static Out ToLatin1(In begin, In end, Out output, char replacement = 0);
+
+    ////////////////////////////////////////////////////////////
+    /// \brief Convert a UTF-32 characters range to UTF-8
+    ///
+    /// \param begin  Iterator pointing to the beginning of the input sequence
+    /// \param end    Iterator pointing to the end of the input sequence
+    /// \param output Iterator pointing to the beginning of the output sequence
+    ///
+    /// \return Iterator to the end of the output sequence which has been written
+    ///
+    ////////////////////////////////////////////////////////////
+    template <typename In, typename Out>
+	static Out toUtf8(In begin, In end, Out output);
+
+    ////////////////////////////////////////////////////////////
+    /// \brief Convert a UTF-32 characters range to UTF-16
+    ///
+    /// \param begin  Iterator pointing to the beginning of the input sequence
+    /// \param end    Iterator pointing to the end of the input sequence
+    /// \param output Iterator pointing to the beginning of the output sequence
+    ///
+    /// \return Iterator to the end of the output sequence which has been written
+    ///
+    ////////////////////////////////////////////////////////////
+    template <typename In, typename Out>
+    static Out ToUtf16(In begin, In end, Out output);
+
+    ////////////////////////////////////////////////////////////
+    /// \brief Convert a UTF-32 characters range to UTF-32
+    ///
+    /// This functions does nothing more than a direct copy;
+    /// it is defined only to provide the same interface as other
+	/// specializations of the efsw::Utf<> template, and allow
+    /// generic code to be written on top of it.
+    ///
+    /// \param begin  Iterator pointing to the beginning of the input sequence
+    /// \param end    Iterator pointing to the end of the input sequence
+    /// \param output Iterator pointing to the beginning of the output sequence
+    ///
+    /// \return Iterator to the end of the output sequence which has been written
+    ///
+    ////////////////////////////////////////////////////////////
+    template <typename In, typename Out>
+    static Out ToUtf32(In begin, In end, Out output);
+
+    ////////////////////////////////////////////////////////////
+    /// \brief Decode a single ANSI character to UTF-32
+    ///
+    /// This function does not exist in other specializations
+	/// of efsw::Utf<>, it is defined for convenience (it is used by
+    /// several other conversion functions).
+    ///
+    /// \param input  Input ANSI character
+    /// \param locale Locale to use for conversion
+    ///
+    /// \return Converted character
+    ///
+    ////////////////////////////////////////////////////////////
+    template <typename In>
+    static Uint32 DecodeAnsi(In input, const std::locale& locale = std::locale());
+
+    ////////////////////////////////////////////////////////////
+    /// \brief Decode a single wide character to UTF-32
+    ///
+    /// This function does not exist in other specializations
+	/// of efsw::Utf<>, it is defined for convenience (it is used by
+    /// several other conversion functions).
+    ///
+    /// \param input Input wide character
+    ///
+    /// \return Converted character
+    ///
+    ////////////////////////////////////////////////////////////
+    template <typename In>
+    static Uint32 DecodeWide(In input);
+
+    ////////////////////////////////////////////////////////////
+    /// \brief Encode a single UTF-32 character to ANSI
+    ///
+    /// This function does not exist in other specializations
+	/// of efsw::Utf<>, it is defined for convenience (it is used by
+    /// several other conversion functions).
+    ///
+    /// \param codepoint   Iterator pointing to the beginning of the input sequence
+    /// \param output      Iterator pointing to the beginning of the output sequence
+    /// \param replacement Replacement if the input character is not convertible to ANSI (use 0 to skip it)
+    /// \param locale      Locale to use for conversion
+    ///
+    /// \return Iterator to the end of the output sequence which has been written
+    ///
+    ////////////////////////////////////////////////////////////
+    template <typename Out>
+    static Out EncodeAnsi(Uint32 codepoint, Out output, char replacement = 0, const std::locale& locale = std::locale());
+
+#ifndef EFSW_NO_WIDECHAR
+    ////////////////////////////////////////////////////////////
+    /// \brief Encode a single UTF-32 character to wide
+    ///
+    /// This function does not exist in other specializations
+	/// of efsw::Utf<>, it is defined for convenience (it is used by
+    /// several other conversion functions).
+    ///
+    /// \param codepoint   Iterator pointing to the beginning of the input sequence
+    /// \param output      Iterator pointing to the beginning of the output sequence
+    /// \param replacement Replacement if the input character is not convertible to wide (use 0 to skip it)
+    ///
+    /// \return Iterator to the end of the output sequence which has been written
+    ///
+    ////////////////////////////////////////////////////////////
+    template <typename Out>
+    static Out EncodeWide(Uint32 codepoint, Out output, wchar_t replacement = 0);
+#endif
+};
+
+#include "Utf.inl"
+
+// Make typedefs to get rid of the template syntax
+typedef Utf<8>  Utf8;
+typedef Utf<16> Utf16;
+typedef Utf<32> Utf32;
+
+}
+#endif
+
+////////////////////////////////////////////////////////////
+/// \class efsw::Utf
+/// \ingroup system
+///
+/// Utility class providing generic functions for UTF conversions.
+///
+/// efsw::Utf is a low-level, generic interface for counting, iterating,
+/// encoding and decoding Unicode characters and strings. It is able
+/// to handle ANSI, wide, UTF-8, UTF-16 and UTF-32 encodings.
+///
+/// efsw::Utf<X> functions are all static, these classes are not meant to
+/// be instanciated. All the functions are template, so that you
+/// can use any character / string type for a given encoding.
+///
+/// It has 3 specializations:
+/// \li efsw::Utf<8> (typedef'd to efsw::Utf8)
+/// \li efsw::Utf<16> (typedef'd to efsw::Utf16)
+/// \li efsw::Utf<32> (typedef'd to efsw::Utf32)
+///
+////////////////////////////////////////////////////////////
diff --git a/dep/efsw/src/efsw/Utf.inl b/dep/efsw/src/efsw/Utf.inl
new file mode 100644
index 00000000000..db8fd5d61ee
--- /dev/null
+++ b/dep/efsw/src/efsw/Utf.inl
@@ -0,0 +1,671 @@
+// References :
+// http://www.unicode.org/
+// http://www.unicode.org/Public/PROGRAMS/CVTUTF/ConvertUTF.c
+// http://www.unicode.org/Public/PROGRAMS/CVTUTF/ConvertUTF.h
+// http://people.w3.org/rishida/scripts/uniview/conversion
+////////////////////////////////////////////////////////////
+
+template <typename In>
+In Utf<8>::Decode(In begin, In end, Uint32& output, Uint32 replacement)
+{
+    // Some useful precomputed data
+    static const int trailing[256] =
+    {
+        0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
+        0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
+        0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
+        0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
+        0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
+        0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
+        1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
+        2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 3, 3, 3, 3, 3, 3, 3, 3, 4, 4, 4, 4, 5, 5, 5, 5
+    };
+    static const Uint32 offsets[6] =
+    {
+        0x00000000, 0x00003080, 0x000E2080, 0x03C82080, 0xFA082080, 0x82082080
+    };
+
+    // Decode the character
+    int trailingBytes = trailing[static_cast<Uint8>(*begin)];
+    if (begin + trailingBytes < end)
+    {
+        output = 0;
+        switch (trailingBytes)
+        {
+            case 5 : output += static_cast<Uint8>(*begin++); output <<= 6;
+            case 4 : output += static_cast<Uint8>(*begin++); output <<= 6;
+            case 3 : output += static_cast<Uint8>(*begin++); output <<= 6;
+            case 2 : output += static_cast<Uint8>(*begin++); output <<= 6;
+            case 1 : output += static_cast<Uint8>(*begin++); output <<= 6;
+            case 0 : output += static_cast<Uint8>(*begin++);
+        }
+        output -= offsets[trailingBytes];
+    }
+    else
+    {
+        // Incomplete character
+        begin = end;
+        output = replacement;
+    }
+
+    return begin;
+}
+
+template <typename Out>
+Out Utf<8>::Encode(Uint32 input, Out output, Uint8 replacement)
+{
+    // Some useful precomputed data
+    static const Uint8 firstBytes[7] =
+    {
+        0x00, 0x00, 0xC0, 0xE0, 0xF0, 0xF8, 0xFC
+    };
+
+    // Encode the character
+    if ((input > 0x0010FFFF) || ((input >= 0xD800) && (input <= 0xDBFF)))
+    {
+        // Invalid character
+        if (replacement)
+            *output++ = replacement;
+    }
+    else
+    {
+        // Valid character
+
+        // Get the number of bytes to write
+        int bytesToWrite = 1;
+        if      (input <  0x80)       bytesToWrite = 1;
+        else if (input <  0x800)      bytesToWrite = 2;
+        else if (input <  0x10000)    bytesToWrite = 3;
+        else if (input <= 0x0010FFFF) bytesToWrite = 4;
+
+        // Extract the bytes to write
+        Uint8 bytes[4];
+        switch (bytesToWrite)
+        {
+            case 4 : bytes[3] = static_cast<Uint8>((input | 0x80) & 0xBF); input >>= 6;
+            case 3 : bytes[2] = static_cast<Uint8>((input | 0x80) & 0xBF); input >>= 6;
+            case 2 : bytes[1] = static_cast<Uint8>((input | 0x80) & 0xBF); input >>= 6;
+            case 1 : bytes[0] = static_cast<Uint8> (input | firstBytes[bytesToWrite]);
+        }
+
+        // Add them to the output
+        const Uint8* currentByte = bytes;
+        switch (bytesToWrite)
+        {
+            case 4 : *output++ = *currentByte++;
+            case 3 : *output++ = *currentByte++;
+            case 2 : *output++ = *currentByte++;
+            case 1 : *output++ = *currentByte++;
+        }
+    }
+
+    return output;
+}
+
+template <typename In>
+In Utf<8>::Next(In begin, In end)
+{
+    Uint32 codepoint;
+    return Decode(begin, end, codepoint);
+}
+
+template <typename In>
+std::size_t Utf<8>::Count(In begin, In end)
+{
+    std::size_t length = 0;
+    while (begin < end)
+    {
+        begin = Next(begin, end);
+        ++length;
+    }
+
+    return length;
+}
+
+template <typename In, typename Out>
+Out Utf<8>::FromAnsi(In begin, In end, Out output, const std::locale& locale)
+{
+    while (begin < end)
+    {
+        Uint32 codepoint = Utf<32>::DecodeAnsi(*begin++, locale);
+        output = Encode(codepoint, output);
+    }
+
+    return output;
+}
+
+template <typename In, typename Out>
+Out Utf<8>::FromWide(In begin, In end, Out output)
+{
+    while (begin < end)
+    {
+        Uint32 codepoint = Utf<32>::DecodeWide(*begin++);
+        output = Encode(codepoint, output);
+    }
+
+    return output;
+}
+
+template <typename In, typename Out>
+Out Utf<8>::FromLatin1(In begin, In end, Out output)
+{
+    // Latin-1 is directly compatible with Unicode encodings,
+    // and can thus be treated as (a sub-range of) UTF-32
+    while (begin < end)
+        output = Encode(*begin++, output);
+
+    return output;
+}
+
+template <typename In, typename Out>
+Out Utf<8>::ToAnsi(In begin, In end, Out output, char replacement, const std::locale& locale)
+{
+    while (begin < end)
+    {
+        Uint32 codepoint;
+        begin = Decode(begin, end, codepoint);
+        output = Utf<32>::EncodeAnsi(codepoint, output, replacement, locale);
+    }
+
+    return output;
+}
+
+#ifndef EFSW_NO_WIDECHAR
+template <typename In, typename Out>
+Out Utf<8>::ToWide(In begin, In end, Out output, wchar_t replacement)
+{
+    while (begin < end)
+    {
+        Uint32 codepoint;
+        begin = Decode(begin, end, codepoint);
+        output = Utf<32>::EncodeWide(codepoint, output, replacement);
+    }
+
+    return output;
+}
+#endif
+
+template <typename In, typename Out>
+Out Utf<8>::ToLatin1(In begin, In end, Out output, char replacement)
+{
+    // Latin-1 is directly compatible with Unicode encodings,
+    // and can thus be treated as (a sub-range of) UTF-32
+    while (begin < end)
+    {
+        Uint32 codepoint;
+        begin = Decode(begin, end, codepoint);
+        *output++ = codepoint < 256 ? static_cast<char>(codepoint) : replacement;
+    }
+
+    return output;
+}
+
+template <typename In, typename Out>
+Out Utf<8>::toUtf8(In begin, In end, Out output)
+{
+    while (begin < end)
+        *output++ = *begin++;
+
+    return output;
+}
+
+template <typename In, typename Out>
+Out Utf<8>::ToUtf16(In begin, In end, Out output)
+{
+    while (begin < end)
+    {
+        Uint32 codepoint;
+        begin = Decode(begin, end, codepoint);
+        output = Utf<16>::Encode(codepoint, output);
+    }
+
+    return output;
+}
+
+template <typename In, typename Out>
+Out Utf<8>::ToUtf32(In begin, In end, Out output)
+{
+    while (begin < end)
+    {
+        Uint32 codepoint;
+        begin = Decode(begin, end, codepoint);
+        *output++ = codepoint;
+    }
+
+    return output;
+}
+
+template <typename In>
+In Utf<16>::Decode(In begin, In end, Uint32& output, Uint32 replacement)
+{
+    Uint16 first = *begin++;
+
+    // If it's a surrogate pair, first convert to a single UTF-32 character
+    if ((first >= 0xD800) && (first <= 0xDBFF))
+    {
+        if (begin < end)
+        {
+            Uint32 second = *begin++;
+            if ((second >= 0xDC00) && (second <= 0xDFFF))
+            {
+                // The second element is valid: convert the two elements to a UTF-32 character
+                output = static_cast<Uint32>(((first - 0xD800) << 10) + (second - 0xDC00) + 0x0010000);
+            }
+            else
+            {
+                // Invalid character
+                output = replacement;
+            }
+        }
+        else
+        {
+            // Invalid character
+            begin = end;
+            output = replacement;
+        }
+    }
+    else
+    {
+        // We can make a direct copy
+        output = first;
+    }
+
+    return begin;
+}
+
+template <typename Out>
+Out Utf<16>::Encode(Uint32 input, Out output, Uint16 replacement)
+{
+    if (input < 0xFFFF)
+    {
+        // The character can be copied directly, we just need to check if it's in the valid range
+        if ((input >= 0xD800) && (input <= 0xDFFF))
+        {
+            // Invalid character (this range is reserved)
+            if (replacement)
+                *output++ = replacement;
+        }
+        else
+        {
+            // Valid character directly convertible to a single UTF-16 character
+            *output++ = static_cast<Uint16>(input);
+        }
+    }
+    else if (input > 0x0010FFFF)
+    {
+        // Invalid character (greater than the maximum unicode value)
+        if (replacement)
+            *output++ = replacement;
+    }
+    else
+    {
+        // The input character will be converted to two UTF-16 elements
+        input -= 0x0010000;
+        *output++ = static_cast<Uint16>((input >> 10)     + 0xD800);
+        *output++ = static_cast<Uint16>((input & 0x3FFUL) + 0xDC00);
+    }
+
+    return output;
+}
+
+template <typename In>
+In Utf<16>::Next(In begin, In end)
+{
+    Uint32 codepoint;
+    return Decode(begin, end, codepoint);
+}
+
+template <typename In>
+std::size_t Utf<16>::Count(In begin, In end)
+{
+    std::size_t length = 0;
+    while (begin < end)
+    {
+        begin = Next(begin, end);
+        ++length;
+    }
+
+    return length;
+}
+
+template <typename In, typename Out>
+Out Utf<16>::FromAnsi(In begin, In end, Out output, const std::locale& locale)
+{
+    while (begin < end)
+    {
+        Uint32 codepoint = Utf<32>::DecodeAnsi(*begin++, locale);
+        output = Encode(codepoint, output);
+    }
+
+    return output;
+}
+
+template <typename In, typename Out>
+Out Utf<16>::FromWide(In begin, In end, Out output)
+{
+    while (begin < end)
+    {
+        Uint32 codepoint = Utf<32>::DecodeWide(*begin++);
+        output = Encode(codepoint, output);
+    }
+
+    return output;
+}
+
+template <typename In, typename Out>
+Out Utf<16>::FromLatin1(In begin, In end, Out output)
+{
+    // Latin-1 is directly compatible with Unicode encodings,
+    // and can thus be treated as (a sub-range of) UTF-32
+    while (begin < end)
+        *output++ = *begin++;
+
+    return output;
+}
+
+template <typename In, typename Out>
+Out Utf<16>::ToAnsi(In begin, In end, Out output, char replacement, const std::locale& locale)
+{
+    while (begin < end)
+    {
+        Uint32 codepoint;
+        begin = Decode(begin, end, codepoint);
+        output = Utf<32>::EncodeAnsi(codepoint, output, replacement, locale);
+    }
+
+    return output;
+}
+
+#ifndef EFSW_NO_WIDECHAR
+template <typename In, typename Out>
+Out Utf<16>::ToWide(In begin, In end, Out output, wchar_t replacement)
+{
+    while (begin < end)
+    {
+        Uint32 codepoint;
+        begin = Decode(begin, end, codepoint);
+        output = Utf<32>::EncodeWide(codepoint, output, replacement);
+    }
+
+    return output;
+}
+#endif
+
+template <typename In, typename Out>
+Out Utf<16>::ToLatin1(In begin, In end, Out output, char replacement)
+{
+    // Latin-1 is directly compatible with Unicode encodings,
+    // and can thus be treated as (a sub-range of) UTF-32
+    while (begin < end)
+    {
+        *output++ = *begin < 256 ? static_cast<char>(*begin) : replacement;
+        begin++;
+    }
+
+    return output;
+}
+
+template <typename In, typename Out>
+Out Utf<16>::toUtf8(In begin, In end, Out output)
+{
+    while (begin < end)
+    {
+        Uint32 codepoint;
+        begin = Decode(begin, end, codepoint);
+        output = Utf<8>::Encode(codepoint, output);
+    }
+
+    return output;
+}
+
+template <typename In, typename Out>
+Out Utf<16>::ToUtf16(In begin, In end, Out output)
+{
+    while (begin < end)
+        *output++ = *begin++;
+
+    return output;
+}
+
+template <typename In, typename Out>
+Out Utf<16>::ToUtf32(In begin, In end, Out output)
+{
+    while (begin < end)
+    {
+        Uint32 codepoint;
+        begin = Decode(begin, end, codepoint);
+        *output++ = codepoint;
+    }
+
+    return output;
+}
+
+template <typename In>
+In Utf<32>::Decode(In begin, In end, Uint32& output, Uint32)
+{
+    output = *begin++;
+    return begin;
+}
+
+template <typename Out>
+Out Utf<32>::Encode(Uint32 input, Out output, Uint32 replacement)
+{
+    *output++ = input;
+    return output;
+}
+
+template <typename In>
+In Utf<32>::Next(In begin, In end)
+{
+    return ++begin;
+}
+
+template <typename In>
+std::size_t Utf<32>::Count(In begin, In end)
+{
+    return begin - end;
+}
+
+template <typename In, typename Out>
+Out Utf<32>::FromAnsi(In begin, In end, Out output, const std::locale& locale)
+{
+    while (begin < end)
+        *output++ = DecodeAnsi(*begin++, locale);
+
+    return output;
+}
+
+template <typename In, typename Out>
+Out Utf<32>::FromWide(In begin, In end, Out output)
+{
+    while (begin < end)
+        *output++ = DecodeWide(*begin++);
+
+    return output;
+}
+
+template <typename In, typename Out>
+Out Utf<32>::FromLatin1(In begin, In end, Out output)
+{
+    // Latin-1 is directly compatible with Unicode encodings,
+    // and can thus be treated as (a sub-range of) UTF-32
+    while (begin < end)
+        *output++ = *begin++;
+
+    return output;
+}
+
+template <typename In, typename Out>
+Out Utf<32>::ToAnsi(In begin, In end, Out output, char replacement, const std::locale& locale)
+{
+    while (begin < end)
+        output = EncodeAnsi(*begin++, output, replacement, locale);
+
+    return output;
+}
+
+#ifndef EFSW_NO_WIDECHAR
+template <typename In, typename Out>
+Out Utf<32>::ToWide(In begin, In end, Out output, wchar_t replacement)
+{
+    while (begin < end)
+        output = EncodeWide(*begin++, output, replacement);
+
+    return output;
+}
+#endif
+
+template <typename In, typename Out>
+Out Utf<32>::ToLatin1(In begin, In end, Out output, char replacement)
+{
+    // Latin-1 is directly compatible with Unicode encodings,
+    // and can thus be treated as (a sub-range of) UTF-32
+    while (begin < end)
+    {
+        *output++ = *begin < 256 ? static_cast<char>(*begin) : replacement;
+        begin++;
+    }
+
+    return output;
+}
+
+template <typename In, typename Out>
+Out Utf<32>::toUtf8(In begin, In end, Out output)
+{
+    while (begin < end)
+        output = Utf<8>::Encode(*begin++, output);
+
+    return output;
+}
+
+template <typename In, typename Out>
+Out Utf<32>::ToUtf16(In begin, In end, Out output)
+{
+    while (begin < end)
+        output = Utf<16>::Encode(*begin++, output);
+
+    return output;
+}
+
+template <typename In, typename Out>
+Out Utf<32>::ToUtf32(In begin, In end, Out output)
+{
+    while (begin < end)
+        *output++ = *begin++;
+
+    return output;
+}
+
+template <typename In>
+Uint32 Utf<32>::DecodeAnsi(In input, const std::locale& locale)
+{
+    // On Windows, gcc's standard library (glibc++) has almost
+    // no support for Unicode stuff. As a consequence, in this
+    // context we can only use the default locale and ignore
+    // the one passed as parameter.
+
+    #if EFSW_PLATFORM == EFSW_PLATFORM_WIN &&                       /* if Windows ... */                          \
+       (defined(__GLIBCPP__) || defined (__GLIBCXX__)) &&     /* ... and standard library is glibc++ ... */ \
+      !(defined(__SGI_STL_PORT) || defined(_STLPORT_VERSION)) /* ... and STLPort is not used on top of it */
+
+        wchar_t character = 0;
+        mbtowc(&character, &input, 1);
+        return static_cast<Uint32>(character);
+
+    #else
+        // Get the facet of the locale which deals with character conversion
+		#ifndef EFSW_NO_WIDECHAR
+        const std::ctype<wchar_t>& facet = std::use_facet< std::ctype<wchar_t> >(locale);
+		#else
+		const std::ctype<char>& facet = std::use_facet< std::ctype<char> >(locale);
+		#endif
+		
+        // Use the facet to convert each character of the input string
+        return static_cast<Uint32>(facet.widen(input));
+
+    #endif
+}
+
+template <typename In>
+Uint32 Utf<32>::DecodeWide(In input)
+{
+    // The encoding of wide characters is not well defined and is left to the system;
+    // however we can safely assume that it is UCS-2 on Windows and
+    // UCS-4 on Unix systems.
+    // In both cases, a simple copy is enough (UCS-2 is a subset of UCS-4,
+    // and UCS-4 *is* UTF-32).
+
+    return input;
+}
+
+template <typename Out>
+Out Utf<32>::EncodeAnsi(Uint32 codepoint, Out output, char replacement, const std::locale& locale)
+{
+    // On Windows, gcc's standard library (glibc++) has almost
+    // no support for Unicode stuff. As a consequence, in this
+    // context we can only use the default locale and ignore
+    // the one passed as parameter.
+
+    #if EFSW_PLATFORM == EFSW_PLATFORM_WIN &&                       /* if Windows ... */                          \
+       (defined(__GLIBCPP__) || defined (__GLIBCXX__)) &&     /* ... and standard library is glibc++ ... */ \
+      !(defined(__SGI_STL_PORT) || defined(_STLPORT_VERSION)) /* ... and STLPort is not used on top of it */
+
+        char character = 0;
+        if (wctomb(&character, static_cast<wchar_t>(codepoint)) >= 0)
+            *output++ = character;
+        else if (replacement)
+            *output++ = replacement;
+
+        return output;
+
+    #else
+        // Get the facet of the locale which deals with character conversion
+		#ifndef EFSW_NO_WIDECHAR
+        const std::ctype<wchar_t>& facet = std::use_facet< std::ctype<wchar_t> >(locale);
+        #else
+        const std::ctype<char>& facet = std::use_facet< std::ctype<char> >(locale);
+        #endif
+
+        // Use the facet to convert each character of the input string
+        *output++ = facet.narrow(static_cast<wchar_t>(codepoint), replacement);
+
+        return output;
+
+    #endif
+}
+
+#ifndef EFSW_NO_WIDECHAR
+template <typename Out>
+Out Utf<32>::EncodeWide(Uint32 codepoint, Out output, wchar_t replacement)
+{
+    // The encoding of wide characters is not well defined and is left to the system;
+    // however we can safely assume that it is UCS-2 on Windows and
+    // UCS-4 on Unix systems.
+    // For UCS-2 we need to check if the source characters fits in (UCS-2 is a subset of UCS-4).
+    // For UCS-4 we can do a direct copy (UCS-4 *is* UTF-32).
+
+    switch (sizeof(wchar_t))
+    {
+        case 4:
+        {
+            *output++ = static_cast<wchar_t>(codepoint);
+            break;
+        }
+
+        default:
+        {
+            if ((codepoint <= 0xFFFF) && ((codepoint < 0xD800) || (codepoint > 0xDFFF)))
+            {
+                *output++ = static_cast<wchar_t>(codepoint);
+            }
+            else if (replacement)
+            {
+                *output++ = replacement;
+            }
+            break;
+        }
+    }
+
+    return output;
+}
+#endif
diff --git a/dep/efsw/src/efsw/Watcher.cpp b/dep/efsw/src/efsw/Watcher.cpp
new file mode 100644
index 00000000000..d81c84228ad
--- /dev/null
+++ b/dep/efsw/src/efsw/Watcher.cpp
@@ -0,0 +1,21 @@
+#include <efsw/Watcher.hpp>
+
+namespace efsw {
+
+Watcher::Watcher() :
+	ID(0),
+	Directory(""),
+	Listener(NULL),
+	Recursive(false)
+{
+}
+
+Watcher::Watcher( WatchID id, std::string directory, FileWatchListener * listener, bool recursive ) :
+	ID( id ),
+	Directory( directory ),
+	Listener( listener ),
+	Recursive( recursive )
+{
+}
+
+} 
diff --git a/dep/efsw/src/efsw/Watcher.hpp b/dep/efsw/src/efsw/Watcher.hpp
new file mode 100644
index 00000000000..5a35cb9a2ac
--- /dev/null
+++ b/dep/efsw/src/efsw/Watcher.hpp
@@ -0,0 +1,30 @@
+#ifndef EFSW_WATCHERIMPL_HPP
+#define EFSW_WATCHERIMPL_HPP
+
+#include <efsw/base.hpp>
+#include <efsw/efsw.hpp>
+
+namespace efsw {
+
+/** @brief Base Watcher class */
+class Watcher
+{
+	public:
+		Watcher();
+
+		Watcher( WatchID id, std::string directory, FileWatchListener * listener, bool recursive );
+
+		virtual ~Watcher() {}
+
+		virtual void watch() {}
+
+		WatchID					ID;
+		std::string				Directory;
+		FileWatchListener	*	Listener;
+		bool					Recursive;
+		std::string				OldFileName;
+};
+
+}
+
+#endif
diff --git a/dep/efsw/src/efsw/WatcherFSEvents.cpp b/dep/efsw/src/efsw/WatcherFSEvents.cpp
new file mode 100644
index 00000000000..3a9700c77ec
--- /dev/null
+++ b/dep/efsw/src/efsw/WatcherFSEvents.cpp
@@ -0,0 +1,264 @@
+#include <efsw/WatcherFSEvents.hpp>
+#include <efsw/FileWatcherFSEvents.hpp>
+#include <efsw/FileSystem.hpp>
+#include <efsw/Debug.hpp>
+
+#if EFSW_PLATFORM == EFSW_PLATFORM_FSEVENTS
+
+namespace efsw {
+
+WatcherFSEvents::WatcherFSEvents() :
+	Watcher(),
+	FWatcher( NULL ),
+	FSStream( NULL ),
+	WatcherGen( NULL ),
+	initializedAsync( false )
+{
+}
+
+WatcherFSEvents::WatcherFSEvents( WatchID id, std::string directory, FileWatchListener * listener, bool recursive, WatcherFSEvents * parent ) :
+	Watcher( id, directory, listener, recursive ),
+	FWatcher( NULL ),
+	FSStream( NULL ),
+	WatcherGen( NULL ),
+	initializedAsync( false )
+{
+}
+
+WatcherFSEvents::~WatcherFSEvents()
+{
+	if ( initializedAsync )
+	{
+		FSEventStreamStop( FSStream );
+		FSEventStreamInvalidate( FSStream );
+	}
+
+	if ( NULL != FSStream )
+	{
+		FSEventStreamRelease( FSStream );
+	}
+
+	efSAFE_DELETE( WatcherGen );
+}
+
+void WatcherFSEvents::init()
+{
+	CFStringRef CFDirectory			= CFStringCreateWithCString( NULL, Directory.c_str(), kCFStringEncodingUTF8 );
+	CFArrayRef	CFDirectoryArray	= CFArrayCreate( NULL, (const void **)&CFDirectory, 1, NULL );
+	
+	Uint32 streamFlags = kFSEventStreamCreateFlagNone;
+	
+	if ( FileWatcherFSEvents::isGranular() )
+	{
+		streamFlags = efswFSEventStreamCreateFlagFileEvents;
+	}
+	else
+	{
+		WatcherGen = new WatcherGeneric( ID, Directory, Listener, FWatcher, Recursive );
+	}
+	
+	FSEventStreamContext ctx;
+	/* Initialize context */
+	ctx.version = 0;
+	ctx.info = this;
+	ctx.retain = NULL;
+	ctx.release = NULL;
+	ctx.copyDescription = NULL;
+
+	FSStream = FSEventStreamCreate( kCFAllocatorDefault, &FileWatcherFSEvents::FSEventCallback, &ctx, CFDirectoryArray, kFSEventStreamEventIdSinceNow, 0.25, streamFlags );
+
+	FWatcher->mNeedInit.push_back( this );
+
+	CFRelease( CFDirectoryArray );
+	CFRelease( CFDirectory );
+}
+
+void WatcherFSEvents::initAsync()
+{
+	FSEventStreamScheduleWithRunLoop( FSStream, FWatcher->mRunLoopRef, kCFRunLoopDefaultMode );
+	FSEventStreamStart( FSStream );
+	initializedAsync = true;
+}
+
+void WatcherFSEvents::sendFileAction( WatchID watchid, const std::string& dir, const std::string& filename, Action action, std::string oldFilename )
+{
+	Listener->handleFileAction( watchid, FileSystem::precomposeFileName( dir ), FileSystem::precomposeFileName( filename ), action, oldFilename );
+}
+
+void WatcherFSEvents::handleAddModDel( const Uint32& flags, const std::string& path, std::string& dirPath, std::string& filePath )
+{
+	if ( flags & efswFSEventStreamEventFlagItemCreated )
+	{
+		if ( FileInfo::exists( path ) )
+		{
+			sendFileAction( ID, dirPath, filePath, Actions::Add );
+		}
+	}
+
+	if ( flags & efswFSEventsModified )
+	{
+		sendFileAction( ID, dirPath, filePath, Actions::Modified );
+	}
+
+	if ( flags & efswFSEventStreamEventFlagItemRemoved )
+	{
+		// Since i don't know the order, at least i try to keep the data consistent with the real state
+		if ( !FileInfo::exists( path ) )
+		{
+			sendFileAction( ID, dirPath, filePath, Actions::Delete );
+		}
+	}
+}
+
+void WatcherFSEvents::handleActions( std::vector<FSEvent>& events )
+{
+	size_t esize = events.size();
+
+	for ( size_t i = 0; i < esize; i++ )
+	{
+		FSEvent& event = events[i];
+
+		if ( event.Flags & (	kFSEventStreamEventFlagUserDropped |
+								kFSEventStreamEventFlagKernelDropped |
+								kFSEventStreamEventFlagEventIdsWrapped |
+								kFSEventStreamEventFlagHistoryDone |
+								kFSEventStreamEventFlagMount |
+								kFSEventStreamEventFlagUnmount |
+								kFSEventStreamEventFlagRootChanged ) )
+		{
+			continue;
+		}
+
+		if ( !Recursive )
+		{
+			/** In case that is not recursive the watcher, ignore the events from subfolders */
+			if ( event.Path.find_last_of( FileSystem::getOSSlash() ) != Directory.size() - 1 )
+			{
+				continue;
+			}
+		}
+
+		if ( FileWatcherFSEvents::isGranular() )
+		{
+			std::string dirPath( FileSystem::pathRemoveFileName( event.Path ) );
+			std::string filePath( FileSystem::fileNameFromPath( event.Path ) );
+
+			if ( event.Flags & (	efswFSEventStreamEventFlagItemCreated |
+									efswFSEventStreamEventFlagItemRemoved |
+									efswFSEventStreamEventFlagItemRenamed )
+			)
+			{
+				if ( dirPath != Directory )
+				{
+					DirsChanged.insert( dirPath );
+				}
+			}
+
+			// This is a mess. But it's FSEvents faults, because shrinks events from the same file in one single event ( so there's no order for them )
+			// For example a file could have been added modified and erased, but i can't know if first was erased and then added and modified, or added, then modified and then erased.
+			// I don't know what they were thinking by doing this...
+			efDEBUG( "Event in: %s - flags: %ld\n", path.c_str(), event.Flags );
+
+			if ( event.Flags & efswFSEventStreamEventFlagItemRenamed )
+			{
+				if (	( i + 1 < esize ) &&
+						( events[ i + 1 ].Flags & efswFSEventStreamEventFlagItemRenamed ) &&
+						( events[ i + 1 ].Id == event.Id + 1 )
+					)
+				{
+					FSEvent& nEvent = events[ i + 1 ];
+					std::string newDir( FileSystem::pathRemoveFileName( nEvent.Path ) );
+					std::string newFilepath( FileSystem::fileNameFromPath( nEvent.Path ) );
+
+					if ( event.Path != nEvent.Path )
+					{
+						if ( dirPath == newDir )
+						{
+							if ( !FileInfo::exists( event.Path ) )
+							{
+								sendFileAction( ID, dirPath, newFilepath, Actions::Moved, filePath );
+							}
+							else
+							{
+								sendFileAction( ID, dirPath, filePath, Actions::Moved, newFilepath );
+							}
+						}
+						else
+						{
+							sendFileAction( ID, dirPath, filePath, Actions::Delete );
+							sendFileAction( ID, newDir, newFilepath, Actions::Add );
+
+							if ( nEvent.Flags & efswFSEventsModified )
+							{
+								sendFileAction( ID, newDir, newFilepath, Actions::Modified );
+							}
+						}
+					}
+					else
+					{
+						handleAddModDel( nEvent.Flags, nEvent.Path, dirPath, filePath );
+					}
+
+					if ( nEvent.Flags & (	efswFSEventStreamEventFlagItemCreated |
+											efswFSEventStreamEventFlagItemRemoved |
+											efswFSEventStreamEventFlagItemRenamed )
+					)
+					{
+						if ( newDir != Directory )
+						{
+							DirsChanged.insert( newDir );
+						}
+					}
+
+					// Skip the renamed file
+					i++;
+				}
+				else if ( FileInfo::exists( event.Path ) )
+				{
+					sendFileAction( ID, dirPath, filePath, Actions::Add );
+
+					if ( event.Flags & efswFSEventsModified )
+					{
+						sendFileAction( ID, dirPath, filePath, Actions::Modified );
+					}
+				}
+				else
+				{
+					sendFileAction( ID, dirPath, filePath, Actions::Delete );
+				}
+			}
+			else
+			{
+				handleAddModDel( event.Flags, event.Path, dirPath, filePath );
+			}
+		}
+		else
+		{
+			efDEBUG( "Directory: %s changed\n", event.Path.c_str() );
+			DirsChanged.insert( event.Path );
+		}
+	}
+}
+
+void WatcherFSEvents::process()
+{
+	std::set<std::string>::iterator it = DirsChanged.begin();
+
+	for ( ; it != DirsChanged.end(); it++ )
+	{
+		if ( !FileWatcherFSEvents::isGranular() )
+		{
+			WatcherGen->watchDir( (*it) );
+		}
+		else
+		{
+			sendFileAction( ID, FileSystem::pathRemoveFileName( (*it) ), FileSystem::fileNameFromPath( (*it) ), Actions::Modified );
+		}
+	}
+
+	DirsChanged.clear();
+}
+
+}
+
+#endif
diff --git a/dep/efsw/src/efsw/WatcherFSEvents.hpp b/dep/efsw/src/efsw/WatcherFSEvents.hpp
new file mode 100644
index 00000000000..d4fc5c9a8d3
--- /dev/null
+++ b/dep/efsw/src/efsw/WatcherFSEvents.hpp
@@ -0,0 +1,70 @@
+#ifndef EFSW_WATCHERINOTIFY_HPP
+#define EFSW_WATCHERINOTIFY_HPP
+
+#include <efsw/FileWatcherImpl.hpp>
+
+#if EFSW_PLATFORM == EFSW_PLATFORM_FSEVENTS
+
+#include <efsw/WatcherGeneric.hpp>
+#include <efsw/FileInfo.hpp>
+#include <CoreFoundation/CoreFoundation.h>
+#include <CoreServices/CoreServices.h>
+#include <set>
+#include <vector>
+
+namespace efsw {
+
+class FileWatcherFSEvents;
+
+class FSEvent
+{
+	public:
+		FSEvent( std::string path, long flags, Uint64 id ) :
+			Path( path ),
+			Flags( flags ),
+			Id ( id )
+		{
+		}
+
+		std::string Path;
+		long Flags;
+		Uint64 Id;
+};
+
+class WatcherFSEvents : public Watcher
+{
+	public:
+		WatcherFSEvents();
+		
+		WatcherFSEvents( WatchID id, std::string directory, FileWatchListener * listener, bool recursive, WatcherFSEvents * parent = NULL );
+		
+		~WatcherFSEvents();
+
+		void init();
+
+		void initAsync();
+
+		void handleActions( std::vector<FSEvent> & events );
+
+		void process();
+
+		FileWatcherFSEvents * FWatcher;
+
+		FSEventStreamRef FSStream;
+	protected:
+		void handleAddModDel( const Uint32 &flags, const std::string &path, std::string &dirPath, std::string &filePath );
+
+		WatcherGeneric * WatcherGen;
+
+		bool initializedAsync;
+
+		std::set<std::string> DirsChanged;
+
+		void sendFileAction( WatchID watchid, const std::string& dir, const std::string& filename, Action action, std::string oldFilename = "" );
+};
+
+}
+
+#endif
+
+#endif
diff --git a/dep/efsw/src/efsw/WatcherGeneric.cpp b/dep/efsw/src/efsw/WatcherGeneric.cpp
new file mode 100644
index 00000000000..94170d3ce87
--- /dev/null
+++ b/dep/efsw/src/efsw/WatcherGeneric.cpp
@@ -0,0 +1,40 @@
+#include <efsw/WatcherGeneric.hpp>
+#include <efsw/FileSystem.hpp>
+#include <efsw/DirWatcherGeneric.hpp>
+
+namespace efsw
+{
+
+WatcherGeneric::WatcherGeneric( WatchID id, const std::string& directory, FileWatchListener * fwl, FileWatcherImpl * fw, bool recursive ) :
+	Watcher( id, directory, fwl, recursive ),
+	WatcherImpl( fw ),
+	DirWatch( NULL )
+{
+	FileSystem::dirAddSlashAtEnd( Directory );
+
+	DirWatch = new DirWatcherGeneric( NULL, this, directory, recursive, false );
+
+	DirWatch->addChilds( false );
+}
+
+WatcherGeneric::~WatcherGeneric()
+{
+	efSAFE_DELETE( DirWatch );
+}
+
+void WatcherGeneric::watch()
+{
+	DirWatch->watch();
+}
+
+void WatcherGeneric::watchDir( std::string dir )
+{
+	DirWatch->watchDir( dir );
+}
+
+bool WatcherGeneric::pathInWatches( std::string path )
+{
+	return DirWatch->pathInWatches( path );
+}
+
+}
diff --git a/dep/efsw/src/efsw/WatcherGeneric.hpp b/dep/efsw/src/efsw/WatcherGeneric.hpp
new file mode 100644
index 00000000000..8794e921b40
--- /dev/null
+++ b/dep/efsw/src/efsw/WatcherGeneric.hpp
@@ -0,0 +1,30 @@
+#ifndef EFSW_WATCHERGENERIC_HPP
+#define EFSW_WATCHERGENERIC_HPP
+
+#include <efsw/FileWatcherImpl.hpp>
+
+namespace efsw
+{
+
+class DirWatcherGeneric;
+
+class WatcherGeneric : public Watcher
+{
+	public:
+		FileWatcherImpl *		WatcherImpl;
+		DirWatcherGeneric *		DirWatch;
+
+		WatcherGeneric( WatchID id, const std::string& directory, FileWatchListener * fwl, FileWatcherImpl * fw, bool recursive );
+
+		~WatcherGeneric();
+
+		void watch();
+
+		void watchDir( std::string dir );
+
+		bool pathInWatches( std::string path );
+};
+
+}
+
+#endif
diff --git a/dep/efsw/src/efsw/WatcherInotify.cpp b/dep/efsw/src/efsw/WatcherInotify.cpp
new file mode 100644
index 00000000000..741641bf43c
--- /dev/null
+++ b/dep/efsw/src/efsw/WatcherInotify.cpp
@@ -0,0 +1,35 @@
+#include <efsw/WatcherInotify.hpp>
+
+namespace efsw {
+
+WatcherInotify::WatcherInotify() :
+	Watcher(),
+	Parent( NULL )
+{
+}
+
+WatcherInotify::WatcherInotify( WatchID id, std::string directory, FileWatchListener * listener, bool recursive, WatcherInotify * parent ) :
+	Watcher( id, directory, listener, recursive ),
+	Parent( parent ),
+	DirInfo( directory )
+{
+}
+
+bool WatcherInotify::inParentTree( WatcherInotify * parent )
+{
+	WatcherInotify * tNext = Parent;
+
+	while ( NULL != tNext )
+	{
+		if ( tNext == parent )
+		{
+			return true;
+		}
+
+		tNext = tNext->Parent;
+	}
+
+	return false;
+}
+	
+} 
diff --git a/dep/efsw/src/efsw/WatcherInotify.hpp b/dep/efsw/src/efsw/WatcherInotify.hpp
new file mode 100644
index 00000000000..1caf399679b
--- /dev/null
+++ b/dep/efsw/src/efsw/WatcherInotify.hpp
@@ -0,0 +1,25 @@
+#ifndef EFSW_WATCHERINOTIFY_HPP 
+#define EFSW_WATCHERINOTIFY_HPP
+
+#include <efsw/FileWatcherImpl.hpp>
+#include <efsw/FileInfo.hpp>
+
+namespace efsw {
+
+class WatcherInotify : public Watcher
+{
+	public:
+		WatcherInotify();
+		
+		WatcherInotify( WatchID id, std::string directory, FileWatchListener * listener, bool recursive, WatcherInotify * parent = NULL );
+
+		bool inParentTree( WatcherInotify * parent );
+
+		WatcherInotify * Parent;
+
+		FileInfo DirInfo;
+};
+
+}
+
+#endif
diff --git a/dep/efsw/src/efsw/WatcherKqueue.cpp b/dep/efsw/src/efsw/WatcherKqueue.cpp
new file mode 100644
index 00000000000..8347fb53439
--- /dev/null
+++ b/dep/efsw/src/efsw/WatcherKqueue.cpp
@@ -0,0 +1,667 @@
+#include <efsw/WatcherKqueue.hpp>
+
+#if EFSW_PLATFORM == EFSW_PLATFORM_KQUEUE || EFSW_PLATFORM == EFSW_PLATFORM_FSEVENTS
+
+#include <sys/stat.h>
+#include <dirent.h>
+#include <unistd.h>
+#include <fcntl.h>
+#include <errno.h>
+#include <cstdio>
+#include <cstdlib>
+#include <cstring>
+#include <efsw/Debug.hpp>
+#include <efsw/String.hpp>
+#include <efsw/System.hpp>
+#include <efsw/FileSystem.hpp>
+#include <efsw/WatcherGeneric.hpp>
+#include <efsw/FileWatcherKqueue.hpp>
+
+#define KEVENT_RESERVE_VALUE (10)
+
+#ifndef O_EVTONLY
+#define O_EVTONLY (O_RDONLY | O_NONBLOCK)
+#endif
+
+namespace efsw {
+
+int comparator(const void* ke1, const void* ke2)
+{
+	const KEvent * kev1	= reinterpret_cast<const KEvent*>( ke1 );
+	const KEvent * kev2	= reinterpret_cast<const KEvent*>( ke2 );
+
+	if ( NULL != kev2->udata )
+	{
+		FileInfo * fi1		= reinterpret_cast<FileInfo*>( kev1->udata );
+		FileInfo * fi2		= reinterpret_cast<FileInfo*>( kev2->udata );
+
+		return strcmp( fi1->Filepath.c_str(), fi2->Filepath.c_str() );
+	}
+
+	return 1;
+}
+
+WatcherKqueue::WatcherKqueue(WatchID watchid, const std::string& dirname, FileWatchListener* listener, bool recursive, FileWatcherKqueue * watcher, WatcherKqueue * parent ) :
+	Watcher( watchid, dirname, listener, recursive ),
+	mLastWatchID(0),
+	mChangeListCount( 0 ),
+	mKqueue( kqueue() ),
+	mWatcher( watcher ),
+	mParent( parent ),
+	mInitOK( true ),
+	mErrno(0)
+{
+	if ( -1 == mKqueue )
+	{
+		efDEBUG( "kqueue() returned invalid descriptor for directory %s. File descriptors count: %ld\n", Directory.c_str(), mWatcher->mFileDescriptorCount );
+		
+		mInitOK = false;
+		mErrno = errno;
+	}
+	else
+	{
+		mWatcher->addFD();
+	}
+}
+
+WatcherKqueue::~WatcherKqueue()
+{
+	// Remove the childs watchers ( sub-folders watches )
+	removeAll();
+
+	for ( size_t i = 0; i < mChangeListCount; i++ )
+	{
+		if ( NULL != mChangeList[i].udata )
+		{
+			FileInfo * fi		= reinterpret_cast<FileInfo*>( mChangeList[i].udata );
+
+			efSAFE_DELETE( fi );
+		}
+	}
+
+	close( mKqueue );
+	
+	mWatcher->removeFD();
+}
+
+void WatcherKqueue::addAll()
+{
+	if ( -1 == mKqueue )
+	{
+		return;
+	}
+
+	// scan directory and call addFile(name, false) on each file
+	FileSystem::dirAddSlashAtEnd( Directory );
+
+	efDEBUG( "addAll(): Added folder: %s\n", Directory.c_str());
+
+	// add base dir
+	int fd = open( Directory.c_str(), O_EVTONLY );
+	
+	if ( -1 == fd )
+	{
+		efDEBUG( "addAll(): Couldn't open folder: %s\n", Directory.c_str() );
+		
+		if ( EACCES != errno )
+		{
+			mInitOK = false;
+		}
+
+		mErrno = errno;
+		
+		return;
+	}
+
+	mDirSnap.setDirectoryInfo( Directory );
+	mDirSnap.scan();
+
+	mChangeList.resize( KEVENT_RESERVE_VALUE );
+
+	// Creates the kevent for the folder
+	EV_SET(
+		&mChangeList[0],
+		fd,
+		EVFILT_VNODE,
+		EV_ADD | EV_ENABLE | EV_ONESHOT,
+		NOTE_DELETE | NOTE_EXTEND | NOTE_WRITE | NOTE_ATTRIB | NOTE_RENAME,
+		0,
+		0
+	);
+
+	mWatcher->addFD();
+
+	// Get the files and directories from the directory
+	FileInfoMap files = FileSystem::filesInfoFromPath( Directory );
+
+	for ( FileInfoMap::iterator it = files.begin(); it != files.end(); it++ )
+	{
+		FileInfo& fi = it->second;
+
+		if ( fi.isRegularFile() )
+		{
+			// Add the regular files kevent
+			addFile( fi.Filepath , false );
+		}
+		else if ( Recursive && fi.isDirectory() && fi.isReadable() )
+		{
+			// Create another watcher for the subfolders ( if recursive )
+			WatchID id = addWatch( fi.Filepath, Listener, Recursive, this );
+
+			// If the watcher is not adding the watcher means that the directory was created
+			if ( id > 0 && !mWatcher->isAddingWatcher() )
+			{
+				handleFolderAction( fi.Filepath, Actions::Add );
+			}
+		}
+	}
+}
+
+void WatcherKqueue::removeAll()
+{
+	efDEBUG( "removeAll(): Removing all child watchers\n" );
+
+	std::list<WatchID> erase;
+
+	for ( WatchMap::iterator it = mWatches.begin(); it != mWatches.end(); it++ )
+	{
+		efDEBUG( "removeAll(): Removed child watcher %s\n", it->second->Directory.c_str() );
+
+		erase.push_back( it->second->ID );
+	}
+
+	for ( std::list<WatchID>::iterator eit = erase.begin(); eit != erase.end(); eit++ )
+	{
+		removeWatch( *eit );
+	}
+}
+
+void WatcherKqueue::addFile(const std::string& name, bool emitEvents)
+{
+	efDEBUG( "addFile(): Added: %s\n", name.c_str() );
+
+	// Open the file to get the file descriptor
+	int fd = open( name.c_str(), O_EVTONLY );
+
+	if( fd == -1 )
+	{
+		efDEBUG( "addFile(): Could open file descriptor for %s. File descriptor count: %ld\n", name.c_str(), mWatcher->mFileDescriptorCount );
+		
+		Errors::Log::createLastError( Errors::FileNotReadable, name );
+		
+		if ( EACCES != errno )
+		{
+			mInitOK = false;
+		}
+
+		mErrno = errno;
+		
+		return;
+	}
+
+	mWatcher->addFD();
+
+	// increase the file kevent file count
+	mChangeListCount++;
+
+	if ( mChangeListCount + KEVENT_RESERVE_VALUE > mChangeList.size() &&
+		 mChangeListCount % KEVENT_RESERVE_VALUE == 0 )
+	{
+		size_t reserve_size = mChangeList.size() + KEVENT_RESERVE_VALUE;
+		mChangeList.resize( reserve_size );
+		efDEBUG( "addFile(): Reserverd more KEvents space for %s, space reserved %ld, list actual size %ld.\n", Directory.c_str(), reserve_size, mChangeListCount );
+	}
+
+	// create entry
+	FileInfo * entry = new FileInfo( name );
+
+	// set the event data at the end of the list
+	EV_SET(
+		&mChangeList[mChangeListCount],
+		fd,
+		EVFILT_VNODE,
+		EV_ADD | EV_ENABLE | EV_ONESHOT,
+		NOTE_DELETE | NOTE_EXTEND | NOTE_WRITE | NOTE_ATTRIB | NOTE_RENAME,
+		0,
+		(void*)entry
+	);
+
+	// qsort sort the list by name
+	qsort(&mChangeList[1], mChangeListCount, sizeof(KEvent), comparator);
+
+	// handle action
+	if( emitEvents )
+	{
+		handleAction(name, Actions::Add);
+	}
+}
+
+void WatcherKqueue::removeFile( const std::string& name, bool emitEvents )
+{
+	efDEBUG( "removeFile(): Trying to remove file: %s\n", name.c_str() );
+
+	// bsearch
+	KEvent target;
+
+	// Create a temporary file info to search the kevent ( searching the directory )
+	FileInfo tempEntry( name );
+
+	target.udata = &tempEntry;
+
+	// Search the kevent
+	KEvent * ke = (KEvent*)bsearch(&target, &mChangeList[0], mChangeListCount + 1, sizeof(KEvent), comparator);
+
+	// Trying to remove a non-existing file?
+	if( !ke )
+	{
+		Errors::Log::createLastError( Errors::FileNotFound, name );
+		efDEBUG( "File not removed\n" );
+		return;
+	}
+
+	efDEBUG( "File removed\n" );
+
+	// handle action
+	if ( emitEvents )
+	{
+		handleAction( name, Actions::Delete );
+	}
+
+	// Delete the user data ( FileInfo ) from the kevent closed
+	FileInfo * del = reinterpret_cast<FileInfo*>( ke->udata );
+
+	efSAFE_DELETE( del );
+
+	// close the file descriptor from the kevent
+	close( ke->ident );
+
+	mWatcher->removeFD();
+
+	memset(ke, 0, sizeof(KEvent));
+
+	// move end to current
+	memcpy(ke, &mChangeList[mChangeListCount], sizeof(KEvent));
+	memset(&mChangeList[mChangeListCount], 0, sizeof(KEvent));
+	--mChangeListCount;
+}
+
+void WatcherKqueue::rescan()
+{
+	efDEBUG( "rescan(): Rescanning: %s\n", Directory.c_str() );
+
+	DirectorySnapshotDiff Diff = mDirSnap.scan();
+
+	if ( Diff.DirChanged )
+	{
+		sendDirChanged();
+	}
+
+	if ( Diff.changed() )
+	{
+		FileInfoList::iterator it;
+		MovedList::iterator mit;
+
+		/// Files
+		DiffIterator( FilesCreated )
+		{
+			addFile( (*it).Filepath );
+		}
+
+		DiffIterator( FilesModified )
+		{
+			handleAction( (*it).Filepath, Actions::Modified );
+		}
+
+		DiffIterator( FilesDeleted )
+		{
+			removeFile( (*it).Filepath );
+		}
+
+		DiffMovedIterator( FilesMoved )
+		{
+			handleAction( (*mit).second.Filepath, Actions::Moved, (*mit).first );
+			removeFile( Directory + (*mit).first, false );
+			addFile( (*mit).second.Filepath, false );
+		}
+
+		/// Directories
+		DiffIterator( DirsCreated )
+		{
+			handleFolderAction( (*it).Filepath, Actions::Add );
+			addWatch( (*it).Filepath, Listener, Recursive, this );
+		}
+
+		DiffIterator( DirsModified )
+		{
+			handleFolderAction( (*it).Filepath, Actions::Modified );
+		}
+
+		DiffIterator( DirsDeleted )
+		{
+			handleFolderAction( (*it).Filepath, Actions::Delete );
+
+			Watcher * watch = findWatcher( (*it).Filepath );
+
+			if ( NULL != watch )
+			{
+				removeWatch( watch->ID );
+
+			}
+		}
+
+		DiffMovedIterator( DirsMoved )
+		{
+			moveDirectory( Directory + (*mit).first, (*mit).second.Filepath );
+		}
+	}
+}
+
+WatchID WatcherKqueue::watchingDirectory( std::string dir )
+{
+	Watcher * watch = findWatcher( dir );
+
+	if ( NULL != watch )
+	{
+		return watch->ID;
+	}
+
+	return Errors::FileNotFound;
+}
+
+void WatcherKqueue::handleAction( const std::string& filename, efsw::Action action, const std::string& oldFilename )
+{
+	Listener->handleFileAction( ID, Directory, FileSystem::fileNameFromPath( filename ), action, FileSystem::fileNameFromPath( oldFilename ) );
+}
+
+void WatcherKqueue::handleFolderAction( std::string filename, efsw::Action action , const std::string &oldFilename )
+{
+	FileSystem::dirRemoveSlashAtEnd( filename );
+
+	handleAction( filename, action, oldFilename );
+}
+
+void WatcherKqueue::sendDirChanged()
+{
+	if ( NULL != mParent )
+	{
+		Listener->handleFileAction( mParent->ID, mParent->Directory, FileSystem::fileNameFromPath( Directory ), Actions::Modified );
+	}
+}
+
+void WatcherKqueue::watch()
+{
+	if ( -1 == mKqueue )
+	{
+		return;
+	}
+
+	int nev = 0;
+	KEvent event;
+
+	// First iterate the childs, to get the events from the deepest folder, to the watcher childs
+	for ( WatchMap::iterator it = mWatches.begin(); it != mWatches.end(); ++it )
+	{
+		it->second->watch();
+	}
+
+	bool needScan = false;
+
+	// Then we get the the events of the current folder
+	while( ( nev = kevent( mKqueue, &mChangeList[0], mChangeListCount + 1, &event, 1, &mWatcher->mTimeOut ) ) != 0 )
+	{
+		// An error ocurred?
+		if( nev == -1 )
+		{
+			efDEBUG( "watch(): Error on directory %s\n", Directory.c_str() );
+			perror("kevent");
+			break;
+		}
+		else
+		{
+			FileInfo * entry = NULL;
+
+			// If udate == NULL means that it is the fisrt element of the change list, the folder.
+			// otherwise it is an event of some file inside the folder
+			if( ( entry = reinterpret_cast<FileInfo*> ( event.udata ) ) != NULL )
+			{
+				efDEBUG( "watch(): File: %s ", entry->Filepath.c_str() );
+
+				// If the event flag is delete... the file was deleted
+				if ( event.fflags & NOTE_DELETE )
+				{
+					efDEBUG( "deleted\n" );
+
+					mDirSnap.removeFile( entry->Filepath );
+
+					removeFile( entry->Filepath );
+				}
+				else if (	event.fflags & NOTE_EXTEND	||
+							event.fflags & NOTE_WRITE	||
+							event.fflags & NOTE_ATTRIB
+				)
+				{
+					// The file was modified
+					efDEBUG( "modified\n" );
+
+					FileInfo fi( entry->Filepath );
+
+					if ( fi != *entry )
+					{
+						*entry = fi;
+
+						mDirSnap.updateFile( entry->Filepath );
+
+						handleAction( entry->Filepath, efsw::Actions::Modified );
+					}
+				}
+				else if ( event.fflags & NOTE_RENAME )
+				{
+					efDEBUG( "moved\n" );
+
+					needScan = true;
+				}
+			}
+			else
+			{
+				needScan = true;
+			}
+		}
+	}
+
+	if ( needScan )
+	{
+		rescan();
+	}
+}
+
+Watcher * WatcherKqueue::findWatcher( const std::string path )
+{
+	WatchMap::iterator it = mWatches.begin();
+
+	for ( ; it != mWatches.end(); it++ )
+	{
+		if ( it->second->Directory == path )
+		{
+			return it->second;
+		}
+	}
+
+	return NULL;
+}
+
+void WatcherKqueue::moveDirectory( std::string oldPath, std::string newPath, bool emitEvents )
+{
+	// Update the directory path if it's a watcher
+	std::string opath2( oldPath );
+	FileSystem::dirAddSlashAtEnd( opath2 );
+
+	Watcher * watch = findWatcher( opath2 );
+
+	if ( NULL != watch )
+	{
+		watch->Directory = opath2;
+	}
+
+	if ( emitEvents )
+	{
+		handleFolderAction( newPath, efsw::Actions::Moved, oldPath );
+	}
+}
+
+WatchID WatcherKqueue::addWatch( const std::string& directory, FileWatchListener* watcher, bool recursive , WatcherKqueue *parent)
+{
+	static long s_fc = 0;
+	static bool s_ug = false;
+
+	std::string dir( directory );
+
+	FileSystem::dirAddSlashAtEnd( dir );
+
+	// This should never happen here
+	if( !FileSystem::isDirectory( dir ) )
+	{
+		return Errors::Log::createLastError( Errors::FileNotFound, dir );
+	}
+	else if ( pathInWatches( dir ) || pathInParent( dir ) )
+	{
+		return Errors::Log::createLastError( Errors::FileRepeated, directory );
+	}
+	else if ( NULL != parent && FileSystem::isRemoteFS( dir ) )
+	{
+		return Errors::Log::createLastError( Errors::FileRemote, dir );
+	}
+
+	std::string curPath;
+	std::string link( FileSystem::getLinkRealPath( dir, curPath ) );
+
+	if ( "" != link )
+	{
+		/// Avoid adding symlinks directories if it's now enabled
+		if ( NULL != parent && !mWatcher->mFileWatcher->followSymlinks() )
+		{
+			return Errors::Log::createLastError( Errors::FileOutOfScope, dir );
+		}
+
+		if ( pathInWatches( link ) || pathInParent( link ) )
+		{
+			return Errors::Log::createLastError( Errors::FileRepeated, link );
+		}
+		else if ( !mWatcher->linkAllowed( curPath, link ) )
+		{
+			return Errors::Log::createLastError( Errors::FileOutOfScope, link );
+		}
+		else
+		{
+			dir = link;
+		}
+	}
+
+	if ( mWatcher->availablesFD() )
+	{
+		WatcherKqueue* watch = new WatcherKqueue( ++mLastWatchID, dir, watcher, recursive, mWatcher, parent );
+
+		mWatches.insert(std::make_pair(mLastWatchID, watch));
+
+		watch->addAll();
+
+		s_fc++;
+
+		// if failed to open the directory... erase the watcher
+		if ( !watch->initOK() )
+		{
+			int le = watch->lastErrno();
+
+			mWatches.erase( watch->ID );
+
+			efSAFE_DELETE( watch );
+
+			mLastWatchID--;
+
+			// Probably the folder has too many files, create a generic watcher
+			if ( EACCES != le )
+			{
+				WatcherGeneric * watch = new WatcherGeneric( ++mLastWatchID, dir, watcher, mWatcher, recursive );
+
+				mWatches.insert(std::make_pair(mLastWatchID, watch));
+			}
+			else
+			{
+				return Errors::Log::createLastError( Errors::Unspecified, link );
+			}
+		}
+	}
+	else
+	{
+		if ( !s_ug )
+		{
+			efDEBUG( "Started using WatcherGeneric, reached file descriptors limit: %ld. Folders added: %ld\n", mWatcher->mFileDescriptorCount, s_fc );
+			s_ug = true;
+		}
+
+		WatcherGeneric * watch = new WatcherGeneric( ++mLastWatchID, dir, watcher, mWatcher, recursive );
+
+		mWatches.insert(std::make_pair(mLastWatchID, watch));
+	}
+
+	return mLastWatchID;
+}
+
+bool WatcherKqueue::initOK()
+{
+	return mInitOK;
+}
+
+void WatcherKqueue::removeWatch( WatchID watchid )
+{
+	WatchMap::iterator iter = mWatches.find(watchid);
+
+	if(iter == mWatches.end())
+		return;
+
+	Watcher * watch = iter->second;
+
+	mWatches.erase(iter);
+
+	efSAFE_DELETE( watch );
+}
+
+bool WatcherKqueue::pathInWatches( const std::string& path )
+{
+	return NULL != findWatcher( path );
+}
+
+bool WatcherKqueue::pathInParent( const std::string &path )
+{
+	WatcherKqueue * pNext = mParent;
+
+	while ( NULL != pNext )
+	{
+		if ( pNext->pathInWatches( path ) )
+		{
+			return true;
+		}
+
+		pNext = pNext->mParent;
+	}
+
+	if ( mWatcher->pathInWatches( path ) )
+	{
+		return true;
+	}
+
+	if ( path == Directory )
+	{
+		return true;
+	}
+
+	return false;
+}
+
+int WatcherKqueue::lastErrno()
+{
+	return mErrno;
+}
+
+}
+
+#endif
diff --git a/dep/efsw/src/efsw/WatcherKqueue.hpp b/dep/efsw/src/efsw/WatcherKqueue.hpp
new file mode 100644
index 00000000000..4babbe73354
--- /dev/null
+++ b/dep/efsw/src/efsw/WatcherKqueue.hpp
@@ -0,0 +1,94 @@
+#ifndef EFSW_WATCHEROSX_HPP
+#define EFSW_WATCHEROSX_HPP
+
+#include <efsw/FileWatcherImpl.hpp>
+
+#if EFSW_PLATFORM == EFSW_PLATFORM_KQUEUE || EFSW_PLATFORM == EFSW_PLATFORM_FSEVENTS
+
+#include <map>
+#include <vector>
+#include <sys/types.h>
+#include <sys/event.h>
+#include <efsw/DirectorySnapshot.hpp>
+
+namespace efsw
+{
+
+class FileWatcherKqueue;
+class WatcherKqueue;
+
+typedef struct kevent KEvent;
+
+/// type for a map from WatchID to WatcherKqueue pointer
+typedef std::map<WatchID, Watcher*> WatchMap;
+
+class WatcherKqueue : public Watcher
+{
+	public:
+		WatcherKqueue( WatchID watchid, const std::string& dirname, FileWatchListener* listener, bool recursive, FileWatcherKqueue * watcher, WatcherKqueue * parent = NULL );
+
+		virtual ~WatcherKqueue();
+
+		void addFile(  const std::string& name, bool emitEvents = true );
+
+		void removeFile( const std::string& name, bool emitEvents = true );
+
+		// called when the directory is actually changed
+		// means a file has been added or removed
+		// rescans the watched directory adding/removing files and sending notices
+		void rescan();
+
+		void handleAction( const std::string& filename, efsw::Action action, const std::string& oldFilename = "" );
+
+		void handleFolderAction( std::string filename, efsw::Action action, const std::string& oldFilename = "" );
+
+		void addAll();
+
+		void removeAll();
+
+		WatchID watchingDirectory( std::string dir );
+
+		void watch();
+
+		WatchID addWatch(const std::string& directory, FileWatchListener* watcher, bool recursive, WatcherKqueue * parent);
+
+		void removeWatch (WatchID watchid );
+		
+		bool initOK();
+
+		int lastErrno();
+	protected:
+		WatchMap			mWatches;
+		int					mLastWatchID;
+
+		// index 0 is always the directory
+		std::vector<KEvent>	mChangeList;
+		size_t				mChangeListCount;
+		DirectorySnapshot	mDirSnap;
+
+		/// The descriptor for the kqueue
+		int					mKqueue;
+
+		FileWatcherKqueue *	mWatcher;
+
+		WatcherKqueue *		mParent;
+		
+		bool				mInitOK;
+		int					mErrno;
+
+		bool pathInWatches( const std::string& path );
+		
+		bool pathInParent( const std::string& path );
+
+		Watcher * findWatcher( const std::string path );
+
+		void moveDirectory( std::string oldPath, std::string newPath, bool emitEvents = true );
+
+		void sendDirChanged();
+};
+
+}
+
+#endif
+
+#endif
diff --git a/dep/efsw/src/efsw/WatcherWin32.cpp b/dep/efsw/src/efsw/WatcherWin32.cpp
new file mode 100644
index 00000000000..01d7b0fcd99
--- /dev/null
+++ b/dep/efsw/src/efsw/WatcherWin32.cpp
@@ -0,0 +1,150 @@
+#include <efsw/WatcherWin32.hpp>
+#include <efsw/String.hpp>
+
+#if EFSW_PLATFORM == EFSW_PLATFORM_WIN32
+
+namespace efsw
+{
+
+/// Unpacks events and passes them to a user defined callback.
+void CALLBACK WatchCallback(DWORD dwErrorCode, DWORD dwNumberOfBytesTransfered, LPOVERLAPPED lpOverlapped)
+{
+	char szFile[MAX_PATH];
+	PFILE_NOTIFY_INFORMATION pNotify;
+	WatcherStructWin32 * tWatch = (WatcherStructWin32*) lpOverlapped;
+	WatcherWin32 * pWatch = tWatch->Watch;
+	size_t offset = 0;
+
+	if (dwNumberOfBytesTransfered == 0)
+	{
+		RefreshWatch(tWatch); // If dwNumberOfBytesTransfered == 0, it means the buffer overflowed (too many changes between GetOverlappedResults calls). Those events are lost, but at least we can refresh so subsequent changes are seen again.
+		return;
+	}
+
+	if (dwErrorCode == ERROR_SUCCESS)
+	{
+		do
+		{
+			bool skip = false;
+
+			pNotify = (PFILE_NOTIFY_INFORMATION) &pWatch->mBuffer[offset];
+			offset += pNotify->NextEntryOffset;
+
+			int count = WideCharToMultiByte(CP_UTF8, 0, pNotify->FileName,
+				pNotify->FileNameLength / sizeof(WCHAR),
+				szFile, MAX_PATH - 1, NULL, NULL);
+			szFile[count] = TEXT('\0');
+
+			std::string nfile( szFile );
+
+			if ( FILE_ACTION_MODIFIED == pNotify->Action )
+			{
+				FileInfo fifile( std::string( pWatch->DirName ) + nfile );
+
+				if ( pWatch->LastModifiedEvent.file.ModificationTime == fifile.ModificationTime && pWatch->LastModifiedEvent.file.Size == fifile.Size && pWatch->LastModifiedEvent.fileName == nfile )
+				{
+					skip = true;
+				}
+
+				pWatch->LastModifiedEvent.fileName	= nfile;
+				pWatch->LastModifiedEvent.file		= fifile;
+			}
+
+			if ( !skip )
+			{
+				pWatch->Watch->handleAction(pWatch, nfile, pNotify->Action);
+			}
+		} while (pNotify->NextEntryOffset != 0);
+	}
+
+	if (!pWatch->StopNow)
+	{
+		RefreshWatch(tWatch);
+	}
+}
+
+/// Refreshes the directory monitoring.
+bool RefreshWatch(WatcherStructWin32* pWatch)
+{
+	return ReadDirectoryChangesW(
+				pWatch->Watch->DirHandle,
+				pWatch->Watch->mBuffer,
+				sizeof(pWatch->Watch->mBuffer),
+				pWatch->Watch->Recursive,
+				pWatch->Watch->NotifyFilter,
+				NULL,
+				&pWatch->Overlapped,
+				NULL
+	) != 0;
+}
+
+/// Stops monitoring a directory.
+void DestroyWatch(WatcherStructWin32* pWatch)
+{
+	if (pWatch)
+	{
+		WatcherWin32 * tWatch = pWatch->Watch;
+
+		tWatch->StopNow = true;
+
+		CancelIo(tWatch->DirHandle);
+
+		RefreshWatch(pWatch);
+
+		if (!HasOverlappedIoCompleted(&pWatch->Overlapped))
+		{
+			SleepEx(5, TRUE);
+		}
+
+		CloseHandle(pWatch->Overlapped.hEvent);
+		CloseHandle(pWatch->Watch->DirHandle);
+		efSAFE_DELETE_ARRAY( pWatch->Watch->DirName );
+		efSAFE_DELETE( pWatch->Watch );
+		HeapFree(GetProcessHeap(), 0, pWatch);
+	}
+}
+
+/// Starts monitoring a directory.
+WatcherStructWin32* CreateWatch(LPCWSTR szDirectory, bool recursive, DWORD NotifyFilter)
+{
+	WatcherStructWin32 * tWatch;
+	size_t ptrsize = sizeof(*tWatch);
+	tWatch = static_cast<WatcherStructWin32*>(HeapAlloc(GetProcessHeap(), HEAP_ZERO_MEMORY, ptrsize));
+
+	WatcherWin32 * pWatch = new WatcherWin32();
+	tWatch->Watch = pWatch;
+
+	pWatch->DirHandle = CreateFileW(
+							szDirectory,
+							GENERIC_READ,
+							FILE_SHARE_READ | FILE_SHARE_WRITE | FILE_SHARE_DELETE,
+							NULL,
+							OPEN_EXISTING,
+							FILE_FLAG_BACKUP_SEMANTICS | FILE_FLAG_OVERLAPPED,
+							NULL
+						);
+
+	if (pWatch->DirHandle != INVALID_HANDLE_VALUE)
+	{
+		tWatch->Overlapped.hEvent = CreateEvent(NULL, TRUE, FALSE, NULL);
+		pWatch->NotifyFilter = NotifyFilter;
+		pWatch->Recursive = recursive;
+
+		if (RefreshWatch(tWatch))
+		{
+			return tWatch;
+		}
+		else
+		{
+			CloseHandle(tWatch->Overlapped.hEvent);
+			CloseHandle(pWatch->DirHandle);
+		}
+	}
+
+	HeapFree(GetProcessHeap(), 0, tWatch);
+	return NULL;
+}
+
+}
+
+ #endif
diff --git a/dep/efsw/src/efsw/WatcherWin32.hpp b/dep/efsw/src/efsw/WatcherWin32.hpp
new file mode 100644
index 00000000000..3c6d988fbf4
--- /dev/null
+++ b/dep/efsw/src/efsw/WatcherWin32.hpp
@@ -0,0 +1,77 @@
+#ifndef EFSW_WATCHERWIN32_HPP
+#define EFSW_WATCHERWIN32_HPP
+
+#include <efsw/FileWatcherImpl.hpp>
+#include <efsw/FileInfo.hpp>
+
+#if EFSW_PLATFORM == EFSW_PLATFORM_WIN32
+
+#include <windows.h>
+
+#ifdef EFSW_COMPILER_MSVC
+	#pragma comment(lib, "comctl32.lib")
+	#pragma comment(lib, "user32.lib")
+	#pragma comment(lib, "ole32.lib")
+
+	// disable secure warnings
+	#pragma warning (disable: 4996)
+#endif
+
+namespace efsw
+{
+
+class WatcherWin32;
+
+/// Internal watch data
+struct WatcherStructWin32
+{
+	OVERLAPPED Overlapped;
+	WatcherWin32 *	Watch;
+};
+
+class cLastModifiedEvent
+{
+	public:
+		cLastModifiedEvent() {}
+		FileInfo	file;
+		std::string fileName;
+};
+
+bool RefreshWatch(WatcherStructWin32* pWatch);
+
+void CALLBACK WatchCallback(DWORD dwErrorCode, DWORD dwNumberOfBytesTransfered, LPOVERLAPPED lpOverlapped);
+
+void DestroyWatch(WatcherStructWin32* pWatch);
+
+WatcherStructWin32* CreateWatch(LPCWSTR szDirectory, bool recursive, DWORD NotifyFilter);
+
+class WatcherWin32 : public Watcher
+{
+	public:
+		WatcherWin32() :
+			Struct( NULL ),
+			DirHandle( NULL ),
+			lParam( 0 ),
+			NotifyFilter( 0 ),
+			StopNow( false ),
+			Watch( NULL ),
+			DirName( NULL )
+		{
+		}
+
+		WatcherStructWin32 * Struct;
+		HANDLE DirHandle;
+		BYTE mBuffer[63 * 1024]; // do NOT make this bigger than 64K because it will fail if the folder being watched is on the network! (see http://msdn.microsoft.com/en-us/library/windows/desktop/aa365465(v=vs.85).aspx)
+		LPARAM lParam;
+		DWORD NotifyFilter;
+		bool StopNow;
+		FileWatcherImpl* Watch;
+		char* DirName;
+		cLastModifiedEvent LastModifiedEvent;
+};
+
+}
+
+#endif
+
+#endif
diff --git a/dep/efsw/src/efsw/base.hpp b/dep/efsw/src/efsw/base.hpp
new file mode 100644
index 00000000000..26108c76985
--- /dev/null
+++ b/dep/efsw/src/efsw/base.hpp
@@ -0,0 +1,110 @@
+#ifndef EFSW_BASE
+#define EFSW_BASE
+
+#include <efsw/sophist.h>
+#include <efsw/efsw.hpp>
+
+namespace efsw {
+
+typedef SOPHIST_int8		Int8;
+typedef SOPHIST_uint8		Uint8;
+typedef SOPHIST_int16		Int16;
+typedef SOPHIST_uint16		Uint16;
+typedef SOPHIST_int32		Int32;
+typedef SOPHIST_uint32		Uint32;
+typedef SOPHIST_int64		Int64;
+typedef SOPHIST_uint64		Uint64;
+
+#define EFSW_OS_WIN		1
+#define EFSW_OS_LINUX	2
+#define EFSW_OS_MACOSX	3
+#define EFSW_OS_BSD		4
+#define EFSW_OS_SOLARIS	5
+#define EFSW_OS_HAIKU	6
+#define EFSW_OS_ANDROID	7
+#define EFSW_OS_IOS		8
+
+#define EFSW_PLATFORM_WIN32		1
+#define EFSW_PLATFORM_INOTIFY	2
+#define EFSW_PLATFORM_KQUEUE	3
+#define EFSW_PLATFORM_FSEVENTS	4
+#define EFSW_PLATFORM_GENERIC	5
+
+#if defined(_WIN32)
+	///	Any Windows platform
+	#define EFSW_OS EFSW_OS_WIN
+	#define EFSW_PLATFORM EFSW_PLATFORM_WIN32
+
+	#if ( defined( _MSCVER ) || defined( _MSC_VER ) )
+		#define EFSW_COMPILER_MSVC
+	#endif
+
+#elif defined( __FreeBSD__ ) || defined(__OpenBSD__) || defined( __NetBSD__ ) || defined( __DragonFly__ )
+	#define EFSW_OS EFSW_OS_BSD
+	#define EFSW_PLATFORM EFSW_PLATFORM_KQUEUE
+
+#elif defined( __APPLE_CC__ ) || defined ( __APPLE__ )
+	#include <TargetConditionals.h>
+
+	#if defined( __IPHONE__ ) || ( defined( TARGET_OS_IPHONE ) && TARGET_OS_IPHONE ) || ( defined( TARGET_IPHONE_SIMULATOR ) && TARGET_IPHONE_SIMULATOR )
+		#define EFSW_OS EFSW_OS_IOS
+		#define EFSW_PLATFORM EFSW_PLATFORM_KQUEUE
+	#else
+		#define EFSW_OS EFSW_OS_MACOSX
+
+		#if defined(EFSW_FSEVENTS_NOT_SUPPORTED)
+			#define EFSW_PLATFORM EFSW_PLATFORM_KQUEUE
+		#else
+			#define EFSW_PLATFORM EFSW_PLATFORM_FSEVENTS
+		#endif
+	#endif
+
+#elif defined(__linux__)
+	///	This includes Linux and Android
+	#ifndef EFSW_KQUEUE
+		#define EFSW_PLATFORM EFSW_PLATFORM_INOTIFY
+	#else
+		/// This is for testing libkqueue, sadly it doesnt work
+		#define EFSW_PLATFORM EFSW_PLATFORM_KQUEUE
+	#endif
+
+	#if defined( __ANDROID__ ) || defined( ANDROID )
+		#define EFSW_OS EFSW_OS_ANDROID
+	#else
+		#define EFSW_OS EFSW_OS_LINUX
+	#endif
+
+#else
+	#if defined( __SVR4 )
+		#define EFSW_OS EFSW_OS_SOLARIS
+	#elif defined( __HAIKU__ ) || defined( __BEOS__ )
+		#define EFSW_OS EFSW_OS_HAIKU
+	#endif
+
+	///	Everything else
+	#define EFSW_PLATFORM EFSW_PLATFORM_GENERIC
+#endif
+
+#if EFSW_PLATFORM != EFSW_PLATFORM_WIN32
+	#define EFSW_PLATFORM_POSIX
+#endif
+
+#if 1 == SOPHIST_pointer64
+	#define EFSW_64BIT
+#else
+	#define EFSW_32BIT
+#endif
+
+#if defined(arm) || defined(__arm__)
+	#define	EFSW_ARM
+#endif
+
+#define efCOMMA ,
+
+#define efSAFE_DELETE(p)		{ if(p) { delete (p);			(p)=NULL; } }
+#define efSAFE_DELETE_ARRAY(p)  { if(p) { delete [] (p);		(p)=NULL; } }
+#define efARRAY_SIZE(__array)	( sizeof(__array) / sizeof(__array[0]) )
+
+}
+
+#endif
diff --git a/dep/efsw/src/efsw/inotify-nosys.h b/dep/efsw/src/efsw/inotify-nosys.h
new file mode 100644
index 00000000000..75df5d3ced3
--- /dev/null
+++ b/dep/efsw/src/efsw/inotify-nosys.h
@@ -0,0 +1,159 @@
+#ifndef _LINUX_INOTIFY_H
+#define _LINUX_INOTIFY_H
+
+#include <stdint.h>
+#include <sys/syscall.h>
+#include <unistd.h>
+
+/*
+ * struct inotify_event - structure read from the inotify device for each event
+ *
+ * When you are watching a directory, you will receive the filename for events
+ * such as IN_CREATE, IN_DELETE, IN_OPEN, IN_CLOSE, ..., relative to the wd.
+ */
+struct inotify_event {
+	int		wd;		/* watch descriptor */
+	uint32_t		mask;		/* watch mask */
+	uint32_t		cookie;		/* cookie to synchronize two events */
+	uint32_t		len;		/* length (including nulls) of name */
+	char		name __flexarr;	/* stub for possible name */
+};
+
+/* the following are legal, implemented events that user-space can watch for */
+#define IN_ACCESS		0x00000001	/* File was accessed */
+#define IN_MODIFY		0x00000002	/* File was modified */
+#define IN_ATTRIB		0x00000004	/* Metadata changed */
+#define IN_CLOSE_WRITE		0x00000008	/* Writtable file was closed */
+#define IN_CLOSE_NOWRITE	0x00000010	/* Unwrittable file closed */
+#define IN_OPEN			0x00000020	/* File was opened */
+#define IN_MOVED_FROM		0x00000040	/* File was moved from X */
+#define IN_MOVED_TO		0x00000080	/* File was moved to Y */
+#define IN_CREATE		0x00000100	/* Subfile was created */
+#define IN_DELETE		0x00000200	/* Subfile was deleted */
+#define IN_DELETE_SELF		0x00000400	/* Self was deleted */
+#define IN_MOVE_SELF		0x00000800	/* Self was moved */
+
+/* the following are legal events.  they are sent as needed to any watch */
+#define IN_UNMOUNT		0x00002000	/* Backing fs was unmounted */
+#define IN_Q_OVERFLOW		0x00004000	/* Event queued overflowed */
+#define IN_IGNORED		0x00008000	/* File was ignored */
+
+/* helper events */
+#define IN_CLOSE		(IN_CLOSE_WRITE | IN_CLOSE_NOWRITE) /* close */
+#define IN_MOVE			(IN_MOVED_FROM | IN_MOVED_TO) /* moves */
+
+/* special flags */
+#define IN_ONLYDIR		0x01000000	/* only watch the path if it is a directory */
+#define IN_DONT_FOLLOW		0x02000000	/* don't follow a sym link */
+#define IN_MASK_ADD		0x20000000	/* add to the mask of an already existing watch */
+#define IN_ISDIR		0x40000000	/* event occurred against dir */
+#define IN_ONESHOT		0x80000000	/* only send event once */
+
+/*
+ * All of the events - we build the list by hand so that we can add flags in
+ * the future and not break backward compatibility.  Apps will get only the
+ * events that they originally wanted.  Be sure to add new events here!
+ */
+#define IN_ALL_EVENTS	(IN_ACCESS | IN_MODIFY | IN_ATTRIB | IN_CLOSE_WRITE | \
+			 IN_CLOSE_NOWRITE | IN_OPEN | IN_MOVED_FROM | \
+			 IN_MOVED_TO | IN_DELETE | IN_CREATE | IN_DELETE_SELF | \
+			 IN_MOVE_SELF)
+
+#if defined (__alpha__)
+# define __NR_inotify_init 444
+# define __NR_inotify_add_watch 445
+# define __NR_inotify_rm_watch 446
+
+#elif defined (__arm__)
+# define __NR_inotify_init (__NR_SYSCALL_BASE+316)
+# define __NR_inotify_add_watch (__NR_SYSCALL_BASE+317)
+# define __NR_inotify_rm_watch (__NR_SYSCALL_BASE+318)
+
+#elif defined (__frv__)
+# define __NR_inotify_init 291
+# define __NR_inotify_add_watch 292
+# define __NR_inotify_rm_watch 293
+
+#elif defined(__i386__)
+# define __NR_inotify_init 291
+# define __NR_inotify_add_watch 292
+# define __NR_inotify_rm_watch 293
+
+#elif defined (__ia64__)
+# define __NR_inotify_init 1277
+# define __NR_inotify_add_watch 1278
+# define __NR_inotify_rm_watch 1279
+
+#elif defined (__mips__)
+# if _MIPS_SIM == _MIPS_SIM_ABI32
+#  define __NR_inotify_init (__NR_Linux + 284)
+#  define __NR_inotify_add_watch (__NR_Linux + 285)
+#  define __NR_inotify_rm_watch (__NR_Linux + 286)
+# endif
+# if _MIPS_SIM == _MIPS_SIM_ABI64
+#  define __NR_inotify_init (__NR_Linux + 243)
+#  define __NR_inotify_add_watch (__NR_Linux + 243)
+#  define __NR_inotify_rm_watch (__NR_Linux + 243)
+# endif
+# if _MIPS_SIM == _MIPS_SIM_NABI32
+#  define __NR_inotify_init (__NR_Linux + 247)
+#  define __NR_inotify_add_watch (__NR_Linux + 248)
+#  define __NR_inotify_rm_watch (__NR_Linux + 249)
+# endif
+
+#elif defined(__parisc__)
+# define __NR_inotify_init (__NR_Linux + 269)
+# define __NR_inotify_add_watch (__NR_Linux + 270)
+# define __NR_inotify_rm_watch (__NR_Linux + 271)
+
+#elif defined(__powerpc__) || defined(__powerpc64__)
+# define __NR_inotify_init 275
+# define __NR_inotify_add_watch 276
+# define __NR_inotify_rm_watch 277
+
+#elif defined (__s390__)
+# define __NR_inotify_init 284
+# define __NR_inotify_add_watch 285
+# define __NR_inotify_rm_watch 286
+
+#elif defined (__sh__)
+# define __NR_inotify_init 290
+# define __NR_inotify_add_watch 291
+# define __NR_inotify_rm_watch 292
+
+#elif defined (__sh64__)
+# define __NR_inotify_init 318
+# define __NR_inotify_add_watch 319
+# define __NR_inotify_rm_watch 320
+
+#elif defined (__sparc__) || defined (__sparc64__)
+# define __NR_inotify_init 151
+# define __NR_inotify_add_watch 152
+# define __NR_inotify_rm_watch 156
+
+#elif defined(__x86_64__)
+# define __NR_inotify_init 253
+# define __NR_inotify_add_watch 254
+# define __NR_inotify_rm_watch 255
+
+#else
+# error "Unsupported architecture!"
+#endif
+
+static inline int inotify_init (void)
+{
+	return syscall (__NR_inotify_init);
+}
+
+static inline int inotify_add_watch (int fd, const char *name, uint32_t mask)
+{
+	return syscall (__NR_inotify_add_watch, fd, name, mask);
+}
+
+static inline int inotify_rm_watch (int fd, uint32_t wd)
+{
+	return syscall (__NR_inotify_rm_watch, fd, wd);
+}
+
+
+#endif	/* _LINUX_INOTIFY_H */
diff --git a/dep/efsw/src/efsw/platform/platformimpl.hpp b/dep/efsw/src/efsw/platform/platformimpl.hpp
new file mode 100644
index 00000000000..86a74eee0c8
--- /dev/null
+++ b/dep/efsw/src/efsw/platform/platformimpl.hpp
@@ -0,0 +1,20 @@
+#ifndef EFSW_PLATFORMIMPL_HPP
+#define EFSW_PLATFORMIMPL_HPP
+
+#include <efsw/base.hpp>
+
+#if defined( EFSW_PLATFORM_POSIX )
+	#include <efsw/platform/posix/ThreadImpl.hpp>
+	#include <efsw/platform/posix/MutexImpl.hpp>
+	#include <efsw/platform/posix/SystemImpl.hpp>
+	#include <efsw/platform/posix/FileSystemImpl.hpp>
+#elif EFSW_PLATFORM == EFSW_PLATFORM_WIN32
+	#include <efsw/platform/win/ThreadImpl.hpp>
+	#include <efsw/platform/win/MutexImpl.hpp>
+	#include <efsw/platform/win/SystemImpl.hpp>
+	#include <efsw/platform/win/FileSystemImpl.hpp>
+#else
+	#error Thread, Mutex, and System not implemented for this platform.
+#endif
+
+#endif
diff --git a/dep/efsw/src/efsw/platform/posix/FileSystemImpl.cpp b/dep/efsw/src/efsw/platform/posix/FileSystemImpl.cpp
new file mode 100644
index 00000000000..e061b25d56b
--- /dev/null
+++ b/dep/efsw/src/efsw/platform/posix/FileSystemImpl.cpp
@@ -0,0 +1,144 @@
+#include <efsw/platform/posix/FileSystemImpl.hpp>
+
+#if defined( EFSW_PLATFORM_POSIX )
+
+#include <efsw/FileInfo.hpp>
+#include <efsw/FileSystem.hpp>
+#include <dirent.h>
+#include <cstring>
+
+#ifndef _DARWIN_FEATURE_64_BIT_INODE
+#define _DARWIN_FEATURE_64_BIT_INODE
+#endif
+
+#ifndef _FILE_OFFSET_BITS
+#define _FILE_OFFSET_BITS 64
+#endif
+
+#include <sys/stat.h>
+#include <cstdlib>
+
+#if EFSW_OS == EFSW_OS_LINUX || EFSW_OS == EFSW_OS_SOLARIS || EFSW_OS == EFSW_OS_ANDROID
+#include <sys/vfs.h>
+#elif EFSW_OS == EFSW_OS_MACOSX || EFSW_OS == EFSW_OS_BSD || EFSW_OS == EFSW_OS_IOS
+#include <sys/param.h>
+#include <sys/mount.h>
+#endif
+
+/** Remote file systems codes */
+#define S_MAGIC_AFS 0x5346414F
+#define S_MAGIC_AUFS 0x61756673
+#define S_MAGIC_CEPH 0x00C36400
+#define S_MAGIC_CIFS 0xFF534D42
+#define S_MAGIC_CODA 0x73757245
+#define S_MAGIC_FHGFS 0x19830326
+#define S_MAGIC_FUSEBLK 0x65735546
+#define S_MAGIC_FUSECTL 0x65735543
+#define S_MAGIC_GFS 0x01161970
+#define S_MAGIC_GPFS 0x47504653
+#define S_MAGIC_KAFS 0x6B414653
+#define S_MAGIC_LUSTRE 0x0BD00BD0
+#define S_MAGIC_NCP 0x564C
+#define S_MAGIC_NFS 0x6969
+#define S_MAGIC_NFSD 0x6E667364
+#define S_MAGIC_OCFS2 0x7461636F
+#define S_MAGIC_PANFS 0xAAD7AAEA
+#define S_MAGIC_PIPEFS 0x50495045
+#define S_MAGIC_SMB 0x517B
+#define S_MAGIC_SNFS 0xBEEFDEAD
+#define S_MAGIC_VMHGFS 0xBACBACBC
+#define S_MAGIC_VXFS 0xA501FCF5
+
+namespace efsw { namespace Platform {
+
+FileInfoMap FileSystem::filesInfoFromPath( const std::string& path )
+{
+	FileInfoMap files;
+
+	DIR *dp;
+	struct dirent *dirp;
+
+	if( ( dp = opendir( path.c_str() ) ) == NULL)
+		return files;
+
+	while ( ( dirp = readdir(dp) ) != NULL)
+	{
+		if ( strcmp( dirp->d_name, ".." ) != 0 && strcmp( dirp->d_name, "." ) != 0 )
+		{
+			std::string name( dirp->d_name );
+			std::string fpath( path + name );
+
+			files[ name ] = FileInfo( fpath );
+		}
+	}
+
+	closedir(dp);
+
+	return files;
+}
+
+char FileSystem::getOSSlash()
+{
+	return '/';
+}
+
+bool FileSystem::isDirectory( const std::string& path )
+{
+	struct stat st;
+	int res = stat( path.c_str(), &st );
+
+	if ( 0 == res )
+	{
+		return static_cast<bool>( S_ISDIR(st.st_mode) );
+	}
+
+	return false;
+}
+
+bool FileSystem::isRemoteFS( const std::string& directory )
+{
+#if EFSW_OS == EFSW_OS_LINUX || EFSW_OS == EFSW_OS_MACOSX || EFSW_OS == EFSW_OS_BSD || EFSW_OS == EFSW_OS_SOLARIS || EFSW_OS == EFSW_OS_ANDROID || EFSW_OS == EFSW_OS_IOS
+	struct statfs statfsbuf;
+
+	statfs( directory.c_str(), &statfsbuf );
+
+	switch ( statfsbuf.f_type | 0UL )
+	{
+		case S_MAGIC_AFS: /* 0x5346414F remote */
+		case S_MAGIC_AUFS: /* 0x61756673 remote */
+		case S_MAGIC_CEPH: /* 0x00C36400 remote */
+		case S_MAGIC_CIFS: /* 0xFF534D42 remote */
+		case S_MAGIC_CODA: /* 0x73757245 remote */
+		case S_MAGIC_FHGFS: /* 0x19830326 remote */
+		case S_MAGIC_FUSEBLK: /* 0x65735546 remote */
+		case S_MAGIC_FUSECTL: /* 0x65735543 remote */
+		case S_MAGIC_GFS: /* 0x01161970 remote */
+		case S_MAGIC_GPFS: /* 0x47504653 remote */
+		case S_MAGIC_KAFS: /* 0x6B414653 remote */
+		case S_MAGIC_LUSTRE: /* 0x0BD00BD0 remote */
+		case S_MAGIC_NCP: /* 0x564C remote */
+		case S_MAGIC_NFS: /* 0x6969 remote */
+		case S_MAGIC_NFSD: /* 0x6E667364 remote */
+		case S_MAGIC_OCFS2: /* 0x7461636F remote */
+		case S_MAGIC_PANFS: /* 0xAAD7AAEA remote */
+		case S_MAGIC_PIPEFS: /* 0x50495045 remote */
+		case S_MAGIC_SMB: /* 0x517B remote */
+		case S_MAGIC_SNFS: /* 0xBEEFDEAD remote */
+		case S_MAGIC_VMHGFS: /* 0xBACBACBC remote */
+		case S_MAGIC_VXFS: /* 0xA501FCF5 remote */
+		{
+			return true;
+		}
+		default:
+		{
+			return false;
+		}
+	}
+#endif
+
+	return false;
+}
+
+}}
+
+#endif
diff --git a/dep/efsw/src/efsw/platform/posix/FileSystemImpl.hpp b/dep/efsw/src/efsw/platform/posix/FileSystemImpl.hpp
new file mode 100644
index 00000000000..865b3f8dfdf
--- /dev/null
+++ b/dep/efsw/src/efsw/platform/posix/FileSystemImpl.hpp
@@ -0,0 +1,27 @@
+#ifndef EFSW_FILESYSTEMIMPLPOSIX_HPP
+#define EFSW_FILESYSTEMIMPLPOSIX_HPP
+
+#include <efsw/base.hpp>
+#include <efsw/FileInfo.hpp>
+
+#if defined( EFSW_PLATFORM_POSIX )
+
+namespace efsw { namespace Platform {
+
+class FileSystem
+{
+	public:
+		static FileInfoMap filesInfoFromPath( const std::string& path );
+
+		static char getOSSlash();
+
+		static bool isDirectory( const std::string& path );
+
+		static bool isRemoteFS( const std::string& directory );
+};
+
+}}
+
+#endif
+
+#endif
diff --git a/dep/efsw/src/efsw/platform/posix/MutexImpl.cpp b/dep/efsw/src/efsw/platform/posix/MutexImpl.cpp
new file mode 100644
index 00000000000..6f2af5abc61
--- /dev/null
+++ b/dep/efsw/src/efsw/platform/posix/MutexImpl.cpp
@@ -0,0 +1,32 @@
+#include <efsw/platform/posix/MutexImpl.hpp>
+
+#if defined( EFSW_PLATFORM_POSIX )
+
+namespace efsw { namespace Platform {
+
+MutexImpl::MutexImpl()
+{
+	pthread_mutexattr_t attributes;
+	pthread_mutexattr_init(&attributes);
+	pthread_mutexattr_settype(&attributes, PTHREAD_MUTEX_RECURSIVE);
+	pthread_mutex_init(&mMutex, &attributes);
+}
+
+MutexImpl::~MutexImpl()
+{
+	pthread_mutex_destroy(&mMutex);
+}
+
+void MutexImpl::lock()
+{
+	pthread_mutex_lock(&mMutex);
+}
+
+void MutexImpl::unlock()
+{
+	pthread_mutex_unlock(&mMutex);
+}
+
+}}
+
+#endif
diff --git a/dep/efsw/src/efsw/platform/posix/MutexImpl.hpp b/dep/efsw/src/efsw/platform/posix/MutexImpl.hpp
new file mode 100644
index 00000000000..d51eecb16aa
--- /dev/null
+++ b/dep/efsw/src/efsw/platform/posix/MutexImpl.hpp
@@ -0,0 +1,31 @@
+#ifndef EFSW_MUTEXIMPLPOSIX_HPP
+#define EFSW_MUTEXIMPLPOSIX_HPP
+
+#include <efsw/base.hpp>
+
+#if defined( EFSW_PLATFORM_POSIX )
+
+#include <pthread.h>
+
+namespace efsw { namespace Platform {
+
+class MutexImpl
+{
+	public:
+		MutexImpl();
+
+		~MutexImpl();
+
+		void lock();
+
+		void unlock();
+	private:
+		pthread_mutex_t mMutex;
+};
+
+}}
+
+#endif
+
+#endif
+ 
diff --git a/dep/efsw/src/efsw/platform/posix/SystemImpl.cpp b/dep/efsw/src/efsw/platform/posix/SystemImpl.cpp
new file mode 100644
index 00000000000..22e37095afd
--- /dev/null
+++ b/dep/efsw/src/efsw/platform/posix/SystemImpl.cpp
@@ -0,0 +1,180 @@
+#include <efsw/platform/posix/SystemImpl.hpp>
+
+#if defined( EFSW_PLATFORM_POSIX )
+
+#include <cstdio>
+#include <pthread.h>
+#include <sys/time.h>
+#include <limits.h>
+#include <sys/resource.h>
+
+#include <efsw/FileSystem.hpp>
+#include <efsw/Debug.hpp>
+
+#if EFSW_OS == EFSW_OS_MACOSX
+	#include <CoreFoundation/CoreFoundation.h>
+#elif EFSW_OS == EFSW_OS_LINUX || EFSW_OS == EFSW_OS_ANDROID
+	#include <libgen.h>
+	#include <unistd.h>
+#elif EFSW_OS == EFSW_OS_HAIKU
+	#include <kernel/OS.h>
+	#include <kernel/image.h>
+#elif EFSW_OS == EFSW_OS_SOLARIS
+	#include <stdlib.h>
+#elif EFSW_OS == EFSW_OS_BSD
+	#include <sys/sysctl.h>
+#endif
+
+namespace efsw { namespace Platform {
+
+void System::sleep( const unsigned long& ms )
+{
+	// usleep( static_cast<unsigned long>( ms * 1000 ) );
+
+	// usleep is not reliable enough (it might block the
+	// whole process instead of just the current thread)
+	// so we must use pthread_cond_timedwait instead
+
+	// this implementation is inspired from Qt
+	// and taken from SFML
+
+	unsigned long long usecs = ms * 1000;
+
+	// get the current time
+	timeval tv;
+	gettimeofday(&tv, NULL);
+
+	// construct the time limit (current time + time to wait)
+	timespec ti;
+	ti.tv_nsec = (tv.tv_usec + (usecs % 1000000)) * 1000;
+	ti.tv_sec = tv.tv_sec + (usecs / 1000000) + (ti.tv_nsec / 1000000000);
+	ti.tv_nsec %= 1000000000;
+
+	// create a mutex and thread condition
+	pthread_mutex_t mutex;
+	pthread_mutex_init(&mutex, 0);
+	pthread_cond_t condition;
+	pthread_cond_init(&condition, 0);
+
+	// wait...
+	pthread_mutex_lock(&mutex);
+	pthread_cond_timedwait(&condition, &mutex, &ti);
+	pthread_mutex_unlock(&mutex);
+
+	// destroy the mutex and condition
+	pthread_cond_destroy(&condition);
+}
+
+std::string System::getProcessPath()
+{
+#if EFSW_OS == EFSW_OS_MACOSX
+	char exe_file[FILENAME_MAX + 1];
+
+	CFBundleRef mainBundle = CFBundleGetMainBundle();
+
+	if (mainBundle)
+	{
+		CFURLRef mainURL = CFBundleCopyBundleURL(mainBundle);
+
+		if (mainURL)
+		{
+			int ok = CFURLGetFileSystemRepresentation ( mainURL, (Boolean) true, (UInt8*)exe_file, FILENAME_MAX );
+
+			if (ok)
+			{
+				return std::string(exe_file) + "/";
+			}
+		}
+	}
+
+	return "./";
+#elif EFSW_OS == EFSW_OS_LINUX
+	char exe_file[FILENAME_MAX + 1];
+
+	int size;
+
+	size = readlink("/proc/self/exe", exe_file, FILENAME_MAX);
+
+	if (size < 0)
+	{
+		return  std::string( "./" );
+	}
+	else
+	{
+		exe_file[size] = '\0';
+		return std::string( dirname( exe_file ) ) + "/";
+	}
+
+#elif EFSW_OS == EFSW_OS_BSD
+	int mib[4];
+	mib[0] = CTL_KERN;
+	mib[1] = KERN_PROC;
+	mib[2] = KERN_PROC_PATHNAME;
+	mib[3] = -1;
+	char buf[1024];
+	size_t cb = sizeof(buf);
+	sysctl(mib, 4, buf, &cb, NULL, 0);
+
+	return FileSystem::pathRemoveFileName( std::string( buf ) );
+
+#elif EFSW_OS == EFSW_OS_SOLARIS
+	return FileSystem::pathRemoveFileName( std::string( getexecname() ) );
+
+#elif EFSW_OS == EFSW_OS_HAIKU
+	image_info info;
+	int32 cookie = 0;
+
+	while ( B_OK == get_next_image_info( 0, &cookie, &info ) )
+	{
+		if ( info.type == B_APP_IMAGE )
+			break;
+	}
+
+	return FileSystem::pathRemoveFileName( std::string( info.name ) );
+
+#elif EFSW_OS == EFSW_OS_ANDROID
+	return "/sdcard/";
+
+#else
+	#warning getProcessPath() not implemented on this platform. ( will return "./" )
+	return "./";
+
+#endif
+}
+
+void System::maxFD()
+{
+	static bool maxed = false;
+
+	if ( !maxed )
+	{
+		struct rlimit limit;
+		getrlimit( RLIMIT_NOFILE, &limit );
+		limit.rlim_cur = limit.rlim_max;
+		setrlimit( RLIMIT_NOFILE, &limit );
+
+		getrlimit( RLIMIT_NOFILE, &limit );
+
+		efDEBUG( "File descriptor limit %ld\n", limit.rlim_cur );
+
+		maxed = true;
+	}
+}
+
+Uint64 System::getMaxFD()
+{
+	static rlim_t max_fd = 0;
+
+	if ( max_fd == 0 )
+	{
+		struct rlimit limit;
+		getrlimit( RLIMIT_NOFILE, &limit );
+		max_fd = limit.rlim_cur;
+	}
+
+	return max_fd;
+}
+
+}}
+
+#endif
diff --git a/dep/efsw/src/efsw/platform/posix/SystemImpl.hpp b/dep/efsw/src/efsw/platform/posix/SystemImpl.hpp
new file mode 100644
index 00000000000..34734104467
--- /dev/null
+++ b/dep/efsw/src/efsw/platform/posix/SystemImpl.hpp
@@ -0,0 +1,26 @@
+#ifndef EFSW_SYSTEMIMPLPOSIX_HPP
+#define EFSW_SYSTEMIMPLPOSIX_HPP
+
+#include <efsw/base.hpp>
+
+#if defined( EFSW_PLATFORM_POSIX )
+
+namespace efsw { namespace Platform {
+
+class System
+{
+	public:
+		static void sleep( const unsigned long& ms );
+
+		static std::string getProcessPath();
+
+		static void maxFD();
+
+		static Uint64 getMaxFD();
+};
+
+}}
+
+#endif
+
+#endif
diff --git a/dep/efsw/src/efsw/platform/posix/ThreadImpl.cpp b/dep/efsw/src/efsw/platform/posix/ThreadImpl.cpp
new file mode 100644
index 00000000000..2d3671db9b8
--- /dev/null
+++ b/dep/efsw/src/efsw/platform/posix/ThreadImpl.cpp
@@ -0,0 +1,68 @@
+#include <efsw/platform/posix/ThreadImpl.hpp>
+#include <efsw/Thread.hpp>
+
+#if defined( EFSW_PLATFORM_POSIX )
+
+#include <cassert>
+#include <iostream>
+#include <efsw/Debug.hpp>
+
+namespace efsw { namespace Platform {
+
+ThreadImpl::ThreadImpl( Thread * owner ) :
+	mIsActive(false)
+{
+	mIsActive = pthread_create( &mThread, NULL, &ThreadImpl::entryPoint, owner ) == 0;
+
+	if ( !mIsActive )
+	{
+		efDEBUG( "Failed to create thread\n" );
+	}
+}
+
+void ThreadImpl::wait()
+{
+	// Wait for the thread to finish, no timeout
+	if ( mIsActive )
+	{
+		assert( pthread_equal( pthread_self(), mThread ) == 0 );
+
+		pthread_join( mThread, NULL );
+
+		mIsActive = false; // Reset the thread state
+	}
+}
+
+void ThreadImpl::terminate()
+{
+	if ( mIsActive )
+	{
+		#if !defined( __ANDROID__ ) && !defined( ANDROID )
+			pthread_cancel( mThread );
+		#else
+			pthread_kill( mThread , SIGUSR1 );
+		#endif
+
+		mIsActive = false;
+	}
+}
+
+void * ThreadImpl::entryPoint( void * userData )
+{
+	// The Thread instance is stored in the user data
+	Thread * owner = static_cast<Thread*>( userData );
+
+	// Tell the thread to handle cancel requests immediatly
+	#ifdef PTHREAD_CANCEL_ASYNCHRONOUS
+		pthread_setcanceltype( PTHREAD_CANCEL_ASYNCHRONOUS, NULL );
+	#endif
+
+	// Forward to the owner
+	owner->run();
+
+	return NULL;
+}
+
+}}
+
+#endif
diff --git a/dep/efsw/src/efsw/platform/posix/ThreadImpl.hpp b/dep/efsw/src/efsw/platform/posix/ThreadImpl.hpp
new file mode 100644
index 00000000000..be6dc1b3e58
--- /dev/null
+++ b/dep/efsw/src/efsw/platform/posix/ThreadImpl.hpp
@@ -0,0 +1,35 @@
+#ifndef EFSW_THREADIMPLPOSIX_HPP
+#define EFSW_THREADIMPLPOSIX_HPP
+
+#include <efsw/base.hpp>
+
+#if defined( EFSW_PLATFORM_POSIX )
+
+#include <pthread.h>
+
+namespace efsw {
+
+class Thread;
+
+namespace Platform {
+
+class ThreadImpl
+{
+	public:
+		ThreadImpl( Thread * owner );
+		
+		void wait();
+		
+		void terminate();
+	protected:
+		static void *	entryPoint( void* userData );
+
+		pthread_t		mThread;
+		bool			mIsActive;
+};
+
+}}
+
+#endif
+
+#endif
diff --git a/dep/efsw/src/efsw/platform/win/FileSystemImpl.cpp b/dep/efsw/src/efsw/platform/win/FileSystemImpl.cpp
new file mode 100644
index 00000000000..376a474e662
--- /dev/null
+++ b/dep/efsw/src/efsw/platform/win/FileSystemImpl.cpp
@@ -0,0 +1,89 @@
+#include <efsw/platform/win/FileSystemImpl.hpp>
+
+#if EFSW_PLATFORM == EFSW_PLATFORM_WIN32
+
+#ifndef WIN32_LEAN_AND_MEAN
+	#define WIN32_LEAN_AND_MEAN
+#endif
+#include <windows.h>
+
+#ifndef EFSW_COMPILER_MSVC
+#include <dirent.h>
+#endif
+
+namespace efsw { namespace Platform {
+
+FileInfoMap FileSystem::filesInfoFromPath( const std::string& path )
+{
+	FileInfoMap files;
+
+	String tpath( path );
+
+	if ( tpath[ tpath.size() - 1 ] == '/' || tpath[ tpath.size() - 1 ] == '\\' )
+	{
+		tpath += "*";
+	}
+	else
+	{
+		tpath += "\\*";
+	}
+
+	WIN32_FIND_DATAW findFileData;
+	HANDLE hFind = FindFirstFileW( (LPCWSTR)tpath.toWideString().c_str(), &findFileData );
+
+	if( hFind != INVALID_HANDLE_VALUE )
+	{
+		std::string name( String( findFileData.cFileName ).toUtf8() );
+		std::string fpath( path + name );
+
+		if ( name != "." && name != ".." )
+		{
+			files[ name ] = FileInfo( fpath );
+		}
+
+		while( FindNextFileW( hFind, &findFileData ) )
+		{
+			name = String( findFileData.cFileName ).toUtf8();
+			fpath = path + name;
+
+			if ( name != "." && name != ".." )
+			{
+				files[ name ] = FileInfo( fpath );
+			}
+		}
+
+		FindClose( hFind );
+	}
+
+	return files;
+}
+
+char FileSystem::getOSSlash()
+{
+	return '\\';
+}
+
+bool FileSystem::isDirectory( const std::string& path )
+{
+	return 0 != ( GetFileAttributesW( String( path ).toWideString().c_str() ) & FILE_ATTRIBUTE_DIRECTORY );
+}
+
+bool FileSystem::isRemoteFS( const std::string& directory )
+{
+	if ((directory[0] == '\\' || directory[0] == '/')  &&
+		(directory[1] == '\\' || directory[1] == '/'))
+	{
+		return true;
+	}
+
+	if ( directory.size() >= 3 )
+	{
+		return 4 == GetDriveTypeA( directory.substr( 0, 3 ).c_str() );
+	}
+
+	return false;
+}
+
+}}
+
+#endif
diff --git a/dep/efsw/src/efsw/platform/win/FileSystemImpl.hpp b/dep/efsw/src/efsw/platform/win/FileSystemImpl.hpp
new file mode 100644
index 00000000000..597edc4a647
--- /dev/null
+++ b/dep/efsw/src/efsw/platform/win/FileSystemImpl.hpp
@@ -0,0 +1,28 @@
+#ifndef EFSW_FILESYSTEMIMPLWIN_HPP
+#define EFSW_FILESYSTEMIMPLWIN_HPP
+
+#include <efsw/base.hpp>
+#include <efsw/String.hpp>
+#include <efsw/FileInfo.hpp>
+
+#if EFSW_PLATFORM == EFSW_PLATFORM_WIN32
+
+namespace efsw { namespace Platform {
+
+class FileSystem
+{
+	public:
+		static FileInfoMap filesInfoFromPath( const std::string& path );
+
+		static char getOSSlash();
+
+		static bool isDirectory( const std::string& path );
+
+		static bool isRemoteFS( const std::string& directory );
+};
+
+}}
+
+#endif
+
+#endif
diff --git a/dep/efsw/src/efsw/platform/win/MutexImpl.cpp b/dep/efsw/src/efsw/platform/win/MutexImpl.cpp
new file mode 100644
index 00000000000..0c8c36d8b39
--- /dev/null
+++ b/dep/efsw/src/efsw/platform/win/MutexImpl.cpp
@@ -0,0 +1,29 @@
+#include <efsw/platform/win/MutexImpl.hpp>
+
+#if EFSW_PLATFORM == EFSW_PLATFORM_WIN32
+
+namespace efsw { namespace Platform {
+
+MutexImpl::MutexImpl()
+{
+	InitializeCriticalSection(&mMutex);
+}
+
+MutexImpl::~MutexImpl()
+{
+	DeleteCriticalSection(&mMutex);
+}
+
+void MutexImpl::lock()
+{
+	EnterCriticalSection(&mMutex);
+}
+
+void MutexImpl::unlock()
+{
+	LeaveCriticalSection(&mMutex);
+}
+
+}}
+
+#endif
diff --git a/dep/efsw/src/efsw/platform/win/MutexImpl.hpp b/dep/efsw/src/efsw/platform/win/MutexImpl.hpp
new file mode 100644
index 00000000000..da1e20c5fa9
--- /dev/null
+++ b/dep/efsw/src/efsw/platform/win/MutexImpl.hpp
@@ -0,0 +1,34 @@
+#ifndef EFSW_MUTEXIMPLWIN_HPP
+#define EFSW_MUTEXIMPLWIN_HPP
+
+#include <efsw/base.hpp>
+
+#if EFSW_PLATFORM == EFSW_PLATFORM_WIN32
+
+#ifndef WIN32_LEAN_AND_MEAN
+	#define WIN32_LEAN_AND_MEAN
+#endif
+#include <windows.h>
+
+namespace efsw { namespace Platform {
+
+class MutexImpl
+{
+	public:
+		MutexImpl();
+
+		~MutexImpl();
+
+		void lock();
+
+		void unlock();
+	private:
+		CRITICAL_SECTION mMutex;
+};
+
+}}
+
+#endif
+
+#endif
+ 
diff --git a/dep/efsw/src/efsw/platform/win/SystemImpl.cpp b/dep/efsw/src/efsw/platform/win/SystemImpl.cpp
new file mode 100644
index 00000000000..ddbe1e5c45c
--- /dev/null
+++ b/dep/efsw/src/efsw/platform/win/SystemImpl.cpp
@@ -0,0 +1,50 @@
+#include <efsw/platform/win/SystemImpl.hpp>
+#include <efsw/String.hpp>
+
+#if EFSW_PLATFORM == EFSW_PLATFORM_WIN32
+
+#ifndef WIN32_LEAN_AND_MEAN
+	#define WIN32_LEAN_AND_MEAN
+#endif
+#include <windows.h>
+#include <cstdlib>
+
+namespace efsw { namespace Platform {
+
+void System::sleep( const unsigned long& ms )
+{
+	::Sleep( ms );
+}
+
+std::string System::getProcessPath()
+{
+	// Get path to executable:
+	WCHAR szDrive[_MAX_DRIVE];
+	WCHAR szDir[_MAX_DIR];
+	WCHAR szFilename[_MAX_DIR];
+	WCHAR szExt[_MAX_DIR];
+	std::wstring dllName( _MAX_DIR, 0 );
+
+	GetModuleFileNameW(0, &dllName[0], _MAX_PATH);
+
+	#ifdef EFSW_COMPILER_MSVC
+	_wsplitpath_s( dllName.c_str(), szDrive, _MAX_DRIVE, szDir, _MAX_DIR, szFilename, _MAX_DIR, szExt, _MAX_DIR );
+	#else
+	_wsplitpath( dllName.c_str(), szDrive, szDir, szFilename, szExt);
+	#endif
+
+	return String( szDrive ).toUtf8() + String( szDir ).toUtf8();
+}
+
+void System::maxFD()
+{
+}
+
+Uint64 System::getMaxFD()
+{	// Number of ReadDirectory per thread
+	return 60;
+}
+
+}}
+
+#endif
diff --git a/dep/efsw/src/efsw/platform/win/SystemImpl.hpp b/dep/efsw/src/efsw/platform/win/SystemImpl.hpp
new file mode 100644
index 00000000000..2f785e3565c
--- /dev/null
+++ b/dep/efsw/src/efsw/platform/win/SystemImpl.hpp
@@ -0,0 +1,26 @@
+#ifndef EFSW_SYSTEMIMPLWIN_HPP
+#define EFSW_SYSTEMIMPLWIN_HPP
+
+#include <efsw/base.hpp>
+
+#if EFSW_PLATFORM == EFSW_PLATFORM_WIN32
+
+namespace efsw { namespace Platform {
+
+class System
+{
+	public:
+		static void sleep( const unsigned long& ms );
+
+		static std::string getProcessPath();
+
+		static void maxFD();
+
+		static Uint64 getMaxFD();
+};
+
+}}
+
+#endif
+
+#endif
diff --git a/dep/efsw/src/efsw/platform/win/ThreadImpl.cpp b/dep/efsw/src/efsw/platform/win/ThreadImpl.cpp
new file mode 100644
index 00000000000..2fa30f30060
--- /dev/null
+++ b/dep/efsw/src/efsw/platform/win/ThreadImpl.cpp
@@ -0,0 +1,56 @@
+#include <efsw/platform/win/ThreadImpl.hpp>
+#include <efsw/Thread.hpp>
+#include <assert.h>
+
+#if EFSW_PLATFORM == EFSW_PLATFORM_WIN32
+
+#include <efsw/Debug.hpp>
+
+namespace efsw { namespace Platform {
+
+ThreadImpl::ThreadImpl( Thread *owner )
+{
+	mThread = reinterpret_cast<HANDLE>( _beginthreadex( NULL, 0, &ThreadImpl::entryPoint, owner, 0, &mThreadId ) );
+
+	if ( !mThread )
+	{
+		efDEBUG( "Failed to create thread\n" );
+	}
+}
+
+void ThreadImpl::wait()
+{
+	// Wait for the thread to finish, no timeout
+	if ( mThread )
+	{
+		assert( mThreadId != GetCurrentThreadId() ); // A thread cannot wait for itself!
+
+		WaitForSingleObject( mThread, INFINITE );
+	}
+}
+
+void ThreadImpl::terminate()
+{
+	if ( mThread )
+	{
+		TerminateThread( mThread, 0 );
+	}
+}
+
+unsigned int __stdcall ThreadImpl::entryPoint( void * userData )
+{
+	// The Thread instance is stored in the user data
+	Thread * owner = static_cast<Thread*>( userData );
+
+	// Forward to the owner
+	owner->run();
+
+	// Optional, but it is cleaner
+	_endthreadex(0);
+
+	return 0;
+}
+
+}}
+
+#endif
diff --git a/dep/efsw/src/efsw/platform/win/ThreadImpl.hpp b/dep/efsw/src/efsw/platform/win/ThreadImpl.hpp
new file mode 100644
index 00000000000..506c659c675
--- /dev/null
+++ b/dep/efsw/src/efsw/platform/win/ThreadImpl.hpp
@@ -0,0 +1,39 @@
+#ifndef EFSW_THREADIMPLWIN_HPP
+#define EFSW_THREADIMPLWIN_HPP
+
+#include <efsw/base.hpp>
+
+#if EFSW_PLATFORM == EFSW_PLATFORM_WIN32
+
+#ifndef WIN32_LEAN_AND_MEAN
+	#define WIN32_LEAN_AND_MEAN
+#endif
+#include <windows.h>
+#include <process.h>
+
+namespace efsw {
+
+class Thread;
+
+namespace Platform {
+
+class ThreadImpl
+{
+	public:
+		ThreadImpl( Thread * owner );
+		
+		void wait();
+		
+		void terminate();
+	protected:
+		static unsigned int __stdcall entryPoint(void* userData);
+
+		HANDLE			mThread;
+		unsigned int	mThreadId;
+};
+
+}}
+
+#endif
+
+#endif
diff --git a/dep/efsw/src/efsw/sophist.h b/dep/efsw/src/efsw/sophist.h
new file mode 100644
index 00000000000..3a645040e2f
--- /dev/null
+++ b/dep/efsw/src/efsw/sophist.h
@@ -0,0 +1,147 @@
+/* sophist.h - 0.3 - public domain - Sean Barrett 2010
+** Knowledge drawn from Brian Hook's posh.h and http://predef.sourceforge.net
+** Sophist provides portable types; you typedef/#define them to your own names
+**
+** defines:
+**   - SOPHIST_endian    - either SOPHIST_little_endian or SOPHIST_big_endian
+**   - SOPHIST_has_64    - either 0 or 1; if 0, int64 types aren't defined
+**   - SOPHIST_pointer64 - either 0 or 1; if 1, pointer is 64-bit
+**
+**   - SOPHIST_intptr, SOPHIST_uintptr - integer same size as pointer
+**   - SOPHIST_int8,  SOPHIST_uint8,  SOPHIST_int16, SOPHIST_uint16
+**   - SOPHIST_int32, SOPHIST_uint32, SOPHIST_int64, SOPHIST_uint64
+**   - SOPHIST_int64_constant(number)      - macros for creating 64-bit
+**   - SOPHIST_uint64_constant(number)       integer constants
+**   - SOPHIST_printf_format64 - string for printf format for int64
+*/
+
+#ifndef __INCLUDE_SOPHIST_H__
+#define __INCLUDE_SOPHIST_H__
+
+#define SOPHIST_compiletime_assert(name,val) \
+            typedef int SOPHIST__assert##name[(val) ? 1 : -1]
+
+/* define a couple synthetic rules to make code more readable */
+#if (defined(__sparc__) || defined(__sparc)) && \
+    (defined(__arch64__) || defined(__sparcv9) || defined(__sparc_v9__))
+  #define SOPHIST_sparc64
+#endif
+
+#if (defined(linux) || defined(__linux__)) &&   \
+    (defined(__alpha)||defined(__alpha__)||defined(__x86_64__)||defined(_M_X64))
+  #define SOPHIST_linux64
+#endif
+
+/* basic types */
+typedef   signed  char SOPHIST_int8;
+typedef unsigned  char SOPHIST_uint8;
+
+typedef   signed short SOPHIST_int16;
+typedef unsigned short SOPHIST_uint16;
+
+#ifdef __palmos__
+  typedef   signed long SOPHIST_int32;
+  typedef unsigned long SOPHIST_uint32;
+#else
+  typedef   signed  int SOPHIST_int32;
+  typedef unsigned  int SOPHIST_uint32;
+#endif
+
+#ifndef SOPHIST_NO_64
+   #if defined(_MSC_VER) || defined(__WATCOMC__) || defined(__BORLANDC__)     \
+       || (defined(__alpha) && defined(__DECC))
+
+     typedef   signed __int64 SOPHIST_int64;
+     typedef unsigned __int64 SOPHIST_uint64;
+     #define SOPHIST_has_64              1
+     #define SOPHIST_int64_constant(x)   (x##i64)
+     #define SOPHIST_uint64_constant(x)  (x##ui64)
+     #define SOPHIST_printf_format64     "I64"
+
+   #elif defined(__LP64__) || defined(__powerpc64__) || defined(SOPHIST_sparc64)
+
+     typedef   signed long    SOPHIST_int64;
+     typedef unsigned long    SOPHIST_uint64;
+
+     #define SOPHIST_has_64              1
+     #define SOPHIST_int64_constant(x)   ((SOPHIST_int64) x)
+     #define SOPHIST_uint64_constant(x)  ((SOPHIST_uint64) x)
+     #define SOPHIST_printf_format64     "l"
+
+   #elif defined(_LONG_LONG) || defined(__SUNPRO_C) || defined(__SUNPRO_CC)  \
+       || defined(__GNUC__)  || defined(__MWERKS__) || defined(__APPLE_CC__) \
+       || defined(sgi)       || defined (__sgi)     || defined(__sgi__)      \
+       || defined(_CRAYC)
+
+     typedef   signed long long SOPHIST_int64;
+     typedef unsigned long long SOPHIST_uint64;
+
+     #define SOPHIST_has_64              1
+     #define SOPHIST_int64_constant(x)   (x##LL)
+     #define SOPHIST_uint64_constant(x)  (x##ULL)
+     #define SOPHIST_printf_format64     "ll"
+   #endif
+#endif
+
+#ifndef SOPHIST_has_64
+#define SOPHIST_has_64 0
+#endif
+
+SOPHIST_compiletime_assert( int8 , sizeof(SOPHIST_int8 ) == 1);
+SOPHIST_compiletime_assert(uint16, sizeof(SOPHIST_int16) == 2);
+SOPHIST_compiletime_assert( int32, sizeof(SOPHIST_int32 ) == 4);
+SOPHIST_compiletime_assert(uint32, sizeof(SOPHIST_uint32) == 4);
+
+#if SOPHIST_has_64
+  SOPHIST_compiletime_assert( int64, sizeof(SOPHIST_int64 ) == 8);
+  SOPHIST_compiletime_assert(uint64, sizeof(SOPHIST_uint64) == 8);
+#endif
+
+/* determine whether pointers are 64-bit */
+
+#if    defined(SOPHIST_linux64) || defined(SOPHIST_sparc64)      \
+    || defined(__osf__) || (defined(_WIN64) && !defined(_XBOX))  \
+    || defined(__64BIT__)                                        \
+    || defined(__LP64)  || defined(__LP64__) || defined(_LP64)   \
+    || defined(_ADDR64) || defined(_CRAYC)                       \
+
+  #define SOPHIST_pointer64 1
+
+  SOPHIST_compiletime_assert(pointer64, sizeof(void*) == 8);  
+
+  typedef SOPHIST_int64  SOPHIST_intptr;
+  typedef SOPHIST_uint64 SOPHIST_uintptr;
+#else
+
+  #define SOPHIST_pointer64 0
+
+  SOPHIST_compiletime_assert(pointer64, sizeof(void*) <= 4);  
+  
+  /* do we care about pointers that are only 16-bit? */
+  typedef SOPHIST_int32  SOPHIST_intptr;
+  typedef SOPHIST_uint32 SOPHIST_uintptr;
+
+#endif 
+
+SOPHIST_compiletime_assert(intptr, sizeof(SOPHIST_intptr) == sizeof(char *));
+
+/* enumerate known little endian cases; fallback to big-endian */
+
+#define SOPHIST_little_endian 1
+#define SOPHIST_big_endian 2
+
+#if    defined(__386__) || defined(i386)    || defined(__i386__)  \
+    || defined(__X86)   || defined(_M_IX86)                       \
+    || defined(_M_X64)  || defined(__x86_64__)                    \
+    || defined(alpha)   || defined(__alpha) || defined(__alpha__) \
+    || defined(_M_ALPHA)                                          \
+    || defined(ARM)     || defined(_ARM)    || defined(__arm__)   \
+    || defined(WIN32)   || defined(_WIN32)  || defined(__WIN32__) \
+    || defined(_WIN32_WCE) || defined(__NT__)                     \
+    || defined(__MIPSEL__)
+  #define SOPHIST_endian  SOPHIST_little_endian
+#else
+  #define SOPHIST_endian  SOPHIST_big_endian
+#endif
+
+#endif /* __INCLUDE_SOPHIST_H__ */
diff --git a/dep/efsw/src/test/efsw-test.cpp b/dep/efsw/src/test/efsw-test.cpp
new file mode 100644
index 00000000000..a49e3414fcc
--- /dev/null
+++ b/dep/efsw/src/test/efsw-test.cpp
@@ -0,0 +1,151 @@
+#include <efsw/efsw.hpp>
+#include <efsw/System.hpp>
+#include <efsw/FileSystem.hpp>
+#include <signal.h>
+#include <iostream>
+
+bool STOP = false;
+
+void sigend(int signal)
+{
+	std::cout << std::endl << "Bye bye" << std::endl;
+	STOP = true;
+}
+
+/// Processes a file action
+class UpdateListener : public efsw::FileWatchListener
+{
+	public:
+		UpdateListener() {}
+
+		std::string getActionName( efsw::Action action )
+		{
+			switch ( action )
+			{
+				case efsw::Actions::Add:		return "Add";
+				case efsw::Actions::Modified:	return "Modified";
+				case efsw::Actions::Delete:		return "Delete";
+				case efsw::Actions::Moved:		return "Moved";
+				default:						return "Bad Action";
+			}
+		}
+
+		void handleFileAction( efsw::WatchID watchid, const std::string& dir, const std::string& filename, efsw::Action action, std::string oldFilename = ""  )
+		{
+			std::cout << "DIR (" << dir + ") FILE (" + ( oldFilename.empty() ? "" : "from file " + oldFilename + " to " ) + filename + ") has event " << getActionName( action ) << std::endl;
+		}
+};
+
+efsw::WatchID handleWatchID( efsw::WatchID watchid )
+{
+	switch ( watchid )
+	{
+		case efsw::Errors::FileNotFound:
+		case efsw::Errors::FileRepeated:
+		case efsw::Errors::FileOutOfScope:
+		case efsw::Errors::FileRemote:
+		case efsw::Errors::Unspecified:
+		{
+			std::cout << efsw::Errors::Log::getLastErrorLog().c_str() << std::endl;
+			break;
+		}
+		default:
+		{
+			std::cout << "Added WatchID: " << watchid << std::endl;
+		}
+	}
+
+	return watchid;
+}
+
+int main(int argc, char **argv)
+{
+	signal( SIGABRT	,	sigend );
+	signal( SIGINT	,	sigend );
+	signal( SIGTERM	,	sigend );
+
+	std::cout << "Press ^C to exit demo" << std::endl;
+
+	bool commonTest = true;
+	bool useGeneric = false;
+	std::string path;
+
+	if ( argc >= 2 )
+	{
+		path = std::string( argv[1] );
+
+		if ( efsw::FileSystem::isDirectory( path ) )
+		{
+			commonTest	= false;
+		}
+
+		if ( argc >= 3 )
+		{
+			if ( std::string( argv[2] ) == "true" )
+			{
+				useGeneric = true;
+			}
+		}
+	}
+
+	UpdateListener * ul = new UpdateListener();
+
+	/// create the file watcher object
+	efsw::FileWatcher fileWatcher( useGeneric );
+
+	fileWatcher.followSymlinks( false );
+	fileWatcher.allowOutOfScopeLinks( false );
+
+	if ( commonTest )
+	{
+		std::string CurPath( efsw::System::getProcessPath() );
+
+		std::cout << "CurPath: " << CurPath.c_str() << std::endl;
+
+		/// add a watch to the system
+		handleWatchID( fileWatcher.addWatch( CurPath + "test", ul, true ) );
+
+		/// starts watching
+		fileWatcher.watch();
+
+		/// adds another watch after started watching...
+		efsw::System::sleep( 100 );
+
+		efsw::WatchID watchID = handleWatchID( fileWatcher.addWatch( CurPath + "test2", ul, true ) );
+
+		/// delete the watch
+		if ( watchID > 0 )
+		{
+			efsw::System::sleep( 1000 );
+			fileWatcher.removeWatch( watchID );
+		}
+	}
+	else
+	{
+		efsw::WatchID err;
+
+		if ( ( err = fileWatcher.addWatch( path, ul, true ) ) > 0 )
+		{
+			fileWatcher.watch();
+
+			std::cout << "Watching directory: " << path.c_str() << std::endl;
+
+			if ( useGeneric )
+			{
+				std::cout << "Using generic backend watcher" << std::endl;
+			}
+		}
+		else
+		{
+			std::cout << "Error trying to watch directory: " << path.c_str() << std::endl;
+			std::cout << efsw::Errors::Log::getLastErrorLog().c_str() << std::endl;
+		}
+	}
+
+	while( !STOP )
+	{
+		efsw::System::sleep( 100 );
+	}
+
+	return 0;
+}
diff --git a/src/server/game/CMakeLists.txt b/src/server/game/CMakeLists.txt
index 06d4c403909..df0d2b7cc2a 100644
--- a/src/server/game/CMakeLists.txt
+++ b/src/server/game/CMakeLists.txt
@@ -43,7 +43,9 @@ target_include_directories(game
 target_link_libraries(game
   PUBLIC
     shared
-    Detour)
+    Detour
+  PRIVATE
+    efsw)
 
 set_target_properties(game
     PROPERTIES