diff --git a/.gitignore b/.gitignore
index c18dd8d..6fe4324 100644
--- a/.gitignore
+++ b/.gitignore
@@ -1 +1,4 @@
 __pycache__/
+/build/
+/.vscode/
+.DS_Store
diff --git a/CMakeLists.txt b/CMakeLists.txt
index e9a699e..bf68ec0 100644
--- a/CMakeLists.txt
+++ b/CMakeLists.txt
@@ -1,5 +1,6 @@
 ######## Project settings
 cmake_minimum_required(VERSION 2.8.5)
+include(CheckCCompilerFlag)
 set (PACKAGE_NAME uchardet)
 project (${PACKAGE_NAME} CXX C)
 enable_testing()
@@ -19,13 +20,8 @@ set (
 	UCHARDET_VERSION
 	${UCHARDET_VERSION_MAJOR}.${UCHARDET_VERSION_MINOR}.${UCHARDET_VERSION_REVISION}${version_suffix}
 )
-
-######## Windows
-
-#if (WIN32)
-#	set(CMAKE_SHARED_LIBRARY_PREFIX ${CMAKE_INSTALL_PREFIX})
-#	set(CMAKE_STATIC_LIBRARY_PREFIX ${CMAKE_INSTALL_PREFIX})
-#endif (WIN32)
+set(CMAKE_CXX_STANDARD 11)
+set(CMAKE_CXX_STANDARD_REQUIRED ON)
 
 ######## Directory
 
@@ -33,13 +29,31 @@ include(GNUInstallDirs)
 
 ######## Configuration
 
-option(BUILD_BINARY "Build executable" ON)
-option(BUILD_SHARED_LIBS "Build shared library and link executable to it" ON)
+option(BUILD_BINARY "Build the CLI tool." ON)
+option(BUILD_SHARED_LIBS "Build shared library and link executable to it." ON)
+option(CHECK_SSE2 "Check and enable SSE2 extensions if supported. Disabling SSE on platforms which support it may decrease performances." ON)
+set(TARGET_ARCHITECTURE "" CACHE STRING "Target CPU architecture. It is autodetected if not specified.")
 
 if (BUILD_SHARED_LIBS)
 	option(BUILD_STATIC "Build static library" ON)
 endif (BUILD_SHARED_LIBS)
 
+if (TARGET_ARCHITECTURE STREQUAL "")
+    string(TOLOWER ${CMAKE_SYSTEM_PROCESSOR} TARGET_ARCHITECTURE)
+endif (TARGET_ARCHITECTURE STREQUAL "")
+
+if (TARGET_ARCHITECTURE MATCHES ".*(x86)|(amd).*")
+    CHECK_C_COMPILER_FLAG(-msse2 SUPPORTS_CFLAG_SSE2)
+    CHECK_C_COMPILER_FLAG(-mfpmath=sse SUPPORTS_CFLAG_SSE_MATH)
+    if (CHECK_SSE2 AND SUPPORTS_CFLAG_SSE2 AND SUPPORTS_CFLAG_SSE_MATH)
+        set(CMAKE_C_FLAGS "${CMAKE_C_FLAGS} -msse2 -mfpmath=sse")
+        set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -msse2 -mfpmath=sse")
+    else (CHECK_SSE2 AND SUPPORTS_CFLAG_SSE2 AND SUPPORTS_CFLAG_SSE_MATH)
+        set(CMAKE_C_FLAGS "${CMAKE_C_FLAGS} -ffloat-store")
+        set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -ffloat-store")
+    endif (CHECK_SSE2 AND SUPPORTS_CFLAG_SSE2 AND SUPPORTS_CFLAG_SSE_MATH)
+endif (TARGET_ARCHITECTURE MATCHES ".*(x86)|(amd).*")
+
 configure_file(
 	uchardet.pc.in
 	uchardet.pc
diff --git a/INSTALL b/INSTALL
index 60b8234..9b058aa 100644
--- a/INSTALL
+++ b/INSTALL
@@ -1,4 +1,4 @@
-# INSTALL uchardet
+# Building uchardet (generic)
 
 `uchardet` uses a typical cmake installation.
 
@@ -14,3 +14,13 @@ Alternatively, use `ccmake`, curses interface to `cmake`.
 * Install with `make install`.
 
 Read `README` for more details on uchardet.
+
+# Building uchardet on Windows
+
+The above procedure is generic, which means it should work on any platform.
+In particular, it works well on Linux.
+
+The procedure is the same on Windows, but if you want more details (for
+instance which tools to use in order to run CMake on Windows, compiler
+information, etc.), the following link may be useful:
+https://github.com/BYVoid/uchardet/issues/39#issuecomment-353873891
diff --git a/README.md b/README.md
index daaf52c..d845a02 100644
--- a/README.md
+++ b/README.md
@@ -1,5 +1,7 @@
 # uchardet
 
+Forked from [freedesktop/uchardet](https://github.com/freedesktop/uchardet)
+
 [uchardet](https://www.freedesktop.org/wiki/Software/uchardet/) is an encoding detector library, which takes a sequence of bytes in an unknown character encoding without any additional information, and attempts to determine the encoding of the text. Returned encoding names are [iconv](https://www.gnu.org/software/libiconv/)-compatible.
 
 uchardet started as a C language binding of the original C++ implementation of the universal charset detection library by Mozilla. It can now detect more charsets, and more reliably than the original implementation.
@@ -32,12 +34,12 @@ Techniques used by universalchardet are described at http://www.mozilla.org/proj
     * ISO-8859-16
     * Windows-1250
     * IBM852
-    * MAC-CENTRALEUROPE
+    * MacCentralEurope
   * Czech
     * Windows-1250
     * ISO-8859-2
     * IBM852
-    * MAC-CENTRALEUROPE
+    * MacCentralEurope
   * Danish
     * ISO-8859-1
     * ISO-8859-15
@@ -109,7 +111,7 @@ Techniques used by universalchardet are described at http://www.mozilla.org/proj
     * ISO-8859-16
     * Windows-1250
     * IBM852
-    * MAC-CENTRALEUROPE
+    * MacCentralEurope
   * Portuguese
     * ISO-8859-1
     * ISO-8859-9
@@ -131,13 +133,13 @@ Techniques used by universalchardet are described at http://www.mozilla.org/proj
     * Windows-1250
     * ISO-8859-2
     * IBM852
-    * MAC-CENTRALEUROPE
+    * MacCentralEurope
   * Slovene
     * ISO-8859-2
     * ISO-8859-16
     * Windows-1250
     * IBM852
-    * MAC-CENTRALEUROPE
+    * MacCentralEurope
   * Spanish
     * ISO-8859-1
     * ISO-8859-15
@@ -162,57 +164,19 @@ Techniques used by universalchardet are described at http://www.mozilla.org/proj
 
 ## Installation
 
-### Debian/Ubuntu/Mint
-
-    apt-get install uchardet libuchardet-dev
-
-### Mageia
-
-    urpmi libuchardet libuchardet-devel
-
-### Fedora
-
-    dnf install uchardet uchardet-devel
-
-### Mac
-
-    brew install uchardet
-
 ### Build from source
 
-Releases are available from:
-https://www.freedesktop.org/software/uchardet/releases/
-
 If you prefer a development version, clone the git repository:
 
-    git clone git://anongit.freedesktop.org/uchardet/uchardet
+    git clone https://github.com/PyYoshi/uchardet.git
 
-The source can be browsed at: https://cgit.freedesktop.org/uchardet/uchardet/
+The source can be browsed at: https://github.com/PyYoshi/uchardet
 
-    cmake .
+    mkdir build/ && cd build/
+    cmake ..
     make
     make install
 
-### Build with flatpak-builder
-
-Here is a working "module" section to include in your Flatpak's json manifest:
-
-```
-"modules": [
-    {
-        "name": "uchardet",
-        "buildsystem": "cmake",
-        "builddir": true,
-        "config-opts": [ "-DCMAKE_INSTALL_LIBDIR=lib" ],
-        "sources": [
-            {
-                ...
-            }
-        ]
-    }
-]
-```
-
 ## Usage
 
 ### Command Line
@@ -233,19 +197,7 @@ Options:
  ```
 ### Library
 
-See [uchardet.h](https://cgit.freedesktop.org/uchardet/uchardet/tree/src/uchardet.h)
-
-## Related Projects
-
-  * [python-chardet](https://github.com/chardet/chardet) Python port
-  * [ruby-rchardet](http://rubyforge.org/projects/chardet/) Ruby port
-  * [juniversalchardet](http://code.google.com/p/juniversalchardet/) Java port of universalchardet
-  * [jchardet](http://jchardet.sourceforge.net/) Java port of chardet
-  * [nuniversalchardet](http://code.google.com/p/nuniversalchardet/) C# port of universalchardet
-  * [nchardet](http://www.conceptdevelopment.net/Localization/NCharDet/) C# port of chardet
-  * [uchardet-enhanced](https://bitbucket.org/medoc/uchardet-enhanced) A fork of mozilla universalchardet
-  * [rust-uchardet](https://github.com/emk/rust-uchardet) Rust language binding of uchardet
-  * [libchardet](https://ftp.oops.org/pub/oops/libchardet/) Another C/C++ API wrapping Mozilla code.
+See [uchardet.h](https://github.com/PyYoshi/uchardet/blob/cchardet/src/uchardet.h)
 
 ## Licenses
 
diff --git a/script/BuildLangModelLogs/LangEstonianModel.log b/script/BuildLangModelLogs/LangEstonianModel.log
index f1095eb..f11652c 100644
--- a/script/BuildLangModelLogs/LangEstonianModel.log
+++ b/script/BuildLangModelLogs/LangEstonianModel.log
@@ -156,4 +156,4 @@ First 512 (typical positive ratio): 0.9972721312183132
 Next 512 (512-1024): 9.687262863877811e-05
 Rest: -5.204170427930421e-18
 
-- Processing end: 2016-09-26 23:47:54.561846
+- Processing end: 2016-09-26 23:47:54.561846
\ No newline at end of file
diff --git a/script/charsets/mac-centraleurope.py b/script/charsets/mac-centraleurope.py
index bf201e6..87cb9c3 100644
--- a/script/charsets/mac-centraleurope.py
+++ b/script/charsets/mac-centraleurope.py
@@ -40,7 +40,7 @@
 
 from codepoints import *
 
-name = 'MAC-CENTRALEUROPE'
+name = 'MacCentralEurope'
 aliases = []
 
 language = \
diff --git a/script/gen.sh b/script/gen.sh
new file mode 100755
index 0000000..27737b3
--- /dev/null
+++ b/script/gen.sh
@@ -0,0 +1,27 @@
+#!/bin/bash
+python ./BuildLangModel.py ar --max-page=100 --max-depth=4
+python ./BuildLangModel.py cs --max-page=100 --max-depth=4
+python ./BuildLangModel.py da --max-page=100 --max-depth=4
+python ./BuildLangModel.py de --max-page=100 --max-depth=4
+python ./BuildLangModel.py el --max-page=100 --max-depth=4
+python ./BuildLangModel.py eo --max-page=100 --max-depth=4
+python ./BuildLangModel.py es --max-page=100 --max-depth=4
+python ./BuildLangModel.py et --max-page=100 --max-depth=4
+python ./BuildLangModel.py fi --max-page=100 --max-depth=4
+python ./BuildLangModel.py fr --max-page=100 --max-depth=4
+python ./BuildLangModel.py ga --max-page=100 --max-depth=4
+python ./BuildLangModel.py hr --max-page=100 --max-depth=4
+python ./BuildLangModel.py hu --max-page=100 --max-depth=4
+python ./BuildLangModel.py it --max-page=100 --max-depth=4
+python ./BuildLangModel.py lt --max-page=100 --max-depth=4
+python ./BuildLangModel.py lv --max-page=100 --max-depth=4
+python ./BuildLangModel.py mt --max-page=100 --max-depth=4
+python ./BuildLangModel.py pl --max-page=100 --max-depth=4
+python ./BuildLangModel.py pt --max-page=100 --max-depth=4
+python ./BuildLangModel.py ro --max-page=100 --max-depth=4
+python ./BuildLangModel.py sk --max-page=100 --max-depth=4
+python ./BuildLangModel.py sl --max-page=100 --max-depth=4
+python ./BuildLangModel.py sv --max-page=100 --max-depth=4
+python ./BuildLangModel.py th --max-page=100 --max-depth=4
+python ./BuildLangModel.py tr --max-page=100 --max-depth=4
+python ./BuildLangModel.py vi --max-page=100 --max-depth=4
diff --git a/script/langs/cs.py b/script/langs/cs.py
index 009ceec..f6865ed 100644
--- a/script/langs/cs.py
+++ b/script/langs/cs.py
@@ -50,7 +50,7 @@
 # ASCII characters are also used in French.
 use_ascii = True
 # The charsets we want to support and create data for.
-charsets = ['ISO-8859-2', 'Windows-1250', 'IBM852', 'MAC-CENTRALEUROPE']
+charsets = ['ISO-8859-2', 'Windows-1250', 'IBM852', 'MacCentralEurope']
 
 ## Optional Properties ##
 
diff --git a/script/langs/hr.py b/script/langs/hr.py
index 3b0dea5..12d9388 100644
--- a/script/langs/hr.py
+++ b/script/langs/hr.py
@@ -46,7 +46,7 @@
 code = 'hr'
 use_ascii = True
 charsets = ['ISO-8859-2', 'ISO-8859-13', 'ISO-8859-16',
-            'Windows-1250', 'IBM852', 'MAC-CENTRALEUROPE']
+            'Windows-1250', 'IBM852', 'MacCentralEurope']
 
 ## Optional Properties ##
 
diff --git a/script/langs/pl.py b/script/langs/pl.py
index 54485a1..7dd1ffb 100644
--- a/script/langs/pl.py
+++ b/script/langs/pl.py
@@ -51,7 +51,7 @@
 use_ascii = True
 # The charsets we want to support and create data for.
 charsets = ['ISO-8859-2', 'ISO-8859-13', 'ISO-8859-16',
-            'Windows-1250', 'IBM852', 'MAC-CENTRALEUROPE']
+            'Windows-1250', 'IBM852', 'MacCentralEurope']
 
 ## Optional Properties ##
 
diff --git a/script/langs/sk.py b/script/langs/sk.py
index c670839..d1e4689 100644
--- a/script/langs/sk.py
+++ b/script/langs/sk.py
@@ -50,7 +50,7 @@
 # ASCII characters are also used in French.
 use_ascii = True
 # The charsets we want to support and create data for.
-charsets = ['ISO-8859-2', 'Windows-1250', 'IBM852', 'MAC-CENTRALEUROPE']
+charsets = ['ISO-8859-2', 'Windows-1250', 'IBM852', 'MacCentralEurope']
 
 ## Optional Properties ##
 
diff --git a/script/langs/sl.py b/script/langs/sl.py
index bf02bf8..5ffe69d 100644
--- a/script/langs/sl.py
+++ b/script/langs/sl.py
@@ -46,7 +46,7 @@
 code = 'sl'
 use_ascii = True
 charsets = ['ISO-8859-2', 'ISO-8859-16',
-            'Windows-1250', 'IBM852', 'MAC-CENTRALEUROPE']
+            'Windows-1250', 'IBM852', 'MacCentralEurope']
 
 ## Optional Properties ##
 
diff --git a/script/requirements.txt b/script/requirements.txt
new file mode 100644
index 0000000..840e6bc
--- /dev/null
+++ b/script/requirements.txt
@@ -0,0 +1,2 @@
+requests==2.31.0
+wikipedia==1.4.0
\ No newline at end of file
diff --git a/src/CMakeLists.txt b/src/CMakeLists.txt
index 952b594..a1db01a 100644
--- a/src/CMakeLists.txt
+++ b/src/CMakeLists.txt
@@ -72,7 +72,13 @@ if (CMAKE_BUILD_TYPE MATCHES Debug)
 	add_definitions(
 		-O0
 		-g3
+		-fsanitize=address
 	)
+	set(CMAKE_CXX_FLAGS "-fsanitize=address")
+	set(CMAKE_C_FLAGS "-fsanitize=address")
+	set(CMAKE_EXE_LINKER_FLAGS "-fsanitize=address")
+	set(CMAKE_MODULE_LINKER_FLAGS "-fsanitize=address")
+	set(CMAKE_SHARED_LINKER_FLAGS "-fsanitize=address")
 endif (CMAKE_BUILD_TYPE MATCHES Debug)
 
 add_library(
@@ -152,4 +158,9 @@ include(symbols.cmake)
 
 if (BUILD_BINARY)
 	add_subdirectory(tools)
+	add_subdirectory(fuzzing)
 endif (BUILD_BINARY)
+
+if (BUILD_FUZZER)
+	add_subdirectory(fuzzing)
+endif (BUILD_FUZZER)
diff --git a/src/EUCTWFreq.tab b/src/EUCTWFreq.tab
index 3a189ae..39f9d97 100644
--- a/src/EUCTWFreq.tab
+++ b/src/EUCTWFreq.tab
@@ -57,7 +57,7 @@
 #define EUCTW_TYPICAL_DISTRIBUTION_RATIO (float)0.75
 
 //Char to FreqOrder table , 
-#define EUCTW_TABLE_SIZE  8102
+#define EUCTW_TABLE_SIZE  5376
 
 static const PRInt16 EUCTWCharToFreqOrder[] =
 {
diff --git a/src/LangModels/LangCroatianModel.cpp b/src/LangModels/LangCroatianModel.cpp
index 58f882e..7151522 100644
--- a/src/LangModels/LangCroatianModel.cpp
+++ b/src/LangModels/LangCroatianModel.cpp
@@ -124,7 +124,7 @@ static const unsigned char Iso_8859_16_CharToOrderMap[] =
 };
 /*X0  X1  X2  X3  X4  X5  X6  X7  X8  X9  XA  XB  XC  XD  XE  XF */
 
-static const unsigned char Mac_Centraleurope_CharToOrderMap[] =
+static const unsigned char Maccentraleurope_CharToOrderMap[] =
 {
   CTR,CTR,CTR,CTR,CTR,CTR,CTR,CTR,CTR,CTR,RET,CTR,CTR,RET,CTR,CTR, /* 0X */
   CTR,CTR,CTR,CTR,CTR,CTR,CTR,CTR,CTR,CTR,CTR,CTR,CTR,CTR,CTR,CTR, /* 1X */
@@ -261,14 +261,14 @@ const SequenceModel Iso_8859_16CroatianModel =
   "ISO-8859-16"
 };
 
-const SequenceModel Mac_CentraleuropeCroatianModel =
+const SequenceModel MaccentraleuropeCroatianModel =
 {
-  Mac_Centraleurope_CharToOrderMap,
+  Maccentraleurope_CharToOrderMap,
   CroatianLangModel,
   31,
   (float)0.9989731099787131,
   PR_TRUE,
-  "MAC-CENTRALEUROPE"
+  "MacCentralEurope"
 };
 
 const SequenceModel Iso_8859_13CroatianModel =
diff --git a/src/LangModels/LangCzechModel.cpp b/src/LangModels/LangCzechModel.cpp
index 4e74ed2..0b8b5bd 100644
--- a/src/LangModels/LangCzechModel.cpp
+++ b/src/LangModels/LangCzechModel.cpp
@@ -82,7 +82,7 @@ static const unsigned char Windows_1250_CharToOrderMap[] =
 };
 /*X0  X1  X2  X3  X4  X5  X6  X7  X8  X9  XA  XB  XC  XD  XE  XF */
 
-static const unsigned char Mac_Centraleurope_CharToOrderMap[] =
+static const unsigned char Maccentraleurope_CharToOrderMap[] =
 {
   CTR,CTR,CTR,CTR,CTR,CTR,CTR,CTR,CTR,CTR,RET,CTR,CTR,RET,CTR,CTR, /* 0X */
   CTR,CTR,CTR,CTR,CTR,CTR,CTR,CTR,CTR,CTR,CTR,CTR,CTR,CTR,CTR,CTR, /* 1X */
@@ -250,14 +250,14 @@ const SequenceModel Windows_1250CzechModel =
   "WINDOWS-1250"
 };
 
-const SequenceModel Mac_CentraleuropeCzechModel =
+const SequenceModel MaccentraleuropeCzechModel =
 {
-  Mac_Centraleurope_CharToOrderMap,
+  Maccentraleurope_CharToOrderMap,
   CzechLangModel,
   41,
   (float)0.9786035192432675,
   PR_TRUE,
-  "MAC-CENTRALEUROPE"
+  "MacCentralEurope"
 };
 
 const SequenceModel Ibm852CzechModel =
@@ -278,4 +278,4 @@ const SequenceModel Iso_8859_2CzechModel =
   (float)0.9786035192432675,
   PR_TRUE,
   "ISO-8859-2"
-};
\ No newline at end of file
+};
diff --git a/src/LangModels/LangEstonianModel.cpp b/src/LangModels/LangEstonianModel.cpp
index c5fa9b3..362f280 100644
--- a/src/LangModels/LangEstonianModel.cpp
+++ b/src/LangModels/LangEstonianModel.cpp
@@ -260,4 +260,4 @@ const SequenceModel Windows_1257EstonianModel =
   (float)0.9972721312183132,
   PR_TRUE,
   "WINDOWS-1257"
-};
+};
\ No newline at end of file
diff --git a/src/LangModels/LangPolishModel.cpp b/src/LangModels/LangPolishModel.cpp
index e7379d1..01370c9 100644
--- a/src/LangModels/LangPolishModel.cpp
+++ b/src/LangModels/LangPolishModel.cpp
@@ -124,7 +124,7 @@ static const unsigned char Iso_8859_2_CharToOrderMap[] =
 };
 /*X0  X1  X2  X3  X4  X5  X6  X7  X8  X9  XA  XB  XC  XD  XE  XF */
 
-static const unsigned char Mac_Centraleurope_CharToOrderMap[] =
+static const unsigned char Maccentraleurope_CharToOrderMap[] =
 {
   CTR,CTR,CTR,CTR,CTR,CTR,CTR,CTR,CTR,CTR,RET,CTR,CTR,RET,CTR,CTR, /* 0X */
   CTR,CTR,CTR,CTR,CTR,CTR,CTR,CTR,CTR,CTR,CTR,CTR,CTR,CTR,CTR,CTR, /* 1X */
@@ -267,14 +267,14 @@ const SequenceModel Iso_8859_2PolishModel =
   "ISO-8859-2"
 };
 
-const SequenceModel Mac_CentraleuropePolishModel =
+const SequenceModel MaccentraleuropePolishModel =
 {
-  Mac_Centraleurope_CharToOrderMap,
+  Maccentraleurope_CharToOrderMap,
   PolishLangModel,
   37,
   (float)0.9894531815946438,
   PR_TRUE,
-  "MAC-CENTRALEUROPE"
+  "MacCentralEurope"
 };
 
 const SequenceModel Iso_8859_13PolishModel =
@@ -295,4 +295,4 @@ const SequenceModel Windows_1250PolishModel =
   (float)0.9894531815946438,
   PR_TRUE,
   "WINDOWS-1250"
-};
\ No newline at end of file
+};
diff --git a/src/LangModels/LangRussianModel.cpp b/src/LangModels/LangRussianModel.cpp
index a532049..e21bfd1 100644
--- a/src/LangModels/LangRussianModel.cpp
+++ b/src/LangModels/LangRussianModel.cpp
@@ -337,7 +337,7 @@ const SequenceModel MacCyrillicRussianModel =
   64,
   (float)0.976601,
   PR_FALSE,
-  "MAC-CYRILLIC"
+  "MacCyrillic"
 };
 
 const SequenceModel Ibm866RussianModel =
diff --git a/src/LangModels/LangSlovakModel.cpp b/src/LangModels/LangSlovakModel.cpp
index cfa94aa..56cee68 100644
--- a/src/LangModels/LangSlovakModel.cpp
+++ b/src/LangModels/LangSlovakModel.cpp
@@ -103,7 +103,7 @@ static const unsigned char Iso_8859_2_CharToOrderMap[] =
 };
 /*X0  X1  X2  X3  X4  X5  X6  X7  X8  X9  XA  XB  XC  XD  XE  XF */
 
-static const unsigned char Mac_Centraleurope_CharToOrderMap[] =
+static const unsigned char Maccentraleurope_CharToOrderMap[] =
 {
   CTR,CTR,CTR,CTR,CTR,CTR,CTR,CTR,CTR,CTR,RET,CTR,CTR,RET,CTR,CTR, /* 0X */
   CTR,CTR,CTR,CTR,CTR,CTR,CTR,CTR,CTR,CTR,CTR,CTR,CTR,CTR,CTR,CTR, /* 1X */
@@ -268,14 +268,14 @@ const SequenceModel Iso_8859_2SlovakModel =
   "ISO-8859-2"
 };
 
-const SequenceModel Mac_CentraleuropeSlovakModel =
+const SequenceModel MaccentraleuropeSlovakModel =
 {
-  Mac_Centraleurope_CharToOrderMap,
+  Maccentraleurope_CharToOrderMap,
   SlovakLangModel,
   45,
   (float)0.9733303573968434,
   PR_TRUE,
-  "MAC-CENTRALEUROPE"
+  "MacCentralEurope"
 };
 
 const SequenceModel Windows_1250SlovakModel =
diff --git a/src/LangModels/LangSloveneModel.cpp b/src/LangModels/LangSloveneModel.cpp
index da28d86..923fcec 100644
--- a/src/LangModels/LangSloveneModel.cpp
+++ b/src/LangModels/LangSloveneModel.cpp
@@ -124,7 +124,7 @@ static const unsigned char Windows_1250_CharToOrderMap[] =
 };
 /*X0  X1  X2  X3  X4  X5  X6  X7  X8  X9  XA  XB  XC  XD  XE  XF */
 
-static const unsigned char Mac_Centraleurope_CharToOrderMap[] =
+static const unsigned char Maccentraleurope_CharToOrderMap[] =
 {
   CTR,CTR,CTR,CTR,CTR,CTR,CTR,CTR,CTR,CTR,RET,CTR,CTR,RET,CTR,CTR, /* 0X */
   CTR,CTR,CTR,CTR,CTR,CTR,CTR,CTR,CTR,CTR,CTR,CTR,CTR,CTR,CTR,CTR, /* 1X */
@@ -238,14 +238,14 @@ const SequenceModel Windows_1250SloveneModel =
   "WINDOWS-1250"
 };
 
-const SequenceModel Mac_CentraleuropeSloveneModel =
+const SequenceModel MaccentraleuropeSloveneModel =
 {
-  Mac_Centraleurope_CharToOrderMap,
+  Maccentraleurope_CharToOrderMap,
   SloveneLangModel,
   29,
   (float)0.9983524317161332,
   PR_TRUE,
-  "MAC-CENTRALEUROPE"
+  "MacCentralEurope"
 };
 
 const SequenceModel Ibm852SloveneModel =
diff --git a/src/fuzzing/CMakeLists.txt b/src/fuzzing/CMakeLists.txt
new file mode 100644
index 0000000..c707ad4
--- /dev/null
+++ b/src/fuzzing/CMakeLists.txt
@@ -0,0 +1,36 @@
+# for macOS: $ cmake -DCMAKE_CXX_COMPILER="`brew --prefix llvm`/bin/clang++" ../
+
+set(
+	UCHARDET_SOURCES
+	uchardet_fuzzer.cpp
+)
+
+add_definitions(
+	-O0
+	-g3
+	-fsanitize=fuzzer,address
+)
+set(CMAKE_CXX_FLAGS "-fsanitize=fuzzer,address")
+set(CMAKE_C_FLAGS "-fsanitize=fuzzer,address")
+set(CMAKE_EXE_LINKER_FLAGS "-fsanitize=fuzzer,address")
+set(CMAKE_MODULE_LINKER_FLAGS "-fsanitize=fuzzer,address")
+set(CMAKE_SHARED_LINKER_FLAGS "-fsanitize=fuzzer,address")
+
+set(FUZZER_BINARY uchardet_fuzzer)
+
+add_executable(
+	${FUZZER_BINARY}
+	${UCHARDET_SOURCES}
+)
+
+target_link_libraries(
+	${FUZZER_BINARY}
+	${UCHARDET_LIBRARY}
+)
+
+# install(
+# 	TARGETS
+# 		${FUZZER_BINARY}
+# 	RUNTIME DESTINATION
+# 		${CMAKE_INSTALL_BINDIR}
+# )
diff --git a/src/fuzzing/uchardet_fuzzer.cpp b/src/fuzzing/uchardet_fuzzer.cpp
new file mode 100644
index 0000000..21da646
--- /dev/null
+++ b/src/fuzzing/uchardet_fuzzer.cpp
@@ -0,0 +1,64 @@
+/* ***** BEGIN LICENSE BLOCK *****
+ * Version: MPL 1.1/GPL 2.0/LGPL 2.1
+ *
+ * The contents of this file are subject to the Mozilla Public License Version
+ * 1.1 (the "License"); you may not use this file except in compliance with
+ * the License. You may obtain a copy of the License at
+ * http://www.mozilla.org/MPL/
+ *
+ * Software distributed under the License is distributed on an "AS IS" basis,
+ * WITHOUT WARRANTY OF ANY KIND, either express or implied. See the License
+ * for the specific language governing rights and limitations under the
+ * License.
+ *
+ * The Original Code is Mozilla Universal charset detector code.
+ *
+ * The Initial Developer of the Original Code is
+ * Netscape Communications Corporation.
+ * Portions created by the Initial Developer are Copyright (C) 2001
+ * the Initial Developer. All Rights Reserved.
+ *
+ * Contributor(s):
+ *          PyYoshi <myoshi321go@gmail.com>
+ *
+ * Alternatively, the contents of this file may be used under the terms of
+ * either the GNU General Public License Version 2 or later (the "GPL"), or
+ * the GNU Lesser General Public License Version 2.1 or later (the "LGPL"),
+ * in which case the provisions of the GPL or the LGPL are applicable instead
+ * of those above. If you wish to allow use of your version of this file only
+ * under the terms of either the GPL or the LGPL, and not to allow others to
+ * use your version of this file under the terms of the MPL, indicate your
+ * decision by deleting the provisions above and replace them with the notice
+ * and other provisions required by the GPL or the LGPL. If you do not delete
+ * the provisions above, a recipient may use your version of this file under
+ * the terms of any one of the MPL, the GPL or the LGPL.
+ *
+ * ***** END LICENSE BLOCK ***** */
+#include "../uchardet.h"
+#include <iostream>
+#include <cstdlib>
+#include <stdio.h>
+#include <stdint.h>
+
+// extern "C" int LLVMFuzzerTestOneInput(const uint8_t *data, size_t size);
+
+extern "C" 
+int LLVMFuzzerTestOneInput(const uint8_t *data, size_t size) {
+    const char* buff = reinterpret_cast<const char*>(data);
+
+    uchardet_t handle = uchardet_new();
+
+    int retval = uchardet_handle_data(handle, buff, size);
+    if (retval == HANDLE_DATA_RESULT_ERROR)
+    {
+        fprintf(stderr, "Handle data error.\n");
+        exit(1);
+    }
+
+    uchardet_data_end(handle);
+    uchardet_get_charset(handle);
+    uchardet_get_confidence(handle);
+    uchardet_delete(handle);
+
+    return 0;
+}
diff --git a/src/nsBig5Prober.cpp b/src/nsBig5Prober.cpp
index 7a85abb..3e47d6e 100644
--- a/src/nsBig5Prober.cpp
+++ b/src/nsBig5Prober.cpp
@@ -46,7 +46,7 @@ void  nsBig5Prober::Reset(void)
 
 nsProbingState nsBig5Prober::HandleData(const char* aBuf, PRUint32 aLen)
 {
-  nsSMState codingState;
+  PRUint32 codingState;
 
   for (PRUint32 i = 0; i < aLen; i++)
   {
diff --git a/src/nsCharSetProber.cpp b/src/nsCharSetProber.cpp
index 0429dd1..79d5ace 100644
--- a/src/nsCharSetProber.cpp
+++ b/src/nsCharSetProber.cpp
@@ -74,7 +74,7 @@ PRBool nsCharSetProber::FilterWithoutEnglishLetters(const char* aBuf, PRUint32 a
   if (meetMSB && curPtr > prevPtr) 
     while (prevPtr < curPtr) *newptr++ = *prevPtr++;  
 
-  newLen = newptr - *newBuf;
+  newLen = (PRUint32) (newptr - *newBuf);
 
   return PR_TRUE;
 }
@@ -119,7 +119,7 @@ PRBool nsCharSetProber::FilterWithEnglishLetters(const char* aBuf, PRUint32 aLen
     while (prevPtr < curPtr)
       *newptr++ = *prevPtr++;  
 
-  newLen = newptr - *newBuf;
+  newLen = (PRUint32) (newptr - *newBuf);
 
   return PR_TRUE;
 }
diff --git a/src/nsCodingStateMachine.h b/src/nsCodingStateMachine.h
index 819f9ab..8861118 100644
--- a/src/nsCodingStateMachine.h
+++ b/src/nsCodingStateMachine.h
@@ -39,11 +39,12 @@
 
 #include "nsPkgInt.h"
 
-typedef enum {
-   eStart = 0,
-   eError = 1,
-   eItsMe = 2 
-} nsSMState;
+/* Apart from these 3 generic states, machine states are specific to
+ * each charset prober.
+ */
+#define eStart 0
+#define eError 1
+#define eItsMe 2
 
 #define GETCLASS(c) GETFROMPCK(((unsigned char)(c)), mModel->classTable)
 
@@ -60,7 +61,7 @@ typedef struct
 class nsCodingStateMachine {
 public:
   nsCodingStateMachine(const SMModel* sm) : mModel(sm) { mCurrentState = eStart; }
-  nsSMState NextState(char c){
+  PRUint32 NextState(char c){
     //for each byte we get its class , if it is first byte, we also get byte length
     PRUint32 byteCls = GETCLASS(c);
     if (mCurrentState == eStart)
@@ -69,8 +70,8 @@ class nsCodingStateMachine {
       mCurrentCharLen = mModel->charLenTable[byteCls];
     }
     //from byte's class and stateTable, we get its next state
-    mCurrentState=(nsSMState)GETFROMPCK(mCurrentState*(mModel->classFactor)+byteCls,
-                                       mModel->stateTable);
+    mCurrentState = GETFROMPCK(mCurrentState * mModel->classFactor + byteCls,
+                               mModel->stateTable);
     mCurrentBytePos++;
     return mCurrentState;
   }
@@ -79,7 +80,7 @@ class nsCodingStateMachine {
   const char * GetCodingStateMachine() {return mModel->name;}
 
 protected:
-  nsSMState mCurrentState;
+  PRUint32 mCurrentState;
   PRUint32 mCurrentCharLen;
   PRUint32 mCurrentBytePos;
 
diff --git a/src/nsEUCJPProber.cpp b/src/nsEUCJPProber.cpp
index 54861b3..f84d154 100644
--- a/src/nsEUCJPProber.cpp
+++ b/src/nsEUCJPProber.cpp
@@ -52,7 +52,7 @@ void  nsEUCJPProber::Reset(void)
 
 nsProbingState nsEUCJPProber::HandleData(const char* aBuf, PRUint32 aLen)
 {
-  nsSMState codingState;
+  PRUint32 codingState;
 
   for (PRUint32 i = 0; i < aLen; i++)
   {
diff --git a/src/nsEUCKRProber.cpp b/src/nsEUCKRProber.cpp
index 3632f1f..6aae8ae 100644
--- a/src/nsEUCKRProber.cpp
+++ b/src/nsEUCKRProber.cpp
@@ -47,7 +47,7 @@ void  nsEUCKRProber::Reset(void)
 
 nsProbingState nsEUCKRProber::HandleData(const char* aBuf, PRUint32 aLen)
 {
-  nsSMState codingState;
+  PRUint32 codingState;
 
   for (PRUint32 i = 0; i < aLen; i++)
   {
diff --git a/src/nsEUCTWProber.cpp b/src/nsEUCTWProber.cpp
index a06e074..7e61ea1 100644
--- a/src/nsEUCTWProber.cpp
+++ b/src/nsEUCTWProber.cpp
@@ -47,7 +47,7 @@ void  nsEUCTWProber::Reset(void)
 
 nsProbingState nsEUCTWProber::HandleData(const char* aBuf, PRUint32 aLen)
 {
-  nsSMState codingState;
+  PRUint32 codingState;
 
   for (PRUint32 i = 0; i < aLen; i++)
   {
diff --git a/src/nsEscCharsetProber.cpp b/src/nsEscCharsetProber.cpp
index 464c753..d093ee4 100644
--- a/src/nsEscCharsetProber.cpp
+++ b/src/nsEscCharsetProber.cpp
@@ -75,7 +75,7 @@ void nsEscCharSetProber::Reset(void)
 
 nsProbingState nsEscCharSetProber::HandleData(const char* aBuf, PRUint32 aLen)
 {
-  nsSMState codingState;
+  PRUint32 codingState;
   PRInt32 j;
   PRUint32 i;
 
diff --git a/src/nsEscSM.cpp b/src/nsEscSM.cpp
index eed1b7c..cad3df8 100644
--- a/src/nsEscSM.cpp
+++ b/src/nsEscSM.cpp
@@ -197,7 +197,7 @@ PCK4BITS(eError,eError,eError,eItsMe,eError,eError,eError,eError),//38-3f
 PCK4BITS(eError,eError,eError,eError,eItsMe,eError,eStart,eStart) //40-47 
 };
 
-static const PRUint32 ISO2022JPCharLenTable[] = {0, 0, 0, 0, 0, 0, 0, 0};
+static const PRUint32 ISO2022JPCharLenTable[] = {0, 0, 0, 0, 0, 0, 0, 0, 0, 0};
 
 const SMModel ISO2022JPSMModel = {
   {eIdxSft4bits, eSftMsk4bits, eBitSft4bits, eUnitMsk4bits, ISO2022JP_cls },
diff --git a/src/nsGB2312Prober.cpp b/src/nsGB2312Prober.cpp
index b6d469c..eac0762 100644
--- a/src/nsGB2312Prober.cpp
+++ b/src/nsGB2312Prober.cpp
@@ -52,7 +52,7 @@ void  nsGB18030Prober::Reset(void)
 
 nsProbingState nsGB18030Prober::HandleData(const char* aBuf, PRUint32 aLen)
 {
-  nsSMState codingState;
+  PRUint32 codingState;
 
   for (PRUint32 i = 0; i < aLen; i++)
   {
diff --git a/src/nsSBCSGroupProber.cpp b/src/nsSBCSGroupProber.cpp
index 66738ea..42b1b34 100644
--- a/src/nsSBCSGroupProber.cpp
+++ b/src/nsSBCSGroupProber.cpp
@@ -128,19 +128,19 @@ nsSBCSGroupProber::nsSBCSGroupProber()
 
   mProbers[46] = new nsSingleByteCharSetProber(&Windows_1250CzechModel);
   mProbers[47] = new nsSingleByteCharSetProber(&Iso_8859_2CzechModel);
-  mProbers[48] = new nsSingleByteCharSetProber(&Mac_CentraleuropeCzechModel);
+  mProbers[48] = new nsSingleByteCharSetProber(&MaccentraleuropeCzechModel);
   mProbers[49] = new nsSingleByteCharSetProber(&Ibm852CzechModel);
 
   mProbers[50] = new nsSingleByteCharSetProber(&Windows_1250SlovakModel);
   mProbers[51] = new nsSingleByteCharSetProber(&Iso_8859_2SlovakModel);
-  mProbers[52] = new nsSingleByteCharSetProber(&Mac_CentraleuropeSlovakModel);
+  mProbers[52] = new nsSingleByteCharSetProber(&MaccentraleuropeSlovakModel);
   mProbers[53] = new nsSingleByteCharSetProber(&Ibm852SlovakModel);
 
   mProbers[54] = new nsSingleByteCharSetProber(&Windows_1250PolishModel);
   mProbers[55] = new nsSingleByteCharSetProber(&Iso_8859_2PolishModel);
   mProbers[56] = new nsSingleByteCharSetProber(&Iso_8859_13PolishModel);
   mProbers[57] = new nsSingleByteCharSetProber(&Iso_8859_16PolishModel);
-  mProbers[58] = new nsSingleByteCharSetProber(&Mac_CentraleuropePolishModel);
+  mProbers[58] = new nsSingleByteCharSetProber(&MaccentraleuropePolishModel);
   mProbers[59] = new nsSingleByteCharSetProber(&Ibm852PolishModel);
 
   mProbers[60] = new nsSingleByteCharSetProber(&Iso_8859_1FinnishModel);
@@ -150,46 +150,46 @@ nsSBCSGroupProber::nsSBCSGroupProber()
   mProbers[64] = new nsSingleByteCharSetProber(&Iso_8859_15FinnishModel);
   mProbers[65] = new nsSingleByteCharSetProber(&Windows_1252FinnishModel);
 
-  mProbers[65] = new nsSingleByteCharSetProber(&Iso_8859_1ItalianModel);
-  mProbers[66] = new nsSingleByteCharSetProber(&Iso_8859_3ItalianModel);
-  mProbers[67] = new nsSingleByteCharSetProber(&Iso_8859_9ItalianModel);
-  mProbers[68] = new nsSingleByteCharSetProber(&Iso_8859_15ItalianModel);
-  mProbers[69] = new nsSingleByteCharSetProber(&Windows_1252ItalianModel);
-
-  mProbers[70] = new nsSingleByteCharSetProber(&Windows_1250CroatianModel);
-  mProbers[71] = new nsSingleByteCharSetProber(&Iso_8859_2CroatianModel);
-  mProbers[72] = new nsSingleByteCharSetProber(&Iso_8859_13CroatianModel);
-  mProbers[73] = new nsSingleByteCharSetProber(&Iso_8859_16CroatianModel);
-  mProbers[74] = new nsSingleByteCharSetProber(&Mac_CentraleuropeCroatianModel);
-  mProbers[75] = new nsSingleByteCharSetProber(&Ibm852CroatianModel);
-
-  mProbers[76] = new nsSingleByteCharSetProber(&Windows_1252EstonianModel);
-  mProbers[77] = new nsSingleByteCharSetProber(&Windows_1257EstonianModel);
-  mProbers[78] = new nsSingleByteCharSetProber(&Iso_8859_4EstonianModel);
-  mProbers[79] = new nsSingleByteCharSetProber(&Iso_8859_13EstonianModel);
-  mProbers[80] = new nsSingleByteCharSetProber(&Iso_8859_15EstonianModel);
-
-  mProbers[81] = new nsSingleByteCharSetProber(&Iso_8859_1IrishModel);
-  mProbers[82] = new nsSingleByteCharSetProber(&Iso_8859_9IrishModel);
-  mProbers[83] = new nsSingleByteCharSetProber(&Iso_8859_15IrishModel);
-  mProbers[84] = new nsSingleByteCharSetProber(&Windows_1252IrishModel);
-
-  mProbers[85] = new nsSingleByteCharSetProber(&Windows_1250RomanianModel);
-  mProbers[86] = new nsSingleByteCharSetProber(&Iso_8859_2RomanianModel);
-  mProbers[87] = new nsSingleByteCharSetProber(&Iso_8859_16RomanianModel);
-  mProbers[88] = new nsSingleByteCharSetProber(&Ibm852RomanianModel);
-
-  mProbers[89] = new nsSingleByteCharSetProber(&Windows_1250SloveneModel);
-  mProbers[90] = new nsSingleByteCharSetProber(&Iso_8859_2SloveneModel);
-  mProbers[91] = new nsSingleByteCharSetProber(&Iso_8859_16SloveneModel);
-  mProbers[92] = new nsSingleByteCharSetProber(&Mac_CentraleuropeSloveneModel);
-  mProbers[93] = new nsSingleByteCharSetProber(&Ibm852SloveneModel);
-
-  mProbers[94] = new nsSingleByteCharSetProber(&Iso_8859_1SwedishModel);
-  mProbers[95] = new nsSingleByteCharSetProber(&Iso_8859_4SwedishModel);
-  mProbers[96] = new nsSingleByteCharSetProber(&Iso_8859_9SwedishModel);
-  mProbers[97] = new nsSingleByteCharSetProber(&Iso_8859_15SwedishModel);
-  mProbers[98] = new nsSingleByteCharSetProber(&Windows_1252SwedishModel);
+  mProbers[66] = new nsSingleByteCharSetProber(&Iso_8859_1ItalianModel);
+  mProbers[67] = new nsSingleByteCharSetProber(&Iso_8859_3ItalianModel);
+  mProbers[68] = new nsSingleByteCharSetProber(&Iso_8859_9ItalianModel);
+  mProbers[69] = new nsSingleByteCharSetProber(&Iso_8859_15ItalianModel);
+  mProbers[70] = new nsSingleByteCharSetProber(&Windows_1252ItalianModel);
+
+  mProbers[71] = new nsSingleByteCharSetProber(&Windows_1250CroatianModel);
+  mProbers[72] = new nsSingleByteCharSetProber(&Iso_8859_2CroatianModel);
+  mProbers[73] = new nsSingleByteCharSetProber(&Iso_8859_13CroatianModel);
+  mProbers[74] = new nsSingleByteCharSetProber(&Iso_8859_16CroatianModel);
+  mProbers[75] = new nsSingleByteCharSetProber(&MaccentraleuropeCroatianModel);
+  mProbers[76] = new nsSingleByteCharSetProber(&Ibm852CroatianModel);
+
+  mProbers[77] = new nsSingleByteCharSetProber(&Windows_1252EstonianModel);
+  mProbers[78] = new nsSingleByteCharSetProber(&Windows_1257EstonianModel);
+  mProbers[79] = new nsSingleByteCharSetProber(&Iso_8859_4EstonianModel);
+  mProbers[80] = new nsSingleByteCharSetProber(&Iso_8859_13EstonianModel);
+  mProbers[81] = new nsSingleByteCharSetProber(&Iso_8859_15EstonianModel);
+
+  mProbers[82] = new nsSingleByteCharSetProber(&Iso_8859_1IrishModel);
+  mProbers[83] = new nsSingleByteCharSetProber(&Iso_8859_9IrishModel);
+  mProbers[84] = new nsSingleByteCharSetProber(&Iso_8859_15IrishModel);
+  mProbers[85] = new nsSingleByteCharSetProber(&Windows_1252IrishModel);
+
+  mProbers[86] = new nsSingleByteCharSetProber(&Windows_1250RomanianModel);
+  mProbers[87] = new nsSingleByteCharSetProber(&Iso_8859_2RomanianModel);
+  mProbers[88] = new nsSingleByteCharSetProber(&Iso_8859_16RomanianModel);
+  mProbers[89] = new nsSingleByteCharSetProber(&Ibm852RomanianModel);
+
+  mProbers[90] = new nsSingleByteCharSetProber(&Windows_1250SloveneModel);
+  mProbers[91] = new nsSingleByteCharSetProber(&Iso_8859_2SloveneModel);
+  mProbers[92] = new nsSingleByteCharSetProber(&Iso_8859_16SloveneModel);
+  mProbers[93] = new nsSingleByteCharSetProber(&MaccentraleuropeSloveneModel);
+  mProbers[94] = new nsSingleByteCharSetProber(&Ibm852SloveneModel);
+
+  mProbers[95] = new nsSingleByteCharSetProber(&Iso_8859_1SwedishModel);
+  mProbers[96] = new nsSingleByteCharSetProber(&Iso_8859_4SwedishModel);
+  mProbers[97] = new nsSingleByteCharSetProber(&Iso_8859_9SwedishModel);
+  mProbers[98] = new nsSingleByteCharSetProber(&Iso_8859_15SwedishModel);
+  mProbers[99] = new nsSingleByteCharSetProber(&Windows_1252SwedishModel);
 
   Reset();
 }
diff --git a/src/nsSBCSGroupProber.h b/src/nsSBCSGroupProber.h
index 64c021b..ec72324 100644
--- a/src/nsSBCSGroupProber.h
+++ b/src/nsSBCSGroupProber.h
@@ -40,7 +40,7 @@
 #define nsSBCSGroupProber_h__
 
 
-#define NUM_OF_SBCS_PROBERS 99
+#define NUM_OF_SBCS_PROBERS 100
 
 class nsCharSetProber;
 class nsSBCSGroupProber: public nsCharSetProber {
diff --git a/src/nsSBCharSetProber.cpp b/src/nsSBCharSetProber.cpp
index 1f7f473..001529f 100644
--- a/src/nsSBCharSetProber.cpp
+++ b/src/nsSBCharSetProber.cpp
@@ -63,7 +63,7 @@ nsProbingState nsSingleByteCharSetProber::HandleData(const char* aBuf, PRUint32
     }
     if (order < mModel->freqCharCount)
     {
-        mFreqChar++;
+      mFreqChar++;
 
       if (mLastOrder < mModel->freqCharCount)
       {
diff --git a/src/nsSBCharSetProber.h b/src/nsSBCharSetProber.h
index 42d21b2..0689369 100644
--- a/src/nsSBCharSetProber.h
+++ b/src/nsSBCharSetProber.h
@@ -190,19 +190,19 @@ extern const SequenceModel Iso_8859_3MalteseModel;
 extern const SequenceModel Windows_1250CzechModel;
 extern const SequenceModel Iso_8859_2CzechModel;
 extern const SequenceModel Ibm852CzechModel;
-extern const SequenceModel Mac_CentraleuropeCzechModel;
+extern const SequenceModel MaccentraleuropeCzechModel;
 
 extern const SequenceModel Windows_1250SlovakModel;
 extern const SequenceModel Iso_8859_2SlovakModel;
 extern const SequenceModel Ibm852SlovakModel;
-extern const SequenceModel Mac_CentraleuropeSlovakModel;
+extern const SequenceModel MaccentraleuropeSlovakModel;
 
 extern const SequenceModel Windows_1250PolishModel;
 extern const SequenceModel Iso_8859_2PolishModel;
 extern const SequenceModel Iso_8859_13PolishModel;
 extern const SequenceModel Iso_8859_16PolishModel;
 extern const SequenceModel Ibm852PolishModel;
-extern const SequenceModel Mac_CentraleuropePolishModel;
+extern const SequenceModel MaccentraleuropePolishModel;
 
 extern const SequenceModel Iso_8859_1FinnishModel;
 extern const SequenceModel Iso_8859_4FinnishModel;
@@ -222,7 +222,7 @@ extern const SequenceModel Iso_8859_2CroatianModel;
 extern const SequenceModel Iso_8859_13CroatianModel;
 extern const SequenceModel Iso_8859_16CroatianModel;
 extern const SequenceModel Ibm852CroatianModel;
-extern const SequenceModel Mac_CentraleuropeCroatianModel;
+extern const SequenceModel MaccentraleuropeCroatianModel;
 
 extern const SequenceModel Windows_1252EstonianModel;
 extern const SequenceModel Windows_1257EstonianModel;
@@ -244,7 +244,7 @@ extern const SequenceModel Windows_1250SloveneModel;
 extern const SequenceModel Iso_8859_2SloveneModel;
 extern const SequenceModel Iso_8859_16SloveneModel;
 extern const SequenceModel Ibm852SloveneModel;
-extern const SequenceModel Mac_CentraleuropeSloveneModel;
+extern const SequenceModel MaccentraleuropeSloveneModel;
 
 extern const SequenceModel Iso_8859_1SwedishModel;
 extern const SequenceModel Iso_8859_4SwedishModel;
diff --git a/src/nsSJISProber.cpp b/src/nsSJISProber.cpp
index c7842f6..1c354a7 100644
--- a/src/nsSJISProber.cpp
+++ b/src/nsSJISProber.cpp
@@ -52,7 +52,7 @@ void  nsSJISProber::Reset(void)
 
 nsProbingState nsSJISProber::HandleData(const char* aBuf, PRUint32 aLen)
 {
-  nsSMState codingState;
+  PRUint32 codingState;
 
   for (PRUint32 i = 0; i < aLen; i++)
   {
diff --git a/src/nsUTF8Prober.cpp b/src/nsUTF8Prober.cpp
index ab8d9f7..93677de 100644
--- a/src/nsUTF8Prober.cpp
+++ b/src/nsUTF8Prober.cpp
@@ -46,7 +46,7 @@ void  nsUTF8Prober::Reset(void)
 
 nsProbingState nsUTF8Prober::HandleData(const char* aBuf, PRUint32 aLen)
 {
-  nsSMState codingState;
+  PRUint32 codingState;
 
   for (PRUint32 i = 0; i < aLen; i++)
   {
diff --git a/src/nsUniversalDetector.cpp b/src/nsUniversalDetector.cpp
index 9711618..29159a8 100644
--- a/src/nsUniversalDetector.cpp
+++ b/src/nsUniversalDetector.cpp
@@ -55,6 +55,7 @@ nsUniversalDetector::nsUniversalDetector(PRUint32 aLanguageFilter)
 
   mStart = PR_TRUE;
   mDetectedCharset = nsnull;
+  mDetectedConfidence = 0.0;
   mGotData = PR_FALSE;
   mInputState = ePureAscii;
   mLastChar = '\0';
@@ -83,6 +84,7 @@ nsUniversalDetector::Reset()
 
   mStart = PR_TRUE;
   mDetectedCharset = nsnull;
+  mDetectedConfidence = 0.0;
   mGotData = PR_FALSE;
   mInputState = ePureAscii;
   mLastChar = '\0';
@@ -117,14 +119,18 @@ nsresult nsUniversalDetector::HandleData(const char* aBuf, PRUint32 aLen)
       switch (aBuf[0])
         {
         case '\xEF':
-          if (('\xBB' == aBuf[1]) && ('\xBF' == aBuf[2]))
+          if (('\xBB' == aBuf[1]) && ('\xBF' == aBuf[2])) {
             /* EF BB BF: UTF-8 encoded BOM. */
-            mDetectedCharset = "UTF-8";
+            mDetectedCharset = "UTF-8-SIG";
+            mDetectedConfidence = 0.99;
+          }
         break;
         case '\xFE':
-          if ('\xFF' == aBuf[1])
+          if ('\xFF' == aBuf[1]) {
             /* FE FF: UTF-16, big endian BOM. */
             mDetectedCharset = "UTF-16";
+            mDetectedConfidence = 0.99;
+          }
         break;
         case '\xFF':
           if ('\xFE' == aBuf[1])
@@ -135,11 +141,13 @@ nsresult nsUniversalDetector::HandleData(const char* aBuf, PRUint32 aLen)
             {
                 /* FF FE 00 00: UTF-32 (LE). */
                 mDetectedCharset = "UTF-32";
+                mDetectedConfidence = 0.99;
             }
             else
             {
                 /* FF FE: UTF-16, little endian BOM. */
                 mDetectedCharset = "UTF-16";
+                mDetectedConfidence = 0.99;
             }
           }
           break;
@@ -151,6 +159,7 @@ nsresult nsUniversalDetector::HandleData(const char* aBuf, PRUint32 aLen)
           {
               /* 00 00 FE FF: UTF-32 (BE). */
               mDetectedCharset = "UTF-32";
+              mDetectedConfidence = 0.99;
           }
           break;
         }
@@ -241,16 +250,7 @@ nsresult nsUniversalDetector::HandleData(const char* aBuf, PRUint32 aLen)
     {
       mDone = PR_TRUE;
       mDetectedCharset = mEscCharSetProber->GetCharSetName();
-    }
-    else if (mNbspFound)
-    {
-      mDetectedCharset = "ISO-8859-1";
-    }
-    else
-    {
-      /* ASCII with the ESC character (or the sequence "~{") is still
-       * ASCII until proven otherwise. */
-      mDetectedCharset = "ASCII";
+      mDetectedConfidence = mEscCharSetProber->GetConfidence();
     }
     break;
   case eHighbyte:
@@ -263,6 +263,7 @@ nsresult nsUniversalDetector::HandleData(const char* aBuf, PRUint32 aLen)
         {
           mDone = PR_TRUE;
           mDetectedCharset = mCharSetProbers[i]->GetCharSetName();
+          mDetectedConfidence = mCharSetProbers[i]->GetConfidence();
           return NS_OK;
         }
       }
@@ -270,17 +271,6 @@ nsresult nsUniversalDetector::HandleData(const char* aBuf, PRUint32 aLen)
     break;
 
   default:
-    if (mNbspFound)
-    {
-      /* ISO-8859-1 is a good result candidate for ASCII + NBSP.
-       * (though it could have been any ISO-8859 encoding). */
-      mDetectedCharset = "ISO-8859-1";
-    }
-    else
-    {
-      /* Pure ASCII */
-      mDetectedCharset = "ASCII";
-    }
     break;
   }
   return NS_OK;
@@ -297,10 +287,35 @@ void nsUniversalDetector::DataEnd()
     return;
   }
 
+  if (! mDetectedCharset)
+  {
+    switch (mInputState)
+    {
+    case eEscAscii:
+    case ePureAscii:
+      if (mNbspFound)
+      {
+          /* ISO-8859-1 is a good result candidate for ASCII + NBSP.
+           * (though it could have been any ISO-8859 encoding). */
+          mDetectedCharset = "ISO-8859-1";
+          mDetectedConfidence = 1.0;
+      }
+      else
+      {
+          /* ASCII with the ESC character (or the sequence "~{") is still
+           * ASCII until proven otherwise. */
+          mDetectedCharset = "ASCII";
+          mDetectedConfidence = 1.0;
+      }
+    default:
+      break;
+    }
+  }
+
   if (mDetectedCharset)
   {
     mDone = PR_TRUE;
-    Report(mDetectedCharset);
+    Report(mDetectedCharset, mDetectedConfidence);
     return;
   }
 
@@ -326,7 +341,7 @@ void nsUniversalDetector::DataEnd()
       }
       //do not report anything because we are not confident of it, that's in fact a negative answer
       if (maxProberConfidence > MINIMUM_THRESHOLD)
-        Report(mCharSetProbers[maxProber]->GetCharSetName());
+        Report(mCharSetProbers[maxProber]->GetCharSetName(), mCharSetProbers[maxProber]->GetConfidence());
     }
     break;
   case eEscAscii:
diff --git a/src/nsUniversalDetector.h b/src/nsUniversalDetector.h
index 9f0a4b1..9ac2db7 100644
--- a/src/nsUniversalDetector.h
+++ b/src/nsUniversalDetector.h
@@ -69,7 +69,7 @@ class nsUniversalDetector {
    virtual void DataEnd(void);
 
 protected:
-   virtual void Report(const char* aCharset) = 0;
+   virtual void Report(const char* aCharset, float aConfidence) = 0;
    virtual void Reset();
    nsInputState  mInputState;
    PRBool  mNbspFound;
@@ -79,6 +79,7 @@ class nsUniversalDetector {
    PRBool  mGotData;
    char    mLastChar;
    const char *  mDetectedCharset;
+   float mDetectedConfidence;
    PRInt32 mBestGuess;
    PRUint32 mLanguageFilter;
 
diff --git a/src/symbols.cmake b/src/symbols.cmake
index c7b1652..b958a35 100644
--- a/src/symbols.cmake
+++ b/src/symbols.cmake
@@ -6,6 +6,7 @@ set(
 	uchardet_data_end
 	uchardet_reset
 	uchardet_get_charset
+	uchardet_get_confidence
 )
 
 set (LINK_FLAGS "")
diff --git a/src/tools/CMakeLists.txt b/src/tools/CMakeLists.txt
index fa15eb2..ab69f33 100644
--- a/src/tools/CMakeLists.txt
+++ b/src/tools/CMakeLists.txt
@@ -3,6 +3,19 @@ set(
 	uchardet.cpp
 )
 
+if (CMAKE_BUILD_TYPE MATCHES Debug)
+	add_definitions(
+		-O0
+		-g3
+		-fsanitize=address
+	)
+	set(CMAKE_CXX_FLAGS "-fsanitize=address")
+	set(CMAKE_C_FLAGS "-fsanitize=address")
+	set(CMAKE_EXE_LINKER_FLAGS "-fsanitize=address")
+	set(CMAKE_MODULE_LINKER_FLAGS "-fsanitize=address")
+	set(CMAKE_SHARED_LINKER_FLAGS "-fsanitize=address")
+endif (CMAKE_BUILD_TYPE MATCHES Debug)
+
 set(UCHARDET_BINARY uchardet)
 
 add_executable(
diff --git a/src/tools/uchardet.cpp b/src/tools/uchardet.cpp
index f2aabfd..670e84a 100644
--- a/src/tools/uchardet.cpp
+++ b/src/tools/uchardet.cpp
@@ -57,7 +57,7 @@ void detect(FILE * fp)
     {
         size_t len = fread(buffer, 1, BUFFER_SIZE, fp);
         int retval = uchardet_handle_data(handle, buffer, len);
-        if (retval != 0)
+        if (retval == HANDLE_DATA_RESULT_ERROR)
         {
             fprintf(stderr, "Handle data error.\n");
             exit(1);
@@ -66,8 +66,9 @@ void detect(FILE * fp)
     uchardet_data_end(handle);
 
     const char * charset = uchardet_get_charset(handle);
+    float confidence = uchardet_get_confidence(handle);
     if (*charset)
-    	printf("%s\n", charset);
+    	printf("{ encoding=%s, confidence=%f }\n", charset, confidence);
 	else
 		printf("unknown\n");
 	
diff --git a/src/uchardet.cpp b/src/uchardet.cpp
index f1951d1..5cfc3bf 100644
--- a/src/uchardet.cpp
+++ b/src/uchardet.cpp
@@ -44,25 +44,29 @@ class HandleUniversalDetector : public nsUniversalDetector
 {
 protected:
     char *m_charset;
-
+    float m_confidence;
 public:
     HandleUniversalDetector()
     : nsUniversalDetector(NS_FILTER_ALL)
     , m_charset(0)
     {
+        m_confidence = 0.0;
     }
 
     virtual ~HandleUniversalDetector()
     {
-        if (m_charset)
+        if (m_charset) {
             free(m_charset);
+            m_confidence = 0.0;
+        }
     }
 
-    virtual void Report(const char* charset)
+    virtual void Report(const char* charset, float confidence)
     {
         if (m_charset)
             free(m_charset);
         m_charset = strdup(charset);
+        m_confidence = confidence;
     }
 
     virtual void Reset()
@@ -71,12 +75,21 @@ class HandleUniversalDetector : public nsUniversalDetector
         if (m_charset)
             free(m_charset);
         m_charset = strdup("");
+        m_confidence = 0.0;
     }
 
     const char* GetCharset() const
     {
         return m_charset? m_charset : "";
     }
+
+    float GetConfidence() {
+        return m_confidence;
+    }
+
+    PRBool HasDone() {
+        return mDone;
+    }
 };
 
 uchardet_t uchardet_new(void)
@@ -92,7 +105,15 @@ void uchardet_delete(uchardet_t ud)
 int uchardet_handle_data(uchardet_t ud, const char * data, size_t len)
 {
     nsresult ret = reinterpret_cast<HandleUniversalDetector*>(ud)->HandleData(data, (PRUint32)len);
-    return (ret != NS_OK);
+    if (ret == NS_ERROR_OUT_OF_MEMORY) {
+        return HANDLE_DATA_RESULT_ERROR;
+    }
+
+    if (reinterpret_cast<HandleUniversalDetector*>(ud)->HasDone()) {
+        return HANDLE_DATA_RESULT_DETECTED;
+    }
+
+    return HANDLE_DATA_RESULT_NEED_MORE_DATA;
 }
 
 void uchardet_data_end(uchardet_t ud)
@@ -109,3 +130,8 @@ const char* uchardet_get_charset(uchardet_t ud)
 {
     return reinterpret_cast<HandleUniversalDetector*>(ud)->GetCharset();
 }
+
+float uchardet_get_confidence(uchardet_t ud)
+{
+    return reinterpret_cast<HandleUniversalDetector*>(ud)->GetConfidence();
+}
diff --git a/src/uchardet.h b/src/uchardet.h
index 0ad1de1..5398cc6 100644
--- a/src/uchardet.h
+++ b/src/uchardet.h
@@ -61,13 +61,17 @@ uchardet_t uchardet_new(void);
  */
 void uchardet_delete(uchardet_t ud);
 
+#define HANDLE_DATA_RESULT_ERROR -1
+#define HANDLE_DATA_RESULT_DETECTED 0
+#define HANDLE_DATA_RESULT_NEED_MORE_DATA 1
+
 /**
  * Feed data to an encoding detector.
  * The detector is able to shortcut processing when it reaches certainty
  * for an encoding, so you should not worry about limiting input data.
  * As far as you should be concerned: the more the better.
  *
- * @param ud [in] handle of a instance of uchardet
+ * @param ud [in] handle of an instance of uchardet
  * @param data [in] data
  * @param len [in] number of byte of data
  * @return non-zero number on failure.
@@ -75,24 +79,26 @@ void uchardet_delete(uchardet_t ud);
 int uchardet_handle_data(uchardet_t ud, const char * data, size_t len);
 
 /**
- * Notify an end of data to an encoding detctor.
- * @param ud [in] handle of a instance of uchardet
+ * Notify an end of data to an encoding detector.
+ * @param ud [in] handle of an instance of uchardet
  */
 void uchardet_data_end(uchardet_t ud);
 
 /**
  * Reset an encoding detector.
- * @param ud [in] handle of a instance of uchardet
+ * @param ud [in] handle of an instance of uchardet
  */
 void uchardet_reset(uchardet_t ud);
 
 /**
  * Get an iconv-compatible name of the encoding that was detected.
- * @param ud [in] handle of a instance of uchardet
+ * @param ud [in] handle of an instance of uchardet
  * @return name of charset on success and "" on failure.
  */
 const char * uchardet_get_charset(uchardet_t ud);
 
+float uchardet_get_confidence(uchardet_t ud);
+
 #ifdef __cplusplus
 }
 #endif
diff --git a/test/CMakeLists.txt b/test/CMakeLists.txt
index a059044..5b27e2d 100644
--- a/test/CMakeLists.txt
+++ b/test/CMakeLists.txt
@@ -3,6 +3,19 @@ set(
 	uchardet-tests.c
 )
 
+if (CMAKE_BUILD_TYPE MATCHES Debug)
+	add_definitions(
+		-O0
+		-g3
+		-fsanitize=address
+	)
+	set(CMAKE_CXX_FLAGS "-fsanitize=address")
+	set(CMAKE_C_FLAGS "-fsanitize=address")
+	set(CMAKE_EXE_LINKER_FLAGS "-fsanitize=address")
+	set(CMAKE_MODULE_LINKER_FLAGS "-fsanitize=address")
+	set(CMAKE_SHARED_LINKER_FLAGS "-fsanitize=address")
+endif (CMAKE_BUILD_TYPE MATCHES Debug)
+
 add_executable(
 	uchardet-tests
 	${UCHARDET_TEST_SOURCES}
diff --git a/test/cs/mac-centraleurope.txt b/test/cs/maccentraleurope.txt
similarity index 100%
rename from test/cs/mac-centraleurope.txt
rename to test/cs/maccentraleurope.txt
diff --git a/test/hr/mac-centraleurope.txt b/test/hr/maccentraleurope.txt
similarity index 100%
rename from test/hr/mac-centraleurope.txt
rename to test/hr/maccentraleurope.txt
diff --git a/test/pl/mac-centraleurope.txt b/test/pl/maccentraleurope.txt
similarity index 100%
rename from test/pl/mac-centraleurope.txt
rename to test/pl/maccentraleurope.txt
diff --git a/test/ru/mac-cyrillic.txt b/test/ru/maccyrillic.txt
similarity index 100%
rename from test/ru/mac-cyrillic.txt
rename to test/ru/maccyrillic.txt
diff --git a/test/sk/mac-centraleurope.txt b/test/sk/maccentraleurope.txt
similarity index 100%
rename from test/sk/mac-centraleurope.txt
rename to test/sk/maccentraleurope.txt
diff --git a/test/sl/mac-centraleurope.txt b/test/sl/maccentraleurope.txt
similarity index 100%
rename from test/sl/mac-centraleurope.txt
rename to test/sl/maccentraleurope.txt
diff --git a/test/uchardet-tests.c b/test/uchardet-tests.c
index 180ac15..c7fa0a0 100644
--- a/test/uchardet-tests.c
+++ b/test/uchardet-tests.c
@@ -56,7 +56,7 @@ detect(FILE *fp)
     {
         size_t len = fread(buffer, 1, BUFFER_SIZE, fp);
         int retval = uchardet_handle_data(handle, buffer, len);
-        if (retval != 0)
+        if (retval == HANDLE_DATA_RESULT_ERROR)
         {
             fprintf(stderr,
                     "uchardet-tests: handle data error.\n");
@@ -120,6 +120,10 @@ main(int argc, char ** argv)
 
     /* In a unit test, 0 means success, other returned values mean failure. */
     success = (strcmp(charset, expected_charset) != 0);
+    
+    // if (success != 0) {
+    //     printf("expected_charset=%s, got=%s\n", expected_charset, charset);
+    // }
 
     free(charset);
     free(filename);