You guys should consider using this branch

DetectorTest.m

@@ -1,11 +1,15 @@
 #import <Foundation/Foundation.h>
+#import <AppKit/AppKit.h>
 
 #import <UniversalDetector/UniversalDetector.h>
 
 int main(int argc,char **argv)
 {
 	NSAutoreleasePool *pool = [[NSAutoreleasePool alloc] init];
 
+	[[NSUserDefaults standardUserDefaults] setBool:YES
+											forKey:UniversalDetectorUseMacRomanHeuristic];
+
 	NSError *error = nil;
 
 	for (int i = 1; i < argc; i++)
@@ -20,6 +24,11 @@ int main(int argc,char **argv)
 											  options:0
 												error:&error];
 
+		if (data == nil) {
+			NSLog(@"%@", error);
+			continue;
+		}
+
 		NSString *str = nil;
 
 		if (data == nil) {
@@ -38,19 +47,65 @@ int main(int argc,char **argv)
 		[detector analyzeData:data];
 		NSString *MIMECharsetName = [detector MIMECharset];
 		NSStringEncoding encoding = [detector encoding];
+		NSStringEncoding appKitEncoding = 0;
+
+		//if (encoding == NSWindowsCP1252StringEncoding || encoding == NSShiftJISStringEncoding)
+		{
+			NSDictionary *documentAttributes = nil;
+
+			// UniversalDetector does not differentiate between Windows Latin 1 and Mac Roman
+			// while AppKit has an apparent Mac Roman bias.
+			NSAttributedString *text = [[NSAttributedString alloc] initWithData:data
+																		options:nil
+															 documentAttributes:&documentAttributes
+																		  error:&error];
+
+			if (text == nil) {
+				NSLog(@"%@", error);
+				continue;
+			}
+			else {
+				[text release];
+
+				NSNumber *encodingNumber = documentAttributes[NSCharacterEncodingDocumentAttribute];
+				appKitEncoding = [encodingNumber intValue];
+			}
+		}
+
+		NSString *appKitResultString = nil;
+		if (appKitEncoding != 0) {
+			if (appKitEncoding != encoding) {
+				appKitResultString = [NSString stringWithFormat:@"\"%@\"",
+									  [NSString localizedNameOfStringEncoding:appKitEncoding]
+									  ];
+			}
+			else {
+				appKitResultString = @"(same result)";
+			}
+		}
 
-		str = [NSString stringWithFormat:@"%@\n\t\"%@\" (%@) confidence: %.1f%%",
+		str = [NSString stringWithFormat:
+			   @"%@\n"
+			   "\t" "\"%@\" (%@)\n"
+			   "\t" "confidence:% 6.1f%%"
+			   @"\n"
+			   "\t" "AppKit: %@",
 			   fileName,
 			   (encoding != 0) ? [NSString localizedNameOfStringEncoding:encoding] : @"UNKNOWN",
 			   (MIMECharsetName != nil) ? MIMECharsetName : @"UNKNOWN",
-			   ([detector confidence] * 100.0f) 
+			   ([detector confidence] * 100.0f),
+			   (appKitResultString != nil) ? appKitResultString : @"UNDEFINED"
 			   ];
+
+
+
 		printf("%s\n\n", [str UTF8String]);
 
 
 		[detector release];
 	}
 
 	[pool release];
-	return 0;
+
+	return EXIT_SUCCESS;
 }

Info.plist

@@ -1,5 +1,5 @@
 <?xml version="1.0" encoding="UTF-8"?>
-<!DOCTYPE plist PUBLIC "-//Apple Computer//DTD PLIST 1.0//EN" "http://www.apple.com/DTDs/PropertyList-1.0.dtd">
+<!DOCTYPE plist PUBLIC "-//Apple//DTD PLIST 1.0//EN" "http://www.apple.com/DTDs/PropertyList-1.0.dtd">
 <plist version="1.0">
 <dict>
 	<key>CFBundleDevelopmentRegion</key>
@@ -9,7 +9,7 @@
 	<key>CFBundleIconFile</key>
 	<string></string>
 	<key>CFBundleIdentifier</key>
-	<string>org.mozilla.universalchardet</string>
+	<string>$(PRODUCT_BUNDLE_IDENTIFIER)</string>
 	<key>CFBundleInfoDictionaryVersion</key>
 	<string>6.0</string>
 	<key>CFBundleName</key>

UniversalDetector.h

@@ -1,10 +1,15 @@
 #import <Foundation/Foundation.h>
 
+// You can enable this heuristic by setting the BOOL with that key in NSUserDefaults -standardUserDefaults to YES.
+// In this case, only -encoding wll be valid and -MIMECharset will be invalid.
+extern NSString * const	UniversalDetectorUseMacRomanHeuristic;
+
 @interface UniversalDetector:NSObject
 {
 	void *detectorPtr;
 	NSString *charsetName;
 	float confidence;
+	BOOL possiblyMacRoman;
 }
 
 -(void)analyzeContentsOfFile:(NSString *)path;

UniversalDetector.m

@@ -1,6 +1,10 @@
 #import "UniversalDetector.h"
 #import "WrappedUniversalDetector.h"
 
+
+NSString * const	UniversalDetectorUseMacRomanHeuristic			= @"UniversalDetectorUseMacRomanHeuristic";
+
+
 @implementation UniversalDetector
 
 -(id)init
@@ -41,6 +45,39 @@ -(void)analyzeData:(NSData *)data
 -(void)analyzeBytes:(const char *)data length:(int)len
 {
 	UniversalDetectorHandleData(detectorPtr, data, len);
+
+	BOOL useMacRomanHeuristic = [[NSUserDefaults standardUserDefaults] boolForKey:UniversalDetectorUseMacRomanHeuristic];
+
+	if (useMacRomanHeuristic) {
+		// Search for a carriage return (cr) without a following newline.
+		// We do this to determine, if the data could possibly be MacRoman.
+		const size_t searchWindowSize = 4096;
+		char *crPtr = memchr(data, '\r', MIN(len, searchWindowSize));
+		if (crPtr == NULL) {
+			possiblyMacRoman = NO;
+		}
+		else {
+			const int lastIndex = len - 1;
+			int crIndex = (crPtr - data);
+
+			// Check, if we are at least one byte before the end.
+			if (crIndex < lastIndex) {
+				if (data[crIndex+1] == '\n') {
+					possiblyMacRoman = NO;
+				}
+				else {
+					possiblyMacRoman = YES;
+				}
+			}
+			else {
+				possiblyMacRoman = YES;
+			}
+		}
+	}
+	else {
+		possiblyMacRoman = NO;
+	}
+
 	[charsetName release];
 	charsetName=nil;
 }
@@ -77,8 +114,18 @@ -(NSStringEncoding)encoding
 	// UniversalDetector detects CP949 but returns "EUC-KR" because CP949 lacks an IANA name.
 	// Kludge to make strings decode properly anyway.
 	if(cfenc==kCFStringEncodingEUC_KR) cfenc=kCFStringEncodingDOSKorean;
+	// Something similar happens with "Shift_JIS".
+	if(cfenc==kCFStringEncodingShiftJIS) cfenc=kCFStringEncodingDOSJapanese;
 
-	return CFStringConvertEncodingToNSStringEncoding(cfenc);
+	NSStringEncoding encoding = CFStringConvertEncodingToNSStringEncoding(cfenc);
+
+	if (possiblyMacRoman &&
+		(encoding == NSWindowsCP1252StringEncoding ||
+		 encoding == NSShiftJISStringEncoding)) {
+			encoding = NSMacOSRomanStringEncoding;
+	}
+
+	return encoding;
 }
 
 -(float)confidence

UniversalDetector.xcodeproj/project.pbxproj

@@ -41,6 +41,7 @@
 		3D4F0C3D11F42E3C00603CE3 /* WrappedUniversalDetector.h in Headers */ = {isa = PBXBuildFile; fileRef = 3D4F0C3B11F42E3C00603CE3 /* WrappedUniversalDetector.h */; };
 		3D724E301174182900CD3CBD /* Foundation.framework in Frameworks */ = {isa = PBXBuildFile; fileRef = 3D724E2F1174182900CD3CBD /* Foundation.framework */; };
 		3D724E4E11741AEE00CD3CBD /* Foundation.framework in Frameworks */ = {isa = PBXBuildFile; fileRef = 3D724E2F1174182900CD3CBD /* Foundation.framework */; };
+		3D81E2B318C5F92C00834BCA /* AppKit.framework in Frameworks */ = {isa = PBXBuildFile; fileRef = 3D81E2B218C5F92C00834BCA /* AppKit.framework */; };
 		3D8275CD16B943C100061FD9 /* nsMemory.h in Headers */ = {isa = PBXBuildFile; fileRef = 3D8275CB16B943C100061FD9 /* nsMemory.h */; };
 		3D8275D616B946F100061FD9 /* nsDebug.h in Headers */ = {isa = PBXBuildFile; fileRef = 3D8275D516B946F100061FD9 /* nsDebug.h */; };
 /* End PBXBuildFile section */
@@ -122,6 +123,7 @@
 		3D4F0C3A11F42E3C00603CE3 /* WrappedUniversalDetector.cpp */ = {isa = PBXFileReference; fileEncoding = 4; lastKnownFileType = sourcecode.cpp.cpp; path = WrappedUniversalDetector.cpp; sourceTree = "<group>"; };
 		3D4F0C3B11F42E3C00603CE3 /* WrappedUniversalDetector.h */ = {isa = PBXFileReference; fileEncoding = 4; lastKnownFileType = sourcecode.c.h; path = WrappedUniversalDetector.h; sourceTree = "<group>"; };
 		3D724E2F1174182900CD3CBD /* Foundation.framework */ = {isa = PBXFileReference; lastKnownFileType = wrapper.framework; name = Foundation.framework; path = System/Library/Frameworks/Foundation.framework; sourceTree = SDKROOT; };
+		3D81E2B218C5F92C00834BCA /* AppKit.framework */ = {isa = PBXFileReference; lastKnownFileType = wrapper.framework; name = AppKit.framework; path = System/Library/Frameworks/AppKit.framework; sourceTree = SDKROOT; };
 		3D8275CB16B943C100061FD9 /* nsMemory.h */ = {isa = PBXFileReference; fileEncoding = 4; lastKnownFileType = sourcecode.c.h; path = nsMemory.h; sourceTree = "<group>"; };
 		3D8275D516B946F100061FD9 /* nsDebug.h */ = {isa = PBXFileReference; fileEncoding = 4; lastKnownFileType = sourcecode.c.h; path = nsDebug.h; sourceTree = "<group>"; };
 		8DC2EF5A0486A6940098B216 /* Info.plist */ = {isa = PBXFileReference; fileEncoding = 4; lastKnownFileType = text.plist; path = Info.plist; sourceTree = "<group>"; };
@@ -133,6 +135,7 @@
 			isa = PBXFrameworksBuildPhase;
 			buildActionMask = 2147483647;
 			files = (
+				3D81E2B318C5F92C00834BCA /* AppKit.framework in Frameworks */,
 				3D724E4E11741AEE00CD3CBD /* Foundation.framework in Frameworks */,
 				1B0DDCB30A2D0B2E0009B697 /* UniversalDetector.framework in Frameworks */,
 			);
@@ -161,6 +164,7 @@
 		0867D691FE84028FC02AAC07 /* UniversalCharDet */ = {
 			isa = PBXGroup;
 			children = (
+				3D81E2B218C5F92C00834BCA /* AppKit.framework */,
 				08FB77AEFE84172EC02AAC07 /* Classes */,
 				3D8275C716B943C100061FD9 /* mozilla-release */,
 				1B0DDB110A2CB1B20009B697 /* universalchardet */,
@@ -417,7 +421,7 @@
 		0867D690FE84028FC02AAC07 /* Project object */ = {
 			isa = PBXProject;
 			attributes = {
-				LastUpgradeCheck = 0500;
+				LastUpgradeCheck = 0720;
 			};
 			buildConfigurationList = 1DEB91B108733DA50010E9CD /* Build configuration list for PBXProject "UniversalDetector" */;
 			compatibilityVersion = "Xcode 3.2";
@@ -558,8 +562,10 @@
 				GCC_PRECOMPILE_PREFIX_HEADER = YES;
 				GCC_PREFIX_HEADER = UniversalDetector_Prefix.pch;
 				INFOPLIST_FILE = Info.plist;
-				INSTALL_PATH = "@executable_path/../Frameworks";
+				INSTALL_PATH = "@rpath";
+				PRODUCT_BUNDLE_IDENTIFIER = org.mozilla.universalchardet;
 				PRODUCT_NAME = UniversalDetector;
+				SKIP_INSTALL = YES;
 				WRAPPER_EXTENSION = framework;
 			};
 			name = Debug;
@@ -574,8 +580,10 @@
 				GCC_PRECOMPILE_PREFIX_HEADER = YES;
 				GCC_PREFIX_HEADER = UniversalDetector_Prefix.pch;
 				INFOPLIST_FILE = Info.plist;
-				INSTALL_PATH = "@executable_path/../Frameworks";
+				INSTALL_PATH = "@rpath";
+				PRODUCT_BUNDLE_IDENTIFIER = org.mozilla.universalchardet;
 				PRODUCT_NAME = UniversalDetector;
+				SKIP_INSTALL = YES;
 				WRAPPER_EXTENSION = framework;
 			};
 			name = Release;
@@ -588,18 +596,23 @@
 				CLANG_WARN_EMPTY_BODY = YES;
 				CLANG_WARN_ENUM_CONVERSION = YES;
 				CLANG_WARN_INT_CONVERSION = YES;
+				CLANG_WARN_UNREACHABLE_CODE = YES;
 				CLANG_WARN__DUPLICATE_METHOD_MATCH = YES;
 				COMBINE_HIDPI_IMAGES = YES;
 				DEBUG_INFORMATION_FORMAT = dwarf;
+				ENABLE_STRICT_OBJC_MSGSEND = YES;
+				ENABLE_TESTABILITY = YES;
 				GCC_C_LANGUAGE_STANDARD = c99;
 				GCC_INLINES_ARE_PRIVATE_EXTERN = YES;
+				GCC_NO_COMMON_BLOCKS = YES;
 				GCC_VERSION = com.apple.compilers.llvm.clang.1_0;
+				GCC_WARN_64_TO_32_BIT_CONVERSION = NO;
 				GCC_WARN_ABOUT_RETURN_TYPE = YES;
 				GCC_WARN_UNDECLARED_SELECTOR = YES;
 				GCC_WARN_UNINITIALIZED_AUTOS = YES;
 				GCC_WARN_UNUSED_FUNCTION = YES;
 				GCC_WARN_UNUSED_VARIABLE = YES;
-				MACOSX_DEPLOYMENT_TARGET = 10.6;
+				MACOSX_DEPLOYMENT_TARGET = 10.9;
 				ONLY_ACTIVE_ARCH = YES;
 				SKIP_INSTALL = YES;
 			};
@@ -613,18 +626,22 @@
 				CLANG_WARN_EMPTY_BODY = YES;
 				CLANG_WARN_ENUM_CONVERSION = YES;
 				CLANG_WARN_INT_CONVERSION = YES;
+				CLANG_WARN_UNREACHABLE_CODE = YES;
 				CLANG_WARN__DUPLICATE_METHOD_MATCH = YES;
 				COMBINE_HIDPI_IMAGES = YES;
 				DEBUG_INFORMATION_FORMAT = dwarf;
+				ENABLE_STRICT_OBJC_MSGSEND = YES;
 				GCC_C_LANGUAGE_STANDARD = c99;
 				GCC_INLINES_ARE_PRIVATE_EXTERN = YES;
+				GCC_NO_COMMON_BLOCKS = YES;
 				GCC_VERSION = com.apple.compilers.llvm.clang.1_0;
+				GCC_WARN_64_TO_32_BIT_CONVERSION = NO;
 				GCC_WARN_ABOUT_RETURN_TYPE = YES;
 				GCC_WARN_UNDECLARED_SELECTOR = YES;
 				GCC_WARN_UNINITIALIZED_AUTOS = YES;
 				GCC_WARN_UNUSED_FUNCTION = YES;
 				GCC_WARN_UNUSED_VARIABLE = YES;
-				MACOSX_DEPLOYMENT_TARGET = 10.6;
+				MACOSX_DEPLOYMENT_TARGET = 10.9;
 				SKIP_INSTALL = YES;
 			};
 			name = Release;