diff --git a/app/Main.cpp b/app/Main.cpp
index 8f5753e..f2333a9 100644
--- a/app/Main.cpp
+++ b/app/Main.cpp
@@ -2644,6 +2644,8 @@ MainWindow::ConfigureReformatFromPreferences(ReformatHolder* pReformat)
 		pPreferences->GetPrefBool(kPrConvGWP));
 	pReformat->SetReformatAllowed(ReformatHolder::kReformatMagicWindow,
 		pPreferences->GetPrefBool(kPrConvText8));
+	pReformat->SetReformatAllowed(ReformatHolder::kReformatGutenberg,
+		pPreferences->GetPrefBool(kPrConvGutenberg));
 	pReformat->SetReformatAllowed(ReformatHolder::kReformatAWP,
 		pPreferences->GetPrefBool(kPrConvAWP));
 	pReformat->SetReformatAllowed(ReformatHolder::kReformatAWP,
@@ -2707,7 +2709,13 @@ MainWindow::ConfigureReformatFromPreferences(ReformatHolder* pReformat)
 /*static*/ ReformatHolder::SourceFormat
 MainWindow::ReformatterSourceFormat(DiskImg::FSFormat format)
 {
-	if (DiskImg::UsesDOSFileStructure(format))
+	/*
+	 * Gutenberg both UsesDOSFileStructure and is formatted with 
+	 * kFormatGutenberg, so check for the latter first.
+	 */
+	if (format == DiskImg::kFormatGutenberg)
+		return ReformatHolder::kSourceFormatGutenberg;
+	else if (DiskImg::UsesDOSFileStructure(format))
 		return ReformatHolder::kSourceFormatDOS;
 	else if (format == DiskImg::kFormatCPM)
 		return ReformatHolder::kSourceFormatCPM;
diff --git a/app/Preferences.cpp b/app/Preferences.cpp
index 900a068..5007a79 100644
--- a/app/Preferences.cpp
+++ b/app/Preferences.cpp
@@ -118,6 +118,7 @@ const Preferences::PrefMap Preferences::fPrefMaps[kPrefNumLastEntry] = {
 	{ kPrConvBusiness,					kBool,	kPrefsSect,		_T("conv-business") },
 	{ kPrConvGWP,						kBool,	kPrefsSect,		_T("conv-gwp") },
 	{ kPrConvText8,						kBool,	kPrefsSect,		_T("conv-text8") },
+	{ kPrConvGutenberg,					kBool,	kPrefsSect,		_T("conv-gutenberg") },
 	{ kPrConvAWP,						kBool,	kPrefsSect,		_T("conv-awp") },
 	{ kPrConvADB,						kBool,	kPrefsSect,		_T("conv-adb") },
 	{ kPrConvASP,						kBool,	kPrefsSect,		_T("conv-asp") },
@@ -227,6 +228,7 @@ Preferences::Preferences(void)
 	SetPrefBool(kPrConvBusiness, true);
 	SetPrefBool(kPrConvGWP, true);
 	SetPrefBool(kPrConvText8, true);
+	SetPrefBool(kPrConvGutenberg, true);
 	SetPrefBool(kPrConvAWP, true);
 	SetPrefBool(kPrConvADB, true);
 	SetPrefBool(kPrConvASP, true);
diff --git a/app/Preferences.h b/app/Preferences.h
index 360c476..d92741e 100644
--- a/app/Preferences.h
+++ b/app/Preferences.h
@@ -171,6 +171,7 @@ typedef enum {
 	kPrConvBusiness,				// bool
 	kPrConvGWP,						// bool
 	kPrConvText8,					// bool
+	kPrConvGutenberg,				// bool
 	kPrConvAWP,						// bool
 	kPrConvADB,						// bool
 	kPrConvASP,						// bool
diff --git a/reformat/Reformat.cpp b/reformat/Reformat.cpp
index 3561736..fda1696 100644
--- a/reformat/Reformat.cpp
+++ b/reformat/Reformat.cpp
@@ -86,6 +86,7 @@ ReformatHolder::GetReformatInstance(ReformatID id)
 	case kReformatSHR_DG3200:		pReformat = new ReformatDG3200SHR;			break;
 	case kReformatPrintShop:		pReformat = new ReformatPrintShop;			break;
 	case kReformatMacPaint:			pReformat = new ReformatMacPaint;			break;
+	case kReformatGutenberg:		pReformat = new ReformatGutenberg;			break;
 	case kReformatUnknown:
 	case kReformatMAX:
 	default:						assert(false);								break;
@@ -191,6 +192,9 @@ ReformatHolder::GetReformatName(ReformatID id)
 	case kReformatMagicWindow:
 		descr = "Magic Window";
 		break;
+	case kReformatGutenberg:
+		descr = "Gutenberg Word Processor";
+		break;
 	case kReformatAWP:
 		descr = "AppleWorks Word Processor";
 		break;
diff --git a/reformat/Reformat.h b/reformat/Reformat.h
index 164846f..464a85c 100644
--- a/reformat/Reformat.h
+++ b/reformat/Reformat.h
@@ -1,5 +1,6 @@
 /*
  * CiderPress
+ * Copyright (C) 2009 by CiderPress authors.  All Rights Reserved.
  * Copyright (C) 2007, 2008 by faddenSoft, LLC.  All Rights Reserved.
  * See the file LICENSE for distribution terms.
  */
@@ -96,6 +97,7 @@ public:
 		kReformatGWP,
 
 		kReformatMagicWindow,
+		kReformatGutenberg,
 
 		kReformatAWP,
 		kReformatADB,
@@ -182,12 +184,14 @@ public:
 	 *
 	 * We want to know if it's DOS so we can relax some file-type checking,
 	 * and we want to know if it's CP/M so we can adjust the way we think
-	 * about text files.
+	 * about text files.  We want to know if it's Gutenberg because they only
+	 * have one type of file, and it's indistingusihable from any other text file!
 	 */
 	typedef enum SourceFormat {
 		kSourceFormatGeneric = 0,
 		kSourceFormatDOS,
 		kSourceFormatCPM,
+		kSourceFormatGutenberg,
 	} SourceFormat;
 
 
@@ -268,6 +272,7 @@ public:
 	friend class ReformatTeach;
 	friend class ReformatGWP;
 	friend class ReformatMagicWindow;
+	friend class ReformatGutenberg;
 	friend class ReformatAWP;
 	friend class ReformatADB;
 	friend class ReformatASP;
diff --git a/reformat/ReformatBase.cpp b/reformat/ReformatBase.cpp
index ac0bd55..26ff84c 100644
--- a/reformat/ReformatBase.cpp
+++ b/reformat/ReformatBase.cpp
@@ -1,1084 +1,1103 @@
-/*
- * CiderPress
- * Copyright (C) 2007 by faddenSoft, LLC.  All Rights Reserved.
- * See the file LICENSE for distribution terms.
- */
-/*
- * Reformatter base class implementation.
- */
-#include "StdAfx.h"
-#include "ReformatBase.h"
-#include <math.h>
-
-
-static float roundf(float val)
-{
-	if (val < 0.5f)
-		return ::floorf(val);
-	else
-		return ::ceilf(val);
-}
-
-/*
- * ==========================================================================
- *		ReformatText
- * ==========================================================================
- */
-
-/*
- * Convert IIgs high-ASCII characters to Windows equivalents (when
- * available).
- *
- * Also found this:
- * http://www.unicode.org/Public/MAPPINGS/VENDORS/APPLE/ROMAN.TXT
- */
-const int kUnk = 0x3f;		// for unmappable chars, use '?'
-
-/*static*/ const unsigned char ReformatText::kGSCharConv[128] = {
-	0xc4,	// 0x80 A + umlaut (diaeresis?)
-	0xc5,	// 0x81 A + overcircle
-	0xc7,	// 0x82 C + cedilla
-	0xc9,	// 0x83 E + acute
-	0xd1,	// 0x84 N + tilde
-	0xd6,	// 0x85 O + umlaut
-	0xdc,	// 0x86 U + umlaut
-	0xe1,	// 0x87 a + acute
-	0xe0,	// 0x88 a + grave
-	0xe2,	// 0x89 a + circumflex
-	0xe4,	// 0x8a a + umlaut
-	0xe3,	// 0x8b a + tilde
-	0xe5,	// 0x8c a + overcircle
-	0xe7,	// 0x8d c + cedilla
-	0xe9,	// 0x8e e + acute
-	0xe8,	// 0x8f e + grave
-	0xea,	// 0x90 e + circumflex
-	0xeb,	// 0x91 e + umlaut
-	0xed,	// 0x92 i + acute
-	0xec,	// 0x93 i + grave
-	0xee,	// 0x94 i + circumflex
-	0xef,	// 0x95 i + umlaut
-	0xf1,	// 0x96 n + tilde
-	0xf3,	// 0x97 o + acute
-	0xf2,	// 0x98 o + grave
-	0xf4,	// 0x99 o + circumflex
-	0xf6,	// 0x9a o + umlaut
-	0xf5,	// 0x9b o + tilde
-	0xfa,	// 0x9c u + acute
-	0xf9,	// 0x9d u + grave
-	0xfb,	// 0x9e u + circumflex
-	0xfc,	// 0x9f u + umlaut
-	0x87,	// 0xa0 double cross (dagger)
-	0xb0,	// 0xa1 degrees
-	0xa2,	// 0xa2 cents
-	0xa3,	// 0xa3 pounds (UK$)
-	0xa7,	// 0xa4 section start
-	0x95,	// 0xa5 small square (bullet)  [using fat bullet]
-	0xb6,	// 0xa6 paragraph (pilcrow)
-	0xdf,	// 0xa7 curly B (latin small letter sharp S)
-	0xae,	// 0xa8 raised 'R' (registered)
-	0xa9,	// 0xa9 raised 'C' (copyright)
-	0x99,	// 0xaa raised 'TM' (trademark)
-	0xb4,	// 0xab acute accent
-	0xa8,	// 0xac umlaut (diaeresis)
-	kUnk,	// 0xad not-equal
-	0xc6,	// 0xae merged AE
-	0xd8,	// 0xaf O + slash (upper-case nil?)
-	kUnk,	// 0xb0 infinity
-	0xb1,	// 0xb1 +/-
-	kUnk,	// 0xb2 <=
-	kUnk,	// 0xb3 >=
-	0xa5,	// 0xb4 Yen (Japan$)
-	0xb5,	// 0xb5 mu (micro)
-	kUnk,	// 0xb6 delta (partial differentiation) [could use D-bar 0xd0]
-	kUnk,	// 0xb7 epsilon (N-ary summation) [could use C-double-bar 0x80]
-	kUnk,	// 0xb8 PI (N-ary product)
-	kUnk,	// 0xb9 pi
-	kUnk,	// 0xba integral
-	0xaa,	// 0xbb a underbar (feminine ordinal)  [using raised a]
-	0xba,	// 0xbc o underbar (masculine ordinal)  [using raised o]
-	kUnk,	// 0xbd omega (Ohm)
-	0xe6,	// 0xbe merged ae
-	0xf8,	// 0xbf o + slash (lower-case nil?)
-	0xbf,	// 0xc0 upside-down question mark
-	0xa1,	// 0xc1 upside-down exclamation point
-	0xac,	// 0xc2 rotated L ("not" sign)
-	0xb7,	// 0xc3 checkmark (square root) [using small bullet]
-	0x83,	// 0xc4 script f
-	kUnk,	// 0xc5 approximately equal
-	kUnk,	// 0xc6 delta (triangle / increment)
-	0xab,	// 0xc7 much less than
-	0xbb,	// 0xc8 much greater than
-	0x85,	// 0xc9 ellipsis
-	0xa0,	// 0xca blank (sticky space)
-	0xc0,	// 0xcb A + grave
-	0xc3,	// 0xcc A + tilde
-	0xd5,	// 0xcd O + tilde
-	0x8c,	// 0xce merged OE
-	0x9c,	// 0xcf merged oe
-	0x96,	// 0xd0 short hyphen (en dash)
-	0x97,	// 0xd1 long hyphen (em dash)
-	0x93,	// 0xd2 smart double-quote start
-	0x94,	// 0xd3 smart double-quote end
-	0x91,	// 0xd4 smart single-quote start
-	0x92,	// 0xd5 smart single-quote end
-	0xf7,	// 0xd6 divide
-	0xa4,	// 0xd7 diamond (lozenge)  [using spiky circle]
-	0xff,	// 0xd8 y + umlaut
-	// [nothing below here is part of standard Windows-ASCII?]
-	// remaining descriptions based on hfsutils' "charset.txt"
-	kUnk,	// 0xd9 Y + umlaut
-	kUnk,	// 0xda fraction slash
-	kUnk,	// 0xdb currency sign
-	kUnk,	// 0xdc single left-pointing angle quotation mark
-	kUnk,	// 0xdd single right-pointing angle quotation mark
-	kUnk,	// 0xde merged fi
-	kUnk,	// 0xdf merged FL
-	kUnk,	// 0xe0 double dagger
-	kUnk,	// 0xe1 middle dot
-	kUnk,	// 0xe2 single low-9 quotation mark
-	kUnk,	// 0xe3 double low-9 quotation mark
-	kUnk,	// 0xe4 per mille sign
-	kUnk,	// 0xe5 A + circumflex
-	kUnk,	// 0xe6 E + circumflex
-	kUnk,	// 0xe7 A + acute accent
-	kUnk,	// 0xe8 E + diaeresis
-	kUnk,	// 0xe9 E + grave accent
-	kUnk,	// 0xea I + acute accent
-	kUnk,	// 0xeb I + circumflex
-	kUnk,	// 0xec I + diaeresis
-	kUnk,	// 0xed I + grave accent
-	kUnk,	// 0xee O + acute accent
-	kUnk,	// 0xef O + circumflex
-	kUnk,	// 0xf0 apple logo
-	kUnk,	// 0xf1 O + grave accent
-	kUnk,	// 0xf2 U + acute accent
-	kUnk,	// 0xf3 U + circumflex
-	kUnk,	// 0xf4 U + grave accent
-	kUnk,	// 0xf5 i without dot
-	kUnk,	// 0xf6 modifier letter circumflex accent
-	kUnk,	// 0xf7 small tilde
-	kUnk,	// 0xf8 macron
-	kUnk,	// 0xf9 breve
-	kUnk,	// 0xfa dot above
-	kUnk,	// 0xfb ring above
-	kUnk,	// 0xfc cedilla
-	kUnk,	// 0xfd double acute accent
-	kUnk,	// 0xfe ogonek
-	kUnk,	// 0xff caron
-};
-
-/*
- * Quick sanity check on contents of array.
- *
- * No two characters should map to the same thing.  This isn't vital, but
- * if we want to have a reversible transformation someday, it'll make our
- * lives easier then.
- */
-void
-ReformatText::CheckGSCharConv(void)
-{
-#ifdef _DEBUG
-	bool test[256];
-	int i;
-
-	memset(test, 0, sizeof(test));
-	for (i = 0; i < sizeof(kGSCharConv); i++) {
-		if (test[kGSCharConv[i]] && kGSCharConv[i] != kUnk) {
-			WMSG3("Character used twice: 0x%02x at %d (0x%02x)\n",
-				kGSCharConv[i], i, i+128);
-			assert(false);
-		}
-		test[kGSCharConv[i]] = true;
-	}
-#endif
-}
-
-
-/*
- * Set the output format and buffer.
- *
- * Clears our work buffer pointer so we don't free it.
- */
-void
-ReformatText::SetResultBuffer(ReformatOutput* pOutput, bool multiFont)
-{
-	char* buf;
-	long len;
-	fExpBuf.SeizeBuffer(&buf, &len);
-	pOutput->SetTextBuf(buf, len, true);
-
-	if (pOutput->GetTextBuf() == nil) {
-		/*
-		 * Force "raw" mode if there's no output.  This can happen if we,
-		 * say, try to format an empty file as a hex dump.  We never
-		 * produce any output, so no buffer gets allocated.
-		 *
-		 * We set the mode to "raw" so that applications can assume that
-		 * results of type "text" actually have text to look at -- though
-		 * it's possible the length will be zero, we promise that there'll
-		 * be a buffer there.  I'm not sure it's important to do this,
-		 * but it does reduce the #of situations in which we have to
-		 * worry about nil pointers.
-		 */
-		pOutput->SetOutputKind(ReformatOutput::kOutputRaw);
-		WMSG0("ReformatText returning a nil pointer\n");
-	} else {
-		if (fUseRTF)
-			pOutput->SetOutputKind(ReformatOutput::kOutputRTF);
-		else
-			pOutput->SetOutputKind(ReformatOutput::kOutputText);
-	}
-
-	if (fUseRTF && multiFont)
-		pOutput->SetMultipleFontsFlag(true);
-}
-
-
-/*
- * Output the RTF header.
- *
- * The color table is the standard MS Word color table, except that entry
- * #17 (dark grey) has been lightened from (51,51,51) because it's nearly
- * indistinguishable from black on the screen.
- *
- * The default font is Courier New (\f0) at 10 points (\fs20).
- */
-void
-ReformatText::RTFBegin(int flags)
-{
-//	static const char* rtfHdr =
-//"{\\rtf1\\ansi\\ansicpg1252\\deff0\\deflang1033{\\fonttbl{\\f0\\fmodern\\fprq1\\fcharset0 Courier New;}}\r\n"
-//"\\viewkind4\\uc1\\pard\\f0\\fs20 ";
-
-	static const char* rtfHdrStart =
-"{\\rtf1\\ansi\\ansicpg1252\\deff0\\deflang1033\\deflangfe1033{\\fonttbl"
-	"{\\f0\\fmodern\\fprq1\\fcharset0 Courier New;}"
-	"{\\f1\\froman\\fprq2\\fcharset0 Times New Roman;}"
-	"{\\f2\\fswiss\\fprq2\\fcharset0 Arial;}"
-	"{\\f3\\froman\\fprq2\\fcharset2 Symbol;}"
-	"}\r\n";
-
-	static const char* rtfColorTable = 
-"{\\colortbl;"
-	"\\red0\\green0\\blue0;\\red0\\green0\\blue255;\\red0\\green255\\blue255;\\red0\\green255\\blue0;"
-	"\\red255\\green0\\blue255;\\red255\\green0\\blue0;\\red255\\green255\\blue0;\\red255\\green255\\blue255;"
-	"\\red0\\green0\\blue128;\\red0\\green128\\blue128;\\red0\\green128\\blue0;\r\n"
-	"\\red128\\green0\\blue128;\\red128\\green0\\blue0;\\red128\\green128\\blue0;\\red128\\green128\\blue128;"
-	"\\red192\\green192\\blue192;\\red64\\green64\\blue64;\\red255\\green153\\blue0;}\r\n";
-
-	static const char* rtfHdrEnd =
-"\\viewkind4\\uc1\\pard\\f0\\fs20 ";
-
-	if (fUseRTF) {
-		BufPrintf("%s", rtfHdrStart);
-		if ((flags & kRTFFlagColorTable) != 0)
-			BufPrintf("%s", rtfColorTable);
-		BufPrintf("%s", rtfHdrEnd);
-	}
-
-	fPointSize = 10;
-}
-
-/*
- * Output the RTF footer.
- */
-void
-ReformatText::RTFEnd(void) { if (fUseRTF) BufPrintf("}\r\n%c", '\0'); }
-
-/*
- * Output RTF paragraph definition marker.  Do this every time we change some
- * aspect of paragraph formatting, such as margins or justification.
- */
-void
-ReformatText::RTFSetPara(void)
-{
-	if (!fUseRTF)
-		return;
-
-	BufPrintf("\\pard\\nowidctlpar");
-
-	if (fLeftMargin != 0 || fRightMargin != 0) {
-		/* looks like RTF thinks we're getting 12 chars per inch? */
-		if (fLeftMargin != 0)
-			BufPrintf("\\li%d",
-				(int) (fLeftMargin * (kRTFUnitsPerInch/12)));
-		if (fLeftMargin != 0)
-			BufPrintf("\\ri%d",
-				(int) (fRightMargin * (kRTFUnitsPerInch/12)));
-	}
-
-	switch (fJustified) {
-	case kJustifyLeft:							break;
-	case kJustifyRight:		BufPrintf("\\qr");	break;
-	case kJustifyCenter:	BufPrintf("\\qc");	break;
-	case kJustifyFull:		BufPrintf("\\qj");	break;
-	default:
-		assert(false);
-		break;
-	}
-
-	// Ideally we'd suppress this if the next thing is an RTF
-	//  formatting command, esp. "\\par".
-	BufPrintf(" ");
-}
-
-/*
- * Output a new paragraph marker.
- *
- * If you're producing RTF output, this is the right way to output an
- * end-of-line character.
- */
-void
-ReformatText::RTFNewPara(void)
-{
-	if (fUseRTF)
-		BufPrintf("\\par\r\n");
-	else
-		BufPrintf("\r\n");
-}
-
-
-/*
- * Insert a page break.  This isn't supported by the Rich Edit control,
- * so it won't appear in CiderPress or WordPad, but it will come out in
- * Microsoft Word if you extract to a file.
- */
-void
-ReformatText::RTFPageBreak(void)
-{
-	if (fUseRTF)
-		BufPrintf("\\page ");
-}
-
-/*
- * RTF tab character.
- */
-void
-ReformatText::RTFTab(void)
-{
-	if (fUseRTF)
-		BufPrintf("\\tab ");
-}
-
-/*
- * Minor formatting.
- */
-void
-ReformatText::RTFBoldOn(void)
-{
-	if (fBoldEnabled)
-		return;
-	if (fUseRTF) {
-		BufPrintf("\\b ");
-		fBoldEnabled = true;
-	}
-}
-void
-ReformatText::RTFBoldOff(void)
-{
-	if (!fBoldEnabled)
-		return;
-	if (fUseRTF) {
-		BufPrintf("\\b0 ");
-		fBoldEnabled = false;
-	}
-}
-void
-ReformatText::RTFItalicOn(void)
-{
-	if (fItalicEnabled)
-		return;
-	if (fUseRTF) {
-		BufPrintf("\\i ");
-		fItalicEnabled = true;
-	}
-}
-void
-ReformatText::RTFItalicOff(void)
-{
-	if (!fItalicEnabled)
-		return;
-	if (fUseRTF) {
-		BufPrintf("\\i0 ");
-		fItalicEnabled = false;
-	}
-}
-void
-ReformatText::RTFUnderlineOn(void)
-{
-	if (fUnderlineEnabled)
-		return;
-	if (fUseRTF) {
-		BufPrintf("\\ul ");
-		fUnderlineEnabled = true;
-	}
-}
-void
-ReformatText::RTFUnderlineOff(void)
-{
-	if (!fUnderlineEnabled)
-		return;
-	if (fUseRTF) {
-		BufPrintf("\\ulnone ");
-		fUnderlineEnabled = false;
-	}
-}
-void
-ReformatText::RTFSubscriptOn(void)
-{
-	if (fSubscriptEnabled)
-		return;
-	if (fUseRTF) {
-		BufPrintf("\\sub ");
-		fSubscriptEnabled = true;
-	}
-}
-void
-ReformatText::RTFSubscriptOff(void)
-{
-	if (!fSubscriptEnabled)
-		return;
-	if (fUseRTF) {
-		BufPrintf("\\nosupersub ");
-		fSubscriptEnabled = false;
-	}
-}
-void
-ReformatText::RTFSuperscriptOn(void)
-{
-	if (fSuperscriptEnabled)
-		return;
-	if (fUseRTF) {
-		BufPrintf("\\super ");
-		fSuperscriptEnabled = true;
-	}
-}
-void
-ReformatText::RTFSuperscriptOff(void)
-{
-	if (!fSuperscriptEnabled)
-		return;
-	if (fUseRTF) {
-		BufPrintf("\\nosupersub ");
-		fSuperscriptEnabled = false;
-	}
-}
-void
-ReformatText::RTFSetColor(TextColor color)
-{
-	if (color == fTextColor)
-		return;
-	if (fUseRTF) {
-		BufPrintf("\\cf%d ", color);
-		fTextColor = color;
-	}
-}
-
-/*
- * Change paragraph formatting.
- */
-void
-ReformatText::RTFParaLeft(void)
-{
-	if (fJustified != kJustifyLeft) {
-		fJustified = kJustifyLeft;
-		RTFSetPara();
-	}
-}
-void
-ReformatText::RTFParaRight(void)
-{
-	if (fJustified != kJustifyRight) {
-		fJustified = kJustifyRight;
-		RTFSetPara();
-	}
-}
-void
-ReformatText::RTFParaCenter(void)
-{
-	if (fJustified != kJustifyCenter) {
-		fJustified = kJustifyCenter;
-		RTFSetPara();
-	}
-}
-void
-ReformatText::RTFParaJustify(void)
-{
-	if (fJustified != kJustifyFull) {
-		fJustified = kJustifyFull;
-		RTFSetPara();
-	}
-}
-
-/*
- * Page margins, in 1/10th inches.
- */
-void
-ReformatText::RTFLeftMargin(int margin)
-{
-	//WMSG1("+++ Left margin now %d\n", margin);
-	fLeftMargin = margin;
-	RTFSetPara();
-}
-void
-ReformatText::RTFRightMargin(int margin)
-{
-	//WMSG1("+++ Right margin now %d\n", margin);
-	fRightMargin = margin;
-	RTFSetPara();
-}
-
-
-/*
- * Switch to a different font size.
- */
-void
-ReformatText::RTFSetFontSize(int points)
-{
-	if (fUseRTF && fPointSize != points)
-		BufPrintf("\\fs%d ", points * 2);
-	fPointSize = points;
-}
-/*
- * Switch to a different font.
- */
-void
-ReformatText::RTFSetFont(RTFFont font)
-{
-	if (fUseRTF)
-		BufPrintf("\\f%d ", font);
-}
-
-/*
- * Set the font by specifying a IIgs QuickDraw II font family number.
- */
-void
-ReformatText::RTFSetGSFont(unsigned short family)
-{
-	float newMult;
-
-	if (!fUseRTF)
-		return;
-
-	/*
-	 * Apple II fonts seem to be about 1.5x in a WYSIWYG way, except
-	 * for Times, which is about 1:1.
-	 */
-	switch (family) {
-	case kGSFontTimes:
-		RTFSetFont(kFontTimesRoman);
-		newMult = 0.9f;
-		break;
-	case kGSFontNewYork:
-		RTFSetFont(kFontTimesRoman);
-		newMult = 1.1f;
-		break;
-
-	case kGSFontSymbol:
-		RTFSetFont(kFontSymbol);
-		newMult = 1.0f;
-		break;
-
-	case kGSFontMonaco:
-		RTFSetFont(kFontCourierNew);
-		newMult = 0.80f;
-		break;
-	case kGSFontCourier:
-	case kGSFontPCMonospace:
-	case kGSFontAppleM:
-	case kGSFontGenesys:
-		RTFSetFont(kFontCourierNew);
-		newMult = 1.5f;
-		break;
-
-	case kGSFontClassical:
-	case kGSFontGenoa:
-	case kGSFontWestern:
-		RTFSetFont(kFontArial);
-		newMult = 0.80f;
-		break;
-	case kGSFontChicago:
-	case kGSFontVenice:
-	case kGSFontGeneva:
-	case kGSFontStarfleet:
-	case kGSFontUnknown1:
-	case kGSFontUnknown2:
-		RTFSetFont(kFontArial);
-		newMult = 1.0f;
-		break;
-	case kGSFontLondon:
-	case kGSFontAthens:
-	case kGSFontSanFran:
-	case kGSFontShaston:
-	case kGSFontToronto:
-	case kGSFontCairo:
-	case kGSFontLosAngeles:
-	case kGSFontHelvetica:
-	case kGSFontTaliesin:
-		RTFSetFont(kFontArial);
-		newMult = 1.5f;
-		break;
-	default:
-		WMSG1("Unrecognized font family 0x%04x\n", family);
-		RTFSetFont(kFontArial);
-		newMult = 1.0f;
-		break;
-	}
-
-	if (newMult != fGSFontSizeMult) {
-		fGSFontSizeMult = newMult;
-		RTFSetGSFontSize(fPreMultPointSize);
-	}
-}
-
-/*
- * Set the font size of a IIgs font.  We factor the size multiplier in.
- *
- * BUG: we should track the state of the "underline" mode, and turn it
- * on and off based on the font size (8-point fonts aren't underlined).
- */
-void
-ReformatText::RTFSetGSFontSize(int points)
-{
-	RTFSetFontSize((int) roundf(points * fGSFontSizeMult));
-
-	fPreMultPointSize = points;
-}
-
-/*
- * Set bold/italic/underline.  "Teach" ignores you if you try to
- * underline text smaller than 8 points, but if you leave the mode
- * on from a previous block it will act like it wants to underline
- * text but not actually do it.  We have to emulate this behavior,
- * or some documents (e.g. "MZ.MANUAL") look terrible.
- *
- * Set the font size before calling here.
- *
- * Some characters, such as '=' in Shaston 8, look the same in
- * bold as they do in plain.  This doesn't hold true for Windows
- * fonts, so we're going to look different in some circumstances.
- */
-void
-ReformatText::RTFSetGSFontStyle(unsigned char qdStyle)
-{
-	if (!fUseRTF)
-		return;
-
-	if ((qdStyle & kQDStyleBold) != 0)
-		RTFBoldOn();
-	else
-		RTFBoldOff();
-	if ((qdStyle & kQDStyleItalic) != 0)
-		RTFItalicOn();
-	else
-		RTFItalicOff();
-	if ((qdStyle & kQDStyleUnderline) != 0 && fPreMultPointSize > 8)
-		RTFUnderlineOn();
-	else
-		RTFUnderlineOff();
-	if ((qdStyle & kQDStyleSuperscript) != 0)
-		RTFSuperscriptOn();
-	else
-		RTFSuperscriptOff();
-	if ((qdStyle & kQDStyleSubscript) != 0)
-		RTFSubscriptOn();
-	else
-		RTFSubscriptOff();
-}
-
-
-
-#if 0
-void
-ReformatText::RTFProportionalOn(void) {
-	if (fUseRTF)
-		BufPrintf("\\f%d ", kFontTimesRoman);
-}
-void
-ReformatText::RTFProportionalOff(void) {
-	if (fUseRTF)
-		BufPrintf("\\f%d ", kFontCourierNew);
-}
-#endif
-
-
-/*
- * Convert the EOL markers in a buffer.  The output is written to the work
- * buffer.  The input buffer may be CR, LF, or CRLF.
- *
- * If "stripHiBits" is set, the high bit of each character is cleared before
- * the value is considered.
- */
-void
-ReformatText::ConvertEOL(const unsigned char* srcBuf, long srcLen,
-	bool stripHiBits)
-{
-	unsigned char ch;
-	int mask;
-
-	assert(!fUseRTF);	// else we have to use RTFPrintChar
-
-	if (stripHiBits)
-		mask = 0x7f;
-	else
-		mask = 0xff;
-
-	/*
-	 * Could probably speed this up by taking things a line at a time,
-	 * but this is fast enough and much more straightforward.
-	 */
-	while (srcLen) {
-		ch = (*srcBuf++) & mask;
-		srcLen--;
-
-		if (ch == '\r') {
-			/* got CR, check for CRLF */
-			if (srcLen != 0 && ((*srcBuf) & mask) == '\n') {
-				srcBuf++;
-				srcLen--;
-			}
-			BufPrintf("\r\n");
-		} else if (ch == '\n') {
-			BufPrintf("\r\n");
-		} else {
-			BufPrintf("%c", ch);
-		}
-	}
-}
-
-/*
- * Write a hex dump into the buffer.
- */
-void
-ReformatText::BufHexDump(const unsigned char* srcBuf, long srcLen)
-{
-	const unsigned char* origSrcBuf = srcBuf;
-	char chBuf[17];
-	int i, remLen;
-
-	ASSERT(srcBuf != nil);
-	ASSERT(srcLen >= 0);
-
-	chBuf[16] = '\0';
-
-	while (srcLen > 0) {
-		BufPrintf("%08lx: ", srcBuf - origSrcBuf);
-
-		if (srcLen >= 16) {
-			if (!fUseRTF) {
-				/* the really easy (and relatively fast) way */
-				BufPrintf("%02x %02x %02x %02x %02x %02x %02x %02x "
-						  "%02x %02x %02x %02x %02x %02x %02x %02x ",
-					srcBuf[0], srcBuf[1], srcBuf[2], srcBuf[3],
-					srcBuf[4], srcBuf[5], srcBuf[6], srcBuf[7],
-					srcBuf[8], srcBuf[9], srcBuf[10], srcBuf[11],
-					srcBuf[12], srcBuf[13], srcBuf[14], srcBuf[15]);
-			} else {
-				/* the fairly easy (and fairly fast) way */
-				RTFBoldOn();
-				BufPrintf("%02x %02x %02x %02x ",
-					srcBuf[0], srcBuf[1], srcBuf[2], srcBuf[3]);
-				RTFBoldOff();
-				BufPrintf("%02x %02x %02x %02x ",
-					srcBuf[4], srcBuf[5], srcBuf[6], srcBuf[7]);
-				RTFBoldOn();
-				BufPrintf("%02x %02x %02x %02x ",
-					srcBuf[8], srcBuf[9], srcBuf[10], srcBuf[11]);
-				RTFBoldOff();
-				BufPrintf("%02x %02x %02x %02x ",
-					srcBuf[12], srcBuf[13], srcBuf[14], srcBuf[15]);
-			}
-		} else {
-			/* the not-so-easy (and not-so-fast) way */
-			remLen = srcLen;
-
-			for (i = 0; i < remLen; i++) {
-				if (i == 0 || i == 8)
-					RTFBoldOn();
-				else if (i == 4 || i == 12)
-					RTFBoldOff();
-				BufPrintf("%02x ", srcBuf[i]);
-			}
-			RTFBoldOff();
-			for ( ; i < 16; i++)
-				BufPrintf("   ");
-
-			/* blank out the char buf, since we're only filling part in */
-			for (i = 0; i < 16; i++)
-				chBuf[i] = ' ';
-		}
-
-		bool hosed = false;
-		remLen = srcLen;
-		if (remLen > 16)
-			remLen = 16;
-		int i;
-		for (i = 0; i < remLen; i++) {
-			chBuf[i] = PrintableChar(srcBuf[i]);
-			if (fUseRTF &&
-				(chBuf[i] == '\\' || chBuf[i] == '{' || chBuf[i] == '}'))
-			{
-				hosed = true;
-				break;
-			}
-		}
-
-		if (!hosed) {
-			BufPrintf(" %s", chBuf);
-		} else {
-			/* escaped chars in RTF mode; have to do this one the hard way */
-			ASSERT(fUseRTF);
-			BufPrintf(" ");
-			for (i = 0; i < remLen; i++) {
-				RTFPrintChar(srcBuf[i]);
-			}
-		}
-
-		RTFNewPara();
-
-		srcBuf += 16;
-		srcLen -= 16;
-	}
-}
-
-
-/*
- * ==========================================================================
- *		ReformatGraphics
- * ==========================================================================
- */
-
-/*
- * Initialize the Apple II color palette, used for Hi-Res and DHR
- * conversions.  Could also be used for lo-res mode.
- */
-void
-ReformatGraphics::InitPalette(void)
-{
-	ASSERT(kPaletteSize == 16);
-
-	static const RGBQUAD stdPalette[kPaletteSize] = {
-		/* blue, green, red, reserved */
-		{ 0x00, 0x00, 0x00 },	// $0 black
-		{ 0x33, 0x00, 0xdd },	// $1 red (magenta)
-		{ 0x99, 0x00, 0x00 },	// $2 dark blue
-		{ 0xdd, 0x22, 0xdd },	// $3 purple (violet)
-		{ 0x22, 0x77, 0x00 },	// $4 dark green
-		{ 0x55, 0x55, 0x55 },	// $5 grey1 (dark)
-		{ 0xff, 0x22, 0x22 },	// $6 medium blue
-		{ 0xff, 0xaa, 0x66 },	// $7 light blue
-		{ 0x00, 0x55, 0x88 },	// $8 brown
-		{ 0x00, 0x66, 0xff },	// $9 orange
-		{ 0xaa, 0xaa, 0xaa },	// $A grey2 (light)
-		{ 0x88, 0x99, 0xff },	// $B pink
-		{ 0x00, 0xdd, 0x11 },	// $C green (a/k/a light green)
-		{ 0x00, 0xff, 0xff },	// $D yellow
-		{ 0x99, 0xff, 0x44 },	// $E aqua
-		{ 0xff, 0xff, 0xff },	// $F white
-	};
-
-	memcpy(fPalette, stdPalette, sizeof(fPalette));
-}
-
-/*
- * Stuff out DIB into the output fields, and set the appropriate flags.
- */
-void
-ReformatGraphics::SetResultBuffer(ReformatOutput* pOutput, MyDIBitmap* pDib)
-{
-	ASSERT(pOutput != nil);
-	ASSERT(pDib != nil);
-	pOutput->SetOutputKind(ReformatOutput::kOutputBitmap);
-	pOutput->SetDIB(pDib);
-}
-
-
-/*
- * Unpack the Apple PackBytes format.
- *
- * Format is:
- *	<flag><data> ...
- *
- * Flag values (first 6 bits of flag byte):
- *	00xxxxxx: (0-63) 1 to 64 bytes follow, all different
- *	01xxxxxx: (0-63) 1 to 64 repeats of next byte
- *	10xxxxxx: (0-63) 1 to 64 repeats of next 4 bytes
- *	11xxxxxx: (0-63) 1 to 64 repeats of next byte taken as 4 bytes
- *				(as in 10xxxxxx case)
- *
- * Pass the destination buffer in "dst", source buffer in "src", source
- * length in "srcLen", and expected sizes of output in "dstRem".
- *
- * Returns 0 on success, nonzero if the buffer is overfilled or underfilled.
- */
-int
-ReformatGraphics::UnpackBytes(unsigned char* dst, const unsigned char* src,
-	long dstRem, long srcLen)
-{
-	while (srcLen > 0) {
-		unsigned char flag = *src++;
-		int count = (flag & 0x3f) +1;
-		unsigned char val;
-		unsigned char valSet[4];
-		int i;
-
-		srcLen--;
-	
-		switch (flag & 0xc0) {
-		case 0x00:
-			for (i = 0; i < count; i++) {
-				if (srcLen == 0 || dstRem == 0) {
-					WMSG2(" SHR unpack overrun1 (srcLen=%ld dstRem=%ld)\n",
-						srcLen, dstRem);
-					return -1;
-				}
-				*dst++ = *src++;
-				srcLen--;
-				dstRem--;
-			}
-			break;
-		case 0x40:
-			//if (count != 3 || count != 5 || count != 6 || count != 7) {
-			//	WMSG1(" SHR unpack funky len %d?\n", count);
-			//}
-			if (srcLen == 0) {
-				WMSG0(" SHR unpack underrun2\n");
-				return -1;
-			}
-			val = *src++;
-			srcLen--;
-			for (i = 0; i < count; i++) {
-				if (dstRem == 0) {
-					WMSG3(" SHR unpack overrun2 (srcLen=%d, i=%d of %d)\n",
-						srcLen, i, count);
-					return -1;
-				}
-				*dst++ = val;
-				dstRem--;
-			}
-			break;
-		case 0x80:
-			if (srcLen < 4) {
-				WMSG0(" SHR unpack underrun3\n");
-				return -1;
-			}
-			valSet[0] = *src++;
-			valSet[1] = *src++;
-			valSet[2] = *src++;
-			valSet[3] = *src++;
-			srcLen -= 4;
-			for (i = 0; i < count; i++) {
-				if (dstRem < 4) {
-					WMSG2(" SHR unpack overrun3 (srcLen=%ld dstRem=%ld)\n",
-						srcLen, dstRem);
-					return -1;
-				}
-				*dst++ = valSet[0];
-				*dst++ = valSet[1];
-				*dst++ = valSet[2];
-				*dst++ = valSet[3];
-				dstRem -= 4;
-			}
-			break;
-		case 0xc0:
-			if (srcLen == 0) {
-				WMSG0(" SHR unpack underrun4\n");
-				return -1;
-			}
-			val = *src++;
-			srcLen--;
-			for (i = 0; i < count; i++) {
-				if (dstRem < 4) {
-					WMSG3(" SHR unpack overrun4 (srcLen=%ld dstRem=%ld count=%d)\n",
-						srcLen, dstRem, count);
-					return -1;
-				}
-				*dst++ = val;
-				*dst++ = val;
-				*dst++ = val;
-				*dst++ = val;
-				dstRem -= 4;
-			}
-			break;
-		default:
-			ASSERT(false);
-			break;
-		}
-	}
-
-	ASSERT(srcLen == 0);
-
-	/* require that we completely fill the buffer */
-	if (dstRem != 0) {
-		WMSG1(" SHR unpack dstRem at %d\n", dstRem);
-		return -1;
-	}
-
-	return 0;
-}
-
-
-/*
- * Unpack Macintosh PackBits format.  See Technical Note TN1023.
- *
- * Read a byte.
- * If the high bit is set, count is 2s complement +1 (i.e. count = (-byte)+1).
- *   Read the next byte, then write that byte 'count' times.
- * If the high bit is clear, count is 1+value (i.e. count = byte+1).  Read and
- *   copy that many bytes.
- * After "destLen" bytes have been written, return (even if in the middle of
- * a run).
- *
- * NOTE: if the count byte is 0x80, Apple says it's an invalid value and
- * should be skipped over.  Use the following byte as the count byte.  This
- * is probably because PackBits is only supposed to crunch 127 bytes, though
- * that suggests 0x81 and 0x7f are also impossible.
- *
- * We have to watch for underruns on the input and overruns on the output.
- */
-void
-ReformatGraphics::UnPackBits(const unsigned char** pSrcBuf, long* pSrcLen,
-	unsigned char** pOutPtr, long dstLen, unsigned char xorVal)
-{
-	const unsigned char* srcBuf = *pSrcBuf;
-	long length = *pSrcLen;
-	unsigned char* outPtr = *pOutPtr;
-	int pixByte = 0;
-
-	while (pixByte < dstLen && length > 0) {
-		unsigned char countByte;
-		int count;
-		
-		countByte = *srcBuf++;
-		length--;
-		if (countByte & 0x80) {
-			/* RLE string */
-			unsigned char ch;
-			count = (countByte ^ 0xff)+1 +1;
-			ch = *srcBuf++;
-			length--;
-			while (count-- && pixByte < dstLen) {
-				*outPtr++ = ch ^ xorVal;
-				pixByte++;
-			}
-		} else {
-			/* series of bytse */
-			count = countByte +1;
-			while (count && pixByte < dstLen && length > 0) {
-				*outPtr++ = *srcBuf++ ^ xorVal;
-				count--;
-				length--;
-				pixByte++;
-			}
-		}
-	}
-	if (pixByte != 72) {
-		/* can happen if we run out of input early */
-		WMSG1("  MP unexpected pixByte=%d\n", pixByte);
-		/* keep going */
-	}
-
-	*pSrcBuf = srcBuf;
-	*pSrcLen = length;
-	*pOutPtr = outPtr;
-}
+/*
+ * CiderPress
+ * Copyright (C) 2007 by faddenSoft, LLC.  All Rights Reserved.
+ * See the file LICENSE for distribution terms.
+ */
+/*
+ * Reformatter base class implementation.
+ */
+#include "StdAfx.h"
+#include "ReformatBase.h"
+#include <math.h>
+
+
+static float roundf(float val)
+{
+	if (val < 0.5f)
+		return ::floorf(val);
+	else
+		return ::ceilf(val);
+}
+
+/*
+ * ==========================================================================
+ *		ReformatText
+ * ==========================================================================
+ */
+
+/*
+ * Convert IIgs high-ASCII characters to Windows equivalents (when
+ * available).
+ *
+ * Also found this:
+ * http://www.unicode.org/Public/MAPPINGS/VENDORS/APPLE/ROMAN.TXT
+ */
+const int kUnk = 0x3f;		// for unmappable chars, use '?'
+
+/*static*/ const unsigned char ReformatText::kGSCharConv[128] = {
+	0xc4,	// 0x80 A + umlaut (diaeresis?)
+	0xc5,	// 0x81 A + overcircle
+	0xc7,	// 0x82 C + cedilla
+	0xc9,	// 0x83 E + acute
+	0xd1,	// 0x84 N + tilde
+	0xd6,	// 0x85 O + umlaut
+	0xdc,	// 0x86 U + umlaut
+	0xe1,	// 0x87 a + acute
+	0xe0,	// 0x88 a + grave
+	0xe2,	// 0x89 a + circumflex
+	0xe4,	// 0x8a a + umlaut
+	0xe3,	// 0x8b a + tilde
+	0xe5,	// 0x8c a + overcircle
+	0xe7,	// 0x8d c + cedilla
+	0xe9,	// 0x8e e + acute
+	0xe8,	// 0x8f e + grave
+	0xea,	// 0x90 e + circumflex
+	0xeb,	// 0x91 e + umlaut
+	0xed,	// 0x92 i + acute
+	0xec,	// 0x93 i + grave
+	0xee,	// 0x94 i + circumflex
+	0xef,	// 0x95 i + umlaut
+	0xf1,	// 0x96 n + tilde
+	0xf3,	// 0x97 o + acute
+	0xf2,	// 0x98 o + grave
+	0xf4,	// 0x99 o + circumflex
+	0xf6,	// 0x9a o + umlaut
+	0xf5,	// 0x9b o + tilde
+	0xfa,	// 0x9c u + acute
+	0xf9,	// 0x9d u + grave
+	0xfb,	// 0x9e u + circumflex
+	0xfc,	// 0x9f u + umlaut
+	0x87,	// 0xa0 double cross (dagger)
+	0xb0,	// 0xa1 degrees
+	0xa2,	// 0xa2 cents
+	0xa3,	// 0xa3 pounds (UK$)
+	0xa7,	// 0xa4 section start
+	0x95,	// 0xa5 small square (bullet)  [using fat bullet]
+	0xb6,	// 0xa6 paragraph (pilcrow)
+	0xdf,	// 0xa7 curly B (latin small letter sharp S)
+	0xae,	// 0xa8 raised 'R' (registered)
+	0xa9,	// 0xa9 raised 'C' (copyright)
+	0x99,	// 0xaa raised 'TM' (trademark)
+	0xb4,	// 0xab acute accent
+	0xa8,	// 0xac umlaut (diaeresis)
+	kUnk,	// 0xad not-equal
+	0xc6,	// 0xae merged AE
+	0xd8,	// 0xaf O + slash (upper-case nil?)
+	kUnk,	// 0xb0 infinity
+	0xb1,	// 0xb1 +/-
+	kUnk,	// 0xb2 <=
+	kUnk,	// 0xb3 >=
+	0xa5,	// 0xb4 Yen (Japan$)
+	0xb5,	// 0xb5 mu (micro)
+	kUnk,	// 0xb6 delta (partial differentiation) [could use D-bar 0xd0]
+	kUnk,	// 0xb7 epsilon (N-ary summation) [could use C-double-bar 0x80]
+	kUnk,	// 0xb8 PI (N-ary product)
+	kUnk,	// 0xb9 pi
+	kUnk,	// 0xba integral
+	0xaa,	// 0xbb a underbar (feminine ordinal)  [using raised a]
+	0xba,	// 0xbc o underbar (masculine ordinal)  [using raised o]
+	kUnk,	// 0xbd omega (Ohm)
+	0xe6,	// 0xbe merged ae
+	0xf8,	// 0xbf o + slash (lower-case nil?)
+	0xbf,	// 0xc0 upside-down question mark
+	0xa1,	// 0xc1 upside-down exclamation point
+	0xac,	// 0xc2 rotated L ("not" sign)
+	0xb7,	// 0xc3 checkmark (square root) [using small bullet]
+	0x83,	// 0xc4 script f
+	kUnk,	// 0xc5 approximately equal
+	kUnk,	// 0xc6 delta (triangle / increment)
+	0xab,	// 0xc7 much less than
+	0xbb,	// 0xc8 much greater than
+	0x85,	// 0xc9 ellipsis
+	0xa0,	// 0xca blank (sticky space)
+	0xc0,	// 0xcb A + grave
+	0xc3,	// 0xcc A + tilde
+	0xd5,	// 0xcd O + tilde
+	0x8c,	// 0xce merged OE
+	0x9c,	// 0xcf merged oe
+	0x96,	// 0xd0 short hyphen (en dash)
+	0x97,	// 0xd1 long hyphen (em dash)
+	0x93,	// 0xd2 smart double-quote start
+	0x94,	// 0xd3 smart double-quote end
+	0x91,	// 0xd4 smart single-quote start
+	0x92,	// 0xd5 smart single-quote end
+	0xf7,	// 0xd6 divide
+	0xa4,	// 0xd7 diamond (lozenge)  [using spiky circle]
+	0xff,	// 0xd8 y + umlaut
+	// [nothing below here is part of standard Windows-ASCII?]
+	// remaining descriptions based on hfsutils' "charset.txt"
+	kUnk,	// 0xd9 Y + umlaut
+	kUnk,	// 0xda fraction slash
+	kUnk,	// 0xdb currency sign
+	kUnk,	// 0xdc single left-pointing angle quotation mark
+	kUnk,	// 0xdd single right-pointing angle quotation mark
+	kUnk,	// 0xde merged fi
+	kUnk,	// 0xdf merged FL
+	kUnk,	// 0xe0 double dagger
+	kUnk,	// 0xe1 middle dot
+	kUnk,	// 0xe2 single low-9 quotation mark
+	kUnk,	// 0xe3 double low-9 quotation mark
+	kUnk,	// 0xe4 per mille sign
+	kUnk,	// 0xe5 A + circumflex
+	kUnk,	// 0xe6 E + circumflex
+	kUnk,	// 0xe7 A + acute accent
+	kUnk,	// 0xe8 E + diaeresis
+	kUnk,	// 0xe9 E + grave accent
+	kUnk,	// 0xea I + acute accent
+	kUnk,	// 0xeb I + circumflex
+	kUnk,	// 0xec I + diaeresis
+	kUnk,	// 0xed I + grave accent
+	kUnk,	// 0xee O + acute accent
+	kUnk,	// 0xef O + circumflex
+	kUnk,	// 0xf0 apple logo
+	kUnk,	// 0xf1 O + grave accent
+	kUnk,	// 0xf2 U + acute accent
+	kUnk,	// 0xf3 U + circumflex
+	kUnk,	// 0xf4 U + grave accent
+	kUnk,	// 0xf5 i without dot
+	kUnk,	// 0xf6 modifier letter circumflex accent
+	kUnk,	// 0xf7 small tilde
+	kUnk,	// 0xf8 macron
+	kUnk,	// 0xf9 breve
+	kUnk,	// 0xfa dot above
+	kUnk,	// 0xfb ring above
+	kUnk,	// 0xfc cedilla
+	kUnk,	// 0xfd double acute accent
+	kUnk,	// 0xfe ogonek
+	kUnk,	// 0xff caron
+};
+
+/*
+ * Quick sanity check on contents of array.
+ *
+ * No two characters should map to the same thing.  This isn't vital, but
+ * if we want to have a reversible transformation someday, it'll make our
+ * lives easier then.
+ */
+void
+ReformatText::CheckGSCharConv(void)
+{
+#ifdef _DEBUG
+	bool test[256];
+	int i;
+
+	memset(test, 0, sizeof(test));
+	for (i = 0; i < sizeof(kGSCharConv); i++) {
+		if (test[kGSCharConv[i]] && kGSCharConv[i] != kUnk) {
+			WMSG3("Character used twice: 0x%02x at %d (0x%02x)\n",
+				kGSCharConv[i], i, i+128);
+			assert(false);
+		}
+		test[kGSCharConv[i]] = true;
+	}
+#endif
+}
+
+
+/*
+ * Set the output format and buffer.
+ *
+ * Clears our work buffer pointer so we don't free it.
+ */
+void
+ReformatText::SetResultBuffer(ReformatOutput* pOutput, bool multiFont)
+{
+	char* buf;
+	long len;
+	fExpBuf.SeizeBuffer(&buf, &len);
+	pOutput->SetTextBuf(buf, len, true);
+
+	if (pOutput->GetTextBuf() == nil) {
+		/*
+		 * Force "raw" mode if there's no output.  This can happen if we,
+		 * say, try to format an empty file as a hex dump.  We never
+		 * produce any output, so no buffer gets allocated.
+		 *
+		 * We set the mode to "raw" so that applications can assume that
+		 * results of type "text" actually have text to look at -- though
+		 * it's possible the length will be zero, we promise that there'll
+		 * be a buffer there.  I'm not sure it's important to do this,
+		 * but it does reduce the #of situations in which we have to
+		 * worry about nil pointers.
+		 */
+		pOutput->SetOutputKind(ReformatOutput::kOutputRaw);
+		WMSG0("ReformatText returning a nil pointer\n");
+	} else {
+		if (fUseRTF)
+			pOutput->SetOutputKind(ReformatOutput::kOutputRTF);
+		else
+			pOutput->SetOutputKind(ReformatOutput::kOutputText);
+	}
+
+	if (fUseRTF && multiFont)
+		pOutput->SetMultipleFontsFlag(true);
+}
+
+
+/*
+ * Output the RTF header.
+ *
+ * The color table is the standard MS Word color table, except that entry
+ * #17 (dark grey) has been lightened from (51,51,51) because it's nearly
+ * indistinguishable from black on the screen.
+ *
+ * The default font is Courier New (\f0) at 10 points (\fs20).
+ */
+void
+ReformatText::RTFBegin(int flags)
+{
+//	static const char* rtfHdr =
+//"{\\rtf1\\ansi\\ansicpg1252\\deff0\\deflang1033{\\fonttbl{\\f0\\fmodern\\fprq1\\fcharset0 Courier New;}}\r\n"
+//"\\viewkind4\\uc1\\pard\\f0\\fs20 ";
+
+	static const char* rtfHdrStart =
+"{\\rtf1\\ansi\\ansicpg1252\\deff0\\deflang1033\\deflangfe1033{\\fonttbl"
+	"{\\f0\\fmodern\\fprq1\\fcharset0 Courier New;}"
+	"{\\f1\\froman\\fprq2\\fcharset0 Times New Roman;}"
+	"{\\f2\\fswiss\\fprq2\\fcharset0 Arial;}"
+	"{\\f3\\froman\\fprq2\\fcharset2 Symbol;}"
+	"}\r\n";
+
+	static const char* rtfColorTable = 
+"{\\colortbl;"
+	"\\red0\\green0\\blue0;\\red0\\green0\\blue255;\\red0\\green255\\blue255;\\red0\\green255\\blue0;"
+	"\\red255\\green0\\blue255;\\red255\\green0\\blue0;\\red255\\green255\\blue0;\\red255\\green255\\blue255;"
+	"\\red0\\green0\\blue128;\\red0\\green128\\blue128;\\red0\\green128\\blue0;\r\n"
+	"\\red128\\green0\\blue128;\\red128\\green0\\blue0;\\red128\\green128\\blue0;\\red128\\green128\\blue128;"
+	"\\red192\\green192\\blue192;\\red64\\green64\\blue64;\\red255\\green153\\blue0;}\r\n";
+
+	static const char* rtfHdrEnd =
+"\\viewkind4\\uc1\\pard\\f0\\fs20 ";
+
+	if (fUseRTF) {
+		BufPrintf("%s", rtfHdrStart);
+		if ((flags & kRTFFlagColorTable) != 0)
+			BufPrintf("%s", rtfColorTable);
+		BufPrintf("%s", rtfHdrEnd);
+	}
+
+	fPointSize = 10;
+}
+
+/*
+ * Output the RTF footer.
+ */
+void
+ReformatText::RTFEnd(void) { if (fUseRTF) BufPrintf("}\r\n%c", '\0'); }
+
+/*
+ * Output RTF paragraph definition marker.  Do this every time we change some
+ * aspect of paragraph formatting, such as margins or justification.
+ */
+void
+ReformatText::RTFSetPara(void)
+{
+	if (!fUseRTF)
+		return;
+
+	BufPrintf("\\pard\\nowidctlpar");
+
+	if (fLeftMargin != 0 || fRightMargin != 0) {
+		/* looks like RTF thinks we're getting 12 chars per inch? */
+		if (fLeftMargin != 0)
+			BufPrintf("\\li%d",
+				(int) (fLeftMargin * (kRTFUnitsPerInch/12)));
+		if (fLeftMargin != 0)
+			BufPrintf("\\ri%d",
+				(int) (fRightMargin * (kRTFUnitsPerInch/12)));
+	}
+
+	switch (fJustified) {
+	case kJustifyLeft:							break;
+	case kJustifyRight:		BufPrintf("\\qr");	break;
+	case kJustifyCenter:	BufPrintf("\\qc");	break;
+	case kJustifyFull:		BufPrintf("\\qj");	break;
+	default:
+		assert(false);
+		break;
+	}
+
+	// Ideally we'd suppress this if the next thing is an RTF
+	//  formatting command, esp. "\\par".
+	BufPrintf(" ");
+}
+
+/*
+ * Output a new paragraph marker.
+ *
+ * If you're producing RTF output, this is the right way to output an
+ * end-of-line character.
+ */
+void
+ReformatText::RTFNewPara(void)
+{
+	if (fUseRTF)
+		BufPrintf("\\par\r\n");
+	else
+		BufPrintf("\r\n");
+}
+
+
+/*
+ * Insert a page break.  This isn't supported by the Rich Edit control,
+ * so it won't appear in CiderPress or WordPad, but it will come out in
+ * Microsoft Word if you extract to a file.
+ */
+void
+ReformatText::RTFPageBreak(void)
+{
+	if (fUseRTF)
+		BufPrintf("\\page ");
+}
+
+/*
+ * RTF tab character.
+ */
+void
+ReformatText::RTFTab(void)
+{
+	if (fUseRTF)
+		BufPrintf("\\tab ");
+}
+
+/*
+ * Minor formatting.
+ */
+void
+ReformatText::RTFBoldOn(void)
+{
+	if (fBoldEnabled)
+		return;
+	if (fUseRTF) {
+		BufPrintf("\\b ");
+		fBoldEnabled = true;
+	}
+}
+void
+ReformatText::RTFBoldOff(void)
+{
+	if (!fBoldEnabled)
+		return;
+	if (fUseRTF) {
+		BufPrintf("\\b0 ");
+		fBoldEnabled = false;
+	}
+}
+void
+ReformatText::RTFItalicOn(void)
+{
+	if (fItalicEnabled)
+		return;
+	if (fUseRTF) {
+		BufPrintf("\\i ");
+		fItalicEnabled = true;
+	}
+}
+void
+ReformatText::RTFItalicOff(void)
+{
+	if (!fItalicEnabled)
+		return;
+	if (fUseRTF) {
+		BufPrintf("\\i0 ");
+		fItalicEnabled = false;
+	}
+}
+void
+ReformatText::RTFUnderlineOn(void)
+{
+	if (fUnderlineEnabled)
+		return;
+	if (fUseRTF) {
+		BufPrintf("\\ul ");
+		fUnderlineEnabled = true;
+	}
+}
+void
+ReformatText::RTFUnderlineOff(void)
+{
+	if (!fUnderlineEnabled)
+		return;
+	if (fUseRTF) {
+		BufPrintf("\\ulnone ");
+		fUnderlineEnabled = false;
+	}
+}
+void
+ReformatText::RTFSubscriptOn(void)
+{
+	if (fSubscriptEnabled)
+		return;
+	if (fUseRTF) {
+		BufPrintf("\\sub ");
+		fSubscriptEnabled = true;
+	}
+}
+void
+ReformatText::RTFSubscriptOff(void)
+{
+	if (!fSubscriptEnabled)
+		return;
+	if (fUseRTF) {
+		BufPrintf("\\nosupersub ");
+		fSubscriptEnabled = false;
+	}
+}
+void
+ReformatText::RTFSuperscriptOn(void)
+{
+	if (fSuperscriptEnabled)
+		return;
+	if (fUseRTF) {
+		BufPrintf("\\super ");
+		fSuperscriptEnabled = true;
+	}
+}
+void
+ReformatText::RTFSuperscriptOff(void)
+{
+	if (!fSuperscriptEnabled)
+		return;
+	if (fUseRTF) {
+		BufPrintf("\\nosupersub ");
+		fSuperscriptEnabled = false;
+	}
+}
+void
+ReformatText::RTFSetColor(TextColor color)
+{
+	if (color == fTextColor)
+		return;
+	if (fUseRTF) {
+		BufPrintf("\\cf%d ", color);
+		fTextColor = color;
+	}
+}
+
+/*
+ * Change paragraph formatting.
+ */
+void
+ReformatText::RTFParaLeft(void)
+{
+	if (fJustified != kJustifyLeft) {
+		fJustified = kJustifyLeft;
+		RTFSetPara();
+	}
+}
+void
+ReformatText::RTFParaRight(void)
+{
+	if (fJustified != kJustifyRight) {
+		fJustified = kJustifyRight;
+		RTFSetPara();
+	}
+}
+void
+ReformatText::RTFParaCenter(void)
+{
+	if (fJustified != kJustifyCenter) {
+		fJustified = kJustifyCenter;
+		RTFSetPara();
+	}
+}
+void
+ReformatText::RTFParaJustify(void)
+{
+	if (fJustified != kJustifyFull) {
+		fJustified = kJustifyFull;
+		RTFSetPara();
+	}
+}
+
+/*
+ * Page margins, in 1/10th inches.
+ */
+void
+ReformatText::RTFLeftMargin(int margin)
+{
+	//WMSG1("+++ Left margin now %d\n", margin);
+	fLeftMargin = margin;
+	RTFSetPara();
+}
+void
+ReformatText::RTFRightMargin(int margin)
+{
+	//WMSG1("+++ Right margin now %d\n", margin);
+	fRightMargin = margin;
+	RTFSetPara();
+}
+
+
+/*
+ * Switch to a different font size.
+ */
+void
+ReformatText::RTFSetFontSize(int points)
+{
+	if (fUseRTF && fPointSize != points)
+		BufPrintf("\\fs%d ", points * 2);
+	fPointSize = points;
+}
+/*
+ * Switch to a different font.
+ */
+void
+ReformatText::RTFSetFont(RTFFont font)
+{
+	if (fUseRTF)
+		BufPrintf("\\f%d ", font);
+}
+
+/*
+ * Set the font by specifying a IIgs QuickDraw II font family number.
+ */
+void
+ReformatText::RTFSetGSFont(unsigned short family)
+{
+	float newMult;
+
+	if (!fUseRTF)
+		return;
+
+	/*
+	 * Apple II fonts seem to be about 1.5x in a WYSIWYG way, except
+	 * for Times, which is about 1:1.
+	 */
+	switch (family) {
+	case kGSFontTimes:
+		RTFSetFont(kFontTimesRoman);
+		newMult = 0.9f;
+		break;
+	case kGSFontNewYork:
+		RTFSetFont(kFontTimesRoman);
+		newMult = 1.1f;
+		break;
+
+	case kGSFontSymbol:
+		RTFSetFont(kFontSymbol);
+		newMult = 1.0f;
+		break;
+
+	case kGSFontMonaco:
+		RTFSetFont(kFontCourierNew);
+		newMult = 0.80f;
+		break;
+	case kGSFontCourier:
+	case kGSFontPCMonospace:
+	case kGSFontAppleM:
+	case kGSFontGenesys:
+		RTFSetFont(kFontCourierNew);
+		newMult = 1.5f;
+		break;
+
+	case kGSFontClassical:
+	case kGSFontGenoa:
+	case kGSFontWestern:
+		RTFSetFont(kFontArial);
+		newMult = 0.80f;
+		break;
+	case kGSFontChicago:
+	case kGSFontVenice:
+	case kGSFontGeneva:
+	case kGSFontStarfleet:
+	case kGSFontUnknown1:
+	case kGSFontUnknown2:
+		RTFSetFont(kFontArial);
+		newMult = 1.0f;
+		break;
+	case kGSFontLondon:
+	case kGSFontAthens:
+	case kGSFontSanFran:
+	case kGSFontShaston:
+	case kGSFontToronto:
+	case kGSFontCairo:
+	case kGSFontLosAngeles:
+	case kGSFontHelvetica:
+	case kGSFontTaliesin:
+		RTFSetFont(kFontArial);
+		newMult = 1.5f;
+		break;
+	default:
+		WMSG1("Unrecognized font family 0x%04x\n", family);
+		RTFSetFont(kFontArial);
+		newMult = 1.0f;
+		break;
+	}
+
+	if (newMult != fGSFontSizeMult) {
+		fGSFontSizeMult = newMult;
+		RTFSetGSFontSize(fPreMultPointSize);
+	}
+}
+
+/*
+ * Set the font size of a IIgs font.  We factor the size multiplier in.
+ *
+ * BUG: we should track the state of the "underline" mode, and turn it
+ * on and off based on the font size (8-point fonts aren't underlined).
+ */
+void
+ReformatText::RTFSetGSFontSize(int points)
+{
+	RTFSetFontSize((int) roundf(points * fGSFontSizeMult));
+
+	fPreMultPointSize = points;
+}
+
+/*
+ * Set bold/italic/underline.  "Teach" ignores you if you try to
+ * underline text smaller than 8 points, but if you leave the mode
+ * on from a previous block it will act like it wants to underline
+ * text but not actually do it.  We have to emulate this behavior,
+ * or some documents (e.g. "MZ.MANUAL") look terrible.
+ *
+ * Set the font size before calling here.
+ *
+ * Some characters, such as '=' in Shaston 8, look the same in
+ * bold as they do in plain.  This doesn't hold true for Windows
+ * fonts, so we're going to look different in some circumstances.
+ */
+void
+ReformatText::RTFSetGSFontStyle(unsigned char qdStyle)
+{
+	if (!fUseRTF)
+		return;
+
+	if ((qdStyle & kQDStyleBold) != 0)
+		RTFBoldOn();
+	else
+		RTFBoldOff();
+	if ((qdStyle & kQDStyleItalic) != 0)
+		RTFItalicOn();
+	else
+		RTFItalicOff();
+	if ((qdStyle & kQDStyleUnderline) != 0 && fPreMultPointSize > 8)
+		RTFUnderlineOn();
+	else
+		RTFUnderlineOff();
+	if ((qdStyle & kQDStyleSuperscript) != 0)
+		RTFSuperscriptOn();
+	else
+		RTFSuperscriptOff();
+	if ((qdStyle & kQDStyleSubscript) != 0)
+		RTFSubscriptOn();
+	else
+		RTFSubscriptOff();
+}
+
+
+
+#if 0
+void
+ReformatText::RTFProportionalOn(void) {
+	if (fUseRTF)
+		BufPrintf("\\f%d ", kFontTimesRoman);
+}
+void
+ReformatText::RTFProportionalOff(void) {
+	if (fUseRTF)
+		BufPrintf("\\f%d ", kFontCourierNew);
+}
+#endif
+
+
+/*
+ * Convert the EOL markers in a buffer.  The output is written to the work
+ * buffer.  The input buffer may be CR, LF, or CRLF.
+ *
+ * If "stripHiBits" is set, the high bit of each character is cleared before
+ * the value is considered.
+ */
+void
+ReformatText::ConvertEOL(const unsigned char* srcBuf, long srcLen,
+	bool stripHiBits)
+{
+	/* Compatibility - assume we're not stripping nulls */
+	ConvertEOL(srcBuf, srcLen, stripHiBits, false);
+}
+
+/*
+ * Convert the EOL markers in a buffer.  The output is written to the work
+ * buffer.  The input buffer may be CR, LF, or CRLF.
+ *
+ * If "stripHiBits" is set, the high bit of each character is cleared before
+ * the value is considered.
+ *
+ * If "stripNulls" is true, no null values will make it through.
+ */
+void
+ReformatText::ConvertEOL(const unsigned char* srcBuf, long srcLen,
+	bool stripHiBits, bool stripNulls)
+{
+	unsigned char ch;
+	int mask;
+
+	assert(!fUseRTF);	// else we have to use RTFPrintChar
+
+	if (stripHiBits)
+		mask = 0x7f;
+	else
+		mask = 0xff;
+
+	/*
+	 * Could probably speed this up by taking things a line at a time,
+	 * but this is fast enough and much more straightforward.
+	 */
+	while (srcLen) {
+		ch = (*srcBuf++) & mask;
+		srcLen--;
+
+		if (ch == '\r') {
+			/* got CR, check for CRLF */
+			if (srcLen != 0 && ((*srcBuf) & mask) == '\n') {
+				srcBuf++;
+				srcLen--;
+			}
+			BufPrintf("\r\n");
+		} else if (ch == '\n') {
+			BufPrintf("\r\n");
+		} else {
+			/* Strip out null bytes if requested */
+			if ((stripNulls && ch != 0x00) || !stripNulls)
+				BufPrintf("%c", ch);
+		}
+	}
+}
+
+/*
+ * Write a hex dump into the buffer.
+ */
+void
+ReformatText::BufHexDump(const unsigned char* srcBuf, long srcLen)
+{
+	const unsigned char* origSrcBuf = srcBuf;
+	char chBuf[17];
+	int i, remLen;
+
+	ASSERT(srcBuf != nil);
+	ASSERT(srcLen >= 0);
+
+	chBuf[16] = '\0';
+
+	while (srcLen > 0) {
+		BufPrintf("%08lx: ", srcBuf - origSrcBuf);
+
+		if (srcLen >= 16) {
+			if (!fUseRTF) {
+				/* the really easy (and relatively fast) way */
+				BufPrintf("%02x %02x %02x %02x %02x %02x %02x %02x "
+						  "%02x %02x %02x %02x %02x %02x %02x %02x ",
+					srcBuf[0], srcBuf[1], srcBuf[2], srcBuf[3],
+					srcBuf[4], srcBuf[5], srcBuf[6], srcBuf[7],
+					srcBuf[8], srcBuf[9], srcBuf[10], srcBuf[11],
+					srcBuf[12], srcBuf[13], srcBuf[14], srcBuf[15]);
+			} else {
+				/* the fairly easy (and fairly fast) way */
+				RTFBoldOn();
+				BufPrintf("%02x %02x %02x %02x ",
+					srcBuf[0], srcBuf[1], srcBuf[2], srcBuf[3]);
+				RTFBoldOff();
+				BufPrintf("%02x %02x %02x %02x ",
+					srcBuf[4], srcBuf[5], srcBuf[6], srcBuf[7]);
+				RTFBoldOn();
+				BufPrintf("%02x %02x %02x %02x ",
+					srcBuf[8], srcBuf[9], srcBuf[10], srcBuf[11]);
+				RTFBoldOff();
+				BufPrintf("%02x %02x %02x %02x ",
+					srcBuf[12], srcBuf[13], srcBuf[14], srcBuf[15]);
+			}
+		} else {
+			/* the not-so-easy (and not-so-fast) way */
+			remLen = srcLen;
+
+			for (i = 0; i < remLen; i++) {
+				if (i == 0 || i == 8)
+					RTFBoldOn();
+				else if (i == 4 || i == 12)
+					RTFBoldOff();
+				BufPrintf("%02x ", srcBuf[i]);
+			}
+			RTFBoldOff();
+			for ( ; i < 16; i++)
+				BufPrintf("   ");
+
+			/* blank out the char buf, since we're only filling part in */
+			for (i = 0; i < 16; i++)
+				chBuf[i] = ' ';
+		}
+
+		bool hosed = false;
+		remLen = srcLen;
+		if (remLen > 16)
+			remLen = 16;
+		int i;
+		for (i = 0; i < remLen; i++) {
+			chBuf[i] = PrintableChar(srcBuf[i]);
+			if (fUseRTF &&
+				(chBuf[i] == '\\' || chBuf[i] == '{' || chBuf[i] == '}'))
+			{
+				hosed = true;
+				break;
+			}
+		}
+
+		if (!hosed) {
+			BufPrintf(" %s", chBuf);
+		} else {
+			/* escaped chars in RTF mode; have to do this one the hard way */
+			ASSERT(fUseRTF);
+			BufPrintf(" ");
+			for (i = 0; i < remLen; i++) {
+				RTFPrintChar(srcBuf[i]);
+			}
+		}
+
+		RTFNewPara();
+
+		srcBuf += 16;
+		srcLen -= 16;
+	}
+}
+
+
+/*
+ * ==========================================================================
+ *		ReformatGraphics
+ * ==========================================================================
+ */
+
+/*
+ * Initialize the Apple II color palette, used for Hi-Res and DHR
+ * conversions.  Could also be used for lo-res mode.
+ */
+void
+ReformatGraphics::InitPalette(void)
+{
+	ASSERT(kPaletteSize == 16);
+
+	static const RGBQUAD stdPalette[kPaletteSize] = {
+		/* blue, green, red, reserved */
+		{ 0x00, 0x00, 0x00 },	// $0 black
+		{ 0x33, 0x00, 0xdd },	// $1 red (magenta)
+		{ 0x99, 0x00, 0x00 },	// $2 dark blue
+		{ 0xdd, 0x22, 0xdd },	// $3 purple (violet)
+		{ 0x22, 0x77, 0x00 },	// $4 dark green
+		{ 0x55, 0x55, 0x55 },	// $5 grey1 (dark)
+		{ 0xff, 0x22, 0x22 },	// $6 medium blue
+		{ 0xff, 0xaa, 0x66 },	// $7 light blue
+		{ 0x00, 0x55, 0x88 },	// $8 brown
+		{ 0x00, 0x66, 0xff },	// $9 orange
+		{ 0xaa, 0xaa, 0xaa },	// $A grey2 (light)
+		{ 0x88, 0x99, 0xff },	// $B pink
+		{ 0x00, 0xdd, 0x11 },	// $C green (a/k/a light green)
+		{ 0x00, 0xff, 0xff },	// $D yellow
+		{ 0x99, 0xff, 0x44 },	// $E aqua
+		{ 0xff, 0xff, 0xff },	// $F white
+	};
+
+	memcpy(fPalette, stdPalette, sizeof(fPalette));
+}
+
+/*
+ * Stuff out DIB into the output fields, and set the appropriate flags.
+ */
+void
+ReformatGraphics::SetResultBuffer(ReformatOutput* pOutput, MyDIBitmap* pDib)
+{
+	ASSERT(pOutput != nil);
+	ASSERT(pDib != nil);
+	pOutput->SetOutputKind(ReformatOutput::kOutputBitmap);
+	pOutput->SetDIB(pDib);
+}
+
+
+/*
+ * Unpack the Apple PackBytes format.
+ *
+ * Format is:
+ *	<flag><data> ...
+ *
+ * Flag values (first 6 bits of flag byte):
+ *	00xxxxxx: (0-63) 1 to 64 bytes follow, all different
+ *	01xxxxxx: (0-63) 1 to 64 repeats of next byte
+ *	10xxxxxx: (0-63) 1 to 64 repeats of next 4 bytes
+ *	11xxxxxx: (0-63) 1 to 64 repeats of next byte taken as 4 bytes
+ *				(as in 10xxxxxx case)
+ *
+ * Pass the destination buffer in "dst", source buffer in "src", source
+ * length in "srcLen", and expected sizes of output in "dstRem".
+ *
+ * Returns 0 on success, nonzero if the buffer is overfilled or underfilled.
+ */
+int
+ReformatGraphics::UnpackBytes(unsigned char* dst, const unsigned char* src,
+	long dstRem, long srcLen)
+{
+	while (srcLen > 0) {
+		unsigned char flag = *src++;
+		int count = (flag & 0x3f) +1;
+		unsigned char val;
+		unsigned char valSet[4];
+		int i;
+
+		srcLen--;
+	
+		switch (flag & 0xc0) {
+		case 0x00:
+			for (i = 0; i < count; i++) {
+				if (srcLen == 0 || dstRem == 0) {
+					WMSG2(" SHR unpack overrun1 (srcLen=%ld dstRem=%ld)\n",
+						srcLen, dstRem);
+					return -1;
+				}
+				*dst++ = *src++;
+				srcLen--;
+				dstRem--;
+			}
+			break;
+		case 0x40:
+			//if (count != 3 || count != 5 || count != 6 || count != 7) {
+			//	WMSG1(" SHR unpack funky len %d?\n", count);
+			//}
+			if (srcLen == 0) {
+				WMSG0(" SHR unpack underrun2\n");
+				return -1;
+			}
+			val = *src++;
+			srcLen--;
+			for (i = 0; i < count; i++) {
+				if (dstRem == 0) {
+					WMSG3(" SHR unpack overrun2 (srcLen=%d, i=%d of %d)\n",
+						srcLen, i, count);
+					return -1;
+				}
+				*dst++ = val;
+				dstRem--;
+			}
+			break;
+		case 0x80:
+			if (srcLen < 4) {
+				WMSG0(" SHR unpack underrun3\n");
+				return -1;
+			}
+			valSet[0] = *src++;
+			valSet[1] = *src++;
+			valSet[2] = *src++;
+			valSet[3] = *src++;
+			srcLen -= 4;
+			for (i = 0; i < count; i++) {
+				if (dstRem < 4) {
+					WMSG2(" SHR unpack overrun3 (srcLen=%ld dstRem=%ld)\n",
+						srcLen, dstRem);
+					return -1;
+				}
+				*dst++ = valSet[0];
+				*dst++ = valSet[1];
+				*dst++ = valSet[2];
+				*dst++ = valSet[3];
+				dstRem -= 4;
+			}
+			break;
+		case 0xc0:
+			if (srcLen == 0) {
+				WMSG0(" SHR unpack underrun4\n");
+				return -1;
+			}
+			val = *src++;
+			srcLen--;
+			for (i = 0; i < count; i++) {
+				if (dstRem < 4) {
+					WMSG3(" SHR unpack overrun4 (srcLen=%ld dstRem=%ld count=%d)\n",
+						srcLen, dstRem, count);
+					return -1;
+				}
+				*dst++ = val;
+				*dst++ = val;
+				*dst++ = val;
+				*dst++ = val;
+				dstRem -= 4;
+			}
+			break;
+		default:
+			ASSERT(false);
+			break;
+		}
+	}
+
+	ASSERT(srcLen == 0);
+
+	/* require that we completely fill the buffer */
+	if (dstRem != 0) {
+		WMSG1(" SHR unpack dstRem at %d\n", dstRem);
+		return -1;
+	}
+
+	return 0;
+}
+
+
+/*
+ * Unpack Macintosh PackBits format.  See Technical Note TN1023.
+ *
+ * Read a byte.
+ * If the high bit is set, count is 2s complement +1 (i.e. count = (-byte)+1).
+ *   Read the next byte, then write that byte 'count' times.
+ * If the high bit is clear, count is 1+value (i.e. count = byte+1).  Read and
+ *   copy that many bytes.
+ * After "destLen" bytes have been written, return (even if in the middle of
+ * a run).
+ *
+ * NOTE: if the count byte is 0x80, Apple says it's an invalid value and
+ * should be skipped over.  Use the following byte as the count byte.  This
+ * is probably because PackBits is only supposed to crunch 127 bytes, though
+ * that suggests 0x81 and 0x7f are also impossible.
+ *
+ * We have to watch for underruns on the input and overruns on the output.
+ */
+void
+ReformatGraphics::UnPackBits(const unsigned char** pSrcBuf, long* pSrcLen,
+	unsigned char** pOutPtr, long dstLen, unsigned char xorVal)
+{
+	const unsigned char* srcBuf = *pSrcBuf;
+	long length = *pSrcLen;
+	unsigned char* outPtr = *pOutPtr;
+	int pixByte = 0;
+
+	while (pixByte < dstLen && length > 0) {
+		unsigned char countByte;
+		int count;
+		
+		countByte = *srcBuf++;
+		length--;
+		if (countByte & 0x80) {
+			/* RLE string */
+			unsigned char ch;
+			count = (countByte ^ 0xff)+1 +1;
+			ch = *srcBuf++;
+			length--;
+			while (count-- && pixByte < dstLen) {
+				*outPtr++ = ch ^ xorVal;
+				pixByte++;
+			}
+		} else {
+			/* series of bytse */
+			count = countByte +1;
+			while (count && pixByte < dstLen && length > 0) {
+				*outPtr++ = *srcBuf++ ^ xorVal;
+				count--;
+				length--;
+				pixByte++;
+			}
+		}
+	}
+	if (pixByte != 72) {
+		/* can happen if we run out of input early */
+		WMSG1("  MP unexpected pixByte=%d\n", pixByte);
+		/* keep going */
+	}
+
+	*pSrcBuf = srcBuf;
+	*pSrcLen = length;
+	*pOutPtr = outPtr;
+}
diff --git a/reformat/ReformatBase.h b/reformat/ReformatBase.h
index 5370c08..f09d709 100644
--- a/reformat/ReformatBase.h
+++ b/reformat/ReformatBase.h
@@ -314,6 +314,8 @@ protected:
 
 	void ConvertEOL(const unsigned char* srcBuf, long srcLen,
 		bool stripHiBits);
+	void ConvertEOL(const unsigned char* srcBuf, long srcLen,
+		bool stripHiBits, bool stripNulls);
 	void BufHexDump(const unsigned char* srcBuf, long srcLen);
 	void SetResultBuffer(ReformatOutput* pOutput, bool multiFont = false);
 
diff --git a/reformat/Text8.cpp b/reformat/Text8.cpp
index af845fd..c33e013 100644
--- a/reformat/Text8.cpp
+++ b/reformat/Text8.cpp
@@ -1,174 +1,225 @@
-/*
- * CiderPress
- * Copyright (C) 2007 by faddenSoft, LLC.  All Rights Reserved.
- * See the file LICENSE for distribution terms.
- */
-/*
- * Convert 8-bit word processor files.
- *
- * Most formats convert reasonably well with "Converted Text", but this
- * allows the files to be handled more transparently (e.g. Magic Window
- * "formatted files", which can be mistaken for code.
- */
-#include "StdAfx.h"
-#include "Text8.h"
-
-
-/*
- * ===========================================================================
- *		Magic Window / Magic Window II
- * ===========================================================================
- */
-
-/*
- * Magic Window and Magic Window II appear to use the same format for their
- * "formatted files".  The files are of type 'B', with a valid address field,
- * and what looks like junk in the length field.  The files have a 256-byte
- * header that seems to hold some sort of title string as well as some
- * binary goodies that I'm not sure what they are.
- *
- * The data from offset 256 on is entirely mixed-case high-ASCII text.  It
- * may contain printer-specific escape codes for bold, italic, etc.
- *
- * A ".MW" filename suffix is enforced by the program.
- */
-
-/*
- * Decide whether or not we want to handle this file.
- */
-void
-ReformatMagicWindow::Examine(ReformatHolder* pHolder)
-{
-	if (pHolder->GetFileType() == kTypeBIN) {
-		bool isMW = ReformatMagicWindow::IsFormatted(pHolder);
-		bool isDotMW = strcasecmp(pHolder->GetNameExt(), ".MW") == 0;
-
-		if (isMW && isDotMW) {
-			/* gotta be */
-			pHolder->SetApplic(ReformatHolder::kReformatMagicWindow,
-				ReformatHolder::kApplicYes,
-				ReformatHolder::kApplicNot, ReformatHolder::kApplicNot);
-		} else if (isDotMW) {
-			/* right type and name; maybe our test is broken? */
-			pHolder->SetApplic(ReformatHolder::kReformatMagicWindow,
-				ReformatHolder::kApplicProbably,
-				ReformatHolder::kApplicNot, ReformatHolder::kApplicNot);
-		} else if (isMW) {
-			/* not likely, but offer it as non-default option */
-			pHolder->SetApplic(ReformatHolder::kReformatMagicWindow,
-				ReformatHolder::kApplicProbablyNot,
-				ReformatHolder::kApplicNot, ReformatHolder::kApplicNot);
-		} else {
-			/* not one of ours */
-			pHolder->SetApplic(ReformatHolder::kReformatMagicWindow,
-				ReformatHolder::kApplicNot,
-				ReformatHolder::kApplicNot, ReformatHolder::kApplicNot);
-		}
-	} else {
-		/* "unformatted" text even if ".MW"; nothing special required */
-		pHolder->SetApplic(ReformatHolder::kReformatMagicWindow,
-			ReformatHolder::kApplicNot,
-			ReformatHolder::kApplicNot, ReformatHolder::kApplicNot);
-	}
-}
-
-/*
- * Figure out if this is a Magic Window "formatted" file.
- *
- * I don't know much about the format, so this is based on the similarities
- * observed between half a dozen documents from different sources.
- */
-/*static*/ bool
-ReformatMagicWindow::IsFormatted(const ReformatHolder* pHolder)
-{
-	const unsigned char* ptr = pHolder->GetSourceBuf(ReformatHolder::kPartData);
-	long srcLen = pHolder->GetSourceLen(ReformatHolder::kPartData);
-	int i, count00, count20;
-
-
-	/* want 256-byte header, plus a few bytes to check text */
-	if (srcLen < kHeaderLen+8)
-		return false;
-
-	/*
-	 * First byte always seems to be 0x8d.
-	 */
-	if (ptr[0x00] != 0x8d)
-		return false;
-
-	/*
-	 * 0x58 - 0xa0 is mostly filled with 0x00 (for Magic Window) or 0x20
-	 * (for Magic Window II).  Both seem to have space for the title in the
-	 * preceeding part, but it's high-ASCII for MW and low-ASCII for MW2.
-	 *
-	 * Expect 50 out of 72 to match.  If this is actually just uninitialized
-	 * data then this test will be bogus.
-	 */
-	count00 = count20 = 0;
-	for (i = 0x58; i < 0xa0; i++) {
-		if (ptr[i] == 0x00)
-			count00++;
-		if (ptr[i] == 0x20)
-			count20++;
-	}
-	if (count00 < 50 && count20 < 50)
-		return false;
-
-	/*
-	 * 0xa2 has some recognizeable bytes; sample values:
-	 *	MW  42 06 36 50 08 40
-	 *	MW2 42 06 36 55 08 40
-	 *  MW2 42 04 3a 50 00 50
-	 * Not really sure what to make of these.  If we can bracket these
-	 * values we might have something.
-	 */
-	if (ptr[0xa2] != 0x42 ||
-		(ptr[0xa3] < 2 && ptr[0xa3] > 10) ||
-		(ptr[0xa4] < 0x30 && ptr[0xa4] > 0x40))
-		return false;
-
-	/*
-	 * Make sure the rest of the file is 100% high ASCII.
-	 */
-	ptr += kHeaderLen;
-	srcLen -= kHeaderLen;
-	while (srcLen--) {
-		if ((*ptr & 0x80) == 0)
-			return false;
-	}
-
-	return true;
-}
-
-
-/*
- * Skip the header and text-convert the reset.
- */
-int
-ReformatMagicWindow::Process(const ReformatHolder* pHolder,
-	ReformatHolder::ReformatID id, ReformatHolder::ReformatPart part,
-	ReformatOutput* pOutput)
-{
-	const unsigned char* srcPtr = pHolder->GetSourceBuf(part);
-	long srcLen = pHolder->GetSourceLen(part);
-	long length = srcLen;
-	int retval = -1;
-
-	fUseRTF = false;
-
-	RTFBegin();
-
-	if (srcLen <= kHeaderLen)
-		goto bail;
-
-	ConvertEOL(srcPtr + kHeaderLen, srcLen - kHeaderLen, true);
-
-	//done:
-	RTFEnd();
-
-	SetResultBuffer(pOutput);
-	retval = 0;
-
-bail:
-	return retval;
-}
+/*
+ * CiderPress
+ * Copyright (C) 2009 by CiderPress authors.  All Rights Reserved.
+ * Copyright (C) 2007 by faddenSoft, LLC.  All Rights Reserved.
+ * See the file LICENSE for distribution terms.
+ */
+/*
+ * Convert 8-bit word processor files.
+ *
+ * Most formats convert reasonably well with "Converted Text", but this
+ * allows the files to be handled more transparently (e.g. Magic Window
+ * "formatted files", which can be mistaken for code.
+ */
+#include "StdAfx.h"
+#include "Text8.h"
+
+
+/*
+ * ===========================================================================
+ *		Magic Window / Magic Window II
+ * ===========================================================================
+ */
+
+/*
+ * Magic Window and Magic Window II appear to use the same format for their
+ * "formatted files".  The files are of type 'B', with a valid address field,
+ * and what looks like junk in the length field.  The files have a 256-byte
+ * header that seems to hold some sort of title string as well as some
+ * binary goodies that I'm not sure what they are.
+ *
+ * The data from offset 256 on is entirely mixed-case high-ASCII text.  It
+ * may contain printer-specific escape codes for bold, italic, etc.
+ *
+ * A ".MW" filename suffix is enforced by the program.
+ */
+
+/*
+ * Decide whether or not we want to handle this file.
+ */
+void
+ReformatMagicWindow::Examine(ReformatHolder* pHolder)
+{
+	if (pHolder->GetFileType() == kTypeBIN) {
+		bool isMW = ReformatMagicWindow::IsFormatted(pHolder);
+		bool isDotMW = strcasecmp(pHolder->GetNameExt(), ".MW") == 0;
+
+		if (isMW && isDotMW) {
+			/* gotta be */
+			pHolder->SetApplic(ReformatHolder::kReformatMagicWindow,
+				ReformatHolder::kApplicYes,
+				ReformatHolder::kApplicNot, ReformatHolder::kApplicNot);
+		} else if (isDotMW) {
+			/* right type and name; maybe our test is broken? */
+			pHolder->SetApplic(ReformatHolder::kReformatMagicWindow,
+				ReformatHolder::kApplicProbably,
+				ReformatHolder::kApplicNot, ReformatHolder::kApplicNot);
+		} else if (isMW) {
+			/* not likely, but offer it as non-default option */
+			pHolder->SetApplic(ReformatHolder::kReformatMagicWindow,
+				ReformatHolder::kApplicProbablyNot,
+				ReformatHolder::kApplicNot, ReformatHolder::kApplicNot);
+		} else {
+			/* not one of ours */
+			pHolder->SetApplic(ReformatHolder::kReformatMagicWindow,
+				ReformatHolder::kApplicNot,
+				ReformatHolder::kApplicNot, ReformatHolder::kApplicNot);
+		}
+	} else {
+		/* "unformatted" text even if ".MW"; nothing special required */
+		pHolder->SetApplic(ReformatHolder::kReformatMagicWindow,
+			ReformatHolder::kApplicNot,
+			ReformatHolder::kApplicNot, ReformatHolder::kApplicNot);
+	}
+}
+
+/*
+ * Figure out if this is a Magic Window "formatted" file.
+ *
+ * I don't know much about the format, so this is based on the similarities
+ * observed between half a dozen documents from different sources.
+ */
+/*static*/ bool
+ReformatMagicWindow::IsFormatted(const ReformatHolder* pHolder)
+{
+	const unsigned char* ptr = pHolder->GetSourceBuf(ReformatHolder::kPartData);
+	long srcLen = pHolder->GetSourceLen(ReformatHolder::kPartData);
+	int i, count00, count20;
+
+
+	/* want 256-byte header, plus a few bytes to check text */
+	if (srcLen < kHeaderLen+8)
+		return false;
+
+	/*
+	 * First byte always seems to be 0x8d.
+	 */
+	if (ptr[0x00] != 0x8d)
+		return false;
+
+	/*
+	 * 0x58 - 0xa0 is mostly filled with 0x00 (for Magic Window) or 0x20
+	 * (for Magic Window II).  Both seem to have space for the title in the
+	 * preceeding part, but it's high-ASCII for MW and low-ASCII for MW2.
+	 *
+	 * Expect 50 out of 72 to match.  If this is actually just uninitialized
+	 * data then this test will be bogus.
+	 */
+	count00 = count20 = 0;
+	for (i = 0x58; i < 0xa0; i++) {
+		if (ptr[i] == 0x00)
+			count00++;
+		if (ptr[i] == 0x20)
+			count20++;
+	}
+	if (count00 < 50 && count20 < 50)
+		return false;
+
+	/*
+	 * 0xa2 has some recognizeable bytes; sample values:
+	 *	MW  42 06 36 50 08 40
+	 *	MW2 42 06 36 55 08 40
+	 *  MW2 42 04 3a 50 00 50
+	 * Not really sure what to make of these.  If we can bracket these
+	 * values we might have something.
+	 */
+	if (ptr[0xa2] != 0x42 ||
+		(ptr[0xa3] < 2 && ptr[0xa3] > 10) ||
+		(ptr[0xa4] < 0x30 && ptr[0xa4] > 0x40))
+		return false;
+
+	/*
+	 * Make sure the rest of the file is 100% high ASCII.
+	 */
+	ptr += kHeaderLen;
+	srcLen -= kHeaderLen;
+	while (srcLen--) {
+		if ((*ptr & 0x80) == 0)
+			return false;
+	}
+
+	return true;
+}
+
+
+/*
+ * Skip the header and text-convert the rest.
+ */
+int
+ReformatMagicWindow::Process(const ReformatHolder* pHolder,
+	ReformatHolder::ReformatID id, ReformatHolder::ReformatPart part,
+	ReformatOutput* pOutput)
+{
+	const unsigned char* srcPtr = pHolder->GetSourceBuf(part);
+	long srcLen = pHolder->GetSourceLen(part);
+	long length = srcLen;
+	int retval = -1;
+
+	fUseRTF = false;
+
+	RTFBegin();
+
+	if (srcLen <= kHeaderLen)
+		goto bail;
+
+	ConvertEOL(srcPtr + kHeaderLen, srcLen - kHeaderLen, true);
+
+	//done:
+	RTFEnd();
+
+	SetResultBuffer(pOutput);
+	retval = 0;
+
+bail:
+	return retval;
+}
+
+
+/*
+ * ===========================================================================
+ *		Gutenberg Word Processor
+ * ===========================================================================
+ */
+
+/*
+ * Decide whether or not we want to handle this file.
+ */
+void
+ReformatGutenberg::Examine(ReformatHolder* pHolder)
+{
+	if ((pHolder->GetFileType() == kTypeTXT)  && 
+		(pHolder->GetSourceFormat() == ReformatHolder::kSourceFormatGutenberg)) {
+
+		pHolder->SetApplic(ReformatHolder::kReformatGutenberg,
+			ReformatHolder::kApplicYes,
+			ReformatHolder::kApplicNot, ReformatHolder::kApplicNot);
+	}
+}
+
+
+/*
+ * Convert the text.
+ */
+int
+ReformatGutenberg::Process(const ReformatHolder* pHolder,
+	ReformatHolder::ReformatID id, ReformatHolder::ReformatPart part,
+	ReformatOutput* pOutput)
+{
+	const unsigned char* srcPtr = pHolder->GetSourceBuf(part);
+	long srcLen = pHolder->GetSourceLen(part);
+	long length = srcLen;
+	int retval = -1;
+
+	fUseRTF = false;
+
+	RTFBegin();
+
+	ConvertEOL(srcPtr, srcLen, true, true);
+
+	RTFEnd();
+
+	SetResultBuffer(pOutput);
+	retval = 0;
+
+	return retval;
+}
diff --git a/reformat/Text8.h b/reformat/Text8.h
index 3ad42b6..8fc9099 100644
--- a/reformat/Text8.h
+++ b/reformat/Text8.h
@@ -1,33 +1,49 @@
-/*
- * CiderPress
- * Copyright (C) 2007 by faddenSoft, LLC.  All Rights Reserved.
- * See the file LICENSE for distribution terms.
- */
-/*
- * Reformat 8-bit word processor files.
- */
-#ifndef __LR_TEXT8__
-#define __LR_TEXT8__
-
-#include "ReformatBase.h"
-
-/*
- * Magic Window / Magic Window II
- */
-class ReformatMagicWindow : public ReformatText {
-public:
-	ReformatMagicWindow(void) {}
-	virtual ~ReformatMagicWindow(void) {}
-
-	virtual void Examine(ReformatHolder* pHolder);
-	virtual int Process(const ReformatHolder* pHolder,
-		ReformatHolder::ReformatID id, ReformatHolder::ReformatPart part,
-		ReformatOutput* pOutput);
-
-private:
-	static bool IsFormatted(const ReformatHolder* pHolder);
-
-	enum { kHeaderLen = 256 };
-};
-
-#endif /*__LR_TEXT8__*/
+/*
+ * CiderPress
+ * Copyright (C) 2009 by Ciderpress authors.  All Rights Reserved.
+ * Copyright (C) 2007 by faddenSoft, LLC.  All Rights Reserved.
+ * See the file LICENSE for distribution terms.
+ */
+/*
+ * Reformat 8-bit word processor files.
+ */
+#ifndef __LR_TEXT8__
+#define __LR_TEXT8__
+
+#include "ReformatBase.h"
+
+/*
+ * Magic Window / Magic Window II
+ */
+class ReformatMagicWindow : public ReformatText {
+public:
+	ReformatMagicWindow(void) {}
+	virtual ~ReformatMagicWindow(void) {}
+
+	virtual void Examine(ReformatHolder* pHolder);
+	virtual int Process(const ReformatHolder* pHolder,
+		ReformatHolder::ReformatID id, ReformatHolder::ReformatPart part,
+		ReformatOutput* pOutput);
+
+private:
+	static bool IsFormatted(const ReformatHolder* pHolder);
+
+	enum { kHeaderLen = 256 };
+};
+
+/*
+ * Guterberg Word Processor
+ */
+class ReformatGutenberg : public ReformatText {
+public:
+	ReformatGutenberg(void) {}
+	virtual ~ReformatGutenberg(void) {}
+
+	virtual void Examine(ReformatHolder* pHolder);
+	virtual int Process(const ReformatHolder* pHolder,
+		ReformatHolder::ReformatID id, ReformatHolder::ReformatPart part,
+		ReformatOutput* pOutput);
+
+};
+
+#endif /*__LR_TEXT8__*/