ciderpress/reformat/Text8.cpp

/*
 * 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;
}