fofi/FoFiType1C.cc

//========================================================================
//
// FoFiType1C.cc
//
// Copyright 1999-2003 Glyph & Cog, LLC
//
//========================================================================

//========================================================================
//
// Modified under the Poppler project - http://poppler.freedesktop.org
//
// All changes made under the Poppler project to this file are licensed
// under GPL version 2 or later
//
// Copyright (C) 2009, 2010, 2017-2019 Albert Astals Cid <aacid@kde.org>
// Copyright (C) 2012 Thomas Freitag <Thomas.Freitag@alfa.de>
// Copyright (C) 2018 Adam Reichold <adam.reichold@t-online.de>
// Copyright (C) 2019 Tomoyuki Kubota <himajin100000@gmail.com>
// Copyright (C) 2019 Volker Krause <vkrause@kde.org>
//
// To see a description of the changes please see the Changelog file that
// came with your tarball or type make ChangeLog if you are building from git
//
//========================================================================

#include <config.h>

#include <stdlib.h>
#include <string.h>
#include <math.h>
#include "goo/gmem.h"
#include "goo/gstrtod.h"
#include "goo/GooLikely.h"
#include "goo/GooString.h"
#include "poppler/Error.h"
#include "FoFiEncodings.h"
#include "FoFiType1C.h"

//------------------------------------------------------------------------

static char hexChars[17] = "0123456789ABCDEF";

//------------------------------------------------------------------------
// FoFiType1C
//------------------------------------------------------------------------

FoFiType1C *FoFiType1C::make(const char *fileA, int lenA) {
  FoFiType1C *ff;

  ff = new FoFiType1C(fileA, lenA, false);
  if (!ff->parse()) {
    delete ff;
    return nullptr;
  }
  return ff;
}

FoFiType1C *FoFiType1C::load(const char *fileName) {
  FoFiType1C *ff;
  char *fileA;
  int lenA;

  if (!(fileA = FoFiBase::readFile(fileName, &lenA))) {
    return nullptr;
  }
  ff = new FoFiType1C(fileA, lenA, true);
  if (!ff->parse()) {
    delete ff;
    return nullptr;
  }
  return ff;
}

FoFiType1C::FoFiType1C(const char *fileA, int lenA, bool freeFileDataA):
  FoFiBase(fileA, lenA, freeFileDataA)
{
  name = nullptr;
  encoding = nullptr;
  privateDicts = nullptr;
  fdSelect = nullptr;
  charset = nullptr;
  charsetLength = 0;
}

FoFiType1C::~FoFiType1C() {
  int i;

  if (name) {
    delete name;
  }
  if (encoding &&
      encoding != fofiType1StandardEncoding &&
      encoding != fofiType1ExpertEncoding) {
    for (i = 0; i < 256; ++i) {
      gfree(encoding[i]);
    }
    gfree(encoding);
  }
  if (privateDicts) {
    gfree(privateDicts);
  }
  if (fdSelect) {
    gfree(fdSelect);
  }
  if (charset &&
      charset != fofiType1CISOAdobeCharset &&
      charset != fofiType1CExpertCharset &&
      charset != fofiType1CExpertSubsetCharset) {
    gfree(const_cast<unsigned short *>(charset));
  }
}

const char *FoFiType1C::getName() const {
  return name ? name->c_str() : nullptr;
}

char **FoFiType1C::getEncoding() const {
  return encoding;
}

GooString *FoFiType1C::getGlyphName(int gid) const {
  char buf[256];
  bool ok;

  ok = true;
  if (gid < 0 || gid >= charsetLength)
    return nullptr;
  getString(charset[gid], buf, &ok);
  if (!ok) {
    return nullptr;
  }
  return new GooString(buf);
}

int *FoFiType1C::getCIDToGIDMap(int *nCIDs) const {
  int *map;
  int n, i;

  // a CID font's top dict has ROS as the first operator
  if (topDict.firstOp != 0x0c1e) {
    *nCIDs = 0;
    return nullptr;
  }

  // in a CID font, the charset data is the GID-to-CID mapping, so all
  // we have to do is reverse it
  n = 0;
  for (i = 0; i < nGlyphs && i < charsetLength; ++i) {
    if (charset[i] > n) {
      n = charset[i];
    }
  }
  ++n;
  map = (int *)gmallocn(n, sizeof(int));
  memset(map, 0, n * sizeof(int));
  for (i = 0; i < nGlyphs; ++i) {
    map[charset[i]] = i;
  }
  *nCIDs = n;
  return map;
}

void FoFiType1C::getFontMatrix(double *mat) const {
  int i;

  if (topDict.firstOp == 0x0c1e && privateDicts[0].hasFontMatrix) {
    if (topDict.hasFontMatrix) {
      mat[0] = topDict.fontMatrix[0] * privateDicts[0].fontMatrix[0] +
	       topDict.fontMatrix[1] * privateDicts[0].fontMatrix[2];
      mat[1] = topDict.fontMatrix[0] * privateDicts[0].fontMatrix[1] +
               topDict.fontMatrix[1] * privateDicts[0].fontMatrix[3];
      mat[2] = topDict.fontMatrix[2] * privateDicts[0].fontMatrix[0] +
	       topDict.fontMatrix[3] * privateDicts[0].fontMatrix[2];
      mat[3] = topDict.fontMatrix[2] * privateDicts[0].fontMatrix[1] +
	       topDict.fontMatrix[3] * privateDicts[0].fontMatrix[3];
      mat[4] = topDict.fontMatrix[4] * privateDicts[0].fontMatrix[0] +
	       topDict.fontMatrix[5] * privateDicts[0].fontMatrix[2];
      mat[5] = topDict.fontMatrix[4] * privateDicts[0].fontMatrix[1] +
	       topDict.fontMatrix[5] * privateDicts[0].fontMatrix[3];
    } else {
      for (i = 0; i < 6; ++i) {
	mat[i] = privateDicts[0].fontMatrix[i];
      }
    }
  } else {
    for (i = 0; i < 6; ++i) {
      mat[i] = topDict.fontMatrix[i];
    }
  }
}

void FoFiType1C::convertToType1(const char *psName, const char **newEncoding, bool ascii,
				FoFiOutputFunc outputFunc,
				void *outputStream) {
  int psNameLen;
  Type1CEexecBuf eb;
  Type1CIndex subrIdx;
  Type1CIndexVal val;
  GooString *buf;
  char buf2[256];
  const char **enc;
  bool ok;
  int i;

  if (psName) {
    psNameLen = strlen(psName);
  } else {
    psName = name->c_str();
    psNameLen = name->getLength();
  }

  // write header and font dictionary, up to encoding
  ok = true;
  (*outputFunc)(outputStream, "%!FontType1-1.0: ", 17);
  (*outputFunc)(outputStream, psName, psNameLen);
  if (topDict.versionSID != 0) {
    getString(topDict.versionSID, buf2, &ok);
    (*outputFunc)(outputStream, buf2, strlen(buf2));
  }
  (*outputFunc)(outputStream, "\n", 1);
  // the dictionary needs room for 12 entries: the following 9, plus
  // Private and CharStrings (in the eexec section) and FID (which is
  // added by definefont)
  (*outputFunc)(outputStream, "12 dict begin\n", 14);
  (*outputFunc)(outputStream, "/FontInfo 10 dict dup begin\n", 28);
  if (topDict.versionSID != 0) {
    (*outputFunc)(outputStream, "/version ", 9);
    writePSString(buf2, outputFunc, outputStream);
    (*outputFunc)(outputStream, " readonly def\n", 14);
  }
  if (topDict.noticeSID != 0) {
    getString(topDict.noticeSID, buf2, &ok);
    (*outputFunc)(outputStream, "/Notice ", 8);
    writePSString(buf2, outputFunc, outputStream);
    (*outputFunc)(outputStream, " readonly def\n", 14);
  }
  if (topDict.copyrightSID != 0) {
    getString(topDict.copyrightSID, buf2, &ok);
    (*outputFunc)(outputStream, "/Copyright ", 11);
    writePSString(buf2, outputFunc, outputStream);
    (*outputFunc)(outputStream, " readonly def\n", 14);
  }
  if (topDict.fullNameSID != 0) {
    getString(topDict.fullNameSID, buf2, &ok);
    (*outputFunc)(outputStream, "/FullName ", 10);
    writePSString(buf2, outputFunc, outputStream);
    (*outputFunc)(outputStream, " readonly def\n", 14);
  }
  if (topDict.familyNameSID != 0) {
    getString(topDict.familyNameSID, buf2, &ok);
    (*outputFunc)(outputStream, "/FamilyName ", 12);
    writePSString(buf2, outputFunc, outputStream);
    (*outputFunc)(outputStream, " readonly def\n", 14);
  }
  if (topDict.weightSID != 0) {
    getString(topDict.weightSID, buf2, &ok);
    (*outputFunc)(outputStream, "/Weight ", 8);
    writePSString(buf2, outputFunc, outputStream);
    (*outputFunc)(outputStream, " readonly def\n", 14);
  }
  if (topDict.isFixedPitch) {
    (*outputFunc)(outputStream, "/isFixedPitch true def\n", 23);
  } else {
    (*outputFunc)(outputStream, "/isFixedPitch false def\n", 24);
  }
  buf = GooString::format("/ItalicAngle {0:.4g} def\n", topDict.italicAngle);
  (*outputFunc)(outputStream, buf->c_str(), buf->getLength());
  delete buf;
  buf = GooString::format("/UnderlinePosition {0:.4g} def\n",
			topDict.underlinePosition);
  (*outputFunc)(outputStream, buf->c_str(), buf->getLength());
  delete buf;
  buf = GooString::format("/UnderlineThickness {0:.4g} def\n",
			topDict.underlineThickness);
  (*outputFunc)(outputStream, buf->c_str(), buf->getLength());
  delete buf;
  (*outputFunc)(outputStream, "end readonly def\n", 17);
  (*outputFunc)(outputStream, "/FontName /", 11);
  (*outputFunc)(outputStream, psName, psNameLen);
  (*outputFunc)(outputStream, " def\n", 5);
  buf = GooString::format("/PaintType {0:d} def\n", topDict.paintType);
  (*outputFunc)(outputStream, buf->c_str(), buf->getLength());
  delete buf;
  (*outputFunc)(outputStream, "/FontType 1 def\n", 16);
  buf = GooString::format("/FontMatrix [{0:.8g} {1:.8g} {2:.8g} {3:.8g} {4:.8g} {5:.8g}] readonly def\n",
			topDict.fontMatrix[0], topDict.fontMatrix[1],
			topDict.fontMatrix[2], topDict.fontMatrix[3],
			topDict.fontMatrix[4], topDict.fontMatrix[5]);
  (*outputFunc)(outputStream, buf->c_str(), buf->getLength());
  delete buf;
  buf = GooString::format("/FontBBox [{0:.4g} {1:.4g} {2:.4g} {3:.4g}] readonly def\n",
			topDict.fontBBox[0], topDict.fontBBox[1],
			topDict.fontBBox[2], topDict.fontBBox[3]);
  (*outputFunc)(outputStream, buf->c_str(), buf->getLength());
  delete buf;
  buf = GooString::format("/StrokeWidth {0:.4g} def\n", topDict.strokeWidth);
  (*outputFunc)(outputStream, buf->c_str(), buf->getLength());
  delete buf;
  if (topDict.uniqueID != 0) {
    buf = GooString::format("/UniqueID {0:d} def\n", topDict.uniqueID);
    (*outputFunc)(outputStream, buf->c_str(), buf->getLength());
    delete buf;
  }

  // write the encoding
  (*outputFunc)(outputStream, "/Encoding ", 10);
  if (!newEncoding && encoding == fofiType1StandardEncoding) {
    (*outputFunc)(outputStream, "StandardEncoding def\n", 21);
  } else {
    (*outputFunc)(outputStream, "256 array\n", 10);
    (*outputFunc)(outputStream,
		  "0 1 255 {1 index exch /.notdef put} for\n", 40);
    enc = newEncoding ? newEncoding : (const char **)encoding;
    for (i = 0; i < 256; ++i) {
      if (enc[i]) {
	buf = GooString::format("dup {0:d} /{1:s} put\n", i, enc[i]);
	(*outputFunc)(outputStream, buf->c_str(), buf->getLength());
	delete buf;
      }
    }
    (*outputFunc)(outputStream, "readonly def\n", 13);
  }
  (*outputFunc)(outputStream, "currentdict end\n", 16);

  // start the binary section
  (*outputFunc)(outputStream, "currentfile eexec\n", 18);
  eb.outputFunc = outputFunc;
  eb.outputStream = outputStream;
  eb.ascii = ascii;
  eb.r1 = 55665;
  eb.line = 0;

  // write the private dictionary
  eexecWrite(&eb, "\x83\xca\x73\xd5");
  eexecWrite(&eb, "dup /Private 32 dict dup begin\n");
  eexecWrite(&eb, "/RD {string currentfile exch readstring pop}"
	     " executeonly def\n");
  eexecWrite(&eb, "/ND {noaccess def} executeonly def\n");
  eexecWrite(&eb, "/NP {noaccess put} executeonly def\n");
  eexecWrite(&eb, "/MinFeature {16 16} def\n");
  eexecWrite(&eb, "/password 5839 def\n");
  if (privateDicts[0].nBlueValues) {
    eexecWrite(&eb, "/BlueValues [");
    for (i = 0; i < privateDicts[0].nBlueValues; ++i) {
      buf = GooString::format("{0:s}{1:d}",
			    i > 0 ? " " : "", privateDicts[0].blueValues[i]);
      eexecWrite(&eb, buf->c_str());
      delete buf;
    }
    eexecWrite(&eb, "] def\n");
  }
  if (privateDicts[0].nOtherBlues) {
    eexecWrite(&eb, "/OtherBlues [");
    for (i = 0; i < privateDicts[0].nOtherBlues; ++i) {
      buf = GooString::format("{0:s}{1:d}",
			    i > 0 ? " " : "", privateDicts[0].otherBlues[i]);
      eexecWrite(&eb, buf->c_str());
      delete buf;
    }
    eexecWrite(&eb, "] def\n");
  }
  if (privateDicts[0].nFamilyBlues) {
    eexecWrite(&eb, "/FamilyBlues [");
    for (i = 0; i < privateDicts[0].nFamilyBlues; ++i) {
      buf = GooString::format("{0:s}{1:d}",
			    i > 0 ? " " : "", privateDicts[0].familyBlues[i]);
      eexecWrite(&eb, buf->c_str());
      delete buf;
    }
    eexecWrite(&eb, "] def\n");
  }
  if (privateDicts[0].nFamilyOtherBlues) {
    eexecWrite(&eb, "/FamilyOtherBlues [");
    for (i = 0; i < privateDicts[0].nFamilyOtherBlues; ++i) {
      buf = GooString::format("{0:s}{1:d}", i > 0 ? " " : "",
			    privateDicts[0].familyOtherBlues[i]);
      eexecWrite(&eb, buf->c_str());
      delete buf;
    }
    eexecWrite(&eb, "] def\n");
  }
  if (privateDicts[0].blueScale != 0.039625) {
    buf = GooString::format("/BlueScale {0:.4g} def\n",
			  privateDicts[0].blueScale);
    eexecWrite(&eb, buf->c_str());
    delete buf;
  }
  if (privateDicts[0].blueShift != 7) {
    buf = GooString::format("/BlueShift {0:d} def\n", privateDicts[0].blueShift);
    eexecWrite(&eb, buf->c_str());
    delete buf;
  }
  if (privateDicts[0].blueFuzz != 1) {
    buf = GooString::format("/BlueFuzz {0:d} def\n", privateDicts[0].blueFuzz);
    eexecWrite(&eb, buf->c_str());
    delete buf;
  }
  if (privateDicts[0].hasStdHW) {
    buf = GooString::format("/StdHW [{0:.4g}] def\n", privateDicts[0].stdHW);
    eexecWrite(&eb, buf->c_str());
    delete buf;
  }
  if (privateDicts[0].hasStdVW) {
    buf = GooString::format("/StdVW [{0:.4g}] def\n", privateDicts[0].stdVW);
    eexecWrite(&eb, buf->c_str());
    delete buf;
  }
  if (privateDicts[0].nStemSnapH) {
    eexecWrite(&eb, "/StemSnapH [");
    for (i = 0; i < privateDicts[0].nStemSnapH; ++i) {
      buf = GooString::format("{0:s}{1:.4g}",
			    i > 0 ? " " : "", privateDicts[0].stemSnapH[i]);
      eexecWrite(&eb, buf->c_str());
      delete buf;
    }
    eexecWrite(&eb, "] def\n");
  }
  if (privateDicts[0].nStemSnapV) {
    eexecWrite(&eb, "/StemSnapV [");
    for (i = 0; i < privateDicts[0].nStemSnapV; ++i) {
      buf = GooString::format("{0:s}{1:.4g}",
			    i > 0 ? " " : "", privateDicts[0].stemSnapV[i]);
      eexecWrite(&eb, buf->c_str());
      delete buf;
    }
    eexecWrite(&eb, "] def\n");
  }
  if (privateDicts[0].hasForceBold) {
    buf = GooString::format("/ForceBold {0:s} def\n",
			  privateDicts[0].forceBold ? "true" : "false");
    eexecWrite(&eb, buf->c_str());
    delete buf;
  }
  if (privateDicts[0].forceBoldThreshold != 0) {
    buf = GooString::format("/ForceBoldThreshold {0:.4g} def\n",
			  privateDicts[0].forceBoldThreshold);
    eexecWrite(&eb, buf->c_str());
    delete buf;
  }
  if (privateDicts[0].languageGroup != 0) {
    buf = GooString::format("/LanguageGroup {0:d} def\n",
			  privateDicts[0].languageGroup);
    eexecWrite(&eb, buf->c_str());
    delete buf;
  }
  if (privateDicts[0].expansionFactor != 0.06) {
    buf = GooString::format("/ExpansionFactor {0:.4g} def\n",
			  privateDicts[0].expansionFactor);
    eexecWrite(&eb, buf->c_str());
    delete buf;
  }

  // set up subroutines
  ok = true;
  getIndex(privateDicts[0].subrsOffset, &subrIdx, &ok);
  if (!ok) {
    subrIdx.pos = -1;
  }

  // write the CharStrings
  buf = GooString::format("2 index /CharStrings {0:d} dict dup begin\n",
			nGlyphs);
  eexecWrite(&eb, buf->c_str());
  delete buf;
  for (i = 0; i < nGlyphs; ++i) {
    ok = true;
    getIndexVal(&charStringsIdx, i, &val, &ok);
    if (ok && i < charsetLength) {
      getString(charset[i], buf2, &ok);
      if (ok) {
	eexecCvtGlyph(&eb, buf2, val.pos, val.len, &subrIdx, &privateDicts[0]);
      }
    }
  }
  eexecWrite(&eb, "end\n");
  eexecWrite(&eb, "end\n");
  eexecWrite(&eb, "readonly put\n");
  eexecWrite(&eb, "noaccess put\n");
  eexecWrite(&eb, "dup /FontName get exch definefont pop\n");
  eexecWrite(&eb, "mark currentfile closefile\n");

  // trailer
  if (ascii && eb.line > 0) {
    (*outputFunc)(outputStream, "\n", 1);
  }
  for (i = 0; i < 8; ++i) {
    (*outputFunc)(outputStream, "0000000000000000000000000000000000000000000000000000000000000000\n", 65);
  }
  (*outputFunc)(outputStream, "cleartomark\n", 12);
}

void FoFiType1C::convertToCIDType0(const char *psName, const int *codeMap, int nCodes,
				   FoFiOutputFunc outputFunc,
				   void *outputStream) {
  int *cidMap;
  GooString *charStrings;
  int *charStringOffsets;
  Type1CIndex subrIdx;
  Type1CIndexVal val;
  int nCIDs, gdBytes;
  GooString *buf;
  char buf2[256];
  bool ok;
  int gid, offset, n, i, j, k;

  // compute the CID count and build the CID-to-GID mapping
  if (codeMap) {
    nCIDs = nCodes;
    cidMap = (int *)gmallocn(nCIDs, sizeof(int));
    for (i = 0; i < nCodes; ++i) {
      if (codeMap[i] >= 0 && codeMap[i] < nGlyphs) {
	cidMap[i] = codeMap[i];
      } else {
	cidMap[i] = -1;
      }
    }
  } else if (topDict.firstOp == 0x0c1e) {
    nCIDs = 0;
    for (i = 0; i < nGlyphs && i < charsetLength; ++i) {
      if (charset[i] >= nCIDs) {
	nCIDs = charset[i] + 1;
      }
    }
    cidMap = (int *)gmallocn(nCIDs, sizeof(int));
    for (i = 0; i < nCIDs; ++i) {
      cidMap[i] = -1;
    }
    for (i = 0; i < nGlyphs && i < charsetLength; ++i) {
      cidMap[charset[i]] = i;
    }
  } else {
    nCIDs = nGlyphs;
    cidMap = (int *)gmallocn(nCIDs, sizeof(int));
    for (i = 0; i < nCIDs; ++i) {
      cidMap[i] = i;
    }
  }

  // build the charstrings
  charStrings = new GooString();
  charStringOffsets = (int *)gmallocn(nCIDs + 1, sizeof(int));
  for (i = 0; i < nCIDs; ++i) {
    charStringOffsets[i] = charStrings->getLength();
    if ((gid = cidMap[i]) >= 0) {
      ok = true;
      getIndexVal(&charStringsIdx, gid, &val, &ok);
      if (ok) {
	getIndex(privateDicts[fdSelect ? fdSelect[gid] : 0].subrsOffset,
		 &subrIdx, &ok);
	if (!ok) {
	  subrIdx.pos = -1;
	}
	cvtGlyph(val.pos, val.len, charStrings,
		 &subrIdx, &privateDicts[fdSelect ? fdSelect[gid] : 0], true);
      }
    }
  }
  charStringOffsets[nCIDs] = charStrings->getLength();

  // compute gdBytes = number of bytes needed for charstring offsets
  // (offset size needs to account for the charstring offset table,
  // with a worst case of five bytes per entry, plus the charstrings
  // themselves)
  i = (nCIDs + 1) * 5 + charStrings->getLength();
  if (i < 0x100) {
    gdBytes = 1;
  } else if (i < 0x10000) {
    gdBytes = 2;
  } else if (i < 0x1000000) {
    gdBytes = 3;
  } else {
    gdBytes = 4;
  }

  // begin the font dictionary
  (*outputFunc)(outputStream, "/CIDInit /ProcSet findresource begin\n", 37);
  (*outputFunc)(outputStream, "20 dict begin\n", 14);
  (*outputFunc)(outputStream, "/CIDFontName /", 14);
  (*outputFunc)(outputStream, psName, strlen(psName));
  (*outputFunc)(outputStream, " def\n", 5);
  (*outputFunc)(outputStream, "/CIDFontType 0 def\n", 19);
  (*outputFunc)(outputStream, "/CIDSystemInfo 3 dict dup begin\n", 32);
  if (topDict.registrySID > 0 && topDict.orderingSID > 0) {
    ok = true;
    getString(topDict.registrySID, buf2, &ok);
    if (ok) {
      (*outputFunc)(outputStream, "  /Registry (", 13);
      (*outputFunc)(outputStream, buf2, strlen(buf2));
      (*outputFunc)(outputStream, ") def\n", 6);
    }
    ok = true;
    getString(topDict.orderingSID, buf2, &ok);
    if (ok) {
      (*outputFunc)(outputStream, "  /Ordering (", 13);
      (*outputFunc)(outputStream, buf2, strlen(buf2));
      (*outputFunc)(outputStream, ") def\n", 6);
    }
  } else {
    (*outputFunc)(outputStream, "  /Registry (Adobe) def\n", 24);
    (*outputFunc)(outputStream, "  /Ordering (Identity) def\n", 27);
  }
  buf = GooString::format("  /Supplement {0:d} def\n", topDict.supplement);
  (*outputFunc)(outputStream, buf->c_str(), buf->getLength());
  delete buf;
  (*outputFunc)(outputStream, "end def\n", 8);
  if (topDict.hasFontMatrix) {
    buf = GooString::format("/FontMatrix [{0:.8g} {1:.8g} {2:.8g} {3:.8g} {4:.8g} {5:.8g}] def\n",
			  topDict.fontMatrix[0], topDict.fontMatrix[1],
			  topDict.fontMatrix[2], topDict.fontMatrix[3],
			  topDict.fontMatrix[4], topDict.fontMatrix[5]);
    (*outputFunc)(outputStream, buf->c_str(), buf->getLength());
    delete buf;
  } else if (privateDicts[0].hasFontMatrix) {
    (*outputFunc)(outputStream, "/FontMatrix [1 0 0 1 0 0] def\n", 30);
  } else {
    (*outputFunc)(outputStream,
		  "/FontMatrix [0.001 0 0 0.001 0 0] def\n", 38);
  }
  buf = GooString::format("/FontBBox [{0:.4g} {1:.4g} {2:.4g} {3:.4g}] def\n",
			topDict.fontBBox[0], topDict.fontBBox[1],
			topDict.fontBBox[2], topDict.fontBBox[3]);
  (*outputFunc)(outputStream, buf->c_str(), buf->getLength());
  delete buf;
  (*outputFunc)(outputStream, "/FontInfo 1 dict dup begin\n", 27);
  (*outputFunc)(outputStream, "  /FSType 8 def\n", 16);
  (*outputFunc)(outputStream, "end def\n", 8);

  // CIDFont-specific entries
  buf = GooString::format("/CIDCount {0:d} def\n", nCIDs);
  (*outputFunc)(outputStream, buf->c_str(), buf->getLength());
  delete buf;
  (*outputFunc)(outputStream, "/FDBytes 1 def\n", 15);
  buf = GooString::format("/GDBytes {0:d} def\n", gdBytes);
  (*outputFunc)(outputStream, buf->c_str(), buf->getLength());
  delete buf;
  (*outputFunc)(outputStream, "/CIDMapOffset 0 def\n", 20);
  if (topDict.paintType != 0) {
    buf = GooString::format("/PaintType {0:d} def\n", topDict.paintType);
    (*outputFunc)(outputStream, buf->c_str(), buf->getLength());
    delete buf;
    buf = GooString::format("/StrokeWidth {0:.4g} def\n", topDict.strokeWidth);
    (*outputFunc)(outputStream, buf->c_str(), buf->getLength());
    delete buf;
  }

  // FDArray entry
  buf = GooString::format("/FDArray {0:d} array\n", nFDs);
  (*outputFunc)(outputStream, buf->c_str(), buf->getLength());
  delete buf;
  for (i = 0; i < nFDs; ++i) {
    buf = GooString::format("dup {0:d} 10 dict begin\n", i);
    (*outputFunc)(outputStream, buf->c_str(), buf->getLength());
    delete buf;
    (*outputFunc)(outputStream, "/FontType 1 def\n", 16);
    if (privateDicts[i].hasFontMatrix) {
      buf = GooString::format("/FontMatrix [{0:.8g} {1:.8g} {2:.8g} {3:.8g} {4:.8g} {5:.8g}] def\n",
			    privateDicts[i].fontMatrix[0],
			    privateDicts[i].fontMatrix[1],
			    privateDicts[i].fontMatrix[2],
			    privateDicts[i].fontMatrix[3],
			    privateDicts[i].fontMatrix[4],
			    privateDicts[i].fontMatrix[5]);
      (*outputFunc)(outputStream, buf->c_str(), buf->getLength());
      delete buf;
    } else {
      (*outputFunc)(outputStream, "/FontMatrix [1 0 0 1 0 0] def\n", 30);
    }
    buf = GooString::format("/PaintType {0:d} def\n", topDict.paintType);
    (*outputFunc)(outputStream, buf->c_str(), buf->getLength());
    delete buf;
    (*outputFunc)(outputStream, "/Private 32 dict begin\n", 23);
    if (privateDicts[i].nBlueValues) {
      (*outputFunc)(outputStream, "/BlueValues [", 13);
      for (j = 0; j < privateDicts[i].nBlueValues; ++j) {
	buf = GooString::format("{0:s}{1:d}",
			      j > 0 ? " " : "", privateDicts[i].blueValues[j]);
	(*outputFunc)(outputStream, buf->c_str(), buf->getLength());
	delete buf;
      }
      (*outputFunc)(outputStream, "] def\n", 6);
    }
    if (privateDicts[i].nOtherBlues) {
      (*outputFunc)(outputStream, "/OtherBlues [", 13);
      for (j = 0; j < privateDicts[i].nOtherBlues; ++j) {
	buf = GooString::format("{0:s}{1:d}",
			      j > 0 ? " " : "", privateDicts[i].otherBlues[j]);
	(*outputFunc)(outputStream, buf->c_str(), buf->getLength());
	delete buf;
      }
      (*outputFunc)(outputStream, "] def\n", 6);
    }
    if (privateDicts[i].nFamilyBlues) {
      (*outputFunc)(outputStream, "/FamilyBlues [", 14);
      for (j = 0; j < privateDicts[i].nFamilyBlues; ++j) {
	buf = GooString::format("{0:s}{1:d}",
			      j > 0 ? " " : "",
			      privateDicts[i].familyBlues[j]);
	(*outputFunc)(outputStream, buf->c_str(), buf->getLength());
	delete buf;
      }
      (*outputFunc)(outputStream, "] def\n", 6);
    }
    if (privateDicts[i].nFamilyOtherBlues) {
      (*outputFunc)(outputStream, "/FamilyOtherBlues [", 19);
      for (j = 0; j < privateDicts[i].nFamilyOtherBlues; ++j) {
	buf = GooString::format("{0:s}{1:d}", j > 0 ? " " : "",
			      privateDicts[i].familyOtherBlues[j]);
	(*outputFunc)(outputStream, buf->c_str(), buf->getLength());
	delete buf;
      }
      (*outputFunc)(outputStream, "] def\n", 6);
    }
    if (privateDicts[i].blueScale != 0.039625) {
      buf = GooString::format("/BlueScale {0:.4g} def\n",
			    privateDicts[i].blueScale);
      (*outputFunc)(outputStream, buf->c_str(), buf->getLength());
      delete buf;
    }
    if (privateDicts[i].blueShift != 7) {
      buf = GooString::format("/BlueShift {0:d} def\n",
			    privateDicts[i].blueShift);
      (*outputFunc)(outputStream, buf->c_str(), buf->getLength());
      delete buf;
    }
    if (privateDicts[i].blueFuzz != 1) {
      buf = GooString::format("/BlueFuzz {0:d} def\n", privateDicts[i].blueFuzz);
      (*outputFunc)(outputStream, buf->c_str(), buf->getLength());
      delete buf;
    }
    if (privateDicts[i].hasStdHW) {
      buf = GooString::format("/StdHW [{0:.4g}] def\n", privateDicts[i].stdHW);
      (*outputFunc)(outputStream, buf->c_str(), buf->getLength());
      delete buf;
    }
    if (privateDicts[i].hasStdVW) {
      buf = GooString::format("/StdVW [{0:.4g}] def\n", privateDicts[i].stdVW);
      (*outputFunc)(outputStream, buf->c_str(), buf->getLength());
      delete buf;
    }
    if (privateDicts[i].nStemSnapH) {
      (*outputFunc)(outputStream, "/StemSnapH [", 12);
      for (j = 0; j < privateDicts[i].nStemSnapH; ++j) {
	buf = GooString::format("{0:s}{1:.4g}",
			      j > 0 ? " " : "", privateDicts[i].stemSnapH[j]);
	(*outputFunc)(outputStream, buf->c_str(), buf->getLength());
	delete buf;
      }
      (*outputFunc)(outputStream, "] def\n", 6);
    }
    if (privateDicts[i].nStemSnapV) {
      (*outputFunc)(outputStream, "/StemSnapV [", 12);
      for (j = 0; j < privateDicts[i].nStemSnapV; ++j) {
	buf = GooString::format("{0:s}{1:.4g}",
			      j > 0 ? " " : "", privateDicts[i].stemSnapV[j]);
	(*outputFunc)(outputStream, buf->c_str(), buf->getLength());
	delete buf;
      }
      (*outputFunc)(outputStream, "] def\n", 6);
    }
    if (privateDicts[i].hasForceBold) {
      buf = GooString::format("/ForceBold {0:s} def\n",
			    privateDicts[i].forceBold ? "true" : "false");
      (*outputFunc)(outputStream, buf->c_str(), buf->getLength());
      delete buf;
    }
    if (privateDicts[i].forceBoldThreshold != 0) {
      buf = GooString::format("/ForceBoldThreshold {0:.4g} def\n",
			    privateDicts[i].forceBoldThreshold);
      (*outputFunc)(outputStream, buf->c_str(), buf->getLength());
      delete buf;
    }
    if (privateDicts[i].languageGroup != 0) {
      buf = GooString::format("/LanguageGroup {0:d} def\n",
			    privateDicts[i].languageGroup);
      (*outputFunc)(outputStream, buf->c_str(), buf->getLength());
      delete buf;
    }
    if (privateDicts[i].expansionFactor != 0.06) {
      buf = GooString::format("/ExpansionFactor {0:.4g} def\n",
			    privateDicts[i].expansionFactor);
      (*outputFunc)(outputStream, buf->c_str(), buf->getLength());
      delete buf;
    }
    (*outputFunc)(outputStream, "currentdict end def\n", 20);
    (*outputFunc)(outputStream, "currentdict end put\n", 20);
  }
  (*outputFunc)(outputStream, "def\n", 4);

  // start the binary section
  offset = (nCIDs + 1) * (1 + gdBytes);
  buf = GooString::format("(Hex) {0:d} StartData\n",
			offset + charStrings->getLength());
  (*outputFunc)(outputStream, buf->c_str(), buf->getLength());
  delete buf;

  // write the charstring offset (CIDMap) table
  for (i = 0; i <= nCIDs; i += 6) {
    for (j = 0; j < 6 && i+j <= nCIDs; ++j) {
      if (i+j < nCIDs && cidMap[i+j] >= 0 && fdSelect) {
	buf2[0] = (char)fdSelect[cidMap[i+j]];
      } else {
	buf2[0] = (char)0;
      }
      n = offset + charStringOffsets[i+j];
      for (k = gdBytes; k >= 1; --k) {
	buf2[k] = (char)(n & 0xff);
	n >>= 8;
      }
      for (k = 0; k <= gdBytes; ++k) {
	buf = GooString::format("{0:02x}", buf2[k] & 0xff);
	(*outputFunc)(outputStream, buf->c_str(), buf->getLength());
	delete buf;
      }
    }
    (*outputFunc)(outputStream, "\n", 1);
  }

  // write the charstring data
  n = charStrings->getLength();
  for (i = 0; i < n; i += 32) {
    for (j = 0; j < 32 && i+j < n; ++j) {
      buf = GooString::format("{0:02x}", charStrings->getChar(i+j) & 0xff);
      (*outputFunc)(outputStream, buf->c_str(), buf->getLength());
      delete buf;
    }
    if (i + 32 >= n) {
      (*outputFunc)(outputStream, ">", 1);
    }
    (*outputFunc)(outputStream, "\n", 1);
  }

  gfree(charStringOffsets);
  delete charStrings;
  gfree(cidMap);
}

void FoFiType1C::convertToType0(const char *psName, const int *codeMap, int nCodes,
				FoFiOutputFunc outputFunc,
				void *outputStream) {
  int *cidMap;
  Type1CIndex subrIdx;
  Type1CIndexVal val;
  int nCIDs;
  GooString *buf;
  Type1CEexecBuf eb;
  bool ok;
  int fd, i, j, k;

  // compute the CID count and build the CID-to-GID mapping
  if (codeMap) {
    nCIDs = nCodes;
    cidMap = (int *)gmallocn(nCIDs, sizeof(int));
    for (i = 0; i < nCodes; ++i) {
      if (codeMap[i] >= 0 && codeMap[i] < nGlyphs) {
	cidMap[i] = codeMap[i];
      } else {
	cidMap[i] = -1;
      }
    }
  } else if (topDict.firstOp == 0x0c1e) {
    nCIDs = 0;
    for (i = 0; i < nGlyphs && i < charsetLength; ++i) {
      if (charset[i] >= nCIDs) {
	nCIDs = charset[i] + 1;
      }
    }
    cidMap = (int *)gmallocn(nCIDs, sizeof(int));
    for (i = 0; i < nCIDs; ++i) {
      cidMap[i] = -1;
    }
    for (i = 0; i < nGlyphs && i < charsetLength; ++i) {
      cidMap[charset[i]] = i;
    }
  } else {
    nCIDs = nGlyphs;
    cidMap = (int *)gmallocn(nCIDs, sizeof(int));
    for (i = 0; i < nCIDs; ++i) {
      cidMap[i] = i;
    }
  }

  if (privateDicts) {
  // write the descendant Type 1 fonts
    for (i = 0; i < nCIDs; i += 256) {

      //~ this assumes that all CIDs in this block have the same FD --
      //~ to handle multiple FDs correctly, need to somehow divide the
      //~ font up by FD; as a kludge we ignore CID 0, which is .notdef
      fd = 0;
      // if fdSelect is NULL, we have an 8-bit font, so just leave fd=0
      if (fdSelect) {
	for (j = i==0 ? 1 : 0; j < 256 && i+j < nCIDs; ++j) {
	  if (cidMap[i+j] >= 0) {
	    fd = fdSelect[cidMap[i+j]];
	    break;
	  }
	}
      }

      if (fd >= nFDs)
	continue;

      // font dictionary (unencrypted section)
      (*outputFunc)(outputStream, "16 dict begin\n", 14);
      (*outputFunc)(outputStream, "/FontName /", 11);
      (*outputFunc)(outputStream, psName, strlen(psName));
      buf = GooString::format("_{0:02x} def\n", i >> 8);
      (*outputFunc)(outputStream, buf->c_str(), buf->getLength());
      delete buf;
      (*outputFunc)(outputStream, "/FontType 1 def\n", 16);
      if (privateDicts[fd].hasFontMatrix) {
	buf = GooString::format("/FontMatrix [{0:.8g} {1:.8g} {2:.8g} {3:.8g} {4:.8g} {5:.8g}] def\n",
			      privateDicts[fd].fontMatrix[0],
			      privateDicts[fd].fontMatrix[1],
			      privateDicts[fd].fontMatrix[2],
			      privateDicts[fd].fontMatrix[3],
			      privateDicts[fd].fontMatrix[4],
			      privateDicts[fd].fontMatrix[5]);
	(*outputFunc)(outputStream, buf->c_str(), buf->getLength());
	delete buf;
      } else if (topDict.hasFontMatrix) {
	(*outputFunc)(outputStream, "/FontMatrix [1 0 0 1 0 0] def\n", 30);
      } else {
	(*outputFunc)(outputStream,
		      "/FontMatrix [0.001 0 0 0.001 0 0] def\n", 38);
      }
      buf = GooString::format("/FontBBox [{0:.4g} {1:.4g} {2:.4g} {3:.4g}] def\n",
			    topDict.fontBBox[0], topDict.fontBBox[1],
			    topDict.fontBBox[2], topDict.fontBBox[3]);
      (*outputFunc)(outputStream, buf->c_str(), buf->getLength());
      delete buf;
      buf = GooString::format("/PaintType {0:d} def\n", topDict.paintType);
      (*outputFunc)(outputStream, buf->c_str(), buf->getLength());
      delete buf;
      if (topDict.paintType != 0) {
	buf = GooString::format("/StrokeWidth {0:.4g} def\n", topDict.strokeWidth);
	(*outputFunc)(outputStream, buf->c_str(), buf->getLength());
	delete buf;
      }
      (*outputFunc)(outputStream, "/Encoding 256 array\n", 20);
      for (j = 0; j < 256 && i+j < nCIDs; ++j) {
	buf = GooString::format("dup {0:d} /c{1:02x} put\n", j, j);
	(*outputFunc)(outputStream, buf->c_str(), buf->getLength());
	delete buf;
      }
      if (j < 256) {
	buf = GooString::format("{0:d} 1 255 {{ 1 index exch /.notdef put }} for\n",
			      j);
	(*outputFunc)(outputStream, buf->c_str(), buf->getLength());
	delete buf;
      }
      (*outputFunc)(outputStream, "readonly def\n", 13);
      (*outputFunc)(outputStream, "currentdict end\n", 16);

      // start the binary section
      (*outputFunc)(outputStream, "currentfile eexec\n", 18);
      eb.outputFunc = outputFunc;
      eb.outputStream = outputStream;
      eb.ascii = true;
      eb.r1 = 55665;
      eb.line = 0;

      // start the private dictionary
      eexecWrite(&eb, "\x83\xca\x73\xd5");
      eexecWrite(&eb, "dup /Private 32 dict dup begin\n");
      eexecWrite(&eb, "/RD {string currentfile exch readstring pop}"
		" executeonly def\n");
      eexecWrite(&eb, "/ND {noaccess def} executeonly def\n");
      eexecWrite(&eb, "/NP {noaccess put} executeonly def\n");
      eexecWrite(&eb, "/MinFeature {16 16} def\n");
      eexecWrite(&eb, "/password 5839 def\n");
      if (privateDicts[fd].nBlueValues) {
	eexecWrite(&eb, "/BlueValues [");
	for (k = 0; k < privateDicts[fd].nBlueValues; ++k) {
	  buf = GooString::format("{0:s}{1:d}",
				k > 0 ? " " : "",
				privateDicts[fd].blueValues[k]);
	  eexecWrite(&eb, buf->c_str());
	  delete buf;
	}
	eexecWrite(&eb, "] def\n");
      }
      if (privateDicts[fd].nOtherBlues) {
	eexecWrite(&eb, "/OtherBlues [");
	for (k = 0; k < privateDicts[fd].nOtherBlues; ++k) {
	  buf = GooString::format("{0:s}{1:d}",
				k > 0 ? " " : "",
				privateDicts[fd].otherBlues[k]);
	  eexecWrite(&eb, buf->c_str());
	  delete buf;
	}
	eexecWrite(&eb, "] def\n");
      }
      if (privateDicts[fd].nFamilyBlues) {
	eexecWrite(&eb, "/FamilyBlues [");
	for (k = 0; k < privateDicts[fd].nFamilyBlues; ++k) {
	  buf = GooString::format("{0:s}{1:d}", k > 0 ? " " : "",
				privateDicts[fd].familyBlues[k]);
	  eexecWrite(&eb, buf->c_str());
	  delete buf;
	}
	eexecWrite(&eb, "] def\n");
      }
      if (privateDicts[fd].nFamilyOtherBlues) {
	eexecWrite(&eb, "/FamilyOtherBlues [");
	for (k = 0; k < privateDicts[fd].nFamilyOtherBlues; ++k) {
	  buf = GooString::format("{0:s}{1:d}", k > 0 ? " " : "",
				privateDicts[fd].familyOtherBlues[k]);
	  eexecWrite(&eb, buf->c_str());
	  delete buf;
	}
	eexecWrite(&eb, "] def\n");
      }
      if (privateDicts[fd].blueScale != 0.039625) {
	buf = GooString::format("/BlueScale {0:.4g} def\n",
			      privateDicts[fd].blueScale);
	eexecWrite(&eb, buf->c_str());
	delete buf;
      }
      if (privateDicts[fd].blueShift != 7) {
	buf = GooString::format("/BlueShift {0:d} def\n",
			      privateDicts[fd].blueShift);
	eexecWrite(&eb, buf->c_str());
	delete buf;
      }
      if (privateDicts[fd].blueFuzz != 1) {
	buf = GooString::format("/BlueFuzz {0:d} def\n",
			      privateDicts[fd].blueFuzz);
	eexecWrite(&eb, buf->c_str());
	delete buf;
      }
      if (privateDicts[fd].hasStdHW) {
	buf = GooString::format("/StdHW [{0:.4g}] def\n", privateDicts[fd].stdHW);
	eexecWrite(&eb, buf->c_str());
	delete buf;
      }
      if (privateDicts[fd].hasStdVW) {
	buf = GooString::format("/StdVW [{0:.4g}] def\n", privateDicts[fd].stdVW);
	eexecWrite(&eb, buf->c_str());
	delete buf;
      }
      if (privateDicts[fd].nStemSnapH) {
	eexecWrite(&eb, "/StemSnapH [");
	for (k = 0; k < privateDicts[fd].nStemSnapH; ++k) {
	  buf = GooString::format("{0:s}{1:.4g}",
				k > 0 ? " " : "", privateDicts[fd].stemSnapH[k]);
	  eexecWrite(&eb, buf->c_str());
	  delete buf;
	}
	eexecWrite(&eb, "] def\n");
      }
      if (privateDicts[fd].nStemSnapV) {
	eexecWrite(&eb, "/StemSnapV [");
	for (k = 0; k < privateDicts[fd].nStemSnapV; ++k) {
	  buf = GooString::format("{0:s}{1:.4g}",
				k > 0 ? " " : "", privateDicts[fd].stemSnapV[k]);
	  eexecWrite(&eb, buf->c_str());
	  delete buf;
	}
	eexecWrite(&eb, "] def\n");
      }
      if (privateDicts[fd].hasForceBold) {
	buf = GooString::format("/ForceBold {0:s} def\n",
			      privateDicts[fd].forceBold ? "true" : "false");
	eexecWrite(&eb, buf->c_str());
	delete buf;
      }
      if (privateDicts[fd].forceBoldThreshold != 0) {
	buf = GooString::format("/ForceBoldThreshold {0:.4g} def\n",
			      privateDicts[fd].forceBoldThreshold);
	eexecWrite(&eb, buf->c_str());
	delete buf;
      }
      if (privateDicts[fd].languageGroup != 0) {
	buf = GooString::format("/LanguageGroup {0:d} def\n",
			      privateDicts[fd].languageGroup);
	eexecWrite(&eb, buf->c_str());
	delete buf;
      }
      if (privateDicts[fd].expansionFactor != 0.06) {
	buf = GooString::format("/ExpansionFactor {0:.4g} def\n",
			      privateDicts[fd].expansionFactor);
	eexecWrite(&eb, buf->c_str());
	delete buf;
      }

      // set up the subroutines
      ok = true;
      getIndex(privateDicts[fd].subrsOffset, &subrIdx, &ok);
      if (!ok) {
	subrIdx.pos = -1;
      }

      // start the CharStrings
      eexecWrite(&eb, "2 index /CharStrings 256 dict dup begin\n");

      // write the .notdef CharString
      ok = true;
      getIndexVal(&charStringsIdx, 0, &val, &ok);
      if (ok) {
	eexecCvtGlyph(&eb, ".notdef", val.pos, val.len,
		      &subrIdx, &privateDicts[fd]);
      }

      // write the CharStrings
      for (j = 0; j < 256 && i+j < nCIDs; ++j) {
	if (cidMap[i+j] >= 0) {
	  ok = true;
	  getIndexVal(&charStringsIdx, cidMap[i+j], &val, &ok);
	  if (ok) {
	    buf = GooString::format("c{0:02x}", j);
	    eexecCvtGlyph(&eb, buf->c_str(), val.pos, val.len,
			  &subrIdx, &privateDicts[fd]);
	    delete buf;
	  }
	}
      }
      eexecWrite(&eb, "end\n");
      eexecWrite(&eb, "end\n");
      eexecWrite(&eb, "readonly put\n");
      eexecWrite(&eb, "noaccess put\n");
      eexecWrite(&eb, "dup /FontName get exch definefont pop\n");
      eexecWrite(&eb, "mark currentfile closefile\n");

      // trailer
      if (eb.line > 0) {
	(*outputFunc)(outputStream, "\n", 1);
      }
      for (j = 0; j < 8; ++j) {
	(*outputFunc)(outputStream, "0000000000000000000000000000000000000000000000000000000000000000\n", 65);
      }
      (*outputFunc)(outputStream, "cleartomark\n", 12);
    }
  } else {
    error(errSyntaxError, -1, "FoFiType1C::convertToType0 without privateDicts");
  }

  // write the Type 0 parent font
  (*outputFunc)(outputStream, "16 dict begin\n", 14);
  (*outputFunc)(outputStream, "/FontName /", 11);
  (*outputFunc)(outputStream, psName, strlen(psName));
  (*outputFunc)(outputStream, " def\n", 5);
  (*outputFunc)(outputStream, "/FontType 0 def\n", 16);
  if (topDict.hasFontMatrix) {
    buf = GooString::format("/FontMatrix [{0:.8g} {1:.8g} {2:.8g} {3:.8g} {4:.8g} {5:.8g}] def\n",
			  topDict.fontMatrix[0], topDict.fontMatrix[1],
			  topDict.fontMatrix[2], topDict.fontMatrix[3],
			  topDict.fontMatrix[4], topDict.fontMatrix[5]);
    (*outputFunc)(outputStream, buf->c_str(), buf->getLength());
    delete buf;
  } else {
    (*outputFunc)(outputStream, "/FontMatrix [1 0 0 1 0 0] def\n", 30);
  }
  (*outputFunc)(outputStream, "/FMapType 2 def\n", 16);
  (*outputFunc)(outputStream, "/Encoding [\n", 12);
  for (i = 0; i < nCIDs; i += 256) {
    buf = GooString::format("{0:d}\n", i >> 8);
    (*outputFunc)(outputStream, buf->c_str(), buf->getLength());
    delete buf;
  }
  (*outputFunc)(outputStream, "] def\n", 6);
  (*outputFunc)(outputStream, "/FDepVector [\n", 14);
  for (i = 0; i < nCIDs; i += 256) {
    (*outputFunc)(outputStream, "/", 1);
    (*outputFunc)(outputStream, psName, strlen(psName));
    buf = GooString::format("_{0:02x} findfont\n", i >> 8);
    (*outputFunc)(outputStream, buf->c_str(), buf->getLength());
    delete buf;
  }
  (*outputFunc)(outputStream, "] def\n", 6);
  (*outputFunc)(outputStream, "FontName currentdict end definefont pop\n", 40);

  gfree(cidMap);
}

void FoFiType1C::eexecCvtGlyph(Type1CEexecBuf *eb, const char *glyphName,
			       int offset, int nBytes,
			       Type1CIndex *subrIdx,
			       Type1CPrivateDict *pDict) {
  GooString *buf;
  GooString *charBuf;

  // generate the charstring
  charBuf = new GooString();
  cvtGlyph(offset, nBytes, charBuf, subrIdx, pDict, true);

  buf = GooString::format("/{0:s} {1:d} RD ", glyphName, charBuf->getLength());
  eexecWrite(eb, buf->c_str());
  delete buf;
  eexecWriteCharstring(eb, (unsigned char *)charBuf->c_str(),
		       charBuf->getLength());
  eexecWrite(eb, " ND\n");

  delete charBuf;
}

void FoFiType1C::cvtGlyph(int offset, int nBytes, GooString *charBuf,
			  Type1CIndex *subrIdx, Type1CPrivateDict *pDict,
			  bool top) {
  Type1CIndexVal val;
  bool ok, dFP;
  double d, dx, dy;
  unsigned short r2;
  unsigned char byte;
  int pos, subrBias, start, i, k;

  start = charBuf->getLength();
  if (top) {
    charBuf->append('\x49'); //73;
    charBuf->append('\x3A'); //58;
    charBuf->append('\x93'); //147;
    charBuf->append('\x86'); //134;
    nOps = 0;
    nHints = 0;
    firstOp = true;
    openPath = false;
  }

  pos = offset;
  while (pos < offset + nBytes) {
    ok = true;
    pos = getOp(pos, true, &ok);
    if (!ok) {
      break;
    }
    if (!ops[nOps - 1].isNum) {
      --nOps; // drop the operator
      switch (ops[nOps].op) {
      case 0x0001:		// hstem
	if (firstOp) {
	  cvtGlyphWidth(nOps & 1, charBuf, pDict);
	  firstOp = false;
	}
	if (nOps & 1) {
	  //~ error(-1, "Wrong number of args (%d) to Type 2 hstem", nOps);
	}
	d = 0;
	dFP = false;
	for (k = 0; k < nOps; k += 2) {
	  // convert Type 2 edge hints (-20 or -21) to Type 1 ghost hints
	  if (ops[k+1].num < 0) {
	    d += ops[k].num + ops[k+1].num;
	    dFP |= ops[k].isFP | ops[k+1].isFP;
	    cvtNum(d, dFP, charBuf);
	    cvtNum(-ops[k+1].num, ops[k+1].isFP, charBuf);
	  } else {
	    d += ops[k].num;
	    dFP |= ops[k].isFP;
	    cvtNum(d, dFP, charBuf);
	    cvtNum(ops[k+1].num, ops[k+1].isFP, charBuf);
	    d += ops[k+1].num;
	    dFP |= ops[k+1].isFP;
	  }
	  charBuf->append((char)1);
	}
	nHints += nOps / 2;
	nOps = 0;
	break;
      case 0x0003:		// vstem
	if (firstOp) {
	  cvtGlyphWidth(nOps & 1, charBuf, pDict);
	  firstOp = false;
	}
	if (nOps & 1) {
	  //~ error(-1, "Wrong number of args (%d) to Type 2 vstem", nOps);
	}
	d = 0;
	dFP = false;
	for (k = 0; k < nOps; k += 2) {
	  // convert Type 2 edge hints (-20 or -21) to Type 1 ghost hints
	  if (ops[k+1].num < 0) {
	    d += ops[k].num + ops[k+1].num;
	    dFP |= ops[k].isFP | ops[k+1].isFP;
	    cvtNum(d, dFP, charBuf);
	    cvtNum(-ops[k+1].num, ops[k+1].isFP, charBuf);
	  } else {
	    d += ops[k].num;
	    dFP |= ops[k].isFP;
	    cvtNum(d, dFP, charBuf);
	    cvtNum(ops[k+1].num, ops[k+1].isFP, charBuf);
	    d += ops[k+1].num;
	    dFP |= ops[k+1].isFP;
	  }
	  charBuf->append((char)3);
	}
	nHints += nOps / 2;
	nOps = 0;
	break;
      case 0x0004:		// vmoveto
	if (firstOp) {
	  cvtGlyphWidth(nOps == 2, charBuf, pDict);
	  firstOp = false;
	}
	if (openPath) {
	  charBuf->append((char)9);
	  openPath = false;
	}
	if (nOps != 1) {
	  //~ error(-1, "Wrong number of args (%d) to Type 2 vmoveto", nOps);
	}
	cvtNum(ops[0].num, ops[0].isFP, charBuf);
	charBuf->append((char)4);
	nOps = 0;
	break;
      case 0x0005:		// rlineto
	if (nOps < 2 || nOps % 2 != 0) {
	  //~ error(-1, "Wrong number of args (%d) to Type 2 rlineto", nOps);
	}
	for (k = 0; k < nOps; k += 2) {
	  cvtNum(ops[k].num, ops[k].isFP, charBuf);
	  cvtNum(ops[k+1].num, ops[k+1].isFP, charBuf);
	  charBuf->append((char)5);
	}
	nOps = 0;
	openPath = true;
	break;
      case 0x0006:		// hlineto
	if (nOps < 1) {
	  //~ error(-1, "Wrong number of args (%d) to Type 2 hlineto", nOps);
	}
	for (k = 0; k < nOps; ++k) {
	  cvtNum(ops[k].num, ops[k].isFP, charBuf);
	  charBuf->append((char)((k & 1) ? 7 : 6));
	}
	nOps = 0;
	openPath = true;
	break;
      case 0x0007:		// vlineto
	if (nOps < 1) {
	  //~ error(-1, "Wrong number of args (%d) to Type 2 vlineto", nOps);
	}
	for (k = 0; k < nOps; ++k) {
	  cvtNum(ops[k].num, ops[k].isFP, charBuf);
	  charBuf->append((char)((k & 1) ? 6 : 7));
	}
	nOps = 0;
	openPath = true;
	break;
      case 0x0008:		// rrcurveto
	if (nOps < 6 || nOps % 6 != 0) {
	  //~ error(-1, "Wrong number of args (%d) to Type 2 rrcurveto", nOps);
	}
	for (k = 0; k < nOps; k += 6) {
	  cvtNum(ops[k].num, ops[k].isFP, charBuf);
	  cvtNum(ops[k+1].num, ops[k+1].isFP, charBuf);
	  cvtNum(ops[k+2].num, ops[k+2].isFP, charBuf);
	  cvtNum(ops[k+3].num, ops[k+3].isFP, charBuf);
	  cvtNum(ops[k+4].num, ops[k+4].isFP, charBuf);
	  cvtNum(ops[k+5].num, ops[k+5].isFP, charBuf);
	  charBuf->append((char)8);
	}
	nOps = 0;
	openPath = true;
	break;
      case 0x000a:		// callsubr
	if (nOps >= 1) {
	  subrBias = (subrIdx->len < 1240)
	               ? 107 : (subrIdx->len < 33900) ? 1131 : 32768;
	  k = subrBias + (int)ops[nOps - 1].num;
	  --nOps;
	  ok = true;
	  getIndexVal(subrIdx, k, &val, &ok);
	  if (likely(ok && val.pos != offset)) {
	    cvtGlyph(val.pos, val.len, charBuf, subrIdx, pDict, false);
	  }
	} else {
	  //~ error(-1, "Too few args to Type 2 callsubr");
	}
	// don't clear the stack
	break;
      case 0x000b:		// return
	// don't clear the stack
	break;
      case 0x000e:		// endchar / seac
	if (firstOp) {
	  cvtGlyphWidth(nOps == 1 || nOps == 5, charBuf, pDict);
	  firstOp = false;
	}
	if (openPath) {
	  charBuf->append((char)9);
	  openPath = false;
	}
	if (nOps == 4) {
	  cvtNum(0, false, charBuf);
	  cvtNum(ops[0].num, ops[0].isFP, charBuf);
	  cvtNum(ops[1].num, ops[1].isFP, charBuf);
	  cvtNum(ops[2].num, ops[2].isFP, charBuf);
	  cvtNum(ops[3].num, ops[3].isFP, charBuf);
	  charBuf->append((char)12)->append((char)6);
	} else if (nOps == 0) {
	  charBuf->append((char)14);
	} else {
	  //~ error(-1, "Wrong number of args (%d) to Type 2 endchar", nOps);
	}
	nOps = 0;
	break;
      case 0x000f:		// (obsolete)
	// this op is ignored, but we need the glyph width
	if (firstOp) {
	  cvtGlyphWidth(nOps > 0, charBuf, pDict);
	  firstOp = false;
	}
	nOps = 0;
	break;
      case 0x0010:		// blend
	//~ error(-1, "Unimplemented Type 2 charstring op: %d", file[i]);
	nOps = 0;
	break;
      case 0x0012:		// hstemhm
	// ignored
	if (firstOp) {
	  cvtGlyphWidth(nOps & 1, charBuf, pDict);
	  firstOp = false;
	}
	if (nOps & 1) {
	  //~ error(-1, "Wrong number of args (%d) to Type 2 hstemhm", nOps);
	}
	nHints += nOps / 2;
	nOps = 0;
	break;
      case 0x0013:		// hintmask
	// ignored
	if (firstOp) {
	  cvtGlyphWidth(nOps & 1, charBuf, pDict);
	  firstOp = false;
	}
	if (nOps > 0) {
	  if (nOps & 1) {
	    //~ error(-1, "Wrong number of args (%d) to Type 2 hintmask/vstemhm",
	    //~       nOps);
	  }
	  nHints += nOps / 2;
	}
	pos += (nHints + 7) >> 3;
	nOps = 0;
	break;
      case 0x0014:		// cntrmask
	// ignored
	if (firstOp) {
	  cvtGlyphWidth(nOps & 1, charBuf, pDict);
	  firstOp = false;
	}
	if (nOps > 0) {
	  if (nOps & 1) {
	    //~ error(-1, "Wrong number of args (%d) to Type 2 cntrmask/vstemhm",
	    //~       nOps);
	  }
	  nHints += nOps / 2;
	}
	pos += (nHints + 7) >> 3;
	nOps = 0;
	break;
      case 0x0015:		// rmoveto
	if (firstOp) {
	  cvtGlyphWidth(nOps == 3, charBuf, pDict);
	  firstOp = false;
	}
	if (openPath) {
	  charBuf->append((char)9);
	  openPath = false;
	}
	if (nOps != 2) {
	  //~ error(-1, "Wrong number of args (%d) to Type 2 rmoveto", nOps);
	}
	cvtNum(ops[0].num, ops[0].isFP, charBuf);
	cvtNum(ops[1].num, ops[1].isFP, charBuf);
	charBuf->append((char)21);
	nOps = 0;
	break;
      case 0x0016:		// hmoveto
	if (firstOp) {
	  cvtGlyphWidth(nOps == 2, charBuf, pDict);
	  firstOp = false;
	}
	if (openPath) {
	  charBuf->append((char)9);
	  openPath = false;
	}
	if (nOps != 1) {
	  //~ error(-1, "Wrong number of args (%d) to Type 2 hmoveto", nOps);
	}
	cvtNum(ops[0].num, ops[0].isFP, charBuf);
	charBuf->append((char)22);
	nOps = 0;
	break;
      case 0x0017:		// vstemhm
	// ignored
	if (firstOp) {
	  cvtGlyphWidth(nOps & 1, charBuf, pDict);
	  firstOp = false;
	}
	if (nOps & 1) {
	  //~ error(-1, "Wrong number of args (%d) to Type 2 vstemhm", nOps);
	}
	nHints += nOps / 2;
	nOps = 0;
	break;
      case 0x0018:		// rcurveline
	if (nOps < 8 || (nOps - 2) % 6 != 0) {
	  //~ error(-1, "Wrong number of args (%d) to Type 2 rcurveline", nOps);
	}
	for (k = 0; k < nOps - 2; k += 6) {
	  cvtNum(ops[k].num, ops[k].isFP, charBuf);
	  cvtNum(ops[k+1].num, ops[k+1].isFP, charBuf);
	  cvtNum(ops[k+2].num, ops[k+2].isFP, charBuf);
	  cvtNum(ops[k+3].num, ops[k+3].isFP, charBuf);
	  cvtNum(ops[k+4].num, ops[k+4].isFP, charBuf);
	  cvtNum(ops[k+5].num, ops[k+5].isFP, charBuf);
	  charBuf->append((char)8);
	}
	cvtNum(ops[k].num, ops[k].isFP, charBuf);
	cvtNum(ops[k+1].num, ops[k].isFP, charBuf);
	charBuf->append((char)5);
	nOps = 0;
	openPath = true;
	break;
      case 0x0019:		// rlinecurve
	if (nOps < 8 || (nOps - 6) % 2 != 0) {
	  //~ error(-1, "Wrong number of args (%d) to Type 2 rlinecurve", nOps);
	}
	for (k = 0; k < nOps - 6; k += 2) {
	  cvtNum(ops[k].num, ops[k].isFP, charBuf);
	  cvtNum(ops[k+1].num, ops[k].isFP, charBuf);
	  charBuf->append((char)5);
	}
	cvtNum(ops[k].num, ops[k].isFP, charBuf);
	cvtNum(ops[k+1].num, ops[k+1].isFP, charBuf);
	cvtNum(ops[k+2].num, ops[k+2].isFP, charBuf);
	cvtNum(ops[k+3].num, ops[k+3].isFP, charBuf);
	cvtNum(ops[k+4].num, ops[k+4].isFP, charBuf);
	cvtNum(ops[k+5].num, ops[k+5].isFP, charBuf);
	charBuf->append((char)8);
	nOps = 0;
	openPath = true;
	break;
      case 0x001a:		// vvcurveto
	if (nOps < 4 || !(nOps % 4 == 0 || (nOps-1) % 4 == 0)) {
	  //~ error(-1, "Wrong number of args (%d) to Type 2 vvcurveto", nOps);
	}
	if (nOps % 2 == 1) {
	  cvtNum(ops[0].num, ops[0].isFP, charBuf);
	  cvtNum(ops[1].num, ops[1].isFP, charBuf);
	  cvtNum(ops[2].num, ops[2].isFP, charBuf);
	  cvtNum(ops[3].num, ops[3].isFP, charBuf);
	  cvtNum(0, false, charBuf);
	  cvtNum(ops[4].num, ops[4].isFP, charBuf);
	  charBuf->append((char)8);
	  k = 5;
	} else {
	  k = 0;
	}
	for (; k < nOps; k += 4) {
	  cvtNum(0, false, charBuf);
	  cvtNum(ops[k].num, ops[k].isFP, charBuf);
	  cvtNum(ops[k+1].num, ops[k+1].isFP, charBuf);
	  cvtNum(ops[k+2].num, ops[k+2].isFP, charBuf);
	  cvtNum(0, false, charBuf);
	  cvtNum(ops[k+3].num, ops[k+3].isFP, charBuf);
	  charBuf->append((char)8);
	}
	nOps = 0;
	openPath = true;
	break;
      case 0x001b:		// hhcurveto
	if (nOps < 4 || !(nOps % 4 == 0 || (nOps-1) % 4 == 0)) {
	  //~ error(-1, "Wrong number of args (%d) to Type 2 hhcurveto", nOps);
	}
	if (nOps % 2 == 1) {
	  cvtNum(ops[1].num, ops[1].isFP, charBuf);
	  cvtNum(ops[0].num, ops[0].isFP, charBuf);
	  cvtNum(ops[2].num, ops[2].isFP, charBuf);
	  cvtNum(ops[3].num, ops[3].isFP, charBuf);
	  cvtNum(ops[4].num, ops[4].isFP, charBuf);
	  cvtNum(0, false, charBuf);
	  charBuf->append((char)8);
	  k = 5;
	} else {
	  k = 0;
	}
	for (; k < nOps; k += 4) {
	  cvtNum(ops[k].num, ops[k].isFP, charBuf);
	  cvtNum(0, false, charBuf);
	  cvtNum(ops[k+1].num, ops[k+1].isFP, charBuf);
	  cvtNum(ops[k+2].num, ops[k+2].isFP, charBuf);
	  cvtNum(ops[k+3].num, ops[k+3].isFP, charBuf);
	  cvtNum(0, false, charBuf);
	  charBuf->append((char)8);
	}
	nOps = 0;
	openPath = true;
	break;
      case 0x001d:		// callgsubr
	if (nOps >= 1) {
	  k = gsubrBias + (int)ops[nOps - 1].num;
	  --nOps;
	  ok = true;
	  getIndexVal(&gsubrIdx, k, &val, &ok);
	  if (likely(ok && val.pos != offset)) {
	    cvtGlyph(val.pos, val.len, charBuf, subrIdx, pDict, false);
	  }
	} else {
	  //~ error(-1, "Too few args to Type 2 callgsubr");
	}
	// don't clear the stack
	break;
      case 0x001e:		// vhcurveto
	if (nOps < 4 || !(nOps % 4 == 0 || (nOps-1) % 4 == 0)) {
	  //~ error(-1, "Wrong number of args (%d) to Type 2 vhcurveto", nOps);
	}
	for (k = 0; k < nOps && k != nOps-5; k += 4) {
	  if (k % 8 == 0) {
	    cvtNum(ops[k].num, ops[k].isFP, charBuf);
	    cvtNum(ops[k+1].num, ops[k+1].isFP, charBuf);
	    cvtNum(ops[k+2].num, ops[k+2].isFP, charBuf);
	    cvtNum(ops[k+3].num, ops[k+3].isFP, charBuf);
	    charBuf->append((char)30);
	  } else {
	    cvtNum(ops[k].num, ops[k].isFP, charBuf);
	    cvtNum(ops[k+1].num, ops[k+1].isFP, charBuf);
	    cvtNum(ops[k+2].num, ops[k+2].isFP, charBuf);
	    cvtNum(ops[k+3].num, ops[k+3].isFP, charBuf);
	    charBuf->append((char)31);
	  }
	}
	if (k == nOps-5) {
	  if (k % 8 == 0) {
	    cvtNum(0, false, charBuf);
	    cvtNum(ops[k].num, ops[k].isFP, charBuf);
	    cvtNum(ops[k+1].num, ops[k+1].isFP, charBuf);
	    cvtNum(ops[k+2].num, ops[k+2].isFP, charBuf);
	    cvtNum(ops[k+3].num, ops[k+3].isFP, charBuf);
	    cvtNum(ops[k+4].num, ops[k+4].isFP, charBuf);
	  } else {
	    cvtNum(ops[k].num, ops[k].isFP, charBuf);
	    cvtNum(0, false, charBuf);
	    cvtNum(ops[k+1].num, ops[k+1].isFP, charBuf);
	    cvtNum(ops[k+2].num, ops[k+2].isFP, charBuf);
	    cvtNum(ops[k+4].num, ops[k+4].isFP, charBuf);
	    cvtNum(ops[k+3].num, ops[k+3].isFP, charBuf);
	  }
	  charBuf->append((char)8);
	}
	nOps = 0;
	openPath = true;
	break;
      case 0x001f:		// hvcurveto
	if (nOps < 4 || !(nOps % 4 == 0 || (nOps-1) % 4 == 0)) {
	  //~ error(-1, "Wrong number of args (%d) to Type 2 hvcurveto", nOps);
	}
	for (k = 0; k < nOps && k != nOps-5; k += 4) {
	  if (k % 8 == 0) {
	    cvtNum(ops[k].num, ops[k].isFP, charBuf);
	    cvtNum(ops[k+1].num, ops[k+1].isFP, charBuf);
	    cvtNum(ops[k+2].num, ops[k+2].isFP, charBuf);
	    cvtNum(ops[k+3].num, ops[k+3].isFP, charBuf);
	    charBuf->append((char)31);
	  } else {
	    cvtNum(ops[k].num, ops[k].isFP, charBuf);
	    cvtNum(ops[k+1].num, ops[k+1].isFP, charBuf);
	    cvtNum(ops[k+2].num, ops[k+2].isFP, charBuf);
	    cvtNum(ops[k+3].num, ops[k+3].isFP, charBuf);
	    charBuf->append((char)30);
	  }
	}
	if (k == nOps-5) {
	  if (k % 8 == 0) {
	    cvtNum(ops[k].num, ops[k].isFP, charBuf);
	    cvtNum(0, false, charBuf);
	    cvtNum(ops[k+1].num, ops[k+1].isFP, charBuf);
	    cvtNum(ops[k+2].num, ops[k+2].isFP, charBuf);
	    cvtNum(ops[k+4].num, ops[k+4].isFP, charBuf);
	    cvtNum(ops[k+3].num, ops[k+3].isFP, charBuf);
	  } else {
	    cvtNum(0, false, charBuf);
	    cvtNum(ops[k].num, ops[k].isFP, charBuf);
	    cvtNum(ops[k+1].num, ops[k+1].isFP, charBuf);
	    cvtNum(ops[k+2].num, ops[k+2].isFP, charBuf);
	    cvtNum(ops[k+3].num, ops[k+3].isFP, charBuf);
	    cvtNum(ops[k+4].num, ops[k+4].isFP, charBuf);
	  }
	  charBuf->append((char)8);
	}
	nOps = 0;
	openPath = true;
	break;
      case 0x0c00:		// dotsection (should be Type 1 only?)
	// ignored
	nOps = 0;
	break;
      case 0x0c03:		// and
      case 0x0c04:		// or
      case 0x0c05:		// not
      case 0x0c08:		// store
      case 0x0c09:		// abs
      case 0x0c0a:		// add
      case 0x0c0b:		// sub
      case 0x0c0c:		// div
      case 0x0c0d:		// load
      case 0x0c0e:		// neg
      case 0x0c0f:		// eq
      case 0x0c12:		// drop
      case 0x0c14:		// put
      case 0x0c15:		// get
      case 0x0c16:		// ifelse
      case 0x0c17:		// random
      case 0x0c18:		// mul
      case 0x0c1a:		// sqrt
      case 0x0c1b:		// dup
      case 0x0c1c:		// exch
      case 0x0c1d:		// index
      case 0x0c1e:		// roll
	//~ error(-1, "Unimplemented Type 2 charstring op: 12.%d", file[i+1]);
	nOps = 0;
	break;
      case 0x0c22:		// hflex
	if (nOps != 7) {
	  //~ error(-1, "Wrong number of args (%d) to Type 2 hflex", nOps);
	}
	cvtNum(ops[0].num, ops[0].isFP, charBuf);
	cvtNum(0, false, charBuf);
	cvtNum(ops[1].num, ops[1].isFP, charBuf);
	cvtNum(ops[2].num, ops[2].isFP, charBuf);
	cvtNum(ops[3].num, ops[3].isFP, charBuf);
	cvtNum(0, false, charBuf);
	charBuf->append((char)8);
	cvtNum(ops[4].num, ops[4].isFP, charBuf);
	cvtNum(0, false, charBuf);
	cvtNum(ops[5].num, ops[5].isFP, charBuf);
	cvtNum(-ops[2].num, ops[2].isFP, charBuf);
	cvtNum(ops[6].num, ops[6].isFP, charBuf);
	cvtNum(0, false, charBuf);
	charBuf->append((char)8);
	nOps = 0;
	openPath = true;
	break;
      case 0x0c23:		// flex
	if (nOps != 13) {
	  //~ error(-1, "Wrong number of args (%d) to Type 2 flex", nOps);
	}
	cvtNum(ops[0].num, ops[0].isFP, charBuf);
	cvtNum(ops[1].num, ops[1].isFP, charBuf);
	cvtNum(ops[2].num, ops[2].isFP, charBuf);
	cvtNum(ops[3].num, ops[3].isFP, charBuf);
	cvtNum(ops[4].num, ops[4].isFP, charBuf);
	cvtNum(ops[5].num, ops[5].isFP, charBuf);
	charBuf->append((char)8);
	cvtNum(ops[6].num, ops[6].isFP, charBuf);
	cvtNum(ops[7].num, ops[7].isFP, charBuf);
	cvtNum(ops[8].num, ops[8].isFP, charBuf);
	cvtNum(ops[9].num, ops[9].isFP, charBuf);
	cvtNum(ops[10].num, ops[10].isFP, charBuf);
	cvtNum(ops[11].num, ops[11].isFP, charBuf);
	charBuf->append((char)8);
	nOps = 0;
	openPath = true;
	break;
      case 0x0c24:		// hflex1
	if (nOps != 9) {
	  //~ error(-1, "Wrong number of args (%d) to Type 2 hflex1", nOps);
	}
	cvtNum(ops[0].num, ops[0].isFP, charBuf);
	cvtNum(ops[1].num, ops[1].isFP, charBuf);
	cvtNum(ops[2].num, ops[2].isFP, charBuf);
	cvtNum(ops[3].num, ops[3].isFP, charBuf);
	cvtNum(ops[4].num, ops[4].isFP, charBuf);
	cvtNum(0, false, charBuf);
	charBuf->append((char)8);
	cvtNum(ops[5].num, ops[5].isFP, charBuf);
	cvtNum(0, false, charBuf);
	cvtNum(ops[6].num, ops[6].isFP, charBuf);
	cvtNum(ops[7].num, ops[7].isFP, charBuf);
	cvtNum(ops[8].num, ops[8].isFP, charBuf);
	cvtNum(-(ops[1].num + ops[3].num + ops[7].num),
	       ops[1].isFP | ops[3].isFP | ops[7].isFP, charBuf);
	charBuf->append((char)8);
	nOps = 0;
	openPath = true;
	break;
      case 0x0c25:		// flex1
	if (nOps != 11) {
	  //~ error(-1, "Wrong number of args (%d) to Type 2 flex1", nOps);
	}
	cvtNum(ops[0].num, ops[0].isFP, charBuf);
	cvtNum(ops[1].num, ops[1].isFP, charBuf);
	cvtNum(ops[2].num, ops[2].isFP, charBuf);
	cvtNum(ops[3].num, ops[3].isFP, charBuf);
	cvtNum(ops[4].num, ops[4].isFP, charBuf);
	cvtNum(ops[5].num, ops[5].isFP, charBuf);
	charBuf->append((char)8);
	cvtNum(ops[6].num, ops[6].isFP, charBuf);
	cvtNum(ops[7].num, ops[7].isFP, charBuf);
	cvtNum(ops[8].num, ops[8].isFP, charBuf);
	cvtNum(ops[9].num, ops[9].isFP, charBuf);
	dx = ops[0].num + ops[2].num + ops[4].num + ops[6].num + ops[8].num;
	dy = ops[1].num + ops[3].num + ops[5].num + ops[7].num + ops[9].num;
	if (fabs(dx) > fabs(dy)) {
	  cvtNum(ops[10].num, ops[10].isFP, charBuf);
	  cvtNum(-dy, ops[1].isFP | ops[3].isFP | ops[5].isFP |
		      ops[7].isFP | ops[9].isFP, charBuf);
	} else {
	  cvtNum(-dx, ops[0].isFP | ops[2].isFP | ops[4].isFP |
		      ops[6].isFP | ops[8].isFP, charBuf);
	  cvtNum(ops[10].num, ops[10].isFP, charBuf);
	}
	charBuf->append((char)8);
	nOps = 0;
	openPath = true;
	break;
      default:
	//~ error(-1, "Illegal Type 2 charstring op: %04x",
	//~       ops[nOps].op);
	nOps = 0;
	break;
      }
    }
  }

  // charstring encryption
  if (top) {
    r2 = 4330;
    for (i = start; i < charBuf->getLength(); ++i) {
      byte = charBuf->getChar(i) ^ (r2 >> 8);
      charBuf->setChar(i, byte);
      r2 = (byte + r2) * 52845 + 22719;
    }
  }
}

void FoFiType1C::cvtGlyphWidth(bool useOp, GooString *charBuf,
			       Type1CPrivateDict *pDict) {
  double w;
  bool wFP;
  int i;

  if (useOp) {
    w = pDict->nominalWidthX + ops[0].num;
    wFP = pDict->nominalWidthXFP | ops[0].isFP;
    for (i = 1; i < nOps; ++i) {
      ops[i-1] = ops[i];
    }
    --nOps;
  } else {
    w = pDict->defaultWidthX;
    wFP = pDict->defaultWidthXFP;
  }
  cvtNum(0, false, charBuf);
  cvtNum(w, wFP, charBuf);
  charBuf->append((char)13);
}

void FoFiType1C::cvtNum(double x, bool isFP, GooString *charBuf) const {
  unsigned char buf[12];
  int y, n;

  n = 0;
  if (isFP) {
    if (x >= -32768 && x < 32768) {
      y = (int)(x * 256.0);
      buf[0] = 255;
      buf[1] = (unsigned char)(y >> 24);
      buf[2] = (unsigned char)(y >> 16);
      buf[3] = (unsigned char)(y >> 8);
      buf[4] = (unsigned char)y;
      buf[5] = 255;
      buf[6] = 0;
      buf[7] = 0;
      buf[8] = 1;
      buf[9] = 0;
      buf[10] = 12;
      buf[11] = 12;
      n = 12;
    } else {
      //~ error(-1, "Type 2 fixed point constant out of range");
    }
  } else {
    y = (int)x;
    if (y >= -107 && y <= 107) {
      buf[0] = (unsigned char)(y + 139);
      n = 1;
    } else if (y > 107 && y <= 1131) {
      y -= 108;
      buf[0] = (unsigned char)((y >> 8) + 247);
      buf[1] = (unsigned char)(y & 0xff);
      n = 2;
    } else if (y < -107 && y >= -1131) {
      y = -y - 108;
      buf[0] = (unsigned char)((y >> 8) + 251);
      buf[1] = (unsigned char)(y & 0xff);
      n = 2;
    } else {
      buf[0] = 255;
      buf[1] = (unsigned char)(y >> 24);
      buf[2] = (unsigned char)(y >> 16);
      buf[3] = (unsigned char)(y >> 8);
      buf[4] = (unsigned char)y;
      n = 5;
    }
  }
  charBuf->append((char *)buf, n);
}

void FoFiType1C::eexecWrite(Type1CEexecBuf *eb, const char *s) const {
  unsigned char *p;
  unsigned char x;

  for (p = (unsigned char *)s; *p; ++p) {
    x = *p ^ (eb->r1 >> 8);
    eb->r1 = (x + eb->r1) * 52845 + 22719;
    if (eb->ascii) {
      (*eb->outputFunc)(eb->outputStream, &hexChars[x >> 4], 1);
      (*eb->outputFunc)(eb->outputStream, &hexChars[x & 0x0f], 1);
      eb->line += 2;
      if (eb->line == 64) {
	(*eb->outputFunc)(eb->outputStream, "\n", 1);
	eb->line = 0;
      }
    } else {
      (*eb->outputFunc)(eb->outputStream, (char *)&x, 1);
    }
  }
}

void FoFiType1C::eexecWriteCharstring(Type1CEexecBuf *eb,
				      const unsigned char *s, int n) const {
  unsigned char x;
  int i;

  // eexec encryption
  for (i = 0; i < n; ++i) {
    x = s[i] ^ (eb->r1 >> 8);
    eb->r1 = (x + eb->r1) * 52845 + 22719;
    if (eb->ascii) {
      (*eb->outputFunc)(eb->outputStream, &hexChars[x >> 4], 1);
      (*eb->outputFunc)(eb->outputStream, &hexChars[x & 0x0f], 1);
      eb->line += 2;
      if (eb->line == 64) {
	(*eb->outputFunc)(eb->outputStream, "\n", 1);
	eb->line = 0;
      }
    } else {
      (*eb->outputFunc)(eb->outputStream, (char *)&x, 1);
    }
  }
}

void FoFiType1C::writePSString(const char *s, FoFiOutputFunc outputFunc,
			       void *outputStream) const {
  char buf[80];
  const char *p;
  int i, c;

  i = 0;
  buf[i++] = '(';
  for (p = s; *p; ++p) {
    c = *p & 0xff;
    if (c == '(' || c == ')' || c == '\\') {
      buf[i++] = '\\';
      buf[i++] = c;
    } else if (c < 0x20 || c >= 0x80) {
      buf[i++] = '\\';
      buf[i++] = '0' + ((c >> 6) & 7);
      buf[i++] = '0' + ((c >> 3) & 7);
      buf[i++] = '0' + (c & 7);
    } else {
      buf[i++] = c;
    }
    if (i >= 64) {
      buf[i++] = '\\';
      buf[i++] = '\n';
      (*outputFunc)(outputStream, buf, i);
      i = 0;
    }
  }
  buf[i++] = ')';
  (*outputFunc)(outputStream, buf, i);
}

bool FoFiType1C::parse() {
  Type1CIndex fdIdx;
  Type1CIndexVal val;
  int i;

  parsedOk = true;

  // some tools embed Type 1C fonts with an extra whitespace char at
  // the beginning
  if (len > 0 && file[0] != '\x01') {
    ++file;
    --len;
  }

  // find the indexes
  getIndex(getU8(2, &parsedOk), &nameIdx, &parsedOk);
  getIndex(nameIdx.endPos, &topDictIdx, &parsedOk);
  getIndex(topDictIdx.endPos, &stringIdx, &parsedOk);
  getIndex(stringIdx.endPos, &gsubrIdx, &parsedOk);
  if (!parsedOk) {
    return false;
  }
  gsubrBias = (gsubrIdx.len < 1240) ? 107
                                    : (gsubrIdx.len < 33900) ? 1131 : 32768;

  // read the first font name
  getIndexVal(&nameIdx, 0, &val, &parsedOk);
  if (!parsedOk) {
    return false;
  }
  name = new GooString((char *)&file[val.pos], val.len);

  // read the top dict for the first font
  readTopDict();

  // for CID fonts: read the FDArray dicts and private dicts
  if (topDict.firstOp == 0x0c1e) {
    if (topDict.fdArrayOffset == 0) {
      nFDs = 1;
      privateDicts = (Type1CPrivateDict *)gmalloc(sizeof(Type1CPrivateDict));
      readPrivateDict(0, 0, &privateDicts[0]);
    } else {
      getIndex(topDict.fdArrayOffset, &fdIdx, &parsedOk);
      if (!parsedOk) {
	return false;
      }
      nFDs = fdIdx.len;
      privateDicts = (Type1CPrivateDict *)
	                 gmallocn(nFDs, sizeof(Type1CPrivateDict));
      for (i = 0; i < nFDs; ++i) {
	getIndexVal(&fdIdx, i, &val, &parsedOk);
	if (!parsedOk) {
	  return false;
	}
	readFD(val.pos, val.len, &privateDicts[i]);
      }
    }

  // for 8-bit fonts: read the private dict
  } else {
    nFDs = 1;
    privateDicts = (Type1CPrivateDict *)gmalloc(sizeof(Type1CPrivateDict));
    readPrivateDict(topDict.privateOffset, topDict.privateSize,
		    &privateDicts[0]);
  }

  // check for parse errors in the private dict(s)
  if (!parsedOk) {
    return false;
  }

  // get the charstrings index
  if (topDict.charStringsOffset <= 0) {
    parsedOk = false;
    return false;
  }
  getIndex(topDict.charStringsOffset, &charStringsIdx, &parsedOk);
  if (!parsedOk) {
    return false;
  }
  nGlyphs = charStringsIdx.len;

  // for CID fonts: read the FDSelect table
  if (topDict.firstOp == 0x0c1e) {
    readFDSelect();
    if (!parsedOk) {
      return false;
    }
  }

  // read the charset
  if (!readCharset()) {
    parsedOk = false;
    return false;
  }

  // for 8-bit fonts: build the encoding
  if (topDict.firstOp != 0x0c14 && topDict.firstOp != 0x0c1e) {
    buildEncoding();
    if (!parsedOk) {
      return false;
    }
  }

  return parsedOk;
}

void FoFiType1C::readTopDict() {
  Type1CIndexVal topDictPtr;
  int pos;

  topDict.firstOp = -1;
  topDict.versionSID = 0;
  topDict.noticeSID = 0;
  topDict.copyrightSID = 0;
  topDict.fullNameSID = 0;
  topDict.familyNameSID = 0;
  topDict.weightSID = 0;
  topDict.isFixedPitch = 0;
  topDict.italicAngle = 0;
  topDict.underlinePosition = -100;
  topDict.underlineThickness = 50;
  topDict.paintType = 0;
  topDict.charstringType = 2;
  topDict.fontMatrix[0] = 0.001;
  topDict.fontMatrix[1] = 0;
  topDict.fontMatrix[2] = 0;
  topDict.fontMatrix[3] = 0.001;
  topDict.fontMatrix[4] = 0;
  topDict.fontMatrix[5] = 0;
  topDict.hasFontMatrix = false;
  topDict.uniqueID = 0;
  topDict.fontBBox[0] = 0;
  topDict.fontBBox[1] = 0;
  topDict.fontBBox[2] = 0;
  topDict.fontBBox[3] = 0;
  topDict.strokeWidth = 0;
  topDict.charsetOffset = 0;
  topDict.encodingOffset = 0;
  topDict.charStringsOffset = 0;
  topDict.privateSize = 0;
  topDict.privateOffset = 0;
  topDict.registrySID = 0;
  topDict.orderingSID = 0;
  topDict.supplement = 0;
  topDict.fdArrayOffset = 0;
  topDict.fdSelectOffset = 0;

  getIndexVal(&topDictIdx, 0, &topDictPtr, &parsedOk);
  if (!parsedOk) {
      return;
  }
  pos = topDictPtr.pos;
  nOps = 0;
  while (pos < topDictPtr.pos + topDictPtr.len) {
    pos = getOp(pos, false, &parsedOk);
    if (!parsedOk) {
      break;
    }
    if (!ops[nOps - 1].isNum) {
      --nOps; // drop the operator
      if (topDict.firstOp < 0) {
	topDict.firstOp = ops[nOps].op;
      }
      switch (ops[nOps].op) {
      case 0x0000: topDict.versionSID = (int)ops[0].num; break;
      case 0x0001: topDict.noticeSID = (int)ops[0].num; break;
      case 0x0c00: topDict.copyrightSID = (int)ops[0].num; break;
      case 0x0002: topDict.fullNameSID = (int)ops[0].num; break;
      case 0x0003: topDict.familyNameSID = (int)ops[0].num; break;
      case 0x0004: topDict.weightSID = (int)ops[0].num; break;
      case 0x0c01: topDict.isFixedPitch = (int)ops[0].num; break;
      case 0x0c02: topDict.italicAngle = ops[0].num; break;
      case 0x0c03: topDict.underlinePosition = ops[0].num; break;
      case 0x0c04: topDict.underlineThickness = ops[0].num; break;
      case 0x0c05: topDict.paintType = (int)ops[0].num; break;
      case 0x0c06: topDict.charstringType = (int)ops[0].num; break;
      case 0x0c07: topDict.fontMatrix[0] = ops[0].num;
	           topDict.fontMatrix[1] = ops[1].num;
	           topDict.fontMatrix[2] = ops[2].num;
	           topDict.fontMatrix[3] = ops[3].num;
	           topDict.fontMatrix[4] = ops[4].num;
	           topDict.fontMatrix[5] = ops[5].num;
		   topDict.hasFontMatrix = true; break;
      case 0x000d: topDict.uniqueID = (int)ops[0].num; break;
      case 0x0005: topDict.fontBBox[0] = ops[0].num;
	           topDict.fontBBox[1] = ops[1].num;
	           topDict.fontBBox[2] = ops[2].num;
	           topDict.fontBBox[3] = ops[3].num; break;
      case 0x0c08: topDict.strokeWidth = ops[0].num; break;
      case 0x000f: topDict.charsetOffset = (int)ops[0].num; break;
      case 0x0010: topDict.encodingOffset = (int)ops[0].num; break;
      case 0x0011: topDict.charStringsOffset = (int)ops[0].num; break;
      case 0x0012: topDict.privateSize = (int)ops[0].num;
	           topDict.privateOffset = (int)ops[1].num; break;
      case 0x0c1e: topDict.registrySID = (int)ops[0].num;
	           topDict.orderingSID = (int)ops[1].num;
		   topDict.supplement = (int)ops[2].num; break;
      case 0x0c24: topDict.fdArrayOffset = (int)ops[0].num; break;
      case 0x0c25: topDict.fdSelectOffset = (int)ops[0].num; break;
      }
      nOps = 0;
    }
  }
}

// Read a CID font dict (FD) - this pulls out the private dict
// pointer, and reads the private dict.  It also pulls the FontMatrix
// (if any) out of the FD.
void FoFiType1C::readFD(int offset, int length, Type1CPrivateDict *pDict) {
  int pSize, pOffset;
  double fontMatrix[6] = {0};
  bool hasFontMatrix;

  hasFontMatrix = false;
  fontMatrix[0] = fontMatrix[1] = fontMatrix[2] = 0; // make gcc happy
  fontMatrix[3] = fontMatrix[4] = fontMatrix[5] = 0;
  pSize = pOffset = 0;

  int posEnd;
  if (checkedAdd(offset, length, &posEnd)) {
    return;
  }

  int pos = offset;
  nOps = 0;
  while (pos < posEnd) {
    pos = getOp(pos, false, &parsedOk);
    if (!parsedOk) {
      return;
    }
    if (!ops[nOps - 1].isNum) {
      if (ops[nOps - 1].op == 0x0012) {
        if (nOps < 3) {
          parsedOk = false;
          return;
        }
        pSize = (int)ops[0].num;
        pOffset = (int)ops[1].num;
        break;
      } else if (ops[nOps - 1].op == 0x0c07) {
        fontMatrix[0] = ops[0].num;
        fontMatrix[1] = ops[1].num;
        fontMatrix[2] = ops[2].num;
        fontMatrix[3] = ops[3].num;
        fontMatrix[4] = ops[4].num;
        fontMatrix[5] = ops[5].num;
        hasFontMatrix = true;
      }
      nOps = 0;
    }
  }
  readPrivateDict(pOffset, pSize, pDict);
  if (hasFontMatrix) {
    pDict->fontMatrix[0] = fontMatrix[0];
    pDict->fontMatrix[1] = fontMatrix[1];
    pDict->fontMatrix[2] = fontMatrix[2];
    pDict->fontMatrix[3] = fontMatrix[3];
    pDict->fontMatrix[4] = fontMatrix[4];
    pDict->fontMatrix[5] = fontMatrix[5];
    pDict->hasFontMatrix = true;
  }
}

void FoFiType1C::readPrivateDict(int offset, int length,
				 Type1CPrivateDict *pDict) {
  pDict->hasFontMatrix = false;
  pDict->nBlueValues = 0;
  pDict->nOtherBlues = 0;
  pDict->nFamilyBlues = 0;
  pDict->nFamilyOtherBlues = 0;
  pDict->blueScale = 0.039625;
  pDict->blueShift = 7;
  pDict->blueFuzz = 1;
  pDict->hasStdHW = false;
  pDict->hasStdVW = false;
  pDict->nStemSnapH = 0;
  pDict->nStemSnapV = 0;
  pDict->hasForceBold = false;
  pDict->forceBoldThreshold = 0;
  pDict->languageGroup = 0;
  pDict->expansionFactor = 0.06;
  pDict->initialRandomSeed = 0;
  pDict->subrsOffset = 0;
  pDict->defaultWidthX = 0;
  pDict->defaultWidthXFP = false;
  pDict->nominalWidthX = 0;
  pDict->nominalWidthXFP = false;

  // no dictionary
  if (offset == 0 || length == 0) {
    return;
  }

  int posEnd;
  if (checkedAdd(offset, length, &posEnd)) {
    return;
  }

  int pos = offset;
  nOps = 0;
  while (pos < posEnd) {
    pos = getOp(pos, false, &parsedOk);
    if (!parsedOk) {
      break;
    }
    if (!ops[nOps - 1].isNum) {
      --nOps; // drop the operator
      switch (ops[nOps].op) {
      case 0x0006:
	pDict->nBlueValues = getDeltaIntArray(pDict->blueValues,
					      type1CMaxBlueValues);
	break;
      case 0x0007:
	pDict->nOtherBlues = getDeltaIntArray(pDict->otherBlues,
					      type1CMaxOtherBlues);
	break;
      case 0x0008:
	pDict->nFamilyBlues = getDeltaIntArray(pDict->familyBlues,
					       type1CMaxBlueValues);
	break;
      case 0x0009:
	pDict->nFamilyOtherBlues = getDeltaIntArray(pDict->familyOtherBlues,
						    type1CMaxOtherBlues);
	break;
      case 0x0c09:
	pDict->blueScale = ops[0].num;
	break;
      case 0x0c0a:
	pDict->blueShift = (int)ops[0].num;
	break;
      case 0x0c0b:
	pDict->blueFuzz = (int)ops[0].num;
	break;
      case 0x000a:
	pDict->stdHW = ops[0].num;
	pDict->hasStdHW = true;
	break;
      case 0x000b:
	pDict->stdVW = ops[0].num;
	pDict->hasStdVW = true;
	break;
      case 0x0c0c:
	pDict->nStemSnapH = getDeltaFPArray(pDict->stemSnapH,
					    type1CMaxStemSnap);
	break;
      case 0x0c0d:
	pDict->nStemSnapV = getDeltaFPArray(pDict->stemSnapV,
					    type1CMaxStemSnap);
	break;
      case 0x0c0e:
	pDict->forceBold = ops[0].num != 0;
	pDict->hasForceBold = true;
	break;
      case 0x0c0f:
	pDict->forceBoldThreshold = ops[0].num;
	break;
      case 0x0c11:
	pDict->languageGroup = (int)ops[0].num;
	break;
      case 0x0c12:
	pDict->expansionFactor = ops[0].num;
	break;
      case 0x0c13:
	pDict->initialRandomSeed = (int)ops[0].num;
	break;
      case 0x0013:
	pDict->subrsOffset = offset + (int)ops[0].num;
	break;
      case 0x0014:
	pDict->defaultWidthX = ops[0].num;
	pDict->defaultWidthXFP = ops[0].isFP;
	break;
      case 0x0015:
	pDict->nominalWidthX = ops[0].num;
	pDict->nominalWidthXFP = ops[0].isFP;
	break;
      }
      nOps = 0;
    }
  }
}

void FoFiType1C::readFDSelect() {
  int fdSelectFmt, pos, nRanges, gid0, gid1, fd, i, j;

  fdSelect = (unsigned char *)gmalloc(nGlyphs);
  if (topDict.fdSelectOffset == 0) {
    for (i = 0; i < nGlyphs; ++i) {
      fdSelect[i] = 0;
    }
  } else {
    pos = topDict.fdSelectOffset;
    fdSelectFmt = getU8(pos++, &parsedOk);
    if (!parsedOk) {
      return;
    }
    if (fdSelectFmt == 0) {
      if (!checkRegion(pos, nGlyphs)) {
	parsedOk = false;
	return;
      }
      memcpy(fdSelect, file + pos, nGlyphs);
    } else if (fdSelectFmt == 3) {
      nRanges = getU16BE(pos, &parsedOk);
      pos += 2;
      gid0 = getU16BE(pos, &parsedOk);
      pos += 2;
      for (i = 1; i <= nRanges; ++i) {
	fd = getU8(pos++, &parsedOk);
	gid1 = getU16BE(pos, &parsedOk);
	if (!parsedOk) {
	  return;
	}
	pos += 2;
	if (gid0 > gid1 || gid1 > nGlyphs) {
	  //~ error(-1, "Bad FDSelect table in CID font");
	  parsedOk = false;
	  return;
	}
	for (j = gid0; j < gid1; ++j) {
	  fdSelect[j] = fd;
	}
	gid0 = gid1;
      }
    } else {
      //~ error(-1, "Unknown FDSelect table format in CID font");
      for (i = 0; i < nGlyphs; ++i) {
	fdSelect[i] = 0;
      }
    }
  }
}

void FoFiType1C::buildEncoding() {
  char buf[256];
  int nCodes, nRanges, encFormat;
  int pos, c, sid, nLeft, nSups, i, j;