private InlineImage ParseInlineImage(
)
{
/*
* NOTE: Inline images use a peculiar syntax that's an exception to the usual rule
* that the data in a content stream is interpreted according to the standard PDF syntax
* for objects.
*/
InlineImageHeader header;
{
List <PdfDirectObject> operands = new List <PdfDirectObject>();
// Parsing the image entries...
while (MoveNext() &&
tokenType != TokenTypeEnum.Keyword) // Not keyword (i.e. end at image data beginning (ID operator)).
{
operands.Add(ParsePdfObject());
}
header = new InlineImageHeader(operands);
}
InlineImageBody body;
{
MoveNext();
bytes::Buffer data = new bytes::Buffer();
byte c1 = 0, c2 = 0;
do
{
try
{
while (true)
{
c1 = (byte)stream.ReadByte();
c2 = (byte)stream.ReadByte();
if (c1 == 'E' && c2 == 'I')
{
break;
}
data.Append(c1);
data.Append(c2);
}
break;
}
catch
{
/* NOTE: Current stream has finished. */
// Move to the next stream!
MoveNextStream();
}
} while(stream != null);
body = new InlineImageBody(data);
}
return(new InlineImage(
header,
body
));
}
private InlineImage ParseInlineImage()
{
/*
* NOTE: Inline images use a peculiar syntax that's an exception to the usual rule
* that the data in a content stream is interpreted according to the standard PDF syntax
* for objects.
*/
InlineImageHeader header;
{
List <PdfDirectObject> operands = new List <PdfDirectObject>();
// Parsing the image entries...
while (MoveNext() && TokenType != TokenTypeEnum.Keyword) // Not keyword (i.e. end at image data beginning (ID operator)).
{
operands.Add((PdfDirectObject)ParsePdfObject());
}
header = new InlineImageHeader(operands);
}
InlineImageBody body;
{
// [FIX:51,74] Wrong 'EI' token handling on inline image parsing.
bytes::IInputStream stream = Stream;
stream.ReadByte(); // Should be the whitespace following the 'ID' token.
bytes::Buffer data = new bytes::Buffer();
while (true)
{
int curByte1 = stream.ReadByte();
if (curByte1 == -1)
{
break;
}
int curByte2 = stream.ReadByte();
if (curByte2 == -1)
{
break;
}
if (((char)curByte1 == 'E' && (char)curByte2 == 'I'))
{
break;
}
if (((char)curByte1 == ' ' && (char)curByte2 == 'E'))
{
break;
}
data.Append((byte)curByte1);
data.Append((byte)curByte2);
}
body = new InlineImageBody(data);
}
return(new InlineImage(header, body));
}
private InlineImage ParseInlineImage(
)
{
/*
* NOTE: Inline images use a peculiar syntax that's an exception to the usual rule
* that the data in a content stream is interpreted according to the standard PDF syntax
* for objects.
*/
InlineImageHeader header;
{
List <PdfDirectObject> operands = new List <PdfDirectObject>();
// Parsing the image entries...
while (MoveNext() &&
TokenType != TokenTypeEnum.Keyword) // Not keyword (i.e. end at image data beginning (ID operator)).
{
operands.Add((PdfDirectObject)ParsePdfObject());
}
header = new InlineImageHeader(operands);
}
InlineImageBody body;
{
bytes::IInputStream stream = Stream;
MoveNext();
bytes::Buffer data = new bytes::Buffer();
byte prevByte = 0;
while (true)
{
byte curByte = (byte)stream.ReadByte();
if (prevByte == 'E' && curByte == 'I')
{
break;
}
data.Append(prevByte = curByte);
}
body = new InlineImageBody(data);
}
return(new InlineImage(
header,
body
));
}
private InlineImage ParseInlineImage(
)
{
/*
NOTE: Inline images use a peculiar syntax that's an exception to the usual rule
that the data in a content stream is interpreted according to the standard PDF syntax
for objects.
*/
InlineImageHeader header;
{
List<PdfDirectObject> operands = new List<PdfDirectObject>();
// Parsing the image entries...
while(MoveNext()
&& TokenType != TokenTypeEnum.Keyword) // Not keyword (i.e. end at image data beginning (ID operator)).
{operands.Add((PdfDirectObject)ParsePdfObject());}
header = new InlineImageHeader(operands);
}
InlineImageBody body;
{
bytes::IInputStream stream = Stream;
MoveNext();
bytes::Buffer data = new bytes::Buffer();
while(true)
{
byte c1 = (byte)stream.ReadByte();
byte c2 = (byte)stream.ReadByte();
if(c1 == 'E' && c2 == 'I')
break;
data.Append(c1);
data.Append(c2);
}
body = new InlineImageBody(data);
}
return new InlineImage(
header,
body
);
}
private InlineImage ParseInlineImage(
)
{
/*
* NOTE: Inline images use a peculiar syntax that's an exception to the usual rule
* that the data in a content stream is interpreted according to the standard PDF syntax
* for objects.
*/
InlineImageHeader header;
{
List <PdfDirectObject> operands = new List <PdfDirectObject>();
// Parsing the image entries...
while (MoveNext() &&
TokenType != TokenTypeEnum.Keyword) // Not keyword (i.e. end at image data beginning (ID operator)).
{
operands.Add((PdfDirectObject)ParsePdfObject());
}
header = new InlineImageHeader(operands);
}
InlineImageBody body;
{
// [FIX:51,74] Wrong 'EI' token handling on inline image parsing.
bytes::IInputStream stream = Stream;
stream.ReadByte(); // Should be the whitespace following the 'ID' token.
bytes::Buffer data = new bytes::Buffer();
var endChunkBuffer = new sysIO::MemoryStream(3);
int endChunkIndex = -1;
while (true)
{
int curByte = stream.ReadByte();
if (curByte == -1)
{
throw new PostScriptParseException("No 'EI' token found to close inline image data stream.");
}
if (endChunkIndex == -1)
{
if (IsWhitespace(curByte))
{
/*
* NOTE: Whitespace characters may announce the beginning of the end image operator.
*/
endChunkBuffer.WriteByte((byte)curByte);
endChunkIndex++;
}
else
{
data.Append((byte)curByte);
}
}
else if (endChunkIndex == 0 && IsWhitespace(curByte))
{
/*
* NOTE: Only the last whitespace character may announce the beginning of the end image
* operator.
*/
data.Append(endChunkBuffer.ToArray());
endChunkBuffer.SetLength(0);
endChunkBuffer.WriteByte((byte)curByte);
}
else if ((endChunkIndex == 0 && curByte == 'E') ||
(endChunkIndex == 1 && curByte == 'I'))
{
/*
* NOTE: End image operator characters.
*/
endChunkBuffer.WriteByte((byte)curByte);
endChunkIndex++;
}
else if (endChunkIndex == 2 && IsWhitespace(curByte))
{
/*
* NOTE: The whitespace character after the end image operator completes the pattern.
*/
break;
}
else
{
if (endChunkIndex > -1)
{
data.Append(endChunkBuffer.ToArray());
endChunkBuffer.SetLength(0);
endChunkIndex = -1;
}
data.Append((byte)curByte);
}
}
body = new InlineImageBody(data);
}
return(new InlineImage(
header,
body
));
}
/**
* <summary>Builds a CMap according to the specified arguments.</summary>
* <param name="entryType"></param>
* <param name="cmapName">CMap name (<code>null</code> in case no custom name is needed).</param>
* <param name="codes"></param>
* <param name="outCodeFunction"></param>
* <returns>Buffer containing the serialized CMap.</returns>
*/
public static bytes::IBuffer Build(
EntryTypeEnum entryType,
string cmapName,
SortedDictionary <ByteArray, int> codes,
GetOutCodeDelegate outCodeFunction
)
{
bytes::IBuffer buffer = new bytes::Buffer();
// Header.
string outCodeFormat;
switch (entryType)
{
case EntryTypeEnum.BaseFont:
{
if (cmapName == null)
{
cmapName = "Adobe-Identity-UCS";
}
buffer.Append(
"/CIDInit /ProcSet findresource begin\n"
+ "12 dict begin\n"
+ "begincmap\n"
+ "/CIDSystemInfo\n"
+ "<< /Registry (Adobe)\n"
+ "/Ordering (UCS)\n"
+ "/Supplement 0\n"
+ ">> def\n"
+ "/CMapName /").Append(cmapName).Append(" def\n"
+ "/CMapVersion 10.001 def\n"
+ "/CMapType 2 def\n"
+ "1 begincodespacerange\n"
+ "<0000> <FFFF>\n"
+ "endcodespacerange\n"
);
outCodeFormat = "<{0:X4}>";
break;
}
case EntryTypeEnum.CID:
{
if (cmapName == null)
{
cmapName = "Custom";
}
buffer.Append(
"%!PS-Adobe-3.0 Resource-CMap\n"
+ "%%DocumentNeededResources: ProcSet (CIDInit)\n"
+ "%%IncludeResource: ProcSet (CIDInit)\n"
+ "%%BeginResource: CMap (").Append(cmapName).Append(")\n"
+ "%%Title: (").Append(cmapName).Append(" Adobe Identity 0)\n"
+ "%%Version: 1\n"
+ "%%EndComments\n"
+ "/CIDInit /ProcSet findresource begin\n"
+ "12 dict begin\n"
+ "begincmap\n"
+ "/CIDSystemInfo 3 dict dup begin\n"
+ "/Registry (Adobe) def\n"
+ "/Ordering (Identity) def\n"
+ "/Supplement 0 def\n"
+ "end def\n"
+ "/CMapVersion 1 def\n"
+ "/CMapType 1 def\n"
+ "/CMapName /").Append(cmapName).Append(" def\n"
+ "/WMode 0 def\n"
+ "1 begincodespacerange\n"
+ "<0000> <FFFF>\n"
+ "endcodespacerange\n"
);
outCodeFormat = "{0}";
break;
}
default:
throw new NotImplementedException();
}
// Entries.
{
IList <KeyValuePair <ByteArray, int> > cidChars = new List <KeyValuePair <ByteArray, int> >();
IList <KeyValuePair <ByteArray, int>[]> cidRanges = new List <KeyValuePair <ByteArray, int>[]>();
{
KeyValuePair <ByteArray, int>? lastCodeEntry = null;
KeyValuePair <ByteArray, int>[] lastCodeRange = null;
foreach (KeyValuePair <ByteArray, int> codeEntry in codes)
{
if (lastCodeEntry.HasValue)
{
int codeLength = codeEntry.Key.Data.Length;
if (codeLength == lastCodeEntry.Value.Key.Data.Length &&
codeEntry.Key.Data[codeLength - 1] - lastCodeEntry.Value.Key.Data[codeLength - 1] == 1 &&
outCodeFunction(codeEntry) - outCodeFunction(lastCodeEntry.Value) == 1) // Contiguous codes.
{
if (lastCodeRange == null)
{
lastCodeRange = new KeyValuePair <ByteArray, int>[] { lastCodeEntry.Value, default(KeyValuePair <ByteArray, int>) };
}
}
else // Separated codes.
{
AddEntry(cidRanges, cidChars, lastCodeEntry.Value, lastCodeRange);
lastCodeRange = null;
}
}
lastCodeEntry = codeEntry;
}
AddEntry(cidRanges, cidChars, lastCodeEntry.Value, lastCodeRange);
}
// Ranges section.
BuildEntriesSection(buffer, entryType, cidRanges, BuildRangeEntry, "range", outCodeFunction, outCodeFormat);
// Chars section.
BuildEntriesSection(buffer, entryType, cidChars, BuildCharEntry, "char", outCodeFunction, outCodeFormat);
}
// Trailer.
switch (entryType)
{
case EntryTypeEnum.BaseFont:
buffer.Append(
"endcmap\n"
+ "CMapName currentdict /CMap defineresource pop\n"
+ "end\n"
+ "end\n"
);
break;
case EntryTypeEnum.CID:
buffer.Append(
"endcmap\n"
+ "CMapName currentdict /CMap defineresource pop\n"
+ "end\n"
+ "end\n"
+ "%%EndResource\n"
+ "%%EOF"
);
break;
default:
throw new NotImplementedException();
}
return(buffer);
}
/**
<summary>Creates the character code mapping for composite fonts.</summary>
*/
private void Load_CreateEncoding(
PdfDictionary font,
PdfDictionary cidFont
)
{
// CMap [PDF:1.6:5.6.4].
bytes::Buffer cmapBuffer = new bytes::Buffer();
cmapBuffer.Append(
"%!PS-Adobe-3.0 Resource-CMap\n"
+ "%%DocumentNeededResources: ProcSet (CIDInit)\n"
+ "%%IncludeResource: ProcSet (CIDInit)\n"
+ "%%BeginResource: CMap (Adobe-Identity-UCS)\n"
+ "%%Title: (Adobe-Identity-UCS Adobe Identity 0)\n"
+ "%%Version: 1\n"
+ "%%EndComments\n"
+ "/CIDInit /ProcSet findresource begin\n"
+ "12 dict begin\n"
+ "begincmap\n"
+ "/CIDSystemInfo\n"
+ "3 dict dup begin\n"
+ "/Registry (Adobe) def\n"
+ "/Ordering (Identity) def\n"
+ "/Supplement 0 def\n"
+ "end def\n"
+ "/CMapName /Adobe-Identity-UCS def\n"
+ "/CMapVersion 1 def\n"
+ "/CMapType 0 def\n"
+ "/WMode 0 def\n"
+ "2 begincodespacerange\n"
+ "<20> <20>\n"
+ "<0000> <19FF>\n"
+ "endcodespacerange\n"
+ glyphIndexes.Count + " begincidchar\n"
);
// ToUnicode [PDF:1.6:5.9.2].
bytes::Buffer toUnicodeBuffer = new bytes::Buffer();
toUnicodeBuffer.Append(
"/CIDInit /ProcSet findresource begin\n"
+ "12 dict begin\n"
+ "begincmap\n"
+ "/CIDSystemInfo\n"
+ "<< /Registry (Adobe)\n"
+ "/Ordering (UCS)\n"
+ "/Supplement 0\n"
+ ">> def\n"
+ "/CMapName /Adobe-Identity-UCS def\n"
+ "/CMapVersion 10.001 def\n"
+ "/CMapType 2 def\n"
+ "2 begincodespacerange\n"
+ "<20> <20>\n"
+ "<0000> <19FF>\n"
+ "endcodespacerange\n"
+ glyphIndexes.Count + " beginbfchar\n"
);
// CIDToGIDMap [PDF:1.6:5.6.3].
bytes::Buffer gIdBuffer = new bytes::Buffer();
gIdBuffer.Append((byte)0);
gIdBuffer.Append((byte)0);
int code = 0;
codes = new BiDictionary<ByteArray,int>(glyphIndexes.Count);
PdfArray widthsObject = new PdfArray(glyphWidths.Count);
foreach(KeyValuePair<int,int> glyphIndexEntry in glyphIndexes)
{
// Character code (unicode to codepoint) entry.
code++;
byte[] charCode = (glyphIndexEntry.Key == 32
? new byte[]{32}
: new byte[]
{
(byte)((code >> 8) & 0xFF),
(byte)(code & 0xFF)
});
codes[new ByteArray(charCode)] = glyphIndexEntry.Key;
// CMap entry.
cmapBuffer.Append("<");
toUnicodeBuffer.Append("<");
for(int charCodeBytesIndex = 0,
charCodeBytesLength = charCode.Length;
charCodeBytesIndex < charCodeBytesLength;
charCodeBytesIndex++
)
{
string hex = ((int)charCode[charCodeBytesIndex]).ToString("X2");
cmapBuffer.Append(hex);
toUnicodeBuffer.Append(hex);
}
cmapBuffer.Append("> " + code + "\n");
toUnicodeBuffer.Append("> <" + glyphIndexEntry.Key.ToString("X4") + ">\n");
// CID-to-GID entry.
int glyphIndex = glyphIndexEntry.Value;
gIdBuffer.Append((byte)((glyphIndex >> 8) & 0xFF));
gIdBuffer.Append((byte)(glyphIndex & 0xFF));
// Width.
int width;
if(!glyphWidths.TryGetValue(glyphIndex, out width))
{width = 0;}
else if(width > 1000)
{width = 1000;}
widthsObject.Add(PdfInteger.Get(width));
}
cmapBuffer.Append(
"endcidchar\n"
+ "endcmap\n"
+ "CMapName currentdict /CMap defineresource pop\n"
+ "end\n"
+ "end\n"
+ "%%EndResource\n"
+ "%%EOF"
);
PdfStream cmapStream = new PdfStream(cmapBuffer);
PdfDictionary cmapHead = cmapStream.Header;
cmapHead[PdfName.Type] = PdfName.CMap;
cmapHead[PdfName.CMapName] = new PdfName("Adobe-Identity-UCS");
cmapHead[PdfName.CIDSystemInfo] = new PdfDictionary(
new PdfName[]
{
PdfName.Registry,
PdfName.Ordering,
PdfName.Supplement
},
new PdfDirectObject[]
{
new PdfTextString("Adobe"),
new PdfTextString("Identity"),
PdfInteger.Get(0)
}
); // Generic predefined CMap (Identity-H/V (Adobe-Identity-0)) [PDF:1.6:5.6.4].
font[PdfName.Encoding] = File.Register(cmapStream);
PdfStream gIdStream = new PdfStream(gIdBuffer);
cidFont[PdfName.CIDToGIDMap] = File.Register(gIdStream);
cidFont[PdfName.W] = new PdfArray(new PdfDirectObject[]{PdfInteger.Get(1),widthsObject});
toUnicodeBuffer.Append(
"endbfchar\n"
+ "endcmap\n"
+ "CMapName currentdict /CMap defineresource pop\n"
+ "end\n"
+ "end\n"
);
PdfStream toUnicodeStream = new PdfStream(toUnicodeBuffer);
font[PdfName.ToUnicode] = File.Register(toUnicodeStream);
}
/**
* <summary>Creates the character code mapping for composite fonts.</summary>
*/
private void Load_CreateEncoding(
PdfDictionary font,
PdfDictionary cidFont
)
{
// CMap [PDF:1.6:5.6.4].
bytes::Buffer cmapBuffer = new bytes::Buffer();
cmapBuffer.Append(
"%!PS-Adobe-3.0 Resource-CMap\n"
+ "%%DocumentNeededResources: ProcSet (CIDInit)\n"
+ "%%IncludeResource: ProcSet (CIDInit)\n"
+ "%%BeginResource: CMap (Adobe-Identity-UCS)\n"
+ "%%Title: (Adobe-Identity-UCS Adobe Identity 0)\n"
+ "%%Version: 1\n"
+ "%%EndComments\n"
+ "/CIDInit /ProcSet findresource begin\n"
+ "12 dict begin\n"
+ "begincmap\n"
+ "/CIDSystemInfo\n"
+ "3 dict dup begin\n"
+ "/Registry (Adobe) def\n"
+ "/Ordering (Identity) def\n"
+ "/Supplement 0 def\n"
+ "end def\n"
+ "/CMapName /Adobe-Identity-UCS def\n"
+ "/CMapVersion 1 def\n"
+ "/CMapType 0 def\n"
+ "/WMode 0 def\n"
+ "2 begincodespacerange\n"
+ "<20> <20>\n"
+ "<0000> <19FF>\n"
+ "endcodespacerange\n"
+ glyphIndexes.Count + " begincidchar\n"
);
// ToUnicode [PDF:1.6:5.9.2].
bytes::Buffer toUnicodeBuffer = new bytes::Buffer();
toUnicodeBuffer.Append(
"/CIDInit /ProcSet findresource begin\n"
+ "12 dict begin\n"
+ "begincmap\n"
+ "/CIDSystemInfo\n"
+ "<< /Registry (Adobe)\n"
+ "/Ordering (UCS)\n"
+ "/Supplement 0\n"
+ ">> def\n"
+ "/CMapName /Adobe-Identity-UCS def\n"
+ "/CMapVersion 10.001 def\n"
+ "/CMapType 2 def\n"
+ "2 begincodespacerange\n"
+ "<20> <20>\n"
+ "<0000> <19FF>\n"
+ "endcodespacerange\n"
+ glyphIndexes.Count + " beginbfchar\n"
);
// CIDToGIDMap [PDF:1.6:5.6.3].
bytes::Buffer gIdBuffer = new bytes::Buffer();
gIdBuffer.Append((byte)0);
gIdBuffer.Append((byte)0);
int code = 0;
codes = new BiDictionary <ByteArray, int>(glyphIndexes.Count);
PdfArray widthsObject = new PdfArray(glyphWidths.Count);
foreach (KeyValuePair <int, int> glyphIndexEntry in glyphIndexes)
{
// Character code (unicode to codepoint) entry.
code++;
byte[] charCode = (glyphIndexEntry.Key == 32
? new byte[] { 32 }
: new byte[]
{
(byte)((code >> 8) & 0xFF),
(byte)(code & 0xFF)
});
codes[new ByteArray(charCode)] = glyphIndexEntry.Key;
// CMap entry.
cmapBuffer.Append("<");
toUnicodeBuffer.Append("<");
for (int charCodeBytesIndex = 0,
charCodeBytesLength = charCode.Length;
charCodeBytesIndex < charCodeBytesLength;
charCodeBytesIndex++
)
{
string hex = ((int)charCode[charCodeBytesIndex]).ToString("X2");
cmapBuffer.Append(hex);
toUnicodeBuffer.Append(hex);
}
cmapBuffer.Append("> " + code + "\n");
toUnicodeBuffer.Append("> <" + glyphIndexEntry.Key.ToString("X4") + ">\n");
// CID-to-GID entry.
int glyphIndex = glyphIndexEntry.Value;
gIdBuffer.Append((byte)((glyphIndex >> 8) & 0xFF));
gIdBuffer.Append((byte)(glyphIndex & 0xFF));
// Width.
int width;
if (!glyphWidths.TryGetValue(glyphIndex, out width))
{
width = 0;
}
else if (width > 1000)
{
width = 1000;
}
widthsObject.Add(PdfInteger.Get(width));
}
cmapBuffer.Append(
"endcidchar\n"
+ "endcmap\n"
+ "CMapName currentdict /CMap defineresource pop\n"
+ "end\n"
+ "end\n"
+ "%%EndResource\n"
+ "%%EOF"
);
PdfStream cmapStream = new PdfStream(cmapBuffer);
PdfDictionary cmapHead = cmapStream.Header;
cmapHead[PdfName.Type] = PdfName.CMap;
cmapHead[PdfName.CMapName] = new PdfName("Adobe-Identity-UCS");
cmapHead[PdfName.CIDSystemInfo] = new PdfDictionary(
new PdfName[]
{
PdfName.Registry,
PdfName.Ordering,
PdfName.Supplement
},
new PdfDirectObject[]
{
new PdfTextString("Adobe"),
new PdfTextString("Identity"),
PdfInteger.Get(0)
}
); // Generic predefined CMap (Identity-H/V (Adobe-Identity-0)) [PDF:1.6:5.6.4].
font[PdfName.Encoding] = File.Register(cmapStream);
PdfStream gIdStream = new PdfStream(gIdBuffer);
cidFont[PdfName.CIDToGIDMap] = File.Register(gIdStream);
cidFont[PdfName.W] = new PdfArray(new PdfDirectObject[] { PdfInteger.Get(1), widthsObject });
toUnicodeBuffer.Append(
"endbfchar\n"
+ "endcmap\n"
+ "CMapName currentdict /CMap defineresource pop\n"
+ "end\n"
+ "end\n"
);
PdfStream toUnicodeStream = new PdfStream(toUnicodeBuffer);
font[PdfName.ToUnicode] = File.Register(toUnicodeStream);
}