public static void ApplyToUnitCode(
DictionaryToken toUnicodeStream,
ref char[] encoding,
out string toUnicodeHeader,
out SortedDictionary <int, char> cMap)
{
var tokeniser = toUnicodeStream.GetStreamBytes();
if (tokeniser.GetStreamOpCode("begincmap") is null)
{
throw tokeniser.StreamException("ToUnicode stream is missing 'begincmap'.");
}
tokeniser.SetStreamMark();
if (tokeniser.GetStreamOpCode("endcodespacerange") is null)
{
throw tokeniser.StreamException("ToUnicode stream is missing 'endcodespacerange'.");
}
toUnicodeHeader = tokeniser.GetStreamMarkedText();
cMap = new SortedDictionary <int, char>();
while (true)
{
var opCode = tokeniser.GetStreamOpCode();
if (opCode is null)
{
throw tokeniser.StreamException("ToUnicode stream incomplete.");
}
var opCodeSpan = opCode.Value.Span;
if (isEqual("beginbfchar", opCodeSpan))
{
tokeniser.StartStreamArgumentReading();
var linesCount = tokeniser.GetStreamInt();
tokeniser.EndStreamArgumentReading();
for (int lineIndex = 0; lineIndex < linesCount; lineIndex++)
{
var cid = tokeniser.GetStreamCid();
var unicodeChar = (char)tokeniser.GetStreamUnicode();
encoding[cid] = unicodeChar;
cMap.Add(cid, unicodeChar);
}
opCode = tokeniser.GetStreamOpCode();
opCodeSpan = opCode !.Value.Span;
if (!isEqual("endbfchar", opCodeSpan))
{
throw tokeniser.StreamException("ToUnicode stream is missing 'endbfchar' after 'beginbfchar'.");
}
}
else if (isEqual("beginbfrange", opCodeSpan))
{
tokeniser.StartStreamArgumentReading();
var linesCount = tokeniser.GetStreamInt();
tokeniser.EndStreamArgumentReading();
for (int lineIndex = 0; lineIndex < linesCount; lineIndex++)
{
var cidStart = tokeniser.GetStreamCid();
var cidEnd = tokeniser.GetStreamCid();
if (cidEnd < cidStart)
{
throw tokeniser.StreamException($"ToUnicode: beginbfrange cid1 '{cidStart} should be smaller than cid2'{cidEnd}'.");
}
var unicodeChar = tokeniser.GetStreamUnicode();
for (ushort cidIndex = cidStart; cidIndex <= cidEnd; cidIndex++)
{
cMap.Add(cidIndex, (char)unicodeChar);
encoding[cidIndex] = (char)unicodeChar++;
}
}
opCode = tokeniser.GetStreamOpCode();
opCodeSpan = opCode !.Value.Span;
if (!isEqual("endbfrange", opCodeSpan))
{
throw tokeniser.StreamException("ToUnicode stream is missing 'endbfrange' after 'beginbfrange'.");
}
}
else if (isEqual("endcmap", opCodeSpan))
{
break;
}
}
}
public PdfContent(DictionaryToken contentsDictionaryToken, IReadOnlyDictionary <string, PdfFont> fonts)
{
contentsDictionaryToken.PdfObject = this;
var tokeniser = contentsDictionaryToken.GetStreamBytes();
Fonts = fonts;
decimal?lastLineOffset = null;
string? newText = null;
try {
//q 0.12 0 0 0.12 0 0 cm
///R7 gs
//0 0 0 rg
//q
//8.33333 0 0 8.33333 0 0 cm BT
//BT
// /F1 24 Tf
// 100 100 Td
// ( Hello World ) Tj
//ET
while (true)
{
//find BT
ReadOnlySpan <byte> opCodeSpan;
do
{
var opCode = tokeniser.GetStreamOpCode();
if (opCode is null)
{
return;
}
opCodeSpan = opCode.Value.Span;
if (opCodeSpan.Length == 2 && opCodeSpan[0] == 'B' && opCodeSpan[1] == 'I')
{
tokeniser.SkipInlineImage();
continue;
}
} while (opCodeSpan.Length != 2 || opCodeSpan[0] != 'B' || opCodeSpan[1] != 'T');
//processes text operation until ET
PdfFont?font = null;
while (true)
{
var opCode = tokeniser.GetStreamOpCode();//cannot return null (end of stream), because opCode ET must follow
if (opCode is null)
{
Error += "Error Content stream: stream end found but 'ET' still missing." + Environment.NewLine;
return;
}
opCodeSpan = opCode.Value.Span;
if (opCodeSpan.Length == 1)
{
if (opCodeSpan[0] == '\'')
{
tokeniser.StartStreamArgumentReading();
newText = tokeniser.GetStreamString(font);
tokeniser.EndStreamArgumentReading();
}
else if (opCodeSpan[0] == '"')
{
tokeniser.StartStreamArgumentReading();
tokeniser.SkipStreamArgument();
tokeniser.SkipStreamArgument();
newText = tokeniser.GetStreamString(font);
tokeniser.EndStreamArgumentReading();
}
else
{
continue;
}
}
else if (opCodeSpan.Length == 2)
{
if (opCodeSpan[0] == 'T')
{
var opCodeChar1 = opCodeSpan[1];
if (opCodeChar1 == 'j')
{
tokeniser.StartStreamArgumentReading();
newText = tokeniser.GetStreamString(font);
tokeniser.EndStreamArgumentReading();
}
else if (opCodeChar1 == 'J')
{
tokeniser.StartStreamArgumentReading();
newText = tokeniser.GetStreamArrayString(font);
tokeniser.EndStreamArgumentReading();
}
else if (opCodeChar1 == 'f')
{
tokeniser.StartStreamArgumentReading();
PdfFontName = tokeniser.GetStreamName();
if (!fonts.TryGetValue(PdfFontName, out font))
{
Error += $"Could not find font '{PdfFontName}'." + Environment.NewLine;
}
tokeniser.EndStreamArgumentReading();
continue;
}
else if (opCodeChar1 == 'd' || opCodeChar1 == 'D' || opCodeChar1 == '*')
{
Text += Environment.NewLine;
continue;
}
else if (opCodeChar1 == 'm')
{
tokeniser.StartStreamArgumentReading();
tokeniser.GetStreamInt();
tokeniser.GetStreamInt();
tokeniser.GetStreamInt();
tokeniser.GetStreamInt();
tokeniser.GetStreamNumber();
var lineOffset = tokeniser.GetStreamNumber();
if (lastLineOffset != lineOffset)
{
lastLineOffset = lineOffset;
if (Text != null)
{
Text += Environment.NewLine;
}
}
tokeniser.EndStreamArgumentReading();
continue;
}
else
{
//skip operants like TL
continue;
}
}
else if (opCodeSpan[0] == 'E' && opCodeSpan[1] == 'T')
{
break;
}
else
{
//other 2 characters opcodes
continue;
}
}
else
{
//opcode is longer than 2 letters
continue;
}
if (newText?.Contains("size") ?? false)
{
}
Text += newText + tokeniser.ContentDelimiter;
}
}
//var s = tokeniser.StreamBytesToString();
} catch (Exception ex) {
if (Exception is null)
{
Exception = "";
}
else
{
Exception += Environment.NewLine + Environment.NewLine;
}
if (ex is PdfStreamException || ex is PdfException)
{
Exception = ex.ToDetailString();
}
else
{
Exception = ex.ToDetailString() + Environment.NewLine + tokeniser.ShowStreamContentAtIndex();
}
}
}
// /CIDInit /ProcSet findresource begin
// 11 dict begin
// begincmap
// /CIDSystemInfo
// << /Registry(Adobe)
// /Ordering(UCS)
// /Supplement 0
// >> def
// /CMapName /Adobe-Identity-UCS def
// /CMapType 2 def
// 1 begincodespacerange
// <0000> <FFFF>
// endcodespacerange
// 2 beginbfchar
// <0003> <0020>
// <00B1> <2013>
// endbfchar
// 1 beginbfrange
// <00B5> <00B6> <2018>
// endbfrange
// endcmap
// CMapName currentdict /CMap defineresource pop
// end
// end
public PdfToUnitCode(DictionaryToken toUnicodeStream)
{
var tokeniser = toUnicodeStream.GetStreamBytes();
if (tokeniser.GetStreamOpCode("begincmap") is null)
{
throw tokeniser.StreamException("ToUnicode stream is missing 'begincmap'.");
}
tokeniser.SetStreamMark();
if (tokeniser.GetStreamOpCode("endcodespacerange") is null)
{
throw tokeniser.StreamException("ToUnicode stream is missing 'endcodespacerange'.");
}
Header = tokeniser.GetStreamMarkedText();
var cidUniCodes = new List <(ushort, char)>();
var cidRangeUniCodes = new List <(ushort, ushort, char)>();
var minCid = int.MaxValue;
var maxCid = int.MinValue;
while (true)
{
var opCode = tokeniser.GetStreamOpCode();
if (opCode is null)
{
throw tokeniser.StreamException("ToUnicode stream incomplete.");
}
var opCodeSpan = opCode.Value.Span;
if (isEqual("beginbfchar", opCodeSpan))
{
tokeniser.StartStreamArgumentReading();
var linesCount = tokeniser.GetStreamInt();
tokeniser.EndStreamArgumentReading();
for (int lineIndex = 0; lineIndex < linesCount; lineIndex++)
{
var cid = tokeniser.GetStreamCid();
minCid = Math.Min(minCid, cid);
maxCid = Math.Max(maxCid, cid);
var unicodeChar = tokeniser.GetStreamUnicode();
cidUniCodes.Add((cid, (char)unicodeChar));
}
opCode = tokeniser.GetStreamOpCode();
opCodeSpan = opCode !.Value.Span;
if (!isEqual("endbfchar", opCodeSpan))
{
throw tokeniser.StreamException("ToUnicode stream is missing 'endbfchar' after 'beginbfchar'.");
}
}
else if (isEqual("beginbfrange", opCodeSpan))
{
tokeniser.StartStreamArgumentReading();
var linesCount = tokeniser.GetStreamInt();
tokeniser.EndStreamArgumentReading();
for (int lineIndex = 0; lineIndex < linesCount; lineIndex++)
{
var cidStart = tokeniser.GetStreamCid();
minCid = Math.Min(minCid, cidStart);
var cidEnd = tokeniser.GetStreamCid();
maxCid = Math.Max(maxCid, cidEnd);
if (cidEnd < cidStart)
{
throw tokeniser.StreamException($"ToUnicode: beginbfrange cid1 '{cidStart} should be smaller than cid2'{cidEnd}'.");
}
var unicodeChar = tokeniser.GetStreamUnicode();
cidRangeUniCodes.Add((cidStart, cidEnd, (char)unicodeChar));
}
opCode = tokeniser.GetStreamOpCode();
opCodeSpan = opCode !.Value.Span;
if (!isEqual("endbfrange", opCodeSpan))
{
throw tokeniser.StreamException("ToUnicode stream is missing 'endbfrange' after 'beginbfrange'.");
}
}
else if (isEqual("endcmap", opCodeSpan))
{
break;
}
}
unicodes = new char[maxCid + 1];
for (int unicodesIndex = 0; unicodesIndex < unicodes.Length; unicodesIndex++)
{
unicodes[unicodesIndex] = (char)unicodesIndex;
}
foreach ((ushort cid, ushort unicodeChar)cidUniCode in cidUniCodes)
{
if (unicodes[cidUniCode.cid] != cidUniCode.cid)
{
throw tokeniser.StreamException($"ToUnicode defines the same cid '{cidUniCode.cid}' twice.");
}
unicodes[cidUniCode.cid] = (char)cidUniCode.unicodeChar;
}
foreach ((ushort cidStart, ushort cidEnd, ushort unicodeChar)cidRangeUniCode in cidRangeUniCodes)
{
var unicodeIndex = cidRangeUniCode.unicodeChar;
for (ushort cidIndex = cidRangeUniCode.cidStart; cidIndex <= cidRangeUniCode.cidEnd; cidIndex++)
{
if (unicodes[cidIndex] != cidIndex)
{
throw tokeniser.StreamException($"ToUnicode defines the same cid '{cidIndex}' twice.");
}
unicodes[cidIndex] = (char)unicodeIndex++;
}
}
//var s = tokeniser.StreamBytesToString();
}