// from < in data
public override void Read(Tokeniser t, CharacterReader r)
{
switch (r.Current)
{
case '!':
t.AdvanceTransition(MarkupDeclarationOpen);
break;
case '/':
t.AdvanceTransition(EndTagOpen);
break;
case '?':
t.AdvanceTransition(BogusComment);
break;
default:
if (r.MatchesLetter())
{
t.CreateTagPending(true);
t.Transition(TagName);
}
else
{
t.Error(this);
t.Emit('<'); // char that got us here
t.Transition(Data);
}
break;
}
}
public override void Read(Tokeniser t, CharacterReader r)
{
if (r.MatchesLetter())
{
string name = r.ConsumeLetterSequence();
t.tagPending.AppendTagName(name.ToLowerInvariant());
t.dataBuffer.Append(name);
return;
}
char c = r.Consume();
switch (c)
{
case '\t':
case '\n':
case '\f':
case ' ':
if (t.IsAppropriateEndTagToken())
{
t.Transition(BeforeAttributeName);
}
else
{
AnythingElse(t, r);
}
break;
case '/':
if (t.IsAppropriateEndTagToken())
{
t.Transition(SelfClosingStartTag);
}
else
{
AnythingElse(t, r);
}
break;
case '>':
if (t.IsAppropriateEndTagToken())
{
t.EmitTagPending();
t.Transition(Data);
}
else
{
AnythingElse(t, r);
}
break;
default:
AnythingElse(t, r);
break;
}
}
public override void Read(Tokeniser t, CharacterReader r)
{
if (r.MatchesLetter())
{
t.CreateTagPending(false);
t.Transition(ScriptDataEndTagName);
}
else
{
t.Emit("</");
t.Transition(ScriptData);
}
}
public override void Read(Tokeniser t, CharacterReader r)
{
if (r.MatchesLetter())
{
t.CreateTagPending(false);
t.tagPending.AppendTagName(char.ToLowerInvariant(r.Current));
t.dataBuffer.Append(r.Current);
t.AdvanceTransition(ScriptDataEscapedEndTagName);
}
else
{
t.Emit("</");
t.Transition(ScriptDataEscaped);
}
}
public override void Read(Tokeniser t, CharacterReader r)
{
if (r.MatchesLetter())
{
string name = r.ConsumeLetterSequence();
t.doctypePending.name.Append(name.ToLowerInvariant());
return;
}
char c = r.Consume();
switch (c)
{
case '>':
t.EmitDoctypePending();
t.Transition(Data);
break;
case '\t':
case '\n':
case '\f':
case ' ':
t.Transition(AfterDoctypeName);
break;
case nullChar:
t.Error(this);
t.doctypePending.name.Append(replacementChar);
break;
case eof:
t.EofError(this);
t.doctypePending.forceQuirks = true;
t.EmitDoctypePending();
t.Transition(Data);
break;
default:
t.doctypePending.name.Append(c);
break;
}
}
public override void Read(Tokeniser t, CharacterReader r)
{
if (r.MatchesLetter())
{
t.CreateTempBuffer();
t.dataBuffer.Append(char.ToLowerInvariant(r.Current));
t.Emit("<" + r.Current);
t.AdvanceTransition(ScriptDataDoubleEscapeStart);
}
else if (r.Matches('/'))
{
t.CreateTempBuffer();
t.AdvanceTransition(ScriptDataEscapedEndTagOpen);
}
else
{
t.Emit('<');
t.Transition(ScriptDataEscaped);
}
}
public override void Read(Tokeniser t, CharacterReader r)
{
if (r.MatchesLetter())
{
t.CreateDoctypePending();
t.Transition(DoctypeName);
return;
}
char c = r.Consume();
switch (c)
{
case '\t':
case '\n':
case '\f':
case ' ':
break; // ignore whitespace
case nullChar:
t.Error(this);
t.doctypePending.name.Append(replacementChar);
t.Transition(DoctypeName);
break;
case eof:
t.EofError(this);
t.CreateDoctypePending();
t.doctypePending.forceQuirks = true;
t.EmitDoctypePending();
t.Transition(Data);
break;
default:
t.CreateDoctypePending();
t.doctypePending.name.Append(c);
t.Transition(DoctypeName);
break;
}
}
// from < in rcdata
public override void Read(Tokeniser t, CharacterReader r)
{
if (r.Matches('/'))
{
t.CreateTempBuffer();
t.AdvanceTransition(RCDATAEndTagOpen);
}
else if (r.MatchesLetter() && !r.ContainsIgnoreCase("</" + t.AppropriateEndTagName()))
{
// diverge from spec: got a start tag, but there's no appropriate end tag (</title>), so rather than
// consuming to EOF; break out here
t.tagPending = new Token.EndTag(t.AppropriateEndTagName());
t.EmitTagPending();
r.Unconsume(); // undo "<"
t.Transition(Data);
}
else
{
t.Emit("<");
t.Transition(Rcdata);
}
}
public override void Read(Tokeniser t, CharacterReader r)
{
if (r.MatchesLetter())
{
string name = r.ConsumeLetterSequence();
t.dataBuffer.Append(name.ToLowerInvariant());
t.Emit(name);
return;
}
char c = r.Consume();
switch (c)
{
case '\t':
case '\n':
case '\f':
case ' ':
case '/':
case '>':
if (t.dataBuffer.ToString().Equals("script"))
{
t.Transition(ScriptDataEscaped);
}
else
{
t.Transition(ScriptDataDoubleEscaped);
}
t.Emit(c);
break;
default:
r.Unconsume();
t.Transition(ScriptDataDoubleEscaped);
break;
}
}
public override void Read(Tokeniser t, CharacterReader r)
{
if (r.IsEmpty)
{
t.EofError(this);
t.Emit("</");
t.Transition(Data);
}
else if (r.MatchesLetter())
{
t.CreateTagPending(false);
t.Transition(TagName);
}
else if (r.Matches('>'))
{
t.Error(this);
t.AdvanceTransition(Data);
}
else
{
t.Error(this);
t.AdvanceTransition(BogusComment);
}
}
public string ConsumeCharacterReference(char?additionalAllowedCharacter, bool inAttribute)
{
if (reader.IsEmpty)
{
return(null);
}
if (additionalAllowedCharacter.HasValue && additionalAllowedCharacter.Value == reader.Current)
{
return(null);
}
if (reader.MatchesAny('\t', '\n', '\f', ' ', '<', '&'))
{
return(null);
}
reader.Mark();
if (reader.MatchConsume("#")) // numbered
{
bool isHexMode = reader.MatchConsumeIgnoreCase("X");
string numRef = isHexMode ? reader.ConsumeHexSequence() : reader.ConsumeDigitSequence();
if (numRef.Length == 0) // didn't match anything
{
ParseError.NumericReferenceWithNoNumerals(this);
reader.RewindToMark();
return(null);
}
if (!reader.MatchConsume(";"))
{
ParseError.MissingSemicolon(this);
}
int charval = -1;
try {
int base2 = isHexMode ? 16 : 10;
charval = Convert.ToInt32(numRef, base2);
} catch (FormatException) {
} // skip
if (charval == -1 || (charval >= 0xD800 && charval <= 0xDFFF) || charval > 0x10FFFF)
{
ParseError.CharOutsideRange(this);
return(replacementStr);
}
else
{
// TODO: implement number replacement table
// TODO: check for extra illegal unicode points as parse errors
return(Char.ConvertFromUtf32(charval));
}
}
else // named
// get as many letters as possible, and look for matching entities. unconsume backwards till a match is found
{
string nameRef = reader.ConsumeLetterThenDigitSequence();
string origNameRef = nameRef; // for error reporting. nameRef gets chomped looking for matches
bool looksLegit = reader.Matches(';');
bool found = false;
while (nameRef.Length > 0 && !found)
{
if (HtmlEncoder.IsNamedEntity(nameRef))
{
found = true;
}
else
{
nameRef = nameRef.Substring(0, nameRef.Length - 1);
reader.Unconsume();
}
}
if (!found)
{
if (looksLegit) // named with semicolon
{
ParseError.InvalidNamedReference(this, origNameRef);
}
reader.RewindToMark();
return(null);
}
if (inAttribute && (reader.MatchesLetter() || reader.MatchesDigit() || reader.MatchesAny('=', '-', '_')))
{
// don't want that to match
reader.RewindToMark();
return(null);
}
if (!reader.MatchConsume(";"))
{
ParseError.MissingSemicolon(this);
}
return(HtmlEncoder.GetCharacterByName(nameRef));
}
}
