void read(Tokeniser t, CharacterReader r) {
if (r.matchesLetter()) {
String name = r.consumeLetterSequence();
t.dataBuffer.append(name.toLowerCase());
t.emit(name);
return;
}
char c = r.consume();
switch (c) {
case '\t':
case '\n':
case '\r':
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);
}
}
void read(Tokeniser t, CharacterReader r) {
char c = r.consume();
switch (c) {
case '\t':
case '\n':
case '\r':
case '\f':
case ' ':
t.transition(BeforeAttributeName);
break;
case '/':
t.transition(SelfClosingStartTag);
break;
case '>':
t.emitTagPending();
t.transition(Data);
break;
case eof:
t.eofError(this);
t.transition(Data);
break;
default:
t.error(this);
r.unconsume();
t.transition(BeforeAttributeName);
}
}
void read(Tokeniser t, CharacterReader r) {
char c = r.consume();
switch (c) {
case '\t':
case '\n':
case '\r':
case '\f':
case ' ':
// ignore
break;
case '"':
t.transition(AttributeValue_doubleQuoted);
break;
case '&':
r.unconsume();
t.transition(AttributeValue_unquoted);
break;
case '\'':
t.transition(AttributeValue_singleQuoted);
break;
case nullChar:
t.error(this);
t.tagPending.appendAttributeValue(replacementChar);
t.transition(AttributeValue_unquoted);
break;
case eof:
t.eofError(this);
t.transition(Data);
break;
case '>':
t.error(this);
t.emitTagPending();
t.transition(Data);
break;
case '<':
case '=':
case '`':
t.error(this);
t.tagPending.appendAttributeValue(c);
t.transition(AttributeValue_unquoted);
break;
default:
r.unconsume();
t.transition(AttributeValue_unquoted);
}
}
void read(Tokeniser t, CharacterReader r) {
// todo: handle bogus comment starting from eof. when does that trigger?
// rewind to capture character that lead us here
r.unconsume();
Token.Comment comment = new Token.Comment();
comment.data.append(r.consumeTo('>'));
// todo: replace nullChar with replaceChar
t.emit(comment);
t.advanceTransition(Data);
}
// from tagname <xxx
void read(Tokeniser t, CharacterReader r) {
char c = r.consume();
switch (c) {
case '\t':
case '\n':
case '\r':
case '\f':
case ' ':
break; // ignore whitespace
case '/':
t.transition(SelfClosingStartTag);
break;
case '>':
t.emitTagPending();
t.transition(Data);
break;
case nullChar:
t.error(this);
t.tagPending.newAttribute();
r.unconsume();
t.transition(AttributeName);
break;
case eof:
t.eofError(this);
t.transition(Data);
break;
case '"':
case '\'':
case '<':
case '=':
t.error(this);
t.tagPending.newAttribute();
t.tagPending.appendAttributeName(c);
t.transition(AttributeName);
break;
default: // A-Z, anything else
t.tagPending.newAttribute();
r.unconsume();
t.transition(AttributeName);
}
}
void read(Tokeniser t, CharacterReader r) {
switch (r.consume()) {
case '/':
t.createTempBuffer();
t.transition(ScriptDataEndTagOpen);
break;
case '!':
t.emit("<!");
t.transition(ScriptDataEscapeStart);
break;
default:
t.emit("<");
r.unconsume();
t.transition(ScriptData);
}
}
// from < in rcdata
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);
}
}