wuggy/Dactyloidae
wuggy/Dactyloidae
1
0
Fork 0
mirror of https://repo.dactyloidae.xyz/Dactyloidae/UXP.git synced 2026-10-09 16:57:30 +09:00
Code Issues Projects Releases Packages Wiki Activity

Fix an issue with the html5 tokenizer and tree builder (java htmlparser)

Browse source
This commit is contained in:
Matt A. Tobin 2020-01-15 22:07:59 -05:00 • committed by Roy Tam
commit 3338198f76
2 changed files with 35 additions and 18 deletions
Download patch file Download diff file Expand all files Collapse all files

32
parser/html/java/htmlparser/src/nu/validator/htmlparser/impl/Tokenizer.java
View file

@ -680,6 +680,22 @@ public class Tokenizer implements Locator {
* *
* @param specialTokenizerState * @param specialTokenizerState
* the tokenizer state to set * the tokenizer state to set
*/
public void setState(int specialTokenizerState) {
this.stateSave = specialTokenizerState;
this.endTagExpectation = null;
this.endTagExpectationAsArray = null;
}
// [NOCPP[
/**
* Sets the tokenizer state and the associated element name. This should
* only ever used to put the tokenizer into one of the states that have
* a special end tag expectation. For use from the tokenizer test harness.
*
* @param specialTokenizerState
* the tokenizer state to set
* @param endTagExpectation * @param endTagExpectation
* the expected end tag for transitioning back to normal * the expected end tag for transitioning back to normal
*/ */
@ -695,6 +711,8 @@ public class Tokenizer implements Locator {
endTagExpectationToArray(); endTagExpectationToArray();
} }
// ]NOCPP]
/** /**
* Sets the tokenizer state and the associated element name. This should * Sets the tokenizer state and the associated element name. This should
* only ever used to put the tokenizer into one of the states that have * only ever used to put the tokenizer into one of the states that have
@ -3749,11 +3767,17 @@ public class Tokenizer implements Locator {
c = checkChar(buf, pos); c = checkChar(buf, pos);
/* /*
* ASSERT! when entering this state, set index to 0 and * ASSERT! when entering this state, set index to 0 and
* call clearStrBufBeforeUse() assert (contentModelElement != * call clearStrBufBeforeUse(); Let's implement the above
* null); Let's implement the above without lookahead. * without lookahead. strBuf is the 'temporary buffer'.
* strBuf is the 'temporary buffer'.
*/ */
if (index < endTagExpectationAsArray.length) { if (endTagExpectationAsArray == null) {
tokenHandler.characters(Tokenizer.LT_SOLIDUS,
0, 2);
cstart = pos;
reconsume = true;
state = transition(state, returnState, reconsume, pos);
continue stateloop;
} else if (index < endTagExpectationAsArray.length) {
char e = endTagExpectationAsArray[index]; char e = endTagExpectationAsArray[index];
char folded = c; char folded = c;
if (c >= 'A' && c <= 'Z') { if (c >= 'A' && c <= 'Z') {

21
parser/html/java/htmlparser/src/nu/validator/htmlparser/impl/TreeBuilder.java
View file

@ -640,8 +640,7 @@ public abstract class TreeBuilder<T> implements TokenHandler,
); );
currentPtr++; currentPtr++;
stack[currentPtr] = node; stack[currentPtr] = node;
tokenizer.setStateAndEndTagExpectation(Tokenizer.DATA, tokenizer.setState(Tokenizer.DATA);
contextName);
// The frameset-ok flag is set even though <frameset> never // The frameset-ok flag is set even though <frameset> never
// ends up being allowed as HTML frameset in the fragment case. // ends up being allowed as HTML frameset in the fragment case.
mode = FRAMESET_OK; mode = FRAMESET_OK;
@ -671,8 +670,7 @@ public abstract class TreeBuilder<T> implements TokenHandler,
); );
currentPtr++; currentPtr++;
stack[currentPtr] = node; stack[currentPtr] = node;
tokenizer.setStateAndEndTagExpectation(Tokenizer.DATA, tokenizer.setState(Tokenizer.DATA);
contextName);
// The frameset-ok flag is set even though <frameset> never // The frameset-ok flag is set even though <frameset> never
// ends up being allowed as HTML frameset in the fragment case. // ends up being allowed as HTML frameset in the fragment case.
mode = FRAMESET_OK; mode = FRAMESET_OK;
@ -691,23 +689,18 @@ public abstract class TreeBuilder<T> implements TokenHandler,
resetTheInsertionMode(); resetTheInsertionMode();
formPointer = getFormPointerForContext(contextNode); formPointer = getFormPointerForContext(contextNode);
if ("title" == contextName || "textarea" == contextName) { if ("title" == contextName || "textarea" == contextName) {
tokenizer.setStateAndEndTagExpectation(Tokenizer.RCDATA, tokenizer.setState(Tokenizer.RCDATA);
contextName);
} else if ("style" == contextName || "xmp" == contextName } else if ("style" == contextName || "xmp" == contextName
|| "iframe" == contextName || "noembed" == contextName || "iframe" == contextName || "noembed" == contextName
|| "noframes" == contextName || "noframes" == contextName
|| (scriptingEnabled && "noscript" == contextName)) { || (scriptingEnabled && "noscript" == contextName)) {
tokenizer.setStateAndEndTagExpectation(Tokenizer.RAWTEXT, tokenizer.setState(Tokenizer.RAWTEXT);
contextName);
} else if ("plaintext" == contextName) { } else if ("plaintext" == contextName) {
tokenizer.setStateAndEndTagExpectation(Tokenizer.PLAINTEXT, tokenizer.setState(Tokenizer.PLAINTEXT);
contextName);
} else if ("script" == contextName) { } else if ("script" == contextName) {
tokenizer.setStateAndEndTagExpectation( tokenizer.setState(Tokenizer.SCRIPT_DATA);
Tokenizer.SCRIPT_DATA, contextName);
} else { } else {
tokenizer.setStateAndEndTagExpectation(Tokenizer.DATA, tokenizer.setState(Tokenizer.DATA);
contextName);
} }
} }
contextName = null; contextName = null;