Magic-Mirror/gecko-dev
1
0
Fork 0
mirror of https://github.com/mozilla/gecko-dev.git synced 2024-10-09 19:35:51 +00:00
Code Issues Actions Packages Projects Releases Wiki Activity

Bug 809403 - Sync Gecko C++ version of the HTML parser with changes made for Rust. rs=smaug.

Browse Source
This commit is contained in:
Henri Sivonen 2012-11-09 11:04:28 +02:00
parent 0bc2b12b46
commit 3c818a1876
3 changed files with 411 additions and 414 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

608
parser/html/javasrc/TreeBuilder.java
View File

@ -203,49 +203,71 @@ public abstract class TreeBuilder<T> implements TokenHandler,
// start insertion modes
private static final int INITIAL = 0;
private static final int IN_ROW = 0;
private static final int BEFORE_HTML = 1;
private static final int IN_TABLE_BODY = 1;
private static final int BEFORE_HEAD = 2;
private static final int IN_TABLE = 2;
private static final int IN_HEAD = 3;
private static final int IN_CAPTION = 3;
private static final int IN_HEAD_NOSCRIPT = 4;
private static final int IN_CELL = 4;
private static final int AFTER_HEAD = 5;
private static final int FRAMESET_OK = 5;
private static final int IN_BODY = 6;
private static final int IN_TABLE = 7;
private static final int IN_HEAD = 7;
private static final int IN_CAPTION = 8;
private static final int IN_HEAD_NOSCRIPT = 8;
// no fall-through
private static final int IN_COLUMN_GROUP = 9;
private static final int IN_TABLE_BODY = 10;
// no fall-through
private static final int IN_SELECT_IN_TABLE = 10;
private static final int IN_ROW = 11;
private static final int IN_SELECT = 11;
private static final int IN_CELL = 12;
// no fall-through
private static final int AFTER_BODY = 12;
private static final int IN_SELECT = 13;
// no fall-through
private static final int IN_FRAMESET = 13;
private static final int IN_SELECT_IN_TABLE = 14;
private static final int AFTER_FRAMESET = 14;
private static final int AFTER_BODY = 15;
// no fall-through
private static final int INITIAL = 15;
// could add fall-through
private static final int IN_FRAMESET = 16;
private static final int BEFORE_HTML = 16;
private static final int AFTER_FRAMESET = 17;
// could add fall-through
private static final int AFTER_AFTER_BODY = 18;
private static final int BEFORE_HEAD = 17;
private static final int AFTER_AFTER_FRAMESET = 19;
// no fall-through
private static final int AFTER_HEAD = 18;
private static final int TEXT = 20;
// no fall-through
private static final int AFTER_AFTER_BODY = 19;
private static final int FRAMESET_OK = 21;
// no fall-through
private static final int AFTER_AFTER_FRAMESET = 20;
// no fall-through
private static final int TEXT = 21;
// start charset states
@ -614,232 +636,214 @@ public abstract class TreeBuilder<T> implements TokenHandler,
public final void doctype(@Local String name, String publicIdentifier,
String systemIdentifier, boolean forceQuirks) throws SAXException {
needToDropLF = false;
if (!isInForeign()) {
switch (mode) {
case INITIAL:
// [NOCPP[
if (reportingDoctype) {
// ]NOCPP]
String emptyString = Portability.newEmptyString();
appendDoctypeToDocument(name == null ? "" : name,
publicIdentifier == null ? emptyString
: publicIdentifier,
systemIdentifier == null ? emptyString
: systemIdentifier);
Portability.releaseString(emptyString);
// [NOCPP[
}
switch (doctypeExpectation) {
case HTML:
// ]NOCPP]
if (isQuirky(name, publicIdentifier,
systemIdentifier, forceQuirks)) {
errQuirkyDoctype();
documentModeInternal(DocumentMode.QUIRKS_MODE,
publicIdentifier, systemIdentifier,
false);
} else if (isAlmostStandards(publicIdentifier,
systemIdentifier)) {
// [NOCPP[
if (firstCommentLocation != null) {
warn("Comments seen before doctype. Internet Explorer will go into the quirks mode.", firstCommentLocation);
}
// ]NOCPP]
errAlmostStandardsDoctype();
documentModeInternal(
DocumentMode.ALMOST_STANDARDS_MODE,
publicIdentifier, systemIdentifier,
false);
} else {
// [NOCPP[
if (firstCommentLocation != null) {
warn("Comments seen before doctype. Internet Explorer will go into the quirks mode.", firstCommentLocation);
}
if ((Portability.literalEqualsString(
"-//W3C//DTD HTML 4.0//EN",
publicIdentifier) && (systemIdentifier == null || Portability.literalEqualsString(
"http://www.w3.org/TR/REC-html40/strict.dtd",
systemIdentifier)))
|| (Portability.literalEqualsString(
"-//W3C//DTD HTML 4.01//EN",
publicIdentifier) && (systemIdentifier == null || Portability.literalEqualsString(
"http://www.w3.org/TR/html4/strict.dtd",
systemIdentifier)))
|| (Portability.literalEqualsString(
"-//W3C//DTD XHTML 1.0 Strict//EN",
publicIdentifier) && Portability.literalEqualsString(
"http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd",
systemIdentifier))
|| (Portability.literalEqualsString(
"-//W3C//DTD XHTML 1.1//EN",
publicIdentifier) && Portability.literalEqualsString(
"http://www.w3.org/TR/xhtml11/DTD/xhtml11.dtd",
systemIdentifier))
) {
warn("Obsolete doctype. Expected \u201C<!DOCTYPE html>\u201D.");
} else if (!((systemIdentifier == null || Portability.literalEqualsString(
"about:legacy-compat", systemIdentifier)) && publicIdentifier == null)) {
err("Legacy doctype. Expected \u201C<!DOCTYPE html>\u201D.");
}
// ]NOCPP]
documentModeInternal(
DocumentMode.STANDARDS_MODE,
publicIdentifier, systemIdentifier,
false);
}
// [NOCPP[
break;
case HTML401_STRICT:
html4 = true;
tokenizer.turnOnAdditionalHtml4Errors();
if (isQuirky(name, publicIdentifier,
systemIdentifier, forceQuirks)) {
err("Quirky doctype. Expected \u201C<!DOCTYPE HTML PUBLIC \"-//W3C//DTD HTML 4.01//EN\" \"http://www.w3.org/TR/html4/strict.dtd\">\u201D.");
documentModeInternal(DocumentMode.QUIRKS_MODE,
publicIdentifier, systemIdentifier,
true);
} else if (isAlmostStandards(publicIdentifier,
systemIdentifier)) {
if (firstCommentLocation != null) {
warn("Comments seen before doctype. Internet Explorer will go into the quirks mode.", firstCommentLocation);
}
err("Almost standards mode doctype. Expected \u201C<!DOCTYPE HTML PUBLIC \"-//W3C//DTD HTML 4.01//EN\" \"http://www.w3.org/TR/html4/strict.dtd\">\u201D.");
documentModeInternal(
DocumentMode.ALMOST_STANDARDS_MODE,
publicIdentifier, systemIdentifier,
true);
} else {
if (firstCommentLocation != null) {
warn("Comments seen before doctype. Internet Explorer will go into the quirks mode.", firstCommentLocation);
}
if ("-//W3C//DTD HTML 4.01//EN".equals(publicIdentifier)) {
if (!"http://www.w3.org/TR/html4/strict.dtd".equals(systemIdentifier)) {
warn("The doctype did not contain the system identifier prescribed by the HTML 4.01 specification. Expected \u201C<!DOCTYPE HTML PUBLIC \"-//W3C//DTD HTML 4.01//EN\" \"http://www.w3.org/TR/html4/strict.dtd\">\u201D.");
}
} else {
err("The doctype was not the HTML 4.01 Strict doctype. Expected \u201C<!DOCTYPE HTML PUBLIC \"-//W3C//DTD HTML 4.01//EN\" \"http://www.w3.org/TR/html4/strict.dtd\">\u201D.");
}
documentModeInternal(
DocumentMode.STANDARDS_MODE,
publicIdentifier, systemIdentifier,
true);
}
break;
case HTML401_TRANSITIONAL:
html4 = true;
tokenizer.turnOnAdditionalHtml4Errors();
if (isQuirky(name, publicIdentifier,
systemIdentifier, forceQuirks)) {
err("Quirky doctype. Expected \u201C<!DOCTYPE HTML PUBLIC \"-//W3C//DTD HTML 4.01 Transitional//EN\" \"http://www.w3.org/TR/html4/loose.dtd\">\u201D.");
documentModeInternal(DocumentMode.QUIRKS_MODE,
publicIdentifier, systemIdentifier,
true);
} else if (isAlmostStandards(publicIdentifier,
systemIdentifier)) {
if (firstCommentLocation != null) {
warn("Comments seen before doctype. Internet Explorer will go into the quirks mode.", firstCommentLocation);
}
if ("-//W3C//DTD HTML 4.01 Transitional//EN".equals(publicIdentifier)
&& systemIdentifier != null) {
if (!"http://www.w3.org/TR/html4/loose.dtd".equals(systemIdentifier)) {
warn("The doctype did not contain the system identifier prescribed by the HTML 4.01 specification. Expected \u201C<!DOCTYPE HTML PUBLIC \"-//W3C//DTD HTML 4.01 Transitional//EN\" \"http://www.w3.org/TR/html4/loose.dtd\">\u201D.");
}
} else {
err("The doctype was not a non-quirky HTML 4.01 Transitional doctype. Expected \u201C<!DOCTYPE HTML PUBLIC \"-//W3C//DTD HTML 4.01 Transitional//EN\" \"http://www.w3.org/TR/html4/loose.dtd\">\u201D.");
}
documentModeInternal(
DocumentMode.ALMOST_STANDARDS_MODE,
publicIdentifier, systemIdentifier,
true);
} else {
if (firstCommentLocation != null) {
warn("Comments seen before doctype. Internet Explorer will go into the quirks mode.", firstCommentLocation);
}
err("The doctype was not the HTML 4.01 Transitional doctype. Expected \u201C<!DOCTYPE HTML PUBLIC \"-//W3C//DTD HTML 4.01 Transitional//EN\" \"http://www.w3.org/TR/html4/loose.dtd\">\u201D.");
documentModeInternal(
DocumentMode.STANDARDS_MODE,
publicIdentifier, systemIdentifier,
true);
}
break;
case AUTO:
html4 = isHtml4Doctype(publicIdentifier);
if (html4) {
tokenizer.turnOnAdditionalHtml4Errors();
}
if (isQuirky(name, publicIdentifier,
systemIdentifier, forceQuirks)) {
err("Quirky doctype. Expected e.g. \u201C<!DOCTYPE html>\u201D.");
documentModeInternal(DocumentMode.QUIRKS_MODE,
publicIdentifier, systemIdentifier,
html4);
} else if (isAlmostStandards(publicIdentifier,
systemIdentifier)) {
if (firstCommentLocation != null) {
warn("Comments seen before doctype. Internet Explorer will go into the quirks mode.", firstCommentLocation);
}
if ("-//W3C//DTD HTML 4.01 Transitional//EN".equals(publicIdentifier)) {
if (!"http://www.w3.org/TR/html4/loose.dtd".equals(systemIdentifier)) {
warn("The doctype did not contain the system identifier prescribed by the HTML 4.01 specification. Expected \u201C<!DOCTYPE HTML PUBLIC \"-//W3C//DTD HTML 4.01 Transitional//EN\" \"http://www.w3.org/TR/html4/loose.dtd\">\u201D.");
}
} else {
err("Almost standards mode doctype. Expected e.g. \u201C<!DOCTYPE html>\u201D.");
}
documentModeInternal(
DocumentMode.ALMOST_STANDARDS_MODE,
publicIdentifier, systemIdentifier,
html4);
} else {
if (firstCommentLocation != null) {
warn("Comments seen before doctype. Internet Explorer will go into the quirks mode.", firstCommentLocation);
}
if ("-//W3C//DTD HTML 4.01//EN".equals(publicIdentifier)) {
if (!"http://www.w3.org/TR/html4/strict.dtd".equals(systemIdentifier)) {
warn("The doctype did not contain the system identifier prescribed by the HTML 4.01 specification. Expected \u201C<!DOCTYPE HTML PUBLIC \"-//W3C//DTD HTML 4.01//EN\" \"http://www.w3.org/TR/html4/strict.dtd\">\u201D.");
}
} else if (!((systemIdentifier == null || Portability.literalEqualsString(
"about:legacy-compat", systemIdentifier)) && publicIdentifier == null)) {
err("Legacy doctype. Expected e.g. \u201C<!DOCTYPE html>\u201D.");
}
documentModeInternal(
DocumentMode.STANDARDS_MODE,
publicIdentifier, systemIdentifier,
html4);
}
break;
case NO_DOCTYPE_ERRORS:
if (isQuirky(name, publicIdentifier,
systemIdentifier, forceQuirks)) {
documentModeInternal(DocumentMode.QUIRKS_MODE,
publicIdentifier, systemIdentifier,
false);
} else if (isAlmostStandards(publicIdentifier,
systemIdentifier)) {
documentModeInternal(
DocumentMode.ALMOST_STANDARDS_MODE,
publicIdentifier, systemIdentifier,
false);
} else {
documentModeInternal(
DocumentMode.STANDARDS_MODE,
publicIdentifier, systemIdentifier,
false);
}
break;
}
if (!isInForeign() && mode == INITIAL) {
// [NOCPP[
if (reportingDoctype) {
// ]NOCPP]
String emptyString = Portability.newEmptyString();
appendDoctypeToDocument(name == null ? "" : name,
publicIdentifier == null ? emptyString
: publicIdentifier,
systemIdentifier == null ? emptyString
: systemIdentifier);
Portability.releaseString(emptyString);
// [NOCPP[
}
switch (doctypeExpectation) {
case HTML:
// ]NOCPP]
if (isQuirky(name, publicIdentifier, systemIdentifier,
forceQuirks)) {
errQuirkyDoctype();
documentModeInternal(DocumentMode.QUIRKS_MODE,
publicIdentifier, systemIdentifier, false);
} else if (isAlmostStandards(publicIdentifier,
systemIdentifier)) {
// [NOCPP[
if (firstCommentLocation != null) {
warn("Comments seen before doctype. Internet Explorer will go into the quirks mode.",
firstCommentLocation);
}
// ]NOCPP]
errAlmostStandardsDoctype();
documentModeInternal(
DocumentMode.ALMOST_STANDARDS_MODE,
publicIdentifier, systemIdentifier, false);
} else {
// [NOCPP[
if (firstCommentLocation != null) {
warn("Comments seen before doctype. Internet Explorer will go into the quirks mode.",
firstCommentLocation);
}
if ((Portability.literalEqualsString(
"-//W3C//DTD HTML 4.0//EN", publicIdentifier) && (systemIdentifier == null || Portability.literalEqualsString(
"http://www.w3.org/TR/REC-html40/strict.dtd",
systemIdentifier)))
|| (Portability.literalEqualsString(
"-//W3C//DTD HTML 4.01//EN",
publicIdentifier) && (systemIdentifier == null || Portability.literalEqualsString(
"http://www.w3.org/TR/html4/strict.dtd",
systemIdentifier)))
|| (Portability.literalEqualsString(
"-//W3C//DTD XHTML 1.0 Strict//EN",
publicIdentifier) && Portability.literalEqualsString(
"http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd",
systemIdentifier))
|| (Portability.literalEqualsString(
"-//W3C//DTD XHTML 1.1//EN",
publicIdentifier) && Portability.literalEqualsString(
"http://www.w3.org/TR/xhtml11/DTD/xhtml11.dtd",
systemIdentifier))
/*
*
* Then, switch to the root element mode of the tree
* construction stage.
*/
mode = BEFORE_HTML;
return;
default:
) {
warn("Obsolete doctype. Expected \u201C<!DOCTYPE html>\u201D.");
} else if (!((systemIdentifier == null || Portability.literalEqualsString(
"about:legacy-compat", systemIdentifier)) && publicIdentifier == null)) {
err("Legacy doctype. Expected \u201C<!DOCTYPE html>\u201D.");
}
// ]NOCPP]
documentModeInternal(DocumentMode.STANDARDS_MODE,
publicIdentifier, systemIdentifier, false);
}
// [NOCPP[
break;
case HTML401_STRICT:
html4 = true;
tokenizer.turnOnAdditionalHtml4Errors();
if (isQuirky(name, publicIdentifier, systemIdentifier,
forceQuirks)) {
err("Quirky doctype. Expected \u201C<!DOCTYPE HTML PUBLIC \"-//W3C//DTD HTML 4.01//EN\" \"http://www.w3.org/TR/html4/strict.dtd\">\u201D.");
documentModeInternal(DocumentMode.QUIRKS_MODE,
publicIdentifier, systemIdentifier, true);
} else if (isAlmostStandards(publicIdentifier,
systemIdentifier)) {
if (firstCommentLocation != null) {
warn("Comments seen before doctype. Internet Explorer will go into the quirks mode.",
firstCommentLocation);
}
err("Almost standards mode doctype. Expected \u201C<!DOCTYPE HTML PUBLIC \"-//W3C//DTD HTML 4.01//EN\" \"http://www.w3.org/TR/html4/strict.dtd\">\u201D.");
documentModeInternal(
DocumentMode.ALMOST_STANDARDS_MODE,
publicIdentifier, systemIdentifier, true);
} else {
if (firstCommentLocation != null) {
warn("Comments seen before doctype. Internet Explorer will go into the quirks mode.",
firstCommentLocation);
}
if ("-//W3C//DTD HTML 4.01//EN".equals(publicIdentifier)) {
if (!"http://www.w3.org/TR/html4/strict.dtd".equals(systemIdentifier)) {
warn("The doctype did not contain the system identifier prescribed by the HTML 4.01 specification. Expected \u201C<!DOCTYPE HTML PUBLIC \"-//W3C//DTD HTML 4.01//EN\" \"http://www.w3.org/TR/html4/strict.dtd\">\u201D.");
}
} else {
err("The doctype was not the HTML 4.01 Strict doctype. Expected \u201C<!DOCTYPE HTML PUBLIC \"-//W3C//DTD HTML 4.01//EN\" \"http://www.w3.org/TR/html4/strict.dtd\">\u201D.");
}
documentModeInternal(DocumentMode.STANDARDS_MODE,
publicIdentifier, systemIdentifier, true);
}
break;
case HTML401_TRANSITIONAL:
html4 = true;
tokenizer.turnOnAdditionalHtml4Errors();
if (isQuirky(name, publicIdentifier, systemIdentifier,
forceQuirks)) {
err("Quirky doctype. Expected \u201C<!DOCTYPE HTML PUBLIC \"-//W3C//DTD HTML 4.01 Transitional//EN\" \"http://www.w3.org/TR/html4/loose.dtd\">\u201D.");
documentModeInternal(DocumentMode.QUIRKS_MODE,
publicIdentifier, systemIdentifier, true);
} else if (isAlmostStandards(publicIdentifier,
systemIdentifier)) {
if (firstCommentLocation != null) {
warn("Comments seen before doctype. Internet Explorer will go into the quirks mode.",
firstCommentLocation);
}
if ("-//W3C//DTD HTML 4.01 Transitional//EN".equals(publicIdentifier)
&& systemIdentifier != null) {
if (!"http://www.w3.org/TR/html4/loose.dtd".equals(systemIdentifier)) {
warn("The doctype did not contain the system identifier prescribed by the HTML 4.01 specification. Expected \u201C<!DOCTYPE HTML PUBLIC \"-//W3C//DTD HTML 4.01 Transitional//EN\" \"http://www.w3.org/TR/html4/loose.dtd\">\u201D.");
}
} else {
err("The doctype was not a non-quirky HTML 4.01 Transitional doctype. Expected \u201C<!DOCTYPE HTML PUBLIC \"-//W3C//DTD HTML 4.01 Transitional//EN\" \"http://www.w3.org/TR/html4/loose.dtd\">\u201D.");
}
documentModeInternal(
DocumentMode.ALMOST_STANDARDS_MODE,
publicIdentifier, systemIdentifier, true);
} else {
if (firstCommentLocation != null) {
warn("Comments seen before doctype. Internet Explorer will go into the quirks mode.",
firstCommentLocation);
}
err("The doctype was not the HTML 4.01 Transitional doctype. Expected \u201C<!DOCTYPE HTML PUBLIC \"-//W3C//DTD HTML 4.01 Transitional//EN\" \"http://www.w3.org/TR/html4/loose.dtd\">\u201D.");
documentModeInternal(DocumentMode.STANDARDS_MODE,
publicIdentifier, systemIdentifier, true);
}
break;
case AUTO:
html4 = isHtml4Doctype(publicIdentifier);
if (html4) {
tokenizer.turnOnAdditionalHtml4Errors();
}
if (isQuirky(name, publicIdentifier, systemIdentifier,
forceQuirks)) {
err("Quirky doctype. Expected e.g. \u201C<!DOCTYPE html>\u201D.");
documentModeInternal(DocumentMode.QUIRKS_MODE,
publicIdentifier, systemIdentifier, html4);
} else if (isAlmostStandards(publicIdentifier,
systemIdentifier)) {
if (firstCommentLocation != null) {
warn("Comments seen before doctype. Internet Explorer will go into the quirks mode.",
firstCommentLocation);
}
if ("-//W3C//DTD HTML 4.01 Transitional//EN".equals(publicIdentifier)) {
if (!"http://www.w3.org/TR/html4/loose.dtd".equals(systemIdentifier)) {
warn("The doctype did not contain the system identifier prescribed by the HTML 4.01 specification. Expected \u201C<!DOCTYPE HTML PUBLIC \"-//W3C//DTD HTML 4.01 Transitional//EN\" \"http://www.w3.org/TR/html4/loose.dtd\">\u201D.");
}
} else {
err("Almost standards mode doctype. Expected e.g. \u201C<!DOCTYPE html>\u201D.");
}
documentModeInternal(
DocumentMode.ALMOST_STANDARDS_MODE,
publicIdentifier, systemIdentifier, html4);
} else {
if (firstCommentLocation != null) {
warn("Comments seen before doctype. Internet Explorer will go into the quirks mode.",
firstCommentLocation);
}
if ("-//W3C//DTD HTML 4.01//EN".equals(publicIdentifier)) {
if (!"http://www.w3.org/TR/html4/strict.dtd".equals(systemIdentifier)) {
warn("The doctype did not contain the system identifier prescribed by the HTML 4.01 specification. Expected \u201C<!DOCTYPE HTML PUBLIC \"-//W3C//DTD HTML 4.01//EN\" \"http://www.w3.org/TR/html4/strict.dtd\">\u201D.");
}
} else if (!((systemIdentifier == null || Portability.literalEqualsString(
"about:legacy-compat", systemIdentifier)) && publicIdentifier == null)) {
err("Legacy doctype. Expected e.g. \u201C<!DOCTYPE html>\u201D.");
}
documentModeInternal(DocumentMode.STANDARDS_MODE,
publicIdentifier, systemIdentifier, html4);
}
break;
case NO_DOCTYPE_ERRORS:
if (isQuirky(name, publicIdentifier, systemIdentifier,
forceQuirks)) {
documentModeInternal(DocumentMode.QUIRKS_MODE,
publicIdentifier, systemIdentifier, false);
} else if (isAlmostStandards(publicIdentifier,
systemIdentifier)) {
documentModeInternal(
DocumentMode.ALMOST_STANDARDS_MODE,
publicIdentifier, systemIdentifier, false);
} else {
documentModeInternal(DocumentMode.STANDARDS_MODE,
publicIdentifier, systemIdentifier, false);
}
break;
}
// ]NOCPP]
/*
*
* Then, switch to the root element mode of the tree construction
* stage.
*/
mode = BEFORE_HTML;
return;
}
/*
* A DOCTYPE token Parse error.
@ -1611,6 +1615,35 @@ public abstract class TreeBuilder<T> implements TokenHandler,
} // foreignObject / annotation-xml
}
switch (mode) {
case IN_ROW:
switch (group) {
case TD_OR_TH:
clearStackBackTo(findLastOrRoot(TreeBuilder.TR));
appendToCurrentNodeAndPushElement(
elementName,
attributes);
mode = IN_CELL;
insertMarker();
attributes = null; // CPP
break starttagloop;
case CAPTION:
case COL:
case COLGROUP:
case TBODY_OR_THEAD_OR_TFOOT:
case TR:
eltPos = findLastOrRoot(TreeBuilder.TR);
if (eltPos == 0) {
assert fragment;
errNoTableRowToClose();
break starttagloop;
}
clearStackBackTo(eltPos);
pop();
mode = IN_TABLE_BODY;
continue;
default:
// fall through to IN_TABLE
}
case IN_TABLE_BODY:
switch (group) {
case TR:
@ -1646,35 +1679,6 @@ public abstract class TreeBuilder<T> implements TokenHandler,
default:
// fall through to IN_TABLE
}
case IN_ROW:
switch (group) {
case TD_OR_TH:
clearStackBackTo(findLastOrRoot(TreeBuilder.TR));
appendToCurrentNodeAndPushElement(
elementName,
attributes);
mode = IN_CELL;
insertMarker();
attributes = null; // CPP
break starttagloop;
case CAPTION:
case COL:
case COLGROUP:
case TBODY_OR_THEAD_OR_TFOOT:
case TR:
eltPos = findLastOrRoot(TreeBuilder.TR);
if (eltPos == 0) {
assert fragment;
errNoTableRowToClose();
break starttagloop;
}
clearStackBackTo(eltPos);
pop();
mode = IN_TABLE_BODY;
continue;
default:
// fall through to IN_TABLE
}
case IN_TABLE:
intableloop: for (;;) {
switch (group) {
@ -3606,6 +3610,37 @@ public abstract class TreeBuilder<T> implements TokenHandler,
eltPos--;
}
}
case IN_HEAD:
switch (group) {
case HEAD:
pop();
mode = AFTER_HEAD;
break endtagloop;
case BR:
case HTML:
case BODY:
pop();
mode = AFTER_HEAD;
continue;
default:
errStrayEndTag(name);
break endtagloop;
}
case IN_HEAD_NOSCRIPT:
switch (group) {
case NOSCRIPT:
pop();
mode = IN_HEAD;
break endtagloop;
case BR:
errStrayEndTag(name);
pop();
mode = IN_HEAD;
continue;
default:
errStrayEndTag(name);
break endtagloop;
}
case IN_COLUMN_GROUP:
switch (group) {
case COLGROUP:
@ -3806,37 +3841,6 @@ public abstract class TreeBuilder<T> implements TokenHandler,
errStrayEndTag(name);
break endtagloop;
}
case IN_HEAD:
switch (group) {
case HEAD:
pop();
mode = AFTER_HEAD;
break endtagloop;
case BR:
case HTML:
case BODY:
pop();
mode = AFTER_HEAD;
continue;
default:
errStrayEndTag(name);
break endtagloop;
}
case IN_HEAD_NOSCRIPT:
switch (group) {
case NOSCRIPT:
pop();
mode = IN_HEAD;
break endtagloop;
case BR:
errStrayEndTag(name);
pop();
mode = IN_HEAD;
continue;
default:
errStrayEndTag(name);
break endtagloop;
}
case AFTER_HEAD:
switch (group) {
case HTML:

176
parser/html/nsHtml5TreeBuilder.cpp
View File

@ -51,6 +51,7 @@
#include "nsAHtml5TreeBuilderState.h"
#include "nsHtml5Highlighter.h"
#include "nsHtml5ViewSourceUtils.h"
#include "mozilla/Likely.h"
#include "nsHtml5Tokenizer.h"
#include "nsHtml5MetaScanner.h"
@ -64,8 +65,6 @@
#include "nsHtml5TreeBuilder.h"
#include "mozilla/Likely.h"
PRUnichar nsHtml5TreeBuilder::REPLACEMENT_CHARACTER[] = { 0xfffd };
static const char* const QUIRKY_PUBLIC_IDS_DATA[] = { "+//silmaril//dtd html pro v0r11 19970101//", "-//advasoft ltd//dtd html 3.0 aswedit + extensions//", "-//as//dtd html 3.0 aswedit + extensions//", "-//ietf//dtd html 2.0 level 1//", "-//ietf//dtd html 2.0 level 2//", "-//ietf//dtd html 2.0 strict level 1//", "-//ietf//dtd html 2.0 strict level 2//", "-//ietf//dtd html 2.0 strict//", "-//ietf//dtd html 2.0//", "-//ietf//dtd html 2.1e//", "-//ietf//dtd html 3.0//", "-//ietf//dtd html 3.2 final//", "-//ietf//dtd html 3.2//", "-//ietf//dtd html 3//", "-//ietf//dtd html level 0//", "-//ietf//dtd html level 1//", "-//ietf//dtd html level 2//", "-//ietf//dtd html level 3//", "-//ietf//dtd html strict level 0//", "-//ietf//dtd html strict level 1//", "-//ietf//dtd html strict level 2//", "-//ietf//dtd html strict level 3//", "-//ietf//dtd html strict//", "-//ietf//dtd html//", "-//metrius//dtd metrius presentational//", "-//microsoft//dtd internet explorer 2.0 html strict//", "-//microsoft//dtd internet explorer 2.0 html//", "-//microsoft//dtd internet explorer 2.0 tables//", "-//microsoft//dtd internet explorer 3.0 html strict//", "-//microsoft//dtd internet explorer 3.0 html//", "-//microsoft//dtd internet explorer 3.0 tables//", "-//netscape comm. corp.//dtd html//", "-//netscape comm. corp.//dtd strict html//", "-//o'reilly and associates//dtd html 2.0//", "-//o'reilly and associates//dtd html extended 1.0//", "-//o'reilly and associates//dtd html extended relaxed 1.0//", "-//softquad software//dtd hotmetal pro 6.0::19990601::extensions to html 4.0//", "-//softquad//dtd hotmetal pro 4.0::19971010::extensions to html 4.0//", "-//spyglass//dtd html 2.0 extended//", "-//sq//dtd html 2.0 hotmetal + extensions//", "-//sun microsystems corp.//dtd hotjava html//", "-//sun microsystems corp.//dtd hotjava strict html//", "-//w3c//dtd html 3 1995-03-24//", "-//w3c//dtd html 3.2 draft//", "-//w3c//dtd html 3.2 final//", "-//w3c//dtd html 3.2//", "-//w3c//dtd html 3.2s draft//", "-//w3c//dtd html 4.0 frameset//", "-//w3c//dtd html 4.0 transitional//", "-//w3c//dtd html experimental 19960712//", "-//w3c//dtd html experimental 970421//", "-//w3c//dtd w3 html//", "-//w3o//dtd w3 html 3.0//", "-//webtechs//dtd mozilla html 2.0//", "-//webtechs//dtd mozilla html//" };
staticJArray<const char*,int32_t> nsHtml5TreeBuilder::QUIRKY_PUBLIC_IDS = { QUIRKY_PUBLIC_IDS_DATA, NS_ARRAY_LENGTH(QUIRKY_PUBLIC_IDS_DATA) };
@ -125,28 +124,21 @@ void
nsHtml5TreeBuilder::doctype(nsIAtom* name, nsString* publicIdentifier, nsString* systemIdentifier, bool forceQuirks)
{
needToDropLF = false;
if (!isInForeign()) {
switch(mode) {
case NS_HTML5TREE_BUILDER_INITIAL: {
nsString* emptyString = nsHtml5Portability::newEmptyString();
appendDoctypeToDocument(!name ? nsHtml5Atoms::emptystring : name, !publicIdentifier ? emptyString : publicIdentifier, !systemIdentifier ? emptyString : systemIdentifier);
nsHtml5Portability::releaseString(emptyString);
if (isQuirky(name, publicIdentifier, systemIdentifier, forceQuirks)) {
errQuirkyDoctype();
documentModeInternal(QUIRKS_MODE, publicIdentifier, systemIdentifier, false);
} else if (isAlmostStandards(publicIdentifier, systemIdentifier)) {
errAlmostStandardsDoctype();
documentModeInternal(ALMOST_STANDARDS_MODE, publicIdentifier, systemIdentifier, false);
} else {
documentModeInternal(STANDARDS_MODE, publicIdentifier, systemIdentifier, false);
}
mode = NS_HTML5TREE_BUILDER_BEFORE_HTML;
return;
}
default: {
break;
}
if (!isInForeign() && mode == NS_HTML5TREE_BUILDER_INITIAL) {
nsString* emptyString = nsHtml5Portability::newEmptyString();
appendDoctypeToDocument(!name ? nsHtml5Atoms::emptystring : name, !publicIdentifier ? emptyString : publicIdentifier, !systemIdentifier ? emptyString : systemIdentifier);
nsHtml5Portability::releaseString(emptyString);
if (isQuirky(name, publicIdentifier, systemIdentifier, forceQuirks)) {
errQuirkyDoctype();
documentModeInternal(QUIRKS_MODE, publicIdentifier, systemIdentifier, false);
} else if (isAlmostStandards(publicIdentifier, systemIdentifier)) {
errAlmostStandardsDoctype();
documentModeInternal(ALMOST_STANDARDS_MODE, publicIdentifier, systemIdentifier, false);
} else {
documentModeInternal(STANDARDS_MODE, publicIdentifier, systemIdentifier, false);
}
mode = NS_HTML5TREE_BUILDER_BEFORE_HTML;
return;
}
errStrayDoctype();
return;
@ -637,6 +629,36 @@ nsHtml5TreeBuilder::startTag(nsHtml5ElementName* elementName, nsHtml5HtmlAttribu
}
}
switch(mode) {
case NS_HTML5TREE_BUILDER_IN_ROW: {
switch(group) {
case NS_HTML5TREE_BUILDER_TD_OR_TH: {
clearStackBackTo(findLastOrRoot(NS_HTML5TREE_BUILDER_TR));
appendToCurrentNodeAndPushElement(elementName, attributes);
mode = NS_HTML5TREE_BUILDER_IN_CELL;
insertMarker();
attributes = nullptr;
NS_HTML5_BREAK(starttagloop);
}
case NS_HTML5TREE_BUILDER_CAPTION:
case NS_HTML5TREE_BUILDER_COL:
case NS_HTML5TREE_BUILDER_COLGROUP:
case NS_HTML5TREE_BUILDER_TBODY_OR_THEAD_OR_TFOOT:
case NS_HTML5TREE_BUILDER_TR: {
eltPos = findLastOrRoot(NS_HTML5TREE_BUILDER_TR);
if (!eltPos) {
MOZ_ASSERT(fragment);
errNoTableRowToClose();
NS_HTML5_BREAK(starttagloop);
}
clearStackBackTo(eltPos);
pop();
mode = NS_HTML5TREE_BUILDER_IN_TABLE_BODY;
continue;
}
default:
; // fall through
}
}
case NS_HTML5TREE_BUILDER_IN_TABLE_BODY: {
switch(group) {
case NS_HTML5TREE_BUILDER_TR: {
@ -672,36 +694,6 @@ nsHtml5TreeBuilder::startTag(nsHtml5ElementName* elementName, nsHtml5HtmlAttribu
; // fall through
}
}
case NS_HTML5TREE_BUILDER_IN_ROW: {
switch(group) {
case NS_HTML5TREE_BUILDER_TD_OR_TH: {
clearStackBackTo(findLastOrRoot(NS_HTML5TREE_BUILDER_TR));
appendToCurrentNodeAndPushElement(elementName, attributes);
mode = NS_HTML5TREE_BUILDER_IN_CELL;
insertMarker();
attributes = nullptr;
NS_HTML5_BREAK(starttagloop);
}
case NS_HTML5TREE_BUILDER_CAPTION:
case NS_HTML5TREE_BUILDER_COL:
case NS_HTML5TREE_BUILDER_COLGROUP:
case NS_HTML5TREE_BUILDER_TBODY_OR_THEAD_OR_TFOOT:
case NS_HTML5TREE_BUILDER_TR: {
eltPos = findLastOrRoot(NS_HTML5TREE_BUILDER_TR);
if (!eltPos) {
MOZ_ASSERT(fragment);
errNoTableRowToClose();
NS_HTML5_BREAK(starttagloop);
}
clearStackBackTo(eltPos);
pop();
mode = NS_HTML5TREE_BUILDER_IN_TABLE_BODY;
continue;
}
default:
; // fall through
}
}
case NS_HTML5TREE_BUILDER_IN_TABLE: {
for (; ; ) {
switch(group) {
@ -2534,6 +2526,45 @@ nsHtml5TreeBuilder::endTag(nsHtml5ElementName* elementName)
}
}
}
case NS_HTML5TREE_BUILDER_IN_HEAD: {
switch(group) {
case NS_HTML5TREE_BUILDER_HEAD: {
pop();
mode = NS_HTML5TREE_BUILDER_AFTER_HEAD;
NS_HTML5_BREAK(endtagloop);
}
case NS_HTML5TREE_BUILDER_BR:
case NS_HTML5TREE_BUILDER_HTML:
case NS_HTML5TREE_BUILDER_BODY: {
pop();
mode = NS_HTML5TREE_BUILDER_AFTER_HEAD;
continue;
}
default: {
errStrayEndTag(name);
NS_HTML5_BREAK(endtagloop);
}
}
}
case NS_HTML5TREE_BUILDER_IN_HEAD_NOSCRIPT: {
switch(group) {
case NS_HTML5TREE_BUILDER_NOSCRIPT: {
pop();
mode = NS_HTML5TREE_BUILDER_IN_HEAD;
NS_HTML5_BREAK(endtagloop);
}
case NS_HTML5TREE_BUILDER_BR: {
errStrayEndTag(name);
pop();
mode = NS_HTML5TREE_BUILDER_IN_HEAD;
continue;
}
default: {
errStrayEndTag(name);
NS_HTML5_BREAK(endtagloop);
}
}
}
case NS_HTML5TREE_BUILDER_IN_COLUMN_GROUP: {
switch(group) {
case NS_HTML5TREE_BUILDER_COLGROUP: {
@ -2718,45 +2749,6 @@ nsHtml5TreeBuilder::endTag(nsHtml5ElementName* elementName)
}
}
}
case NS_HTML5TREE_BUILDER_IN_HEAD: {
switch(group) {
case NS_HTML5TREE_BUILDER_HEAD: {
pop();
mode = NS_HTML5TREE_BUILDER_AFTER_HEAD;
NS_HTML5_BREAK(endtagloop);
}
case NS_HTML5TREE_BUILDER_BR:
case NS_HTML5TREE_BUILDER_HTML:
case NS_HTML5TREE_BUILDER_BODY: {
pop();
mode = NS_HTML5TREE_BUILDER_AFTER_HEAD;
continue;
}
default: {
errStrayEndTag(name);
NS_HTML5_BREAK(endtagloop);
}
}
}
case NS_HTML5TREE_BUILDER_IN_HEAD_NOSCRIPT: {
switch(group) {
case NS_HTML5TREE_BUILDER_NOSCRIPT: {
pop();
mode = NS_HTML5TREE_BUILDER_IN_HEAD;
NS_HTML5_BREAK(endtagloop);
}
case NS_HTML5TREE_BUILDER_BR: {
errStrayEndTag(name);
pop();
mode = NS_HTML5TREE_BUILDER_IN_HEAD;
continue;
}
default: {
errStrayEndTag(name);
NS_HTML5_BREAK(endtagloop);
}
}
}
case NS_HTML5TREE_BUILDER_AFTER_HEAD: {
switch(group) {
case NS_HTML5TREE_BUILDER_HTML:

41
parser/html/nsHtml5TreeBuilder.h
View File

@ -52,6 +52,7 @@
#include "nsAHtml5TreeBuilderState.h"
#include "nsHtml5Highlighter.h"
#include "nsHtml5ViewSourceUtils.h"
#include "mozilla/Likely.h"
class nsHtml5StreamParser;
@ -318,28 +319,28 @@ class nsHtml5TreeBuilder : public nsAHtml5TreeBuilderState
#define NS_HTML5TREE_BUILDER_FONT 64
#define NS_HTML5TREE_BUILDER_KEYGEN 65
#define NS_HTML5TREE_BUILDER_MENUITEM 66
#define NS_HTML5TREE_BUILDER_INITIAL 0
#define NS_HTML5TREE_BUILDER_BEFORE_HTML 1
#define NS_HTML5TREE_BUILDER_BEFORE_HEAD 2
#define NS_HTML5TREE_BUILDER_IN_HEAD 3
#define NS_HTML5TREE_BUILDER_IN_HEAD_NOSCRIPT 4
#define NS_HTML5TREE_BUILDER_AFTER_HEAD 5
#define NS_HTML5TREE_BUILDER_IN_ROW 0
#define NS_HTML5TREE_BUILDER_IN_TABLE_BODY 1
#define NS_HTML5TREE_BUILDER_IN_TABLE 2
#define NS_HTML5TREE_BUILDER_IN_CAPTION 3
#define NS_HTML5TREE_BUILDER_IN_CELL 4
#define NS_HTML5TREE_BUILDER_FRAMESET_OK 5
#define NS_HTML5TREE_BUILDER_IN_BODY 6
#define NS_HTML5TREE_BUILDER_IN_TABLE 7
#define NS_HTML5TREE_BUILDER_IN_CAPTION 8
#define NS_HTML5TREE_BUILDER_IN_HEAD 7
#define NS_HTML5TREE_BUILDER_IN_HEAD_NOSCRIPT 8
#define NS_HTML5TREE_BUILDER_IN_COLUMN_GROUP 9
#define NS_HTML5TREE_BUILDER_IN_TABLE_BODY 10
#define NS_HTML5TREE_BUILDER_IN_ROW 11
#define NS_HTML5TREE_BUILDER_IN_CELL 12
#define NS_HTML5TREE_BUILDER_IN_SELECT 13
#define NS_HTML5TREE_BUILDER_IN_SELECT_IN_TABLE 14
#define NS_HTML5TREE_BUILDER_AFTER_BODY 15
#define NS_HTML5TREE_BUILDER_IN_FRAMESET 16
#define NS_HTML5TREE_BUILDER_AFTER_FRAMESET 17
#define NS_HTML5TREE_BUILDER_AFTER_AFTER_BODY 18
#define NS_HTML5TREE_BUILDER_AFTER_AFTER_FRAMESET 19
#define NS_HTML5TREE_BUILDER_TEXT 20
#define NS_HTML5TREE_BUILDER_FRAMESET_OK 21
#define NS_HTML5TREE_BUILDER_IN_SELECT_IN_TABLE 10
#define NS_HTML5TREE_BUILDER_IN_SELECT 11
#define NS_HTML5TREE_BUILDER_AFTER_BODY 12
#define NS_HTML5TREE_BUILDER_IN_FRAMESET 13
#define NS_HTML5TREE_BUILDER_AFTER_FRAMESET 14
#define NS_HTML5TREE_BUILDER_INITIAL 15
#define NS_HTML5TREE_BUILDER_BEFORE_HTML 16
#define NS_HTML5TREE_BUILDER_BEFORE_HEAD 17
#define NS_HTML5TREE_BUILDER_AFTER_HEAD 18
#define NS_HTML5TREE_BUILDER_AFTER_AFTER_BODY 19
#define NS_HTML5TREE_BUILDER_AFTER_AFTER_FRAMESET 20
#define NS_HTML5TREE_BUILDER_TEXT 21
#define NS_HTML5TREE_BUILDER_CHARSET_INITIAL 0
#define NS_HTML5TREE_BUILDER_CHARSET_C 1
#define NS_HTML5TREE_BUILDER_CHARSET_H 2