mozilla
/
gecko-dev
зеркало из https://github.com/mozilla/gecko-dev.git
1
0
Форкнуть 0
Код Задачи Пакеты Проекты Релизы Вики Активность

final features for parsing engine (WIP); lots of bug fixes

This commit is contained in:
rickg%netscape.com 1999-05-16 05:27:23 +00:00
Родитель fb8e38b9ea
Коммит 8e9ee1a381
44 изменённых файлов: 2860 добавлений и 2062 удалений
Показать статистику Скачать Patch файл Скачать Diff файл Показать все файлы Свернуть все файлы

956
htmlparser/src/CNavDTD.cpp
Просмотреть файл

Разница между файлами не показана из-за своего большого размера Загрузить разницу

31
htmlparser/src/CNavDTD.h
Просмотреть файл

@ -296,17 +296,6 @@ CLASS_EXPORT_HTMLPARS CNavDTD : public nsIDTD {
*/
virtual PRBool CanOmit(eHTMLTags aParent,eHTMLTags aChild)const;
/**
* This is called to determine if the given parent can omit the
* given child (end tag).
*
* @update gess 3/25/98
* @param aParent -- tag type of parent
* @param aChild -- tag type of child
* @return PR_TRUE if given tag can contain omit child (end tag)
*/
virtual PRBool CanOmitEndTag(eHTMLTags aParent,eHTMLTags aChild)const;
/**
* This method gets called to determine whether a given
* tag is itself a container
@ -405,7 +394,6 @@ CLASS_EXPORT_HTMLPARS CNavDTD : public nsIDTD {
nsresult HandleEndToken(CToken* aToken);
nsresult HandleEntityToken(CToken* aToken);
nsresult HandleCommentToken(CToken* aToken);
nsresult HandleSkippedContentToken(CToken* aToken);
nsresult HandleAttributeToken(CToken* aToken);
nsresult HandleScriptToken(nsCParserNode& aNode);
nsresult HandleStyleToken(CToken* aToken);
@ -431,7 +419,7 @@ CLASS_EXPORT_HTMLPARS CNavDTD : public nsIDTD {
nsresult OpenForm(const nsIParserNode& aNode);
nsresult OpenMap(const nsIParserNode& aNode);
nsresult OpenFrameset(const nsIParserNode& aNode);
nsresult OpenContainer(const nsIParserNode& aNode,PRBool aUpdateStyleStack);
nsresult OpenContainer(const nsIParserNode& aNode,PRBool aClosedByStartTag);
/**
* The next set of methods close the given HTML element.
@ -446,7 +434,7 @@ CLASS_EXPORT_HTMLPARS CNavDTD : public nsIDTD {
nsresult CloseForm(const nsIParserNode& aNode);
nsresult CloseMap(const nsIParserNode& aNode);
nsresult CloseFrameset(const nsIParserNode& aNode);
nsresult CloseContainer(const nsIParserNode& aNode,eHTMLTags anActualTag,PRBool aUpdateStyles);
nsresult CloseContainer(const nsIParserNode& aNode,eHTMLTags anActualTag,PRBool aClosedByStartTag);
/**
* The special purpose methods automatically close
@ -455,8 +443,8 @@ CLASS_EXPORT_HTMLPARS CNavDTD : public nsIDTD {
* @return error code - 0 if all went well.
*/
nsresult CloseTopmostContainer();
nsresult CloseContainersTo(eHTMLTags aTag,PRBool aUpdateStyles);
nsresult CloseContainersTo(PRInt32 anIndex,eHTMLTags aTag,PRBool aUpdateStyles);
nsresult CloseContainersTo(eHTMLTags aTag,PRBool aClosedByStartTag);
nsresult CloseContainersTo(PRInt32 anIndex,eHTMLTags aTag,PRBool aClosedByStartTag);
/**
* Causes leaf to be added to sink at current vector pos.
@ -465,7 +453,7 @@ CLASS_EXPORT_HTMLPARS CNavDTD : public nsIDTD {
* @return error code - 0 if all went well.
*/
nsresult AddLeaf(const nsIParserNode& aNode);
nsresult AddHeadLeaf(const nsIParserNode& aNode);
nsresult AddHeadLeaf(nsIParserNode& aNode);
/**
* Causes auto-closures of context vector stack in order to find a
@ -495,17 +483,16 @@ CLASS_EXPORT_HTMLPARS CNavDTD : public nsIDTD {
* @param aTag -- represents the transient style tag to be handled.
* @return error code -- usually 0
*/
nsresult OpenTransientStyles(eHTMLTags aTag);
nsresult CloseTransientStyles(eHTMLTags aTag);
nsresult OpenTransientStyles(eHTMLTags aChildTag);
nsresult CloseTransientStyles(eHTMLTags aChildTag);
nsresult UpdateStyleStackForOpenTag(eHTMLTags aTag,eHTMLTags aActualTag);
nsresult UpdateStyleStackForCloseTag(eHTMLTags aTag,eHTMLTags aActualTag);
PRBool CanContainStyles(eHTMLTags aTag) const;
nsresult DoFragment(PRBool aFlag);
protected:
nsresult CollectAttributes(nsCParserNode& aNode,PRInt32 aCount);
nsresult CollectAttributes(nsCParserNode& aNode,eHTMLTags aTag,PRInt32 aCount);
nsresult CollectSkippedContent(nsCParserNode& aNode,PRInt32& aCount);
nsresult WillHandleStartTag(CToken* aToken,eHTMLTags aChildTag,nsCParserNode& aNode);
nsresult DidHandleStartTag(nsCParserNode& aNode,eHTMLTags aChildTag);
@ -531,8 +518,10 @@ protected:
nsParser* mParser;
nsITokenizer* mTokenizer;
nsDeque mMisplacedContent;
nsDeque mSkippedContent;
PRBool mHasOpenScript;
PRBool mSaveBadTokens;
eHTMLTags mSkipTarget;
PRUint32 mComputedCRC32;
PRUint32 mExpectedCRC32;

27
htmlparser/src/COtherDTD.cpp
Просмотреть файл

@ -254,20 +254,6 @@ nsresult COtherDTD::HandleCommentToken(CToken* aToken) {
return CNavDTD::HandleCommentToken(aToken);
}
/**
* This method gets called when a skippedcontent token has
* been encountered in the parse process. After verifying
* that the topmost container can contain text, we call
* AddLeaf to store this token in the top container.
*
* @update gess 3/25/98
* @param aToken -- next (start) token to be handled
* @return PR_TRUE if all went well; PR_FALSE if error occured
*/
nsresult COtherDTD::HandleSkippedContentToken(CToken* aToken) {
return CNavDTD::HandleSkippedContentToken(aToken);
}
/**
* This method gets called when an attribute token has been
* encountered in the parse process. This is an error, since
@ -334,19 +320,6 @@ PRBool COtherDTD::CanOmit(eHTMLTags aParent,eHTMLTags aChild) const {
}
/**
* This method gets called to determine whether a given
* ENDtag can be omitted. Admittedly,this is a gross simplification.
*
* @update gess 3/25/98
* @param aTag -- tag to test for containership
* @return PR_TRUE if given tag can contain other tags
*/
PRBool COtherDTD::CanOmitEndTag(eHTMLTags aParent,eHTMLTags aChild) const {
return CNavDTD::CanOmitEndTag(aParent,aChild);
}
/*********************************************
Here comes code that handles the interface
to our content sink.

20
htmlparser/src/COtherDTD.h
Просмотреть файл

@ -121,17 +121,6 @@ class COtherDTD : public CNavDTD {
*/
virtual PRBool CanOmit(eHTMLTags aParent,eHTMLTags aChild)const;
/**
* This method gets called to determine whether a given
* tag can contain newlines. Most do not.
*
* @update gess 3/25/98
* @param aParent -- tag type of parent
* @param aChild -- tag type of child
* @return PR_TRUE if given tag can contain other tags
*/
virtual PRBool CanOmitEndTag(eHTMLTags aParent,eHTMLTags aChild)const;
/**
* This method gets called when a start token has been consumed and needs
* to be handled (possibly added to content model via sink).
@ -181,15 +170,6 @@ class COtherDTD : public CNavDTD {
*/
nsresult HandleCommentToken(CToken* aToken);
/**
* This method gets called when a skipped-content token has been consumed and needs
* to be handled (possibly added to content model via sink).
* @update gess5/11/98
* @param aToken is the skipped-content token to be handled
* @return TRUE if the token was handled.
*/
nsresult HandleSkippedContentToken(CToken* aToken);
/**
* This method gets called when an attribute token has been consumed and needs
* to be handled (possibly added to content model via sink).

79
htmlparser/src/nsDTDUtils.cpp
Просмотреть файл

@ -19,9 +19,9 @@
#include "nsDTDUtils.h"
#include "CNavDTD.h"
#include "nsIObserverService.h"
#include "nsIServiceManager.h"
#include "nsIElementObserver.h"
static NS_DEFINE_IID(kIObserverServiceIID, NS_IOBSERVERSERVICE_IID);
static NS_DEFINE_IID(kObserverServiceCID, NS_OBSERVERSERVICE_CID);
@ -51,6 +51,8 @@ nsEntryStack::nsEntryStack() {
nsEntryStack::~nsEntryStack() {
if(mEntries)
delete [] mEntries;
mCount=mCapacity=0;
mEntries=0;
}
/**
@ -182,7 +184,6 @@ PRInt32 nsEntryStack::GetTopmostIndexOf(eHTMLTags aTag) const {
/**
*
* @update gess9/10/98
* @update gess 4/26/99
*/
nsDTDContext::nsDTDContext() : mStack(), mSkipped(0), mStyles(0) {
#ifdef NS_DEBUG
@ -194,7 +195,6 @@ nsDTDContext::nsDTDContext() : mStack(), mSkipped(0), mStyles(0) {
/**
*
* @update gess9/10/98
* @update gess 4/26/99
*/
nsDTDContext::~nsDTDContext() {
}
@ -202,16 +202,14 @@ nsDTDContext::~nsDTDContext() {
/**
*
* @update gess7/9/98, harishd 04/04/99
* @update gess 4/26/99
*/
PRInt32 nsDTDContext::GetCount(void) {
return mStack.GetSize();
return mStack.GetCount();
}
/**
*
* @update gess7/9/98, harishd 04/04/99
* @update gess 4/26/99
*/
void nsDTDContext::Push(eHTMLTags aTag) {
#ifdef NS_DEBUG
@ -224,7 +222,6 @@ void nsDTDContext::Push(eHTMLTags aTag) {
/**
* @update gess7/9/98, harishd 04/04/99
* @update gess 4/26/99
*/
eHTMLTags nsDTDContext::Pop() {
#ifdef NS_DEBUG
@ -232,6 +229,14 @@ eHTMLTags nsDTDContext::Pop() {
mTags[mStack.mCount-1]=eHTMLTag_unknown;
#endif
nsEntryStack* theStyles=0;
nsTagEntry& theEntry=mStack.EntryAt(mStack.mCount-1);
PRInt32 theIndex=theEntry.mStyleIndex;
if(-1<theIndex){
theStyles=(nsEntryStack*)mStyles.ObjectAt(theIndex);
delete theStyles;
}
eHTMLTags result=mStack.Pop();
return result;
}
@ -239,7 +244,6 @@ eHTMLTags nsDTDContext::Pop() {
/**
*
* @update gess7/9/98
* @update gess 4/26/99
*/
eHTMLTags nsDTDContext::First() const {
return mStack.First();
@ -248,7 +252,6 @@ eHTMLTags nsDTDContext::First() const {
/**
*
* @update gess7/9/98
* @update gess 4/26/99
*/
eHTMLTags nsDTDContext::TagAt(PRInt32 anIndex) const {
return mStack.TagAt(anIndex);
@ -258,7 +261,6 @@ eHTMLTags nsDTDContext::TagAt(PRInt32 anIndex) const {
/**
*
* @update gess7/9/98
* @update gess 4/26/99
*/
eHTMLTags nsDTDContext::operator[](PRInt32 anIndex) const {
return mStack[anIndex];
@ -267,7 +269,6 @@ eHTMLTags nsDTDContext::operator[](PRInt32 anIndex) const {
/**
*
* @update gess7/9/98
* @update gess 4/26/99
*/
eHTMLTags nsDTDContext::Last() const {
return mStack.Last();
@ -276,12 +277,12 @@ eHTMLTags nsDTDContext::Last() const {
/**
*
* @update gess7/9/98
* @update gess 4/26/99
*/
nsEntryStack* nsDTDContext::GetStyles(void) const {
nsEntryStack* nsDTDContext::GetStylesAt(PRUint32 anIndex) const {
nsEntryStack* result=0;
if(0<mStack.mCount){
PRInt32 theIndex=mStack.mEntries[mStack.mCount-1].mStyleIndex;
if(anIndex<mStack.mCount){
nsTagEntry& theEntry=mStack.EntryAt(anIndex);
PRInt32 theIndex=theEntry.mStyleIndex;
if(-1<theIndex){
result=(nsEntryStack*)mStyles.ObjectAt(theIndex);
}
@ -289,6 +290,46 @@ nsEntryStack* nsDTDContext::GetStyles(void) const {
return result;
}
/**
*
* @update gess 04/28/99
*/
void nsDTDContext::PushStyle(eHTMLTags aTag){
nsTagEntry& theEntry=mStack.EntryAt(mStack.mCount-1);
//ok, now go get the right tokenbank deque...
nsEntryStack* theStack=0;
if(-1<theEntry.mStyleIndex)
theStack=(nsEntryStack*)mStyles.ObjectAt(theEntry.mStyleIndex);
if(!theStack){
//time to make a databank for this element...
theStack=new nsEntryStack();
mStyles.Push(theStack);
theEntry.mStyleIndex=mStyles.GetSize()-1;
}
if(theStack){
theStack->Push(aTag);
}
}
/**
*
* @update gess 04/28/99
*/
eHTMLTags nsDTDContext::PopStyle(void){
eHTMLTags result=eHTMLTag_unknown;
nsTagEntry& theEntry=mStack.EntryAt(mStack.mCount-1);
//ok, now go get the right tokenbank deque...
nsEntryStack* theStack=0;
if(-1<theEntry.mStyleIndex)
theStack=(nsEntryStack*)mStyles.ObjectAt(theEntry.mStyleIndex);
if(theStack){
result=theStack->Pop();
}
return result;
}
/**
*
* @update harishd 04/04/99
@ -296,7 +337,7 @@ nsEntryStack* nsDTDContext::GetStyles(void) const {
*/
void nsDTDContext::SaveToken(CToken* aToken, PRInt32 aID)
{
NS_PRECONDITION(aID <= mStack.GetSize() && aID > -1,"Out of bounds");
NS_PRECONDITION(aID <= mStack.GetCount() && aID > -1,"Out of bounds");
if(aToken) {
nsTagEntry& theEntry=mStack.EntryAt(aID);
@ -321,9 +362,9 @@ void nsDTDContext::SaveToken(CToken* aToken, PRInt32 aID)
*/
CToken* nsDTDContext::RestoreTokenFrom(PRInt32 aID)
{
NS_PRECONDITION(aID <= mStack.GetSize() && aID > -1,"Out of bounds");
NS_PRECONDITION(aID <= mStack.GetCount() && aID > -1,"Out of bounds");
CToken* result=0;
if(0<mStack.GetSize()) {
if(0<mStack.GetCount()) {
nsTagEntry theEntry=mStack.EntryAt(aID);
nsDeque* theDeque=(nsDeque*)mSkipped.ObjectAt(theEntry.mBankIndex);
if(theDeque){
@ -340,7 +381,7 @@ CToken* nsDTDContext::RestoreTokenFrom(PRInt32 aID)
*/
PRInt32 nsDTDContext::TokenCountAt(PRInt32 aID)
{
NS_PRECONDITION(aID <= mStack.GetSize(),"Out of bounds");
NS_PRECONDITION(aID <= mStack.GetCount(),"Out of bounds");
nsTagEntry theEntry=mStack.EntryAt(aID);
nsDeque* theDeque=(nsDeque*)mSkipped.ObjectAt(theEntry.mBankIndex);

7
htmlparser/src/nsDTDUtils.h
Просмотреть файл

@ -37,6 +37,7 @@
#include <fstream.h>
#include "nsITokenizer.h"
#include "nsString.h"
#include "nsIElementObserver.h"
/***************************************************************
Before digging into the NavDTD, we'll define a helper
@ -79,7 +80,7 @@ public:
eHTMLTags operator[](PRUint32 anIndex) const;
eHTMLTags Last() const;
void Empty(void);
PRInt32 GetSize(void) const {return mCount;}
PRInt32 GetCount(void) const {return mCount;}
nsTagEntry* mEntries;
PRUint32 mCount;
@ -106,7 +107,9 @@ public:
eHTMLTags Last() const;
void Empty(void);
PRInt32 GetCount(void);
nsEntryStack* GetStyles(void) const;
nsEntryStack* GetStylesAt(PRUint32 anIndex) const;
void PushStyle(eHTMLTags aTag);
eHTMLTags PopStyle(void);
void SaveToken(CToken* aToken, PRInt32 aID);
CToken* RestoreTokenFrom(PRInt32 aID);

825
htmlparser/src/nsElementTable.cpp
Просмотреть файл

Разница между файлами не показана из-за своего большого размера Загрузить разницу

61
htmlparser/src/nsElementTable.h
Просмотреть файл

@ -46,13 +46,14 @@ public:
mTagList=aTagList;
}
PRInt32 GetTopmostIndexOf(nsEntryStack& aTagStack);
PRInt32 GetBottommostIndexOf(nsEntryStack& aTagStack,PRInt32 aStartOffset);
PRBool Contains(eHTMLTags aTag);
PRInt32 GetTopmostIndexOf(nsEntryStack& aTagStack);
PRInt32 GetBottommostIndexOf(nsEntryStack& aTagStack,PRInt32 aStartOffset);
PRBool Contains(eHTMLTags aTag);
eHTMLTags GetTagAt(PRUint32 anIndex) const;
eHTMLTags mTags[5];
eHTMLTags* mTagList;
int mCount;
PRUint32 mCount;
};
//*********************************************************************************************
@ -68,11 +69,13 @@ public:
*/
struct nsHTMLElement {
static PRBool IsContainerType(eHTMLTags aTag,int aType);
static void DebugDumpMembership(const char* aFilename);
static void DebugDumpContainment(const char* aFilename,const char* aTitle);
static void DebugDumpContainType(const char* aFilename);
static PRBool IsBlockElement(eHTMLTags aTag);
static PRBool IsInlineElement(eHTMLTags aTag);
static PRBool IsFlowElement(eHTMLTags aTag);
static PRBool IsBlockEntity(eHTMLTags aTag);
static PRBool IsInlineEntity(eHTMLTags aTag);
static PRBool IsFlowEntity(eHTMLTags aTag);
static PRBool IsBlockCloser(eHTMLTags aTag);
CTagList* GetRootTags(void) const {return mRootNodes;}
@ -80,6 +83,7 @@ struct nsHTMLElement {
CTagList* GetAutoCloseStartTags(void) const {return mAutocloseStart;}
CTagList* GetAutoCloseEndTags(void) const {return mAutocloseEnd;}
CTagList* GetSynonymousTags(void) const {return mSynonymousTags;}
eHTMLTags GetCloseTargetForEndTag(nsEntryStack& aTagStack,PRInt32 anIndex) const;
CTagList* GetSpecialChildren(void) const {return mSpecialKids;}
CTagList* GetSpecialParents(void) const {return mSpecialParents;}
@ -89,9 +93,10 @@ struct nsHTMLElement {
eHTMLTags GetTag(void) const {return mTagID;}
PRBool CanContain(eHTMLTags aChild) const;
PRBool CanExclude(eHTMLTags aChild) const;
PRBool CanOmitStartTag(eHTMLTags aChild) const;
PRBool CanOmitEndTag(eHTMLTags aParent) const;
PRBool CanContainSelf() const;
PRBool CanOmitEndTag(void) const;
PRBool CanContainSelf(void) const;
PRBool HasSpecialProperty(PRInt32 aProperty) const;
PRBool SectionContains(eHTMLTags aTag,PRBool allowDepthSearch);
@ -120,6 +125,7 @@ struct nsHTMLElement {
int mInclusionBits; //defines parental and containment rules
int mExclusionBits; //defines things you CANNOT contain
int mSpecialProperties; //used for various special purposes...
int mPropagateRange; //tells us how far a parent is willing to prop. badly formed children
CTagList* mSpecialParents; //These are the special tags that contain this tag (directly)
CTagList* mSpecialKids; //These are the extra things you can contain
eHTMLTags mSkipTarget; //If set, then we skip all content until this tag is seen
@ -134,7 +140,10 @@ static const int kDiscardTag = 0x0001; //tells us to toss this tag
static const int kOmitEndTag = 0x0002; //safely ignore end tag
static const int kLegalOpen = 0x0004; //Lets BODY, TITLE, SCRIPT to reopen
static const int kOmitWS = 0x0008; //If set, the tag can omit all ws and newlines
static const int kBadContentWatch = 0x0014; //Used in handling illegal contents
static const int kNoPropagate = 0x0010; //If set, this tag won't propagate as a child
static const int kBadContentWatch = 0x0020;
static const int kNoStyleLeaksIn = 0x0040;
static const int kNoStyleLeaksOut = 0x0080;
//*********************************************************************************************
// The following ints define the standard groups of HTML elements...
@ -150,21 +159,23 @@ static const int kSpecial = 0x0008; // A, IMG, APPLET, OBJECT, FONT,
// MAP, Q, SUB, SUP, SPAN, BDO, IFRAME
static const int kFormControl = 0x0010; // INPUT SELECT TEXTAREA LABEL BUTTON
static const int kPreformatted = 0x0011; // PRE
static const int kPreExclusion = 0x0012; // IMG, OBJECT, APPLET, BIG, SMALL, SUB, SUP, FONT, BASEFONT
static const int kFontStyle = 0x0014; // TT, I, B, U, S, STRIKE, BIG, SMALL
static const int kPhrase = 0x0018; // EM, STRONG, DFN, CODE, SAMP, KBD, VAR, CITE, ABBR, ACRONYM
static const int kHeading = 0x0020; // H1..H6
static const int kBlockMisc = 0x0021; // P, DL, DIV, CENTER, NOSCRIPT, NOFRAMES, BLOCKQUOTE
// FORM, ISINDEX, HR, TABLE, FIELDSET, ADDRESS
static const int kPreformatted = 0x0020; // PRE
static const int kPreExclusion = 0x0040; // IMG, OBJECT, APPLET, BIG, SMALL, SUB, SUP, FONT, BASEFONT
static const int kFontStyle = 0x0080; // TT, I, B, U, S, STRIKE, BIG, SMALL, BLINK
static const int kPhrase = 0x0100; // EM, STRONG, DFN, CODE, SAMP, KBD, VAR, CITE, ABBR, ACRONYM
static const int kHeading = 0x0200; // H1..H6
static const int kBlockMisc = 0x0400; // OBJECT, SCRIPT
static const int kBlock = 0x0800; // ADDRESS, BLOCKQUOTE, CENTER, DIV, DL, FIELDSET, FORM,
// ISINDEX, HR, NOSCRIPT, NOFRAMES, P, TABLE
static const int kList = 0x1000; // UL, OL, DIR, MENU
static const int kPCDATA = 0x2000; // just plain text...
static const int kSelf = 0x4000; // whatever THIS tag is...
static const int kExtensions = 0x8000; // BGSOUND, WBR, NOBR
static const int kTable = 0x10000;// TR,TD,THEAD,TBODY,TFOOT,CAPTION,TH
static const int kList = 0x0024; // UL, OL, DIR, MENU
static const int kPCDATA = 0x0028; // just plain text...
static const int kSelf = 0x0040; // whatever THIS tag is...
static const int kInline = (kPCDATA|kFontStyle|kPhrase|kSpecial|kFormControl); // #PCDATA, %fontstyle, %phrase, %special, %formctrl
static const int kBlock = (kHeading|kList|kPreformatted|kBlockMisc); // %heading, %list, %preformatted, %blockmisc
static const int kFlow = (kBlock|kInline); // %block, %inline
static const int kInlineEntity = (kPCDATA|kFontStyle|kPhrase|kSpecial|kFormControl|kExtensions); // #PCDATA, %fontstyle, %phrase, %special, %formctrl
static const int kBlockEntity = (kHeading|kList|kPreformatted|kBlock); // %heading, %list, %preformatted, %blockmisc
static const int kFlowEntity = (kBlockEntity|kInlineEntity); // %block, %inline
#endif

84
htmlparser/src/nsHTMLTokenizer.cpp
Просмотреть файл

@ -101,7 +101,7 @@ NS_IMPL_RELEASE(nsHTMLTokenizer)
* @param
* @return
*/
nsHTMLTokenizer::nsHTMLTokenizer() : nsITokenizer(), mTokenDeque(new CTokenDeallocator()){
nsHTMLTokenizer::nsHTMLTokenizer() : nsITokenizer(), mTokenDeque(0){
NS_INIT_REFCNT();
mDoXMLEmptyTags=PR_FALSE;
}
@ -114,6 +114,10 @@ nsHTMLTokenizer::nsHTMLTokenizer() : nsITokenizer(), mTokenDeque(new CTokenDeall
* @return
*/
nsHTMLTokenizer::~nsHTMLTokenizer(){
if(mTokenDeque.GetSize()){
CTokenDeallocator theDeallocator;
mTokenDeque.ForEach(theDeallocator);
}
}
@ -277,10 +281,10 @@ nsresult nsHTMLTokenizer::ConsumeToken(nsScanner& aScanner) {
case kNotFound:
break;
case 0: //preceeds a EOF...
break;
default:
if(!nsString::IsSpace(theChar)) {
nsAutoString temp(theChar);
@ -411,66 +415,15 @@ nsresult nsHTMLTokenizer::ConsumeAttributes(PRUnichar aChar,CStartToken* aToken,
}
/**
* This is a special case method. It's job is to consume
* all of the given tag up to an including the end tag.
*
* @param aChar: last char read
* @param aScanner: see nsScanner.h
* @param anErrorCode: arg that will hold error condition
* @return new token or null
*/
nsresult nsHTMLTokenizer::ConsumeContentToEndTag(PRUnichar aChar,
eHTMLTags aChildTag,
nsScanner& aScanner,
CToken*& aToken){
//In the case that we just read the given tag, we should go and
//consume all the input until we find a matching end tag.
nsAutoString endTag("</");
endTag.Append(NS_EnumToTag(aChildTag));
endTag.Append(">");
CTokenRecycler* theRecycler=(CTokenRecycler*)GetTokenRecycler();
aToken=theRecycler->CreateTokenOfType(eToken_skippedcontent,aChildTag,endTag);
return aToken->Consume(aChar,aScanner); //tell new token to finish consuming text...
}
/**
*
* In the case that we just read the given tag, we should go and
* consume all the input until we find a matching end tag.
* @update gess12/28/98
* @param
* @return
*/
nsresult nsHTMLTokenizer::HandleSkippedContent(nsScanner& aScanner,CToken*& aToken) {
nsresult nsHTMLTokenizer::ConsumeScriptContent(nsScanner& aScanner,CToken*& aToken) {
nsresult result=NS_OK;
eHTMLTags theTag=(eHTMLTags)aToken->GetTypeID();
if(eHTMLTag_unknown!=gHTMLElements[theTag].mSkipTarget) {
//Do special case handling for <script>, <style>, <title> or <textarea>...
CToken* skippedToken=0;
PRUnichar theChar=0;
result=ConsumeContentToEndTag(theChar,gHTMLElements[theTag].mSkipTarget,aScanner,skippedToken);
CTokenRecycler* theRecycler=(CTokenRecycler*)GetTokenRecycler();
AddToken(skippedToken,result,mTokenDeque,theRecycler);
if(NS_SUCCEEDED(result) && skippedToken){
//In the case that we just read a given tag, we should go and
//consume all the tag content itself (and throw it all away).
nsString& theTagStr=skippedToken->GetStringValueXXX();
CToken* endtoken=theRecycler->CreateTokenOfType(eToken_end,theTag,theTagStr);
if(endtoken){
nsAutoString temp;
theTagStr.Mid(temp,2,theTagStr.Length()-3);
//now strip the leading and trailing delimiters...
endtoken->Reinitialize(theTag,temp);
AddToken(endtoken,result,mTokenDeque,theRecycler);
}
} //if
} //if
return result;
}
@ -498,11 +451,18 @@ nsresult nsHTMLTokenizer::ConsumeStartTag(PRUnichar aChar,CToken*& aToken,nsScan
result=ConsumeAttributes(aChar,(CStartToken*)aToken,aScanner);
}
//now that that's over with, we have one more problem to solve.
//In the case that we just read a <SCRIPT> or <STYLE> tags, we should go and
//consume all the content itself.
if(NS_SUCCEEDED(result)) {
result=HandleSkippedContent(aScanner,aToken);
/* Now that that's over with, we have one more problem to solve.
In the case that we just read a <SCRIPT> or <STYLE> tags, we should go and
consume all the content itself.
*/
if(NS_SUCCEEDED(result) && (eHTMLTag_script==theTag)) {
nsAutoString endTag("</script>");
CTokenRecycler* theRecycler=(CTokenRecycler*)GetTokenRecycler();
CToken* textToken=theRecycler->CreateTokenOfType(eToken_text,theTag,endTag);
result=((CTextToken*)textToken)->ConsumeUntil(0,aScanner,endTag); //tell new token to finish consuming text...
AddToken(textToken,result,mTokenDeque,theRecycler);
CToken* endToken=theRecycler->CreateTokenOfType(eToken_end,theTag);
AddToken(endToken,result,mTokenDeque,theRecycler);
}
//EEEEECCCCKKKK!!!

3
htmlparser/src/nsHTMLTokenizer.h
Просмотреть файл

@ -69,7 +69,7 @@ public:
protected:
virtual nsresult HandleSkippedContent(nsScanner& aScanner,CToken*& aToken);
virtual nsresult ConsumeScriptContent(nsScanner& aScanner,CToken*& aToken);
virtual nsresult ConsumeTag(PRUnichar aChar,CToken*& aToken,nsScanner& aScanner);
virtual nsresult ConsumeStartTag(PRUnichar aChar,CToken*& aToken,nsScanner& aScanner);
virtual nsresult ConsumeEndTag(PRUnichar aChar,CToken*& aToken,nsScanner& aScanner);
@ -79,7 +79,6 @@ protected:
virtual nsresult ConsumeComment(PRUnichar aChar,CToken*& aToken,nsScanner& aScanner);
virtual nsresult ConsumeNewline(PRUnichar aChar,CToken*& aToken,nsScanner& aScanner);
virtual nsresult ConsumeText(const nsString& aString,CToken*& aToken,nsScanner& aScanner);
virtual nsresult ConsumeContentToEndTag(PRUnichar aChar,eHTMLTags aChildTag,nsScanner& aScanner,CToken*& aToken);
virtual nsresult ConsumeProcessingInstruction(PRUnichar aChar,CToken*& aToken,nsScanner& aScanner);
static void AddToken(CToken*& aToken,nsresult aResult,nsDeque& aDeque,CTokenRecycler* aRecycler);

160
htmlparser/src/nsHTMLTokens.cpp
Просмотреть файл

@ -29,7 +29,6 @@
#include "nsHTMLTags.h"
#include "nsHTMLEntities.h"
#include "nsCRT.h"
#include "nsStr.h"
//#define GESS_MACHINE
#ifdef GESS_MACHINE
@ -79,7 +78,6 @@ void CHTMLToken::SetStringValue(const char* name){
}
}
/*
* constructor from tag id
*
@ -272,6 +270,19 @@ void CStartToken::DebugDumpSource(ostream& out) {
out << ">";
}
/*
*
*
* @update gess 3/25/98
* @param anOutputString will recieve the result
* @return nada
*/
void CStartToken::GetSource(nsString& anOutputString){
anOutputString="<";
anOutputString+=mTextValue;
if(!mAttributed)
anOutputString+=">";
}
/*
* constructor from tag id
@ -303,7 +314,6 @@ CEndToken::CEndToken(const nsString& aName) : CHTMLToken(aName) {
* @return error result
*/
nsresult CEndToken::Consume(PRUnichar aChar, nsScanner& aScanner) {
//if you're here, we've already Consumed the <! chars, and are
//ready to Consume the rest of the open tag identifier.
//Stop consuming as soon as you see a space or a '>'.
@ -388,6 +398,19 @@ void CEndToken::DebugDumpSource(ostream& out) {
out << "</" << buffer << ">";
}
/*
*
*
* @update gess 3/25/98
* @param anOutputString will recieve the result
* @return nada
*/
void CEndToken::GetSource(nsString& anOutputString){
anOutputString="</";
anOutputString+=mTextValue;
anOutputString+=">";
}
/*
* default constructor
*
@ -474,6 +497,70 @@ nsresult CTextToken::Consume(PRUnichar aChar, nsScanner& aScanner) {
return result;
}
/*
* Consume as much clear text from scanner as possible.
*
* @update gess 3/25/98
* @param aChar -- last char consumed from stream
* @param aScanner -- controller of underlying input source
* @return error result
*/
nsresult CTextToken::ConsumeUntil(PRUnichar aChar,nsScanner& aScanner,nsString& aTerminalString){
PRBool done=PR_FALSE;
nsresult result=NS_OK;
nsString temp;
PRUnichar theChar;
//We're going to try a new algorithm here. Rather than scan for the matching
//end tag like we used to do, we're now going to scan for whitespace and comments.
//If we find either, just eat them. If we find text or a tag, then go to the
//target endtag, or the start of another comment.
static nsAutoString theWhitespace2("\b\t ");
while((!done) && (NS_OK==result)) {
result=aScanner.GetChar(aChar);
if((NS_OK==result) && (kLessThan==aChar)) {
//we're reading a tag or a comment...
result=aScanner.GetChar(theChar);
if((NS_OK==result) && (kExclamation==theChar)) {
//read a comment...
static CCommentToken theComment;
result=theComment.Consume(aChar,aScanner);
if(NS_OK==result) {
//result=aScanner.SkipWhitespace();
//temp.Append("<!");
temp.Append(theComment.GetStringValueXXX());
//temp.Append(">");
}
} else {
//read a tag...
temp+=aChar;
temp+=theChar;
result=aScanner.ReadUntil(temp,kGreaterThan,PR_TRUE);
}
}
else if(0<=theWhitespace2.BinarySearch(aChar)) {
static CWhitespaceToken theWS;
result=theWS.Consume(aChar,aScanner);
if(NS_OK==result) {
temp.Append(theWS.GetStringValueXXX());
}
}
else {
temp+=aChar;
result=aScanner.ReadUntil(temp,kLessThan,PR_FALSE);
}
nsAutoString theRight;
temp.Right(theRight,aTerminalString.Length());
done=PRBool(0==theRight.Compare(aTerminalString,PR_TRUE));
} //while
int len=temp.Length();
temp.Truncate(len-aTerminalString.Length());
mTextValue=temp;
return result;
}
/*
* default constructor
*
@ -759,14 +846,16 @@ nsresult CCommentToken::Consume(PRUnichar aChar, nsScanner& aScanner) {
PRBool theStrictForm=PR_FALSE;
nsresult result=(theStrictForm) ? ConsumeStrictComment(aChar,aScanner,mTextValue) : ConsumeComment(aChar,aScanner,mTextValue);
/*
//this change is here to make the editor teams' life easier.
//I'm removing the leading and trailing markup...
/*
if(0==mTextValue.Find("<!"))
mTextValue.Cut(0,2); //trim off 1st 2 chars...
if(kGreaterThan==mTextValue.Last())
mTextValue.Truncate(mTextValue.Length()-1); //trim off last char
*/
return result;
}
@ -860,21 +949,21 @@ nsresult CNewlineToken::Consume(PRUnichar aChar, nsScanner& aScanner) {
mTextValue=aChar;
//we already read the \r or \n, let's see what's next!
PRUnichar nextChar;
nsresult result=aScanner.Peek(nextChar);
PRUnichar theChar;
nsresult result=aScanner.Peek(theChar);
if(NS_OK==result) {
switch(aChar) {
case kNewLine:
if(kCR==nextChar) {
result=aScanner.GetChar(nextChar);
mTextValue+=nextChar;
if(kCR==theChar) {
result=aScanner.GetChar(theChar);
mTextValue+=theChar;
}
break;
case kCR:
if(kNewLine==nextChar) {
result=aScanner.GetChar(nextChar);
mTextValue+=nextChar;
if(kNewLine==theChar) {
result=aScanner.GetChar(theChar);
mTextValue+=theChar;
}
break;
default:
@ -968,6 +1057,19 @@ void CAttributeToken::DebugDumpToken(ostream& out) {
out << buffer << ": " << mTypeID << endl;
}
/*
*
*
* @update gess 3/25/98
* @param anOutputString will recieve the result
* @return nada
*/
void CAttributeToken::GetSource(nsString& anOutputString){
anOutputString=mTextKey;
anOutputString+="=";
anOutputString+=mTextValue;
anOutputString+=";";
}
/*
* This general purpose method is used when you want to
@ -997,8 +1099,8 @@ nsresult ConsumeQuotedString(PRUnichar aChar,nsString& aString,nsScanner& aScann
PRUnichar ch=aString.Last();
if(ch!=aChar)
aString+=aChar;
// aString.ReplaceChar(PRUnichar('\n'),PRUnichar(' '));
aString.StripChars("\r"); //per the HTML spec, ignore linefeeds...
//aString.ReplaceChar(PRUnichar('\n'),PRUnichar(' '));
aString.StripChars("\r\n"); //per the HTML spec, ignore linefeeds...
return result;
}
@ -1270,6 +1372,7 @@ const char* CEntityToken::GetClassName(void) {
return "&entity";
}
/*
*
*
@ -1439,6 +1542,19 @@ void CEntityToken::DebugDumpSource(ostream& out) {
delete[] cp;
}
/*
*
*
* @update gess 3/25/98
* @param anOutputString will recieve the result
* @return nada
*/
void CEntityToken::GetSource(nsString& anOutputString){
anOutputString="&";
anOutputString+=anOutputString;
anOutputString+=";";
}
/*
* default constructor
*
@ -1591,9 +1707,9 @@ nsresult CSkippedContentToken::Consume(PRUnichar aChar,nsScanner& aScanner) {
temp+=aChar;
result=aScanner.ReadUntil(temp,kLessThan,PR_FALSE);
}
nsAutoString temp2;
temp.Right(temp2,mTextValue.Length());
done=PRBool(0==temp2.Compare(mTextValue.GetUnicode(),PR_TRUE,mTextValue.Length()));
nsAutoString theRight;
temp.Right(theRight,mTextValue.Length());
done=PRBool(0==theRight.Compare(mTextValue,PR_TRUE));
}
int len=temp.Length();
temp.Truncate(len-mTextValue.Length());
@ -1616,6 +1732,16 @@ void CSkippedContentToken::DebugDumpSource(ostream& out) {
out<<">";
}
/*
*
*
* @update gess 3/25/98
* @param anOutputString will recieve the result
* @return nada
*/
void CSkippedContentToken::GetSource(nsString& anOutputString){
anOutputString="$skipped-content";
}
/**
*

12
htmlparser/src/nsHTMLTokens.h
Просмотреть файл

@ -43,6 +43,7 @@
#include "nsHTMLTags.h"
#include "nsParserError.h"
#include <iostream.h>
#include "nsString.h"
class nsScanner;
@ -121,6 +122,8 @@ class CStartToken: public CHTMLToken {
PRBool IsEmpty(void);
void SetEmpty(PRBool aValue);
virtual void DebugDumpSource(ostream& out);
virtual void GetSource(nsString& anOutputString);
virtual void Reinitialize(PRInt32 aTag, const nsString& aString);
protected:
@ -145,6 +148,7 @@ class CEndToken: public CHTMLToken {
virtual const char* GetClassName(void);
virtual PRInt32 GetTokenType(void);
virtual void DebugDumpSource(ostream& out);
virtual void GetSource(nsString& anOutputString);
};
@ -184,11 +188,8 @@ class CEntityToken : public CHTMLToken {
virtual nsresult Consume(PRUnichar aChar,nsScanner& aScanner);
static PRInt32 ConsumeEntity(PRUnichar aChar,nsString& aString,nsScanner& aScanner);
static PRInt32 TranslateToUnicodeStr(PRInt32 aValue,nsString& aString);
// static PRInt32 FindEntityIndex(nsString& aString);
// static PRInt32 FindEntityIndexMax(const char* aBuffer,PRInt32 aCount=-1);
// static PRBool VerifyEntityTable(void);
// static PRInt32 ReduceEntities(nsString& aString);
virtual void DebugDumpSource(ostream& out);
virtual void GetSource(nsString& anOutputString);
};
@ -220,6 +221,7 @@ class CTextToken: public CHTMLToken {
CTextToken();
CTextToken(const nsString& aString);
virtual nsresult Consume(PRUnichar aChar,nsScanner& aScanner);
nsresult ConsumeUntil(PRUnichar aChar,nsScanner& aScanner,nsString& aTerminalString);
virtual const char* GetClassName(void);
virtual PRInt32 GetTokenType(void);
};
@ -261,6 +263,7 @@ class CAttributeToken: public CHTMLToken {
virtual PRInt32 GetTokenType(void);
virtual nsString& GetKey(void) {return mTextKey;}
virtual void DebugDumpToken(ostream& out);
virtual void GetSource(nsString& anOutputString);
virtual void DebugDumpSource(ostream& out);
PRBool mLastAttribute;
virtual void Reinitialize(PRInt32 aTag, const nsString& aString);
@ -337,6 +340,7 @@ class CSkippedContentToken: public CAttributeToken {
virtual const char* GetClassName(void);
virtual PRInt32 GetTokenType(void);
virtual void DebugDumpSource(ostream& out);
virtual void GetSource(nsString& anOutputString);
protected:
};

1
htmlparser/src/nsIParser.h
Просмотреть файл

@ -186,6 +186,7 @@ class nsIParser : public nsISupports {
* @return ptr to scanner
*/
virtual eParseMode GetParseMode(void)=0;
};
/* ===========================================================*

67
htmlparser/src/nsParser.cpp
Просмотреть файл

@ -47,16 +47,8 @@ static const char* kNullURL = "Error: Null URL given";
static const char* kOnStartNotCalled = "Error: OnStartBinding() must be called before OnDataAvailable()";
static const char* kBadListenerInit = "Error: Parser's IStreamListener API was not setup correctly in constructor.";
//-------------------------------------------------------------------
class CTokenDeallocator: public nsDequeFunctor{
public:
virtual void* operator()(void* anObject) {
CToken* aToken = (CToken*)anObject;
delete aToken;
return 0;
}
};
class CDTDDeallocator: public nsDequeFunctor{
public:
@ -67,6 +59,8 @@ public:
}
};
//-------------------------------------------------------------------
class CDTDFinder: public nsDequeFunctor{
public:
CDTDFinder(nsIDTD* aDTD) {
@ -83,6 +77,8 @@ public:
nsIDTD* mTargetDTD;
};
//-------------------------------------------------------------------
class CSharedParserObjects {
public:
@ -128,14 +124,6 @@ public:
nsString nsParser::gHackMetaCharset = "";
nsString nsParser::gHackMetaCharsetURL = "";
//-------------------------------------------------------------------------
CSharedParserObjects& GetSharedObjects() {
static CSharedParserObjects gSharedParserObjects;
return gSharedParserObjects;
}
/**********************************************************************************
This class is used as an interface between an external agent (like the DOM) and
the parser. It will contain a stack full of tagnames, which is used in our
@ -173,8 +161,13 @@ public:
nsDeque mTags; //will hold a deque of prunichars...
};
//-------------------------------------------------------------------
//-------------------------------------------------------------------------
CSharedParserObjects& GetSharedObjects() {
static CSharedParserObjects gSharedParserObjects;
return gSharedParserObjects;
}
/**
* default constructor
@ -429,18 +422,28 @@ eParseMode DetermineParseMode(nsParser& aParser) {
nsScanner* theScanner=aParser.GetScanner();
if(theScanner){
nsString theBufCopy;
nsString& theBuffer=theScanner->GetBuffer();
PRInt32 theIndex=theBuffer.Find("HTML 4.0");
if(kNotFound==theIndex)
theIndex=theBuffer.Find("html 4.0");
if(kNotFound<theIndex)
return eParseMode_raptor;
else {
PRInt32 theIndex=theBuffer.Find("noquirks");
if(kNotFound==theIndex)
theIndex=theBuffer.Find("NOQUIRKS");
if(kNotFound<theIndex)
return eParseMode_noquirks;
theBuffer.Left(theBufCopy,300);
theBufCopy.ToUpperCase();
PRInt32 theIndex=theBufCopy.Find("<!DOCTYPE");
if(kNotFound==theIndex){
theIndex=theBufCopy.Find("<DOCTYPE");
}
if(kNotFound<theIndex) {
//good, we found "DOCTYPE" -- now go find it's end delimiter '>'
PRInt32 theSubIndex=theBufCopy.Find(kGreaterThan,theIndex+1);
theBufCopy.Truncate(theSubIndex);
theSubIndex=theBufCopy.Find("HTML 4.0");
if(kNotFound<theSubIndex) {
return eParseMode_raptor;
}
}
theIndex=theBufCopy.Find("NOQUIRKS");
if(kNotFound<theIndex) {
return eParseMode_noquirks;
}
}
@ -712,7 +715,7 @@ nsresult nsParser::Parse(nsIInputStream& aStream,PRBool aVerifyEnabled, void* aK
* @param aContentType tells us what type of content to expect in the given string
* @return error code -- 0 if ok, non-zero if error.
*/
nsresult nsParser::Parse(nsString& aSourceBuffer,void* aKey,const nsString& aContentType,PRBool aEnableVerify,PRBool aLastCall){
nsresult nsParser::Parse(nsString& aSourceBuffer,void* aKey,const nsString& aContentType,PRBool aVerifyEnabled,PRBool aLastCall){
#ifdef _rickgdebug
{
@ -751,7 +754,7 @@ nsresult nsParser::Parse(nsString& aSourceBuffer,void* aKey,const nsString& aCon
// till we're completely done.
NS_ADDREF(me);
if(aSourceBuffer.Length() || mUnusedInput.Length()) {
mDTDVerification=aEnableVerify;
mDTDVerification=aVerifyEnabled;
CParserContext* pc=0;
if((!mParserContext) || (mParserContext->mKey!=aKey)) {
@ -1176,6 +1179,7 @@ void nsParser::DebugDumpSource(ostream& aStream) {
}
}
/**
* Call this to get a newly constructed tagstack
* @update gess 5/05/99
@ -1188,3 +1192,4 @@ nsresult nsParser::CreateTagStack(nsITagStack** aTagStack){
return NS_OK;
return NS_ERROR_HTMLPARSER_MEMORYFAILURE;
}

7
htmlparser/src/nsParser.h
Просмотреть файл

@ -69,7 +69,7 @@ class nsIHTMLContentSink;
class nsIDTD;
class nsScanner;
class nsIParserFilter;
class nsTagStack;
#include <fstream.h>
@ -165,7 +165,7 @@ friend class CTokenHandler;
* @param aListener is a listener to forward notifications to
* @return TRUE if all went well -- FALSE otherwise
*/
virtual nsresult Parse(nsIURL* aURL,nsIStreamObserver* aListener,PRBool aEnableVerify=PR_FALSE, void* aKey=0);
virtual nsresult Parse(nsIURL* aURL,nsIStreamObserver* aListener,PRBool aEnableVerify=PR_FALSE,void* aKey=0);
/**
* Cause parser to parse input from given stream
@ -173,7 +173,7 @@ friend class CTokenHandler;
* @param aStream is the i/o source
* @return TRUE if all went well -- FALSE otherwise
*/
virtual nsresult Parse(nsIInputStream& aStream,PRBool aEnableVerify=PR_FALSE, void* aKey=0);
virtual nsresult Parse(nsIInputStream& aStream,PRBool aEnableVerify=PR_FALSE,void* aKey=0);
/**
* @update gess5/11/98
@ -183,7 +183,6 @@ friend class CTokenHandler;
*/
virtual nsresult Parse(nsString& aSourceBuffer,void* aKey,const nsString& aContentType,PRBool aEnableVerify=PR_FALSE,PRBool aLastCall=PR_FALSE);
virtual PRBool IsValidFragment(const nsString& aSourceBuffer,nsITagStack& aStack,PRUint32 anInsertPos,const nsString& aContentType);
virtual nsresult ParseFragment(nsString& aSourceBuffer,void* aKey,nsITagStack& aStack,PRUint32 anInsertPos,const nsString& aContentType);

40
htmlparser/src/nsParserNode.cpp
Просмотреть файл

@ -40,13 +40,13 @@ nsAutoString& GetEmptyString() {
* @param aToken -- token to init internal token
* @return
*/
nsCParserNode::nsCParserNode(CToken* aToken,PRInt32 aLineNumber,nsITokenRecycler* aRecycler): nsIParserNode() {
nsCParserNode::nsCParserNode(CToken* aToken,PRInt32 aLineNumber,nsITokenRecycler* aRecycler):
nsIParserNode(), mSkippedContent("") {
NS_INIT_REFCNT();
mAttributeCount=0;
mLineNumber=aLineNumber;
mToken=aToken;
memset(mAttributes,0,sizeof(mAttributes));
mSkippedContent=0;
mRecycler=aRecycler;
}
@ -65,8 +65,6 @@ nsCParserNode::~nsCParserNode() {
for(index=0;index<mAttributeCount;index++){
mRecycler->RecycleToken(mAttributes[index]);
}
if(mSkippedContent)
mRecycler->RecycleToken(mSkippedContent);
}
}
@ -141,21 +139,6 @@ void nsCParserNode::AddAttribute(CToken* aToken) {
}
}
/**
* This method gets called when the parser encounters
* skipped content after a start token.
*
* @update gess 3/26/98
* @param aToken -- really a skippedcontent token
* @return nada
*/
void nsCParserNode::SetSkippedContent(CToken* aToken){
if(aToken) {
NS_ASSERTION(eToken_skippedcontent == aToken->GetTokenType(), "not a skipped content token");
mSkippedContent = aToken;
}
}
/**
* Gets the name of this node. Currently unused.
@ -191,10 +174,19 @@ const nsString& nsCParserNode::GetText() const {
* @return string ref of text from internal token
*/
const nsString& nsCParserNode::GetSkippedContent() const {
if (nsnull != mSkippedContent) {
return ((CSkippedContentToken*)mSkippedContent)->GetKey();
}
return GetEmptyString();
return mSkippedContent;
}
/**
* Get text value of this node, which translates into
* getting the text value of the underlying token
*
* @update gess 3/25/98
* @param
* @return string ref of text from internal token
*/
void nsCParserNode::SetSkippedContent(nsString& aString) {
mSkippedContent=aString;
}
/**
@ -303,5 +295,3 @@ CToken* nsCParserNode::PopAttributeToken()
}
return nsnull;
}

30
htmlparser/src/nsParserNode.h
Просмотреть файл

@ -90,6 +90,13 @@ class nsCParserNode : public nsIParserNode {
*/
virtual const nsString& GetSkippedContent() const;
/**
* Retrieve skipped context from node
* @update gess5/11/98
* @return string containing skipped content
*/
virtual void SetSkippedContent(nsString& aString);
/**
* Retrieve the type of the parser node.
* @update gess5/11/98
@ -148,14 +155,6 @@ class nsCParserNode : public nsIParserNode {
*/
virtual void AddAttribute(CToken* aToken);
/**
*
* @update gess5/11/98
* @param
* @return
*/
virtual void SetSkippedContent(CToken* aToken);
/**
* This getter retrieves the line number from the input source where
* the token occured. Lines are interpreted as occuring between \n characters.
@ -169,15 +168,14 @@ class nsCParserNode : public nsIParserNode {
* @return token at anIndex
*/
virtual CToken* PopAttributeToken();
protected:
PRUint32 mAttributeCount;
PRInt32 mLineNumber;
CToken* mToken;
CToken* mAttributes[eMaxAttr]; // XXX Ack! This needs to be dynamic!
CToken* mSkippedContent;
nsITokenRecycler* mRecycler;
protected:
PRUint32 mAttributeCount;
PRInt32 mLineNumber;
CToken* mToken;
CToken* mAttributes[eMaxAttr]; // XXX Ack! This needs to be dynamic!
nsAutoString mSkippedContent;
nsITokenRecycler* mRecycler;
};
#endif

10
htmlparser/src/nsToken.cpp
Просмотреть файл

@ -153,6 +153,16 @@ nsString& CToken::GetStringValueXXX(void) {
return mTextValue;
}
/**
* Get string of full contents, suitable for debug dump.
* It should look exactly like the input source.
* @update gess5/11/98
* @return reference to string containing string value
*/
void CToken::GetSource(nsString& anOutputString){
anOutputString=mTextValue;
}
/**
* This method retrieves the value of this internal string
* as a cstring.

8
htmlparser/src/nsToken.h
Просмотреть файл

@ -98,6 +98,14 @@ class CToken {
*/
virtual nsString& GetStringValueXXX(void);
/**
* Get string of full contents, suitable for debug dump.
* It should look exactly like the input source.
* @update gess5/11/98
* @return reference to string containing string value
*/
virtual void GetSource(nsString& anOutputString);
/**
* Setter method that changes the string value of this token
* @update gess5/11/98

25
htmlparser/src/nsViewSourceHTML.cpp
Просмотреть файл

@ -526,14 +526,15 @@ nsresult CViewSourceHTML::WriteText(const nsString& aTextString,nsIContentSink&
temp="";
break;
case kSpace:
if((PR_TRUE==aPreserveSpace) && (kSpace==aTextString.CharAt(theOffset+1))) {
if(temp.Length())
result=aSink.AddLeaf(theTextNode); //just dump the whole string...
WriteNBSP(1,aSink);
temp="";
break;
}
//fall through...
if((PR_TRUE==aPreserveSpace)) {
if(aTextString.Length() > theOffset+1 && (kSpace==aTextString.CharAt(theOffset+1))) {
if(temp.Length())
result=aSink.AddLeaf(theTextNode); //just dump the whole string...
WriteNBSP(1,aSink);
temp="";
break;
}
} //fall through...
default:
//scan ahead looking for valid chars...
temp+=aTextString.CharAt(theOffset);
@ -751,7 +752,7 @@ NS_IMETHODIMP CViewSourceHTML::HandleToken(CToken* aToken,nsIParser* aParser) {
case eToken_text:
{
nsString& theText=aToken->GetStringValueXXX();
WriteText(theText,*mSink,PR_FALSE);
WriteText(theText,*mSink,PR_TRUE);
}
break;
@ -801,8 +802,8 @@ NS_IMETHODIMP CViewSourceHTML::HandleToken(CToken* aToken,nsIParser* aParser) {
CToken* theNextToken = mTokenizer->PopToken();
if(theNextToken) {
theType=eHTMLTokenTypes(theNextToken->GetTokenType());
if(eToken_skippedcontent==theType) {
attrNode.SetSkippedContent(theNextToken);
if(eToken_text==theType) {
attrNode.SetSkippedContent(theNextToken->GetStringValueXXX());
}
}
result= OpenHead(attrNode);
@ -813,7 +814,7 @@ NS_IMETHODIMP CViewSourceHTML::HandleToken(CToken* aToken,nsIParser* aParser) {
if(NS_OK==result)
result=CloseHead(attrNode);
}
nsString& theText=((CAttributeToken*)theNextToken)->GetKey();
const nsString& theText=attrNode.GetSkippedContent();
WriteText(theText,*mSink,PR_FALSE);
}
break;

2
htmlparser/src/nsWellFormedDTD.cpp
Просмотреть файл

@ -114,8 +114,6 @@ NS_IMPL_ADDREF(CWellFormedDTD)
NS_IMPL_RELEASE(CWellFormedDTD)
//static CTokenDeallocator gTokenKiller;
/**
* Default constructor
*

6
htmlparser/src/nsXIFDTD.cpp
Просмотреть файл

@ -1736,11 +1736,7 @@ void nsXIFDTD::EndCSSStyleSheet(const nsIParserNode& aNode)
mBuffer.Append("</");
mBuffer.Append(tagName);
mBuffer.Append(">");
CSkippedContentToken* skipped = new CSkippedContentToken(mBuffer);
nsString& key = skipped->GetKey();
key = mBuffer;
startNode.SetSkippedContent(skipped);
startNode.SetSkippedContent(mBuffer);
mSink->AddLeaf(startNode);
}

956
parser/htmlparser/src/CNavDTD.cpp
Просмотреть файл

Разница между файлами не показана из-за своего большого размера Загрузить разницу

31
parser/htmlparser/src/CNavDTD.h
Просмотреть файл

@ -296,17 +296,6 @@ CLASS_EXPORT_HTMLPARS CNavDTD : public nsIDTD {
*/
virtual PRBool CanOmit(eHTMLTags aParent,eHTMLTags aChild)const;
/**
* This is called to determine if the given parent can omit the
* given child (end tag).
*
* @update gess 3/25/98
* @param aParent -- tag type of parent
* @param aChild -- tag type of child
* @return PR_TRUE if given tag can contain omit child (end tag)
*/
virtual PRBool CanOmitEndTag(eHTMLTags aParent,eHTMLTags aChild)const;
/**
* This method gets called to determine whether a given
* tag is itself a container
@ -405,7 +394,6 @@ CLASS_EXPORT_HTMLPARS CNavDTD : public nsIDTD {
nsresult HandleEndToken(CToken* aToken);
nsresult HandleEntityToken(CToken* aToken);
nsresult HandleCommentToken(CToken* aToken);
nsresult HandleSkippedContentToken(CToken* aToken);
nsresult HandleAttributeToken(CToken* aToken);
nsresult HandleScriptToken(nsCParserNode& aNode);
nsresult HandleStyleToken(CToken* aToken);
@ -431,7 +419,7 @@ CLASS_EXPORT_HTMLPARS CNavDTD : public nsIDTD {
nsresult OpenForm(const nsIParserNode& aNode);
nsresult OpenMap(const nsIParserNode& aNode);
nsresult OpenFrameset(const nsIParserNode& aNode);
nsresult OpenContainer(const nsIParserNode& aNode,PRBool aUpdateStyleStack);
nsresult OpenContainer(const nsIParserNode& aNode,PRBool aClosedByStartTag);
/**
* The next set of methods close the given HTML element.
@ -446,7 +434,7 @@ CLASS_EXPORT_HTMLPARS CNavDTD : public nsIDTD {
nsresult CloseForm(const nsIParserNode& aNode);
nsresult CloseMap(const nsIParserNode& aNode);
nsresult CloseFrameset(const nsIParserNode& aNode);
nsresult CloseContainer(const nsIParserNode& aNode,eHTMLTags anActualTag,PRBool aUpdateStyles);
nsresult CloseContainer(const nsIParserNode& aNode,eHTMLTags anActualTag,PRBool aClosedByStartTag);
/**
* The special purpose methods automatically close
@ -455,8 +443,8 @@ CLASS_EXPORT_HTMLPARS CNavDTD : public nsIDTD {
* @return error code - 0 if all went well.
*/
nsresult CloseTopmostContainer();
nsresult CloseContainersTo(eHTMLTags aTag,PRBool aUpdateStyles);
nsresult CloseContainersTo(PRInt32 anIndex,eHTMLTags aTag,PRBool aUpdateStyles);
nsresult CloseContainersTo(eHTMLTags aTag,PRBool aClosedByStartTag);
nsresult CloseContainersTo(PRInt32 anIndex,eHTMLTags aTag,PRBool aClosedByStartTag);
/**
* Causes leaf to be added to sink at current vector pos.
@ -465,7 +453,7 @@ CLASS_EXPORT_HTMLPARS CNavDTD : public nsIDTD {
* @return error code - 0 if all went well.
*/
nsresult AddLeaf(const nsIParserNode& aNode);
nsresult AddHeadLeaf(const nsIParserNode& aNode);
nsresult AddHeadLeaf(nsIParserNode& aNode);
/**
* Causes auto-closures of context vector stack in order to find a
@ -495,17 +483,16 @@ CLASS_EXPORT_HTMLPARS CNavDTD : public nsIDTD {
* @param aTag -- represents the transient style tag to be handled.
* @return error code -- usually 0
*/
nsresult OpenTransientStyles(eHTMLTags aTag);
nsresult CloseTransientStyles(eHTMLTags aTag);
nsresult OpenTransientStyles(eHTMLTags aChildTag);
nsresult CloseTransientStyles(eHTMLTags aChildTag);
nsresult UpdateStyleStackForOpenTag(eHTMLTags aTag,eHTMLTags aActualTag);
nsresult UpdateStyleStackForCloseTag(eHTMLTags aTag,eHTMLTags aActualTag);
PRBool CanContainStyles(eHTMLTags aTag) const;
nsresult DoFragment(PRBool aFlag);
protected:
nsresult CollectAttributes(nsCParserNode& aNode,PRInt32 aCount);
nsresult CollectAttributes(nsCParserNode& aNode,eHTMLTags aTag,PRInt32 aCount);
nsresult CollectSkippedContent(nsCParserNode& aNode,PRInt32& aCount);
nsresult WillHandleStartTag(CToken* aToken,eHTMLTags aChildTag,nsCParserNode& aNode);
nsresult DidHandleStartTag(nsCParserNode& aNode,eHTMLTags aChildTag);
@ -531,8 +518,10 @@ protected:
nsParser* mParser;
nsITokenizer* mTokenizer;
nsDeque mMisplacedContent;
nsDeque mSkippedContent;
PRBool mHasOpenScript;
PRBool mSaveBadTokens;
eHTMLTags mSkipTarget;
PRUint32 mComputedCRC32;
PRUint32 mExpectedCRC32;

27
parser/htmlparser/src/COtherDTD.cpp
Просмотреть файл

@ -254,20 +254,6 @@ nsresult COtherDTD::HandleCommentToken(CToken* aToken) {
return CNavDTD::HandleCommentToken(aToken);
}
/**
* This method gets called when a skippedcontent token has
* been encountered in the parse process. After verifying
* that the topmost container can contain text, we call
* AddLeaf to store this token in the top container.
*
* @update gess 3/25/98
* @param aToken -- next (start) token to be handled
* @return PR_TRUE if all went well; PR_FALSE if error occured
*/
nsresult COtherDTD::HandleSkippedContentToken(CToken* aToken) {
return CNavDTD::HandleSkippedContentToken(aToken);
}
/**
* This method gets called when an attribute token has been
* encountered in the parse process. This is an error, since
@ -334,19 +320,6 @@ PRBool COtherDTD::CanOmit(eHTMLTags aParent,eHTMLTags aChild) const {
}
/**
* This method gets called to determine whether a given
* ENDtag can be omitted. Admittedly,this is a gross simplification.
*
* @update gess 3/25/98
* @param aTag -- tag to test for containership
* @return PR_TRUE if given tag can contain other tags
*/
PRBool COtherDTD::CanOmitEndTag(eHTMLTags aParent,eHTMLTags aChild) const {
return CNavDTD::CanOmitEndTag(aParent,aChild);
}
/*********************************************
Here comes code that handles the interface
to our content sink.

20
parser/htmlparser/src/COtherDTD.h
Просмотреть файл

@ -121,17 +121,6 @@ class COtherDTD : public CNavDTD {
*/
virtual PRBool CanOmit(eHTMLTags aParent,eHTMLTags aChild)const;
/**
* This method gets called to determine whether a given
* tag can contain newlines. Most do not.
*
* @update gess 3/25/98
* @param aParent -- tag type of parent
* @param aChild -- tag type of child
* @return PR_TRUE if given tag can contain other tags
*/
virtual PRBool CanOmitEndTag(eHTMLTags aParent,eHTMLTags aChild)const;
/**
* This method gets called when a start token has been consumed and needs
* to be handled (possibly added to content model via sink).
@ -181,15 +170,6 @@ class COtherDTD : public CNavDTD {
*/
nsresult HandleCommentToken(CToken* aToken);
/**
* This method gets called when a skipped-content token has been consumed and needs
* to be handled (possibly added to content model via sink).
* @update gess5/11/98
* @param aToken is the skipped-content token to be handled
* @return TRUE if the token was handled.
*/
nsresult HandleSkippedContentToken(CToken* aToken);
/**
* This method gets called when an attribute token has been consumed and needs
* to be handled (possibly added to content model via sink).

79
parser/htmlparser/src/nsDTDUtils.cpp
Просмотреть файл

@ -19,9 +19,9 @@
#include "nsDTDUtils.h"
#include "CNavDTD.h"
#include "nsIObserverService.h"
#include "nsIServiceManager.h"
#include "nsIElementObserver.h"
static NS_DEFINE_IID(kIObserverServiceIID, NS_IOBSERVERSERVICE_IID);
static NS_DEFINE_IID(kObserverServiceCID, NS_OBSERVERSERVICE_CID);
@ -51,6 +51,8 @@ nsEntryStack::nsEntryStack() {
nsEntryStack::~nsEntryStack() {
if(mEntries)
delete [] mEntries;
mCount=mCapacity=0;
mEntries=0;
}
/**
@ -182,7 +184,6 @@ PRInt32 nsEntryStack::GetTopmostIndexOf(eHTMLTags aTag) const {
/**
*
* @update gess9/10/98
* @update gess 4/26/99
*/
nsDTDContext::nsDTDContext() : mStack(), mSkipped(0), mStyles(0) {
#ifdef NS_DEBUG
@ -194,7 +195,6 @@ nsDTDContext::nsDTDContext() : mStack(), mSkipped(0), mStyles(0) {
/**
*
* @update gess9/10/98
* @update gess 4/26/99
*/
nsDTDContext::~nsDTDContext() {
}
@ -202,16 +202,14 @@ nsDTDContext::~nsDTDContext() {
/**
*
* @update gess7/9/98, harishd 04/04/99
* @update gess 4/26/99
*/
PRInt32 nsDTDContext::GetCount(void) {
return mStack.GetSize();
return mStack.GetCount();
}
/**
*
* @update gess7/9/98, harishd 04/04/99
* @update gess 4/26/99
*/
void nsDTDContext::Push(eHTMLTags aTag) {
#ifdef NS_DEBUG
@ -224,7 +222,6 @@ void nsDTDContext::Push(eHTMLTags aTag) {
/**
* @update gess7/9/98, harishd 04/04/99
* @update gess 4/26/99
*/
eHTMLTags nsDTDContext::Pop() {
#ifdef NS_DEBUG
@ -232,6 +229,14 @@ eHTMLTags nsDTDContext::Pop() {
mTags[mStack.mCount-1]=eHTMLTag_unknown;
#endif
nsEntryStack* theStyles=0;
nsTagEntry& theEntry=mStack.EntryAt(mStack.mCount-1);
PRInt32 theIndex=theEntry.mStyleIndex;
if(-1<theIndex){
theStyles=(nsEntryStack*)mStyles.ObjectAt(theIndex);
delete theStyles;
}
eHTMLTags result=mStack.Pop();
return result;
}
@ -239,7 +244,6 @@ eHTMLTags nsDTDContext::Pop() {
/**
*
* @update gess7/9/98
* @update gess 4/26/99
*/
eHTMLTags nsDTDContext::First() const {
return mStack.First();
@ -248,7 +252,6 @@ eHTMLTags nsDTDContext::First() const {
/**
*
* @update gess7/9/98
* @update gess 4/26/99
*/
eHTMLTags nsDTDContext::TagAt(PRInt32 anIndex) const {
return mStack.TagAt(anIndex);
@ -258,7 +261,6 @@ eHTMLTags nsDTDContext::TagAt(PRInt32 anIndex) const {
/**
*
* @update gess7/9/98
* @update gess 4/26/99
*/
eHTMLTags nsDTDContext::operator[](PRInt32 anIndex) const {
return mStack[anIndex];
@ -267,7 +269,6 @@ eHTMLTags nsDTDContext::operator[](PRInt32 anIndex) const {
/**
*
* @update gess7/9/98
* @update gess 4/26/99
*/
eHTMLTags nsDTDContext::Last() const {
return mStack.Last();
@ -276,12 +277,12 @@ eHTMLTags nsDTDContext::Last() const {
/**
*
* @update gess7/9/98
* @update gess 4/26/99
*/
nsEntryStack* nsDTDContext::GetStyles(void) const {
nsEntryStack* nsDTDContext::GetStylesAt(PRUint32 anIndex) const {
nsEntryStack* result=0;
if(0<mStack.mCount){
PRInt32 theIndex=mStack.mEntries[mStack.mCount-1].mStyleIndex;
if(anIndex<mStack.mCount){
nsTagEntry& theEntry=mStack.EntryAt(anIndex);
PRInt32 theIndex=theEntry.mStyleIndex;
if(-1<theIndex){
result=(nsEntryStack*)mStyles.ObjectAt(theIndex);
}
@ -289,6 +290,46 @@ nsEntryStack* nsDTDContext::GetStyles(void) const {
return result;
}
/**
*
* @update gess 04/28/99
*/
void nsDTDContext::PushStyle(eHTMLTags aTag){
nsTagEntry& theEntry=mStack.EntryAt(mStack.mCount-1);
//ok, now go get the right tokenbank deque...
nsEntryStack* theStack=0;
if(-1<theEntry.mStyleIndex)
theStack=(nsEntryStack*)mStyles.ObjectAt(theEntry.mStyleIndex);
if(!theStack){
//time to make a databank for this element...
theStack=new nsEntryStack();
mStyles.Push(theStack);
theEntry.mStyleIndex=mStyles.GetSize()-1;
}
if(theStack){
theStack->Push(aTag);
}
}
/**
*
* @update gess 04/28/99
*/
eHTMLTags nsDTDContext::PopStyle(void){
eHTMLTags result=eHTMLTag_unknown;
nsTagEntry& theEntry=mStack.EntryAt(mStack.mCount-1);
//ok, now go get the right tokenbank deque...
nsEntryStack* theStack=0;
if(-1<theEntry.mStyleIndex)
theStack=(nsEntryStack*)mStyles.ObjectAt(theEntry.mStyleIndex);
if(theStack){
result=theStack->Pop();
}
return result;
}
/**
*
* @update harishd 04/04/99
@ -296,7 +337,7 @@ nsEntryStack* nsDTDContext::GetStyles(void) const {
*/
void nsDTDContext::SaveToken(CToken* aToken, PRInt32 aID)
{
NS_PRECONDITION(aID <= mStack.GetSize() && aID > -1,"Out of bounds");
NS_PRECONDITION(aID <= mStack.GetCount() && aID > -1,"Out of bounds");
if(aToken) {
nsTagEntry& theEntry=mStack.EntryAt(aID);
@ -321,9 +362,9 @@ void nsDTDContext::SaveToken(CToken* aToken, PRInt32 aID)
*/
CToken* nsDTDContext::RestoreTokenFrom(PRInt32 aID)
{
NS_PRECONDITION(aID <= mStack.GetSize() && aID > -1,"Out of bounds");
NS_PRECONDITION(aID <= mStack.GetCount() && aID > -1,"Out of bounds");
CToken* result=0;
if(0<mStack.GetSize()) {
if(0<mStack.GetCount()) {
nsTagEntry theEntry=mStack.EntryAt(aID);
nsDeque* theDeque=(nsDeque*)mSkipped.ObjectAt(theEntry.mBankIndex);
if(theDeque){
@ -340,7 +381,7 @@ CToken* nsDTDContext::RestoreTokenFrom(PRInt32 aID)
*/
PRInt32 nsDTDContext::TokenCountAt(PRInt32 aID)
{
NS_PRECONDITION(aID <= mStack.GetSize(),"Out of bounds");
NS_PRECONDITION(aID <= mStack.GetCount(),"Out of bounds");
nsTagEntry theEntry=mStack.EntryAt(aID);
nsDeque* theDeque=(nsDeque*)mSkipped.ObjectAt(theEntry.mBankIndex);

7
parser/htmlparser/src/nsDTDUtils.h
Просмотреть файл

@ -37,6 +37,7 @@
#include <fstream.h>
#include "nsITokenizer.h"
#include "nsString.h"
#include "nsIElementObserver.h"
/***************************************************************
Before digging into the NavDTD, we'll define a helper
@ -79,7 +80,7 @@ public:
eHTMLTags operator[](PRUint32 anIndex) const;
eHTMLTags Last() const;
void Empty(void);
PRInt32 GetSize(void) const {return mCount;}
PRInt32 GetCount(void) const {return mCount;}
nsTagEntry* mEntries;
PRUint32 mCount;
@ -106,7 +107,9 @@ public:
eHTMLTags Last() const;
void Empty(void);
PRInt32 GetCount(void);
nsEntryStack* GetStyles(void) const;
nsEntryStack* GetStylesAt(PRUint32 anIndex) const;
void PushStyle(eHTMLTags aTag);
eHTMLTags PopStyle(void);
void SaveToken(CToken* aToken, PRInt32 aID);
CToken* RestoreTokenFrom(PRInt32 aID);

825
parser/htmlparser/src/nsElementTable.cpp
Просмотреть файл

Разница между файлами не показана из-за своего большого размера Загрузить разницу

61
parser/htmlparser/src/nsElementTable.h
Просмотреть файл

@ -46,13 +46,14 @@ public:
mTagList=aTagList;
}
PRInt32 GetTopmostIndexOf(nsEntryStack& aTagStack);
PRInt32 GetBottommostIndexOf(nsEntryStack& aTagStack,PRInt32 aStartOffset);
PRBool Contains(eHTMLTags aTag);
PRInt32 GetTopmostIndexOf(nsEntryStack& aTagStack);
PRInt32 GetBottommostIndexOf(nsEntryStack& aTagStack,PRInt32 aStartOffset);
PRBool Contains(eHTMLTags aTag);
eHTMLTags GetTagAt(PRUint32 anIndex) const;
eHTMLTags mTags[5];
eHTMLTags* mTagList;
int mCount;
PRUint32 mCount;
};
//*********************************************************************************************
@ -68,11 +69,13 @@ public:
*/
struct nsHTMLElement {
static PRBool IsContainerType(eHTMLTags aTag,int aType);
static void DebugDumpMembership(const char* aFilename);
static void DebugDumpContainment(const char* aFilename,const char* aTitle);
static void DebugDumpContainType(const char* aFilename);
static PRBool IsBlockElement(eHTMLTags aTag);
static PRBool IsInlineElement(eHTMLTags aTag);
static PRBool IsFlowElement(eHTMLTags aTag);
static PRBool IsBlockEntity(eHTMLTags aTag);
static PRBool IsInlineEntity(eHTMLTags aTag);
static PRBool IsFlowEntity(eHTMLTags aTag);
static PRBool IsBlockCloser(eHTMLTags aTag);
CTagList* GetRootTags(void) const {return mRootNodes;}
@ -80,6 +83,7 @@ struct nsHTMLElement {
CTagList* GetAutoCloseStartTags(void) const {return mAutocloseStart;}
CTagList* GetAutoCloseEndTags(void) const {return mAutocloseEnd;}
CTagList* GetSynonymousTags(void) const {return mSynonymousTags;}
eHTMLTags GetCloseTargetForEndTag(nsEntryStack& aTagStack,PRInt32 anIndex) const;
CTagList* GetSpecialChildren(void) const {return mSpecialKids;}
CTagList* GetSpecialParents(void) const {return mSpecialParents;}
@ -89,9 +93,10 @@ struct nsHTMLElement {
eHTMLTags GetTag(void) const {return mTagID;}
PRBool CanContain(eHTMLTags aChild) const;
PRBool CanExclude(eHTMLTags aChild) const;
PRBool CanOmitStartTag(eHTMLTags aChild) const;
PRBool CanOmitEndTag(eHTMLTags aParent) const;
PRBool CanContainSelf() const;
PRBool CanOmitEndTag(void) const;
PRBool CanContainSelf(void) const;
PRBool HasSpecialProperty(PRInt32 aProperty) const;
PRBool SectionContains(eHTMLTags aTag,PRBool allowDepthSearch);
@ -120,6 +125,7 @@ struct nsHTMLElement {
int mInclusionBits; //defines parental and containment rules
int mExclusionBits; //defines things you CANNOT contain
int mSpecialProperties; //used for various special purposes...
int mPropagateRange; //tells us how far a parent is willing to prop. badly formed children
CTagList* mSpecialParents; //These are the special tags that contain this tag (directly)
CTagList* mSpecialKids; //These are the extra things you can contain
eHTMLTags mSkipTarget; //If set, then we skip all content until this tag is seen
@ -134,7 +140,10 @@ static const int kDiscardTag = 0x0001; //tells us to toss this tag
static const int kOmitEndTag = 0x0002; //safely ignore end tag
static const int kLegalOpen = 0x0004; //Lets BODY, TITLE, SCRIPT to reopen
static const int kOmitWS = 0x0008; //If set, the tag can omit all ws and newlines
static const int kBadContentWatch = 0x0014; //Used in handling illegal contents
static const int kNoPropagate = 0x0010; //If set, this tag won't propagate as a child
static const int kBadContentWatch = 0x0020;
static const int kNoStyleLeaksIn = 0x0040;
static const int kNoStyleLeaksOut = 0x0080;
//*********************************************************************************************
// The following ints define the standard groups of HTML elements...
@ -150,21 +159,23 @@ static const int kSpecial = 0x0008; // A, IMG, APPLET, OBJECT, FONT,
// MAP, Q, SUB, SUP, SPAN, BDO, IFRAME
static const int kFormControl = 0x0010; // INPUT SELECT TEXTAREA LABEL BUTTON
static const int kPreformatted = 0x0011; // PRE
static const int kPreExclusion = 0x0012; // IMG, OBJECT, APPLET, BIG, SMALL, SUB, SUP, FONT, BASEFONT
static const int kFontStyle = 0x0014; // TT, I, B, U, S, STRIKE, BIG, SMALL
static const int kPhrase = 0x0018; // EM, STRONG, DFN, CODE, SAMP, KBD, VAR, CITE, ABBR, ACRONYM
static const int kHeading = 0x0020; // H1..H6
static const int kBlockMisc = 0x0021; // P, DL, DIV, CENTER, NOSCRIPT, NOFRAMES, BLOCKQUOTE
// FORM, ISINDEX, HR, TABLE, FIELDSET, ADDRESS
static const int kPreformatted = 0x0020; // PRE
static const int kPreExclusion = 0x0040; // IMG, OBJECT, APPLET, BIG, SMALL, SUB, SUP, FONT, BASEFONT
static const int kFontStyle = 0x0080; // TT, I, B, U, S, STRIKE, BIG, SMALL, BLINK
static const int kPhrase = 0x0100; // EM, STRONG, DFN, CODE, SAMP, KBD, VAR, CITE, ABBR, ACRONYM
static const int kHeading = 0x0200; // H1..H6
static const int kBlockMisc = 0x0400; // OBJECT, SCRIPT
static const int kBlock = 0x0800; // ADDRESS, BLOCKQUOTE, CENTER, DIV, DL, FIELDSET, FORM,
// ISINDEX, HR, NOSCRIPT, NOFRAMES, P, TABLE
static const int kList = 0x1000; // UL, OL, DIR, MENU
static const int kPCDATA = 0x2000; // just plain text...
static const int kSelf = 0x4000; // whatever THIS tag is...
static const int kExtensions = 0x8000; // BGSOUND, WBR, NOBR
static const int kTable = 0x10000;// TR,TD,THEAD,TBODY,TFOOT,CAPTION,TH
static const int kList = 0x0024; // UL, OL, DIR, MENU
static const int kPCDATA = 0x0028; // just plain text...
static const int kSelf = 0x0040; // whatever THIS tag is...
static const int kInline = (kPCDATA|kFontStyle|kPhrase|kSpecial|kFormControl); // #PCDATA, %fontstyle, %phrase, %special, %formctrl
static const int kBlock = (kHeading|kList|kPreformatted|kBlockMisc); // %heading, %list, %preformatted, %blockmisc
static const int kFlow = (kBlock|kInline); // %block, %inline
static const int kInlineEntity = (kPCDATA|kFontStyle|kPhrase|kSpecial|kFormControl|kExtensions); // #PCDATA, %fontstyle, %phrase, %special, %formctrl
static const int kBlockEntity = (kHeading|kList|kPreformatted|kBlock); // %heading, %list, %preformatted, %blockmisc
static const int kFlowEntity = (kBlockEntity|kInlineEntity); // %block, %inline
#endif

84
parser/htmlparser/src/nsHTMLTokenizer.cpp
Просмотреть файл

@ -101,7 +101,7 @@ NS_IMPL_RELEASE(nsHTMLTokenizer)
* @param
* @return
*/
nsHTMLTokenizer::nsHTMLTokenizer() : nsITokenizer(), mTokenDeque(new CTokenDeallocator()){
nsHTMLTokenizer::nsHTMLTokenizer() : nsITokenizer(), mTokenDeque(0){
NS_INIT_REFCNT();
mDoXMLEmptyTags=PR_FALSE;
}
@ -114,6 +114,10 @@ nsHTMLTokenizer::nsHTMLTokenizer() : nsITokenizer(), mTokenDeque(new CTokenDeall
* @return
*/
nsHTMLTokenizer::~nsHTMLTokenizer(){
if(mTokenDeque.GetSize()){
CTokenDeallocator theDeallocator;
mTokenDeque.ForEach(theDeallocator);
}
}
@ -277,10 +281,10 @@ nsresult nsHTMLTokenizer::ConsumeToken(nsScanner& aScanner) {
case kNotFound:
break;
case 0: //preceeds a EOF...
break;
default:
if(!nsString::IsSpace(theChar)) {
nsAutoString temp(theChar);
@ -411,66 +415,15 @@ nsresult nsHTMLTokenizer::ConsumeAttributes(PRUnichar aChar,CStartToken* aToken,
}
/**
* This is a special case method. It's job is to consume
* all of the given tag up to an including the end tag.
*
* @param aChar: last char read
* @param aScanner: see nsScanner.h
* @param anErrorCode: arg that will hold error condition
* @return new token or null
*/
nsresult nsHTMLTokenizer::ConsumeContentToEndTag(PRUnichar aChar,
eHTMLTags aChildTag,
nsScanner& aScanner,
CToken*& aToken){
//In the case that we just read the given tag, we should go and
//consume all the input until we find a matching end tag.
nsAutoString endTag("</");
endTag.Append(NS_EnumToTag(aChildTag));
endTag.Append(">");
CTokenRecycler* theRecycler=(CTokenRecycler*)GetTokenRecycler();
aToken=theRecycler->CreateTokenOfType(eToken_skippedcontent,aChildTag,endTag);
return aToken->Consume(aChar,aScanner); //tell new token to finish consuming text...
}
/**
*
* In the case that we just read the given tag, we should go and
* consume all the input until we find a matching end tag.
* @update gess12/28/98
* @param
* @return
*/
nsresult nsHTMLTokenizer::HandleSkippedContent(nsScanner& aScanner,CToken*& aToken) {
nsresult nsHTMLTokenizer::ConsumeScriptContent(nsScanner& aScanner,CToken*& aToken) {
nsresult result=NS_OK;
eHTMLTags theTag=(eHTMLTags)aToken->GetTypeID();
if(eHTMLTag_unknown!=gHTMLElements[theTag].mSkipTarget) {
//Do special case handling for <script>, <style>, <title> or <textarea>...
CToken* skippedToken=0;
PRUnichar theChar=0;
result=ConsumeContentToEndTag(theChar,gHTMLElements[theTag].mSkipTarget,aScanner,skippedToken);
CTokenRecycler* theRecycler=(CTokenRecycler*)GetTokenRecycler();
AddToken(skippedToken,result,mTokenDeque,theRecycler);
if(NS_SUCCEEDED(result) && skippedToken){
//In the case that we just read a given tag, we should go and
//consume all the tag content itself (and throw it all away).
nsString& theTagStr=skippedToken->GetStringValueXXX();
CToken* endtoken=theRecycler->CreateTokenOfType(eToken_end,theTag,theTagStr);
if(endtoken){
nsAutoString temp;
theTagStr.Mid(temp,2,theTagStr.Length()-3);
//now strip the leading and trailing delimiters...
endtoken->Reinitialize(theTag,temp);
AddToken(endtoken,result,mTokenDeque,theRecycler);
}
} //if
} //if
return result;
}
@ -498,11 +451,18 @@ nsresult nsHTMLTokenizer::ConsumeStartTag(PRUnichar aChar,CToken*& aToken,nsScan
result=ConsumeAttributes(aChar,(CStartToken*)aToken,aScanner);
}
//now that that's over with, we have one more problem to solve.
//In the case that we just read a <SCRIPT> or <STYLE> tags, we should go and
//consume all the content itself.
if(NS_SUCCEEDED(result)) {
result=HandleSkippedContent(aScanner,aToken);
/* Now that that's over with, we have one more problem to solve.
In the case that we just read a <SCRIPT> or <STYLE> tags, we should go and
consume all the content itself.
*/
if(NS_SUCCEEDED(result) && (eHTMLTag_script==theTag)) {
nsAutoString endTag("</script>");
CTokenRecycler* theRecycler=(CTokenRecycler*)GetTokenRecycler();
CToken* textToken=theRecycler->CreateTokenOfType(eToken_text,theTag,endTag);
result=((CTextToken*)textToken)->ConsumeUntil(0,aScanner,endTag); //tell new token to finish consuming text...
AddToken(textToken,result,mTokenDeque,theRecycler);
CToken* endToken=theRecycler->CreateTokenOfType(eToken_end,theTag);
AddToken(endToken,result,mTokenDeque,theRecycler);
}
//EEEEECCCCKKKK!!!

3
parser/htmlparser/src/nsHTMLTokenizer.h
Просмотреть файл

@ -69,7 +69,7 @@ public:
protected:
virtual nsresult HandleSkippedContent(nsScanner& aScanner,CToken*& aToken);
virtual nsresult ConsumeScriptContent(nsScanner& aScanner,CToken*& aToken);
virtual nsresult ConsumeTag(PRUnichar aChar,CToken*& aToken,nsScanner& aScanner);
virtual nsresult ConsumeStartTag(PRUnichar aChar,CToken*& aToken,nsScanner& aScanner);
virtual nsresult ConsumeEndTag(PRUnichar aChar,CToken*& aToken,nsScanner& aScanner);
@ -79,7 +79,6 @@ protected:
virtual nsresult ConsumeComment(PRUnichar aChar,CToken*& aToken,nsScanner& aScanner);
virtual nsresult ConsumeNewline(PRUnichar aChar,CToken*& aToken,nsScanner& aScanner);
virtual nsresult ConsumeText(const nsString& aString,CToken*& aToken,nsScanner& aScanner);
virtual nsresult ConsumeContentToEndTag(PRUnichar aChar,eHTMLTags aChildTag,nsScanner& aScanner,CToken*& aToken);
virtual nsresult ConsumeProcessingInstruction(PRUnichar aChar,CToken*& aToken,nsScanner& aScanner);
static void AddToken(CToken*& aToken,nsresult aResult,nsDeque& aDeque,CTokenRecycler* aRecycler);

160
parser/htmlparser/src/nsHTMLTokens.cpp
Просмотреть файл

@ -29,7 +29,6 @@
#include "nsHTMLTags.h"
#include "nsHTMLEntities.h"
#include "nsCRT.h"
#include "nsStr.h"
//#define GESS_MACHINE
#ifdef GESS_MACHINE
@ -79,7 +78,6 @@ void CHTMLToken::SetStringValue(const char* name){
}
}
/*
* constructor from tag id
*
@ -272,6 +270,19 @@ void CStartToken::DebugDumpSource(ostream& out) {
out << ">";
}
/*
*
*
* @update gess 3/25/98
* @param anOutputString will recieve the result
* @return nada
*/
void CStartToken::GetSource(nsString& anOutputString){
anOutputString="<";
anOutputString+=mTextValue;
if(!mAttributed)
anOutputString+=">";
}
/*
* constructor from tag id
@ -303,7 +314,6 @@ CEndToken::CEndToken(const nsString& aName) : CHTMLToken(aName) {
* @return error result
*/
nsresult CEndToken::Consume(PRUnichar aChar, nsScanner& aScanner) {
//if you're here, we've already Consumed the <! chars, and are
//ready to Consume the rest of the open tag identifier.
//Stop consuming as soon as you see a space or a '>'.
@ -388,6 +398,19 @@ void CEndToken::DebugDumpSource(ostream& out) {
out << "</" << buffer << ">";
}
/*
*
*
* @update gess 3/25/98
* @param anOutputString will recieve the result
* @return nada
*/
void CEndToken::GetSource(nsString& anOutputString){
anOutputString="</";
anOutputString+=mTextValue;
anOutputString+=">";
}
/*
* default constructor
*
@ -474,6 +497,70 @@ nsresult CTextToken::Consume(PRUnichar aChar, nsScanner& aScanner) {
return result;
}
/*
* Consume as much clear text from scanner as possible.
*
* @update gess 3/25/98
* @param aChar -- last char consumed from stream
* @param aScanner -- controller of underlying input source
* @return error result
*/
nsresult CTextToken::ConsumeUntil(PRUnichar aChar,nsScanner& aScanner,nsString& aTerminalString){
PRBool done=PR_FALSE;
nsresult result=NS_OK;
nsString temp;
PRUnichar theChar;
//We're going to try a new algorithm here. Rather than scan for the matching
//end tag like we used to do, we're now going to scan for whitespace and comments.
//If we find either, just eat them. If we find text or a tag, then go to the
//target endtag, or the start of another comment.
static nsAutoString theWhitespace2("\b\t ");
while((!done) && (NS_OK==result)) {
result=aScanner.GetChar(aChar);
if((NS_OK==result) && (kLessThan==aChar)) {
//we're reading a tag or a comment...
result=aScanner.GetChar(theChar);
if((NS_OK==result) && (kExclamation==theChar)) {
//read a comment...
static CCommentToken theComment;
result=theComment.Consume(aChar,aScanner);
if(NS_OK==result) {
//result=aScanner.SkipWhitespace();
//temp.Append("<!");
temp.Append(theComment.GetStringValueXXX());
//temp.Append(">");
}
} else {
//read a tag...
temp+=aChar;
temp+=theChar;
result=aScanner.ReadUntil(temp,kGreaterThan,PR_TRUE);
}
}
else if(0<=theWhitespace2.BinarySearch(aChar)) {
static CWhitespaceToken theWS;
result=theWS.Consume(aChar,aScanner);
if(NS_OK==result) {
temp.Append(theWS.GetStringValueXXX());
}
}
else {
temp+=aChar;
result=aScanner.ReadUntil(temp,kLessThan,PR_FALSE);
}
nsAutoString theRight;
temp.Right(theRight,aTerminalString.Length());
done=PRBool(0==theRight.Compare(aTerminalString,PR_TRUE));
} //while
int len=temp.Length();
temp.Truncate(len-aTerminalString.Length());
mTextValue=temp;
return result;
}
/*
* default constructor
*
@ -759,14 +846,16 @@ nsresult CCommentToken::Consume(PRUnichar aChar, nsScanner& aScanner) {
PRBool theStrictForm=PR_FALSE;
nsresult result=(theStrictForm) ? ConsumeStrictComment(aChar,aScanner,mTextValue) : ConsumeComment(aChar,aScanner,mTextValue);
/*
//this change is here to make the editor teams' life easier.
//I'm removing the leading and trailing markup...
/*
if(0==mTextValue.Find("<!"))
mTextValue.Cut(0,2); //trim off 1st 2 chars...
if(kGreaterThan==mTextValue.Last())
mTextValue.Truncate(mTextValue.Length()-1); //trim off last char
*/
return result;
}
@ -860,21 +949,21 @@ nsresult CNewlineToken::Consume(PRUnichar aChar, nsScanner& aScanner) {
mTextValue=aChar;
//we already read the \r or \n, let's see what's next!
PRUnichar nextChar;
nsresult result=aScanner.Peek(nextChar);
PRUnichar theChar;
nsresult result=aScanner.Peek(theChar);
if(NS_OK==result) {
switch(aChar) {
case kNewLine:
if(kCR==nextChar) {
result=aScanner.GetChar(nextChar);
mTextValue+=nextChar;
if(kCR==theChar) {
result=aScanner.GetChar(theChar);
mTextValue+=theChar;
}
break;
case kCR:
if(kNewLine==nextChar) {
result=aScanner.GetChar(nextChar);
mTextValue+=nextChar;
if(kNewLine==theChar) {
result=aScanner.GetChar(theChar);
mTextValue+=theChar;
}
break;
default:
@ -968,6 +1057,19 @@ void CAttributeToken::DebugDumpToken(ostream& out) {
out << buffer << ": " << mTypeID << endl;
}
/*
*
*
* @update gess 3/25/98
* @param anOutputString will recieve the result
* @return nada
*/
void CAttributeToken::GetSource(nsString& anOutputString){
anOutputString=mTextKey;
anOutputString+="=";
anOutputString+=mTextValue;
anOutputString+=";";
}
/*
* This general purpose method is used when you want to
@ -997,8 +1099,8 @@ nsresult ConsumeQuotedString(PRUnichar aChar,nsString& aString,nsScanner& aScann
PRUnichar ch=aString.Last();
if(ch!=aChar)
aString+=aChar;
// aString.ReplaceChar(PRUnichar('\n'),PRUnichar(' '));
aString.StripChars("\r"); //per the HTML spec, ignore linefeeds...
//aString.ReplaceChar(PRUnichar('\n'),PRUnichar(' '));
aString.StripChars("\r\n"); //per the HTML spec, ignore linefeeds...
return result;
}
@ -1270,6 +1372,7 @@ const char* CEntityToken::GetClassName(void) {
return "&entity";
}
/*
*
*
@ -1439,6 +1542,19 @@ void CEntityToken::DebugDumpSource(ostream& out) {
delete[] cp;
}
/*
*
*
* @update gess 3/25/98
* @param anOutputString will recieve the result
* @return nada
*/
void CEntityToken::GetSource(nsString& anOutputString){
anOutputString="&";
anOutputString+=anOutputString;
anOutputString+=";";
}
/*
* default constructor
*
@ -1591,9 +1707,9 @@ nsresult CSkippedContentToken::Consume(PRUnichar aChar,nsScanner& aScanner) {
temp+=aChar;
result=aScanner.ReadUntil(temp,kLessThan,PR_FALSE);
}
nsAutoString temp2;
temp.Right(temp2,mTextValue.Length());
done=PRBool(0==temp2.Compare(mTextValue.GetUnicode(),PR_TRUE,mTextValue.Length()));
nsAutoString theRight;
temp.Right(theRight,mTextValue.Length());
done=PRBool(0==theRight.Compare(mTextValue,PR_TRUE));
}
int len=temp.Length();
temp.Truncate(len-mTextValue.Length());
@ -1616,6 +1732,16 @@ void CSkippedContentToken::DebugDumpSource(ostream& out) {
out<<">";
}
/*
*
*
* @update gess 3/25/98
* @param anOutputString will recieve the result
* @return nada
*/
void CSkippedContentToken::GetSource(nsString& anOutputString){
anOutputString="$skipped-content";
}
/**
*

12
parser/htmlparser/src/nsHTMLTokens.h
Просмотреть файл

@ -43,6 +43,7 @@
#include "nsHTMLTags.h"
#include "nsParserError.h"
#include <iostream.h>
#include "nsString.h"
class nsScanner;
@ -121,6 +122,8 @@ class CStartToken: public CHTMLToken {
PRBool IsEmpty(void);
void SetEmpty(PRBool aValue);
virtual void DebugDumpSource(ostream& out);
virtual void GetSource(nsString& anOutputString);
virtual void Reinitialize(PRInt32 aTag, const nsString& aString);
protected:
@ -145,6 +148,7 @@ class CEndToken: public CHTMLToken {
virtual const char* GetClassName(void);
virtual PRInt32 GetTokenType(void);
virtual void DebugDumpSource(ostream& out);
virtual void GetSource(nsString& anOutputString);
};
@ -184,11 +188,8 @@ class CEntityToken : public CHTMLToken {
virtual nsresult Consume(PRUnichar aChar,nsScanner& aScanner);
static PRInt32 ConsumeEntity(PRUnichar aChar,nsString& aString,nsScanner& aScanner);
static PRInt32 TranslateToUnicodeStr(PRInt32 aValue,nsString& aString);
// static PRInt32 FindEntityIndex(nsString& aString);
// static PRInt32 FindEntityIndexMax(const char* aBuffer,PRInt32 aCount=-1);
// static PRBool VerifyEntityTable(void);
// static PRInt32 ReduceEntities(nsString& aString);
virtual void DebugDumpSource(ostream& out);
virtual void GetSource(nsString& anOutputString);
};
@ -220,6 +221,7 @@ class CTextToken: public CHTMLToken {
CTextToken();
CTextToken(const nsString& aString);
virtual nsresult Consume(PRUnichar aChar,nsScanner& aScanner);
nsresult ConsumeUntil(PRUnichar aChar,nsScanner& aScanner,nsString& aTerminalString);
virtual const char* GetClassName(void);
virtual PRInt32 GetTokenType(void);
};
@ -261,6 +263,7 @@ class CAttributeToken: public CHTMLToken {
virtual PRInt32 GetTokenType(void);
virtual nsString& GetKey(void) {return mTextKey;}
virtual void DebugDumpToken(ostream& out);
virtual void GetSource(nsString& anOutputString);
virtual void DebugDumpSource(ostream& out);
PRBool mLastAttribute;
virtual void Reinitialize(PRInt32 aTag, const nsString& aString);
@ -337,6 +340,7 @@ class CSkippedContentToken: public CAttributeToken {
virtual const char* GetClassName(void);
virtual PRInt32 GetTokenType(void);
virtual void DebugDumpSource(ostream& out);
virtual void GetSource(nsString& anOutputString);
protected:
};

1
parser/htmlparser/src/nsIParser.h
Просмотреть файл

@ -186,6 +186,7 @@ class nsIParser : public nsISupports {
* @return ptr to scanner
*/
virtual eParseMode GetParseMode(void)=0;
};
/* ===========================================================*

67
parser/htmlparser/src/nsParser.cpp
Просмотреть файл

@ -47,16 +47,8 @@ static const char* kNullURL = "Error: Null URL given";
static const char* kOnStartNotCalled = "Error: OnStartBinding() must be called before OnDataAvailable()";
static const char* kBadListenerInit = "Error: Parser's IStreamListener API was not setup correctly in constructor.";
//-------------------------------------------------------------------
class CTokenDeallocator: public nsDequeFunctor{
public:
virtual void* operator()(void* anObject) {
CToken* aToken = (CToken*)anObject;
delete aToken;
return 0;
}
};
class CDTDDeallocator: public nsDequeFunctor{
public:
@ -67,6 +59,8 @@ public:
}
};
//-------------------------------------------------------------------
class CDTDFinder: public nsDequeFunctor{
public:
CDTDFinder(nsIDTD* aDTD) {
@ -83,6 +77,8 @@ public:
nsIDTD* mTargetDTD;
};
//-------------------------------------------------------------------
class CSharedParserObjects {
public:
@ -128,14 +124,6 @@ public:
nsString nsParser::gHackMetaCharset = "";
nsString nsParser::gHackMetaCharsetURL = "";
//-------------------------------------------------------------------------
CSharedParserObjects& GetSharedObjects() {
static CSharedParserObjects gSharedParserObjects;
return gSharedParserObjects;
}
/**********************************************************************************
This class is used as an interface between an external agent (like the DOM) and
the parser. It will contain a stack full of tagnames, which is used in our
@ -173,8 +161,13 @@ public:
nsDeque mTags; //will hold a deque of prunichars...
};
//-------------------------------------------------------------------
//-------------------------------------------------------------------------
CSharedParserObjects& GetSharedObjects() {
static CSharedParserObjects gSharedParserObjects;
return gSharedParserObjects;
}
/**
* default constructor
@ -429,18 +422,28 @@ eParseMode DetermineParseMode(nsParser& aParser) {
nsScanner* theScanner=aParser.GetScanner();
if(theScanner){
nsString theBufCopy;
nsString& theBuffer=theScanner->GetBuffer();
PRInt32 theIndex=theBuffer.Find("HTML 4.0");
if(kNotFound==theIndex)
theIndex=theBuffer.Find("html 4.0");
if(kNotFound<theIndex)
return eParseMode_raptor;
else {
PRInt32 theIndex=theBuffer.Find("noquirks");
if(kNotFound==theIndex)
theIndex=theBuffer.Find("NOQUIRKS");
if(kNotFound<theIndex)
return eParseMode_noquirks;
theBuffer.Left(theBufCopy,300);
theBufCopy.ToUpperCase();
PRInt32 theIndex=theBufCopy.Find("<!DOCTYPE");
if(kNotFound==theIndex){
theIndex=theBufCopy.Find("<DOCTYPE");
}
if(kNotFound<theIndex) {
//good, we found "DOCTYPE" -- now go find it's end delimiter '>'
PRInt32 theSubIndex=theBufCopy.Find(kGreaterThan,theIndex+1);
theBufCopy.Truncate(theSubIndex);
theSubIndex=theBufCopy.Find("HTML 4.0");
if(kNotFound<theSubIndex) {
return eParseMode_raptor;
}
}
theIndex=theBufCopy.Find("NOQUIRKS");
if(kNotFound<theIndex) {
return eParseMode_noquirks;
}
}
@ -712,7 +715,7 @@ nsresult nsParser::Parse(nsIInputStream& aStream,PRBool aVerifyEnabled, void* aK
* @param aContentType tells us what type of content to expect in the given string
* @return error code -- 0 if ok, non-zero if error.
*/
nsresult nsParser::Parse(nsString& aSourceBuffer,void* aKey,const nsString& aContentType,PRBool aEnableVerify,PRBool aLastCall){
nsresult nsParser::Parse(nsString& aSourceBuffer,void* aKey,const nsString& aContentType,PRBool aVerifyEnabled,PRBool aLastCall){
#ifdef _rickgdebug
{
@ -751,7 +754,7 @@ nsresult nsParser::Parse(nsString& aSourceBuffer,void* aKey,const nsString& aCon
// till we're completely done.
NS_ADDREF(me);
if(aSourceBuffer.Length() || mUnusedInput.Length()) {
mDTDVerification=aEnableVerify;
mDTDVerification=aVerifyEnabled;
CParserContext* pc=0;
if((!mParserContext) || (mParserContext->mKey!=aKey)) {
@ -1176,6 +1179,7 @@ void nsParser::DebugDumpSource(ostream& aStream) {
}
}
/**
* Call this to get a newly constructed tagstack
* @update gess 5/05/99
@ -1188,3 +1192,4 @@ nsresult nsParser::CreateTagStack(nsITagStack** aTagStack){
return NS_OK;
return NS_ERROR_HTMLPARSER_MEMORYFAILURE;
}

7
parser/htmlparser/src/nsParser.h
Просмотреть файл

@ -69,7 +69,7 @@ class nsIHTMLContentSink;
class nsIDTD;
class nsScanner;
class nsIParserFilter;
class nsTagStack;
#include <fstream.h>
@ -165,7 +165,7 @@ friend class CTokenHandler;
* @param aListener is a listener to forward notifications to
* @return TRUE if all went well -- FALSE otherwise
*/
virtual nsresult Parse(nsIURL* aURL,nsIStreamObserver* aListener,PRBool aEnableVerify=PR_FALSE, void* aKey=0);
virtual nsresult Parse(nsIURL* aURL,nsIStreamObserver* aListener,PRBool aEnableVerify=PR_FALSE,void* aKey=0);
/**
* Cause parser to parse input from given stream
@ -173,7 +173,7 @@ friend class CTokenHandler;
* @param aStream is the i/o source
* @return TRUE if all went well -- FALSE otherwise
*/
virtual nsresult Parse(nsIInputStream& aStream,PRBool aEnableVerify=PR_FALSE, void* aKey=0);
virtual nsresult Parse(nsIInputStream& aStream,PRBool aEnableVerify=PR_FALSE,void* aKey=0);
/**
* @update gess5/11/98
@ -183,7 +183,6 @@ friend class CTokenHandler;
*/
virtual nsresult Parse(nsString& aSourceBuffer,void* aKey,const nsString& aContentType,PRBool aEnableVerify=PR_FALSE,PRBool aLastCall=PR_FALSE);
virtual PRBool IsValidFragment(const nsString& aSourceBuffer,nsITagStack& aStack,PRUint32 anInsertPos,const nsString& aContentType);
virtual nsresult ParseFragment(nsString& aSourceBuffer,void* aKey,nsITagStack& aStack,PRUint32 anInsertPos,const nsString& aContentType);

40
parser/htmlparser/src/nsParserNode.cpp
Просмотреть файл

@ -40,13 +40,13 @@ nsAutoString& GetEmptyString() {
* @param aToken -- token to init internal token
* @return
*/
nsCParserNode::nsCParserNode(CToken* aToken,PRInt32 aLineNumber,nsITokenRecycler* aRecycler): nsIParserNode() {
nsCParserNode::nsCParserNode(CToken* aToken,PRInt32 aLineNumber,nsITokenRecycler* aRecycler):
nsIParserNode(), mSkippedContent("") {
NS_INIT_REFCNT();
mAttributeCount=0;
mLineNumber=aLineNumber;
mToken=aToken;
memset(mAttributes,0,sizeof(mAttributes));
mSkippedContent=0;
mRecycler=aRecycler;
}
@ -65,8 +65,6 @@ nsCParserNode::~nsCParserNode() {
for(index=0;index<mAttributeCount;index++){
mRecycler->RecycleToken(mAttributes[index]);
}
if(mSkippedContent)
mRecycler->RecycleToken(mSkippedContent);
}
}
@ -141,21 +139,6 @@ void nsCParserNode::AddAttribute(CToken* aToken) {
}
}
/**
* This method gets called when the parser encounters
* skipped content after a start token.
*
* @update gess 3/26/98
* @param aToken -- really a skippedcontent token
* @return nada
*/
void nsCParserNode::SetSkippedContent(CToken* aToken){
if(aToken) {
NS_ASSERTION(eToken_skippedcontent == aToken->GetTokenType(), "not a skipped content token");
mSkippedContent = aToken;
}
}
/**
* Gets the name of this node. Currently unused.
@ -191,10 +174,19 @@ const nsString& nsCParserNode::GetText() const {
* @return string ref of text from internal token
*/
const nsString& nsCParserNode::GetSkippedContent() const {
if (nsnull != mSkippedContent) {
return ((CSkippedContentToken*)mSkippedContent)->GetKey();
}
return GetEmptyString();
return mSkippedContent;
}
/**
* Get text value of this node, which translates into
* getting the text value of the underlying token
*
* @update gess 3/25/98
* @param
* @return string ref of text from internal token
*/
void nsCParserNode::SetSkippedContent(nsString& aString) {
mSkippedContent=aString;
}
/**
@ -303,5 +295,3 @@ CToken* nsCParserNode::PopAttributeToken()
}
return nsnull;
}

30
parser/htmlparser/src/nsParserNode.h
Просмотреть файл

@ -90,6 +90,13 @@ class nsCParserNode : public nsIParserNode {
*/
virtual const nsString& GetSkippedContent() const;
/**
* Retrieve skipped context from node
* @update gess5/11/98
* @return string containing skipped content
*/
virtual void SetSkippedContent(nsString& aString);
/**
* Retrieve the type of the parser node.
* @update gess5/11/98
@ -148,14 +155,6 @@ class nsCParserNode : public nsIParserNode {
*/
virtual void AddAttribute(CToken* aToken);
/**
*
* @update gess5/11/98
* @param
* @return
*/
virtual void SetSkippedContent(CToken* aToken);
/**
* This getter retrieves the line number from the input source where
* the token occured. Lines are interpreted as occuring between \n characters.
@ -169,15 +168,14 @@ class nsCParserNode : public nsIParserNode {
* @return token at anIndex
*/
virtual CToken* PopAttributeToken();
protected:
PRUint32 mAttributeCount;
PRInt32 mLineNumber;
CToken* mToken;
CToken* mAttributes[eMaxAttr]; // XXX Ack! This needs to be dynamic!
CToken* mSkippedContent;
nsITokenRecycler* mRecycler;
protected:
PRUint32 mAttributeCount;
PRInt32 mLineNumber;
CToken* mToken;
CToken* mAttributes[eMaxAttr]; // XXX Ack! This needs to be dynamic!
nsAutoString mSkippedContent;
nsITokenRecycler* mRecycler;
};
#endif

10
parser/htmlparser/src/nsToken.cpp
Просмотреть файл

@ -153,6 +153,16 @@ nsString& CToken::GetStringValueXXX(void) {
return mTextValue;
}
/**
* Get string of full contents, suitable for debug dump.
* It should look exactly like the input source.
* @update gess5/11/98
* @return reference to string containing string value
*/
void CToken::GetSource(nsString& anOutputString){
anOutputString=mTextValue;
}
/**
* This method retrieves the value of this internal string
* as a cstring.

8
parser/htmlparser/src/nsToken.h
Просмотреть файл

@ -98,6 +98,14 @@ class CToken {
*/
virtual nsString& GetStringValueXXX(void);
/**
* Get string of full contents, suitable for debug dump.
* It should look exactly like the input source.
* @update gess5/11/98
* @return reference to string containing string value
*/
virtual void GetSource(nsString& anOutputString);
/**
* Setter method that changes the string value of this token
* @update gess5/11/98

25
parser/htmlparser/src/nsViewSourceHTML.cpp
Просмотреть файл

@ -526,14 +526,15 @@ nsresult CViewSourceHTML::WriteText(const nsString& aTextString,nsIContentSink&
temp="";
break;
case kSpace:
if((PR_TRUE==aPreserveSpace) && (kSpace==aTextString.CharAt(theOffset+1))) {
if(temp.Length())
result=aSink.AddLeaf(theTextNode); //just dump the whole string...
WriteNBSP(1,aSink);
temp="";
break;
}
//fall through...
if((PR_TRUE==aPreserveSpace)) {
if(aTextString.Length() > theOffset+1 && (kSpace==aTextString.CharAt(theOffset+1))) {
if(temp.Length())
result=aSink.AddLeaf(theTextNode); //just dump the whole string...
WriteNBSP(1,aSink);
temp="";
break;
}
} //fall through...
default:
//scan ahead looking for valid chars...
temp+=aTextString.CharAt(theOffset);
@ -751,7 +752,7 @@ NS_IMETHODIMP CViewSourceHTML::HandleToken(CToken* aToken,nsIParser* aParser) {
case eToken_text:
{
nsString& theText=aToken->GetStringValueXXX();
WriteText(theText,*mSink,PR_FALSE);
WriteText(theText,*mSink,PR_TRUE);
}
break;
@ -801,8 +802,8 @@ NS_IMETHODIMP CViewSourceHTML::HandleToken(CToken* aToken,nsIParser* aParser) {
CToken* theNextToken = mTokenizer->PopToken();
if(theNextToken) {
theType=eHTMLTokenTypes(theNextToken->GetTokenType());
if(eToken_skippedcontent==theType) {
attrNode.SetSkippedContent(theNextToken);
if(eToken_text==theType) {
attrNode.SetSkippedContent(theNextToken->GetStringValueXXX());
}
}
result= OpenHead(attrNode);
@ -813,7 +814,7 @@ NS_IMETHODIMP CViewSourceHTML::HandleToken(CToken* aToken,nsIParser* aParser) {
if(NS_OK==result)
result=CloseHead(attrNode);
}
nsString& theText=((CAttributeToken*)theNextToken)->GetKey();
const nsString& theText=attrNode.GetSkippedContent();
WriteText(theText,*mSink,PR_FALSE);
}
break;

2
parser/htmlparser/src/nsWellFormedDTD.cpp
Просмотреть файл

@ -114,8 +114,6 @@ NS_IMPL_ADDREF(CWellFormedDTD)
NS_IMPL_RELEASE(CWellFormedDTD)
//static CTokenDeallocator gTokenKiller;
/**
* Default constructor
*

6
parser/htmlparser/src/nsXIFDTD.cpp
Просмотреть файл

@ -1736,11 +1736,7 @@ void nsXIFDTD::EndCSSStyleSheet(const nsIParserNode& aNode)
mBuffer.Append("</");
mBuffer.Append(tagName);
mBuffer.Append(">");
CSkippedContentToken* skipped = new CSkippedContentToken(mBuffer);
nsString& key = skipped->GetKey();
key = mBuffer;
startNode.SetSkippedContent(skipped);
startNode.SetSkippedContent(mBuffer);
mSink->AddLeaf(startNode);
}