Bommels05/Mozilla
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

major improvements to parser API's; fixed a few bugs

Browse Source 
git-svn-id: svn://10.0.0.236/trunk@4894 18797224-902f-48f8-a5cc-f745e15eee43
This commit is contained in:
rickg
1998-07-02 08:14:22 +00:00
parent 5984f25221
commit 5ec2e184eb
40 changed files with 1438 additions and 4660 deletions
Download Patch File Download Diff File Expand all files Collapse all files

13
mozilla/base/src/nsDeque.cpp
View File

@@ -219,13 +219,15 @@ nsDequeIterator nsDeque::End(void) const{
* @param aFunctor object to call for each member
* @return *this
*/
const nsDeque& nsDeque::ForEach(nsDequeFunctor& aFunctor) const{
const void* nsDeque::ForEach(nsDequeFunctor& aFunctor) const{
int i=0;
for(i=0;i<mSize;i++){
void* obj=ObjectAt(i);
aFunctor(obj);
obj=aFunctor(obj);
if(obj)
return obj;
}
return *this;
return 0;
}
/******************************************************
@@ -379,9 +381,8 @@ void* nsDequeIterator::GetCurrent(void) {
* @param aFunctor object to call for each member
* @return *this
*/
const nsDequeIterator& nsDequeIterator::ForEach(nsDequeFunctor& aFunctor) const{
mDeque.ForEach(aFunctor);
return *this;
const void* nsDequeIterator::ForEach(nsDequeFunctor& aFunctor) const{
return mDeque.ForEach(aFunctor);
}

6
mozilla/base/src/nsDeque.h
View File

@@ -50,7 +50,7 @@
*/
class NS_BASE nsDequeFunctor{
public:
virtual void operator()(void* anObject)=0;
virtual void* operator()(void* anObject)=0;
};
@@ -157,7 +157,7 @@ friend class nsDequeIterator;
* @param aFunctor object to call for each member
* @return *this
*/
const nsDeque& ForEach(nsDequeFunctor& aFunctor) const;
const void* ForEach(nsDequeFunctor& aFunctor) const;
/**
* Perform automated selftest on the deque
@@ -346,7 +346,7 @@ public:
* @param aFunctor object to call for each member
* @return *this
*/
const nsDequeIterator& ForEach(nsDequeFunctor& aFunctor) const;
const void* ForEach(nsDequeFunctor& aFunctor) const;
protected:
PRInt32 mIndex;

12
mozilla/content/html/document/src/nsHTMLDocument.cpp
View File

@@ -34,6 +34,7 @@
#include "nsIStreamListener.h"
#include "nsIURL.h"
#include "nsIWebWidget.h"
#include "CNavDTD.h"
static NS_DEFINE_IID(kIDocumentIID, NS_IDOCUMENT_IID);
static NS_DEFINE_IID(kIDOMElementIID, NS_IDOMELEMENT_IID);
@@ -119,6 +120,17 @@ nsHTMLDocument::StartDocumentLoad(nsIURL *aURL,
rv = parser->QueryInterface(kIStreamListenerIID, (void**)aDocListener);
if (NS_OK == rv) {
//The following lines were added by Rick.
//These perform "dynamic" DTD registration, allowing
//the caller total control over process, and decoupling
//parser from any given grammar.
nsIDTD* theDTD=0;
NS_NewNavHTMLDTD(&theDTD);
parser->RegisterDTD(theDTD);
parser->RegisterDTD(theDTD);
parser->SetContentSink(sink);
parser->BeginParse(aURL);
}

168
mozilla/htmlparser/src/CNavDTD.cpp
View File

@@ -40,6 +40,7 @@
#include "nsScanner.h"
#include "nsParserTypes.h"
#include "nsVoidArray.h"
#include "nsTokenHandler.h"
#include "prenv.h" //this is here for debug reasons...
#include "prtypes.h" //this is here for debug reasons...
@@ -61,6 +62,7 @@ static const char* kNullFilename= "Error: Null filename given";
static const char* kNullTokenizer = "Error: Unable to construct tokenizer";
static const char* kNullToken = "Error: Null token given";
static const char* kInvalidTagStackPos = "Error: invalid tag stack position";
static const char* kHTMLTextContentType = "text/html";
static nsAutoString gEmpty;
@@ -78,7 +80,8 @@ static char gStyleTags[]={
eHTMLTag_a, eHTMLTag_bold, eHTMLTag_big,
eHTMLTag_blink, eHTMLTag_cite, eHTMLTag_em,
eHTMLTag_font, eHTMLTag_italic, eHTMLTag_kbd,
eHTMLTag_small, eHTMLTag_spell, eHTMLTag_strike,
eHTMLTag_s, eHTMLTag_small,
eHTMLTag_spell, eHTMLTag_strike,
eHTMLTag_strong, eHTMLTag_sub, eHTMLTag_sup,
eHTMLTag_tt, eHTMLTag_u, eHTMLTag_var,
0};
@@ -215,9 +218,10 @@ void CNavDTD::InitializeDefaultTokenHandlers() {
class CNavTokenDeallocator: public nsDequeFunctor{
public:
virtual void operator()(void* anObject) {
virtual void* operator()(void* anObject) {
CToken* aToken = (CToken*)anObject;
delete aToken;
return 0;
}
};
@@ -277,6 +281,20 @@ void CNavDTD::SetDTDDebug(nsIDTDDebug * aDTDDebug)
NS_ADDREF(mDTDDebug);
}
/**
* This method gets called by the parser to determine if this DTD can handle
* the requested process for the requested content type.
*
* @update gess6/24/98
* @param
* @return TRUE if the DTD can handle the process for this type; FALSE otherwise.
*/
PRBool CNavDTD::IsCapableOf(eProcessType aProcessType, nsString& aContentType,PRInt32 aVersion){
PRBool result=aContentType.Equals(kHTMLTextContentType);
return result;
}
/**
*
* @update gess5/18/98
@@ -337,10 +355,10 @@ PRInt32 CNavDTD::HandleToken(CToken* aToken){
if(aToken) {
CHTMLToken* theToken= (CHTMLToken*)(aToken);
eHTMLTokenTypes theType=eHTMLTokenTypes(theToken->GetTokenType());
CTokenHandler* aHandler=GetTokenHandler(theType);
CITokenHandler* theHandler=GetTokenHandler(theType);
if(aHandler) {
result=(*aHandler)(theToken,this);
if(theHandler) {
result=(*theHandler)(theToken,this);
if (mDTDDebug)
mDTDDebug->Verify(this, mParser, mContextStackPos, mContextStack, mFilename);
}
@@ -365,10 +383,10 @@ PRInt32 CNavDTD::HandleToken(CToken* aToken){
* @param aNode -- CParserNode representing this start token
* @return PR_TRUE if all went well; PR_FALSE if error occured
*/
PRInt32 CNavDTD::HandleDefaultStartToken(CToken* aToken,eHTMLTags aChildTag,nsCParserNode& aNode) {
PRInt32 CNavDTD::HandleDefaultStartToken(CToken* aToken,eHTMLTags aChildTag,nsIParserNode& aNode) {
NS_PRECONDITION(0!=aToken,kNullToken);
eHTMLTags parentTag=(eHTMLTags)GetTopNode();
eHTMLTags parentTag=GetTopNode();
PRInt32 result=kNoError;
PRBool contains=CanContain(parentTag,aChildTag);
@@ -487,30 +505,7 @@ PRInt32 CNavDTD::HandleStartToken(CToken* aToken) {
break;
case eHTMLTag_script:
{
PRInt32 pos=GetTopmostIndexOf(eHTMLTag_body);
nsCParserNode theNode((CHTMLToken*)aToken);
if (kNotFound == pos) {
// We're in the HEAD
result=OpenHead(theNode);
if(kNoError==result) {
mParser->CollectSkippedContent(attrNode,theCount);
if(kNoError==result) {
result=AddLeaf(attrNode);
if(kNoError==result)
result=CloseHead(theNode);
}
}
}
else {
// We're in the BODY
mParser->CollectSkippedContent(attrNode,theCount);
if(kNoError==result) {
result=AddLeaf(attrNode);
}
}
break;
}
result=HandleScriptToken(st); break;
case eHTMLTag_head:
break; //ignore head tags...
@@ -570,7 +565,7 @@ PRInt32 CNavDTD::HandleEndToken(CToken* aToken) {
/*
if(0!=strchr(gStyleTags,tokenTagType)){
eHTMLTags topTag=(eHTMLTags)GetTopNode();
eHTMLTags topTag=GetTopNode();
if(0!=strchr(gStyleTags,topTag)){
tokenTagType=topTag;
}
@@ -704,10 +699,32 @@ PRInt32 CNavDTD::HandleAttributeToken(CToken* aToken) {
*/
PRInt32 CNavDTD::HandleScriptToken(CToken* aToken) {
NS_PRECONDITION(0!=aToken,kNullToken);
PRInt32 result=kNoError;
PRInt32 pos=GetTopmostIndexOf(eHTMLTag_body);
nsCParserNode theNode((CHTMLToken*)aToken);
PRInt32 theCount=0;
PRInt32 result=mParser->CollectSkippedContent(theNode,theCount);
nsCParserNode attrNode((CHTMLToken*)aToken);
PRInt32 attrCount=aToken->GetAttributeCount();
if (kNotFound == pos) {
// We're in the HEAD
result=OpenHead(theNode);
if(kNoError==result) {
mParser->CollectSkippedContent(attrNode,attrCount);
if(kNoError==result) {
result=AddLeaf(attrNode);
if(kNoError==result)
result=CloseHead(theNode);
}
}
}
else {
// We're in the BODY
mParser->CollectSkippedContent(attrNode,attrCount);
if(kNoError==result) {
result=AddLeaf(attrNode);
}
}
return result;
}
@@ -753,8 +770,8 @@ void CNavDTD::DeleteTokenHandlers(void) {
* @param aTagType type of tag to be handled
* @return valid tag handler (if found) or null
*/
CTokenHandler* CNavDTD::GetTokenHandler(eHTMLTokenTypes aType) const {
CTokenHandler* result=0;
CITokenHandler* CNavDTD::GetTokenHandler(eHTMLTokenTypes aType) const {
CITokenHandler* result=0;
if((aType>0) && (aType<eToken_last)) {
result=mTokenHandlers[aType];
}
@@ -771,13 +788,13 @@ CTokenHandler* CNavDTD::GetTokenHandler(eHTMLTokenTypes aType) const {
* @param
* @return
*/
CTokenHandler* CNavDTD::AddTokenHandler(CTokenHandler* aHandler) {
CITokenHandler* CNavDTD::AddTokenHandler(CITokenHandler* aHandler) {
NS_ASSERTION(0!=aHandler,"Error: Null handler");
if(aHandler) {
eHTMLTokenTypes type=(eHTMLTokenTypes)aHandler->GetTokenType();
if(type<eToken_last) {
CTokenHandler* old=mTokenHandlers[type];
CITokenHandler* old=mTokenHandlers[type];
mTokenHandlers[type]=aHandler;
}
else {
@@ -836,8 +853,8 @@ PRBool CNavDTD::CanContainStyles(eHTMLTags aParent) const {
}
/**
* This method is called to determine whether or not a tag
* of one type can contain a tag of another type.
* This method is called to determine whether or not a
* form-type tag can contain a tag of another form-type tag.
*
* @update gess 4/8/98
* @param aParent -- tag enum of parent container
@@ -845,7 +862,18 @@ PRBool CNavDTD::CanContainStyles(eHTMLTags aParent) const {
* @return PR_TRUE if parent can contain child
*/
PRBool CNavDTD::CanContainFormElement(eHTMLTags aParent,eHTMLTags aChild) const {
PRBool result=(mParser) ? HasOpenContainer(eHTMLTag_form) : PR_FALSE;
PRBool result=PR_FALSE;
if(mParser && HasOpenContainer(eHTMLTag_form)) {
eHTMLTags topTag=GetTopNode();
switch(aChild) {
case eHTMLTag_option:
result=PRBool(eHTMLTag_select==topTag); break;
default:
result=PR_TRUE;
break;
} //switch
}//if
return result;
}
@@ -880,7 +908,8 @@ PRBool CNavDTD::CanContain(PRInt32 aParent,PRInt32 aChild) {
eHTMLTag_map, eHTMLTag_menu, eHTMLTag_newline, eHTMLTag_nobr,
eHTMLTag_noframes, eHTMLTag_noscript,
eHTMLTag_object, eHTMLTag_ol, eHTMLTag_paragraph, eHTMLTag_pre,
eHTMLTag_quotation, eHTMLTag_strike, eHTMLTag_samp, eHTMLTag_script,
eHTMLTag_quotation, eHTMLTag_s, eHTMLTag_strike,
eHTMLTag_samp, eHTMLTag_script,
eHTMLTag_select, eHTMLTag_small, eHTMLTag_span, eHTMLTag_strong,
eHTMLTag_sub, eHTMLTag_sup, eHTMLTag_table, eHTMLTag_text,
@@ -901,7 +930,8 @@ PRBool CNavDTD::CanContain(PRInt32 aParent,PRInt32 aChild) {
eHTMLTag_label, eHTMLTag_map, eHTMLTag_newline, eHTMLTag_nobr,
eHTMLTag_object, eHTMLTag_paragraph,
eHTMLTag_quotation, eHTMLTag_strike, eHTMLTag_samp, eHTMLTag_script,
eHTMLTag_quotation, eHTMLTag_s, eHTMLTag_strike,
eHTMLTag_samp, eHTMLTag_script,
eHTMLTag_select, eHTMLTag_small, eHTMLTag_span, eHTMLTag_strong,
eHTMLTag_sub, eHTMLTag_sup, eHTMLTag_text, eHTMLTag_textarea,
eHTMLTag_tt, eHTMLTag_u, eHTMLTag_userdefined, eHTMLTag_var,
@@ -925,7 +955,8 @@ PRBool CNavDTD::CanContain(PRInt32 aParent,PRInt32 aChild) {
eHTMLTag_noframes,
eHTMLTag_noscript, eHTMLTag_object, eHTMLTag_paragraph, eHTMLTag_pre,
eHTMLTag_quotation, eHTMLTag_strike, eHTMLTag_samp, eHTMLTag_small,
eHTMLTag_quotation, eHTMLTag_s, eHTMLTag_strike,
eHTMLTag_samp, eHTMLTag_small,
eHTMLTag_span, eHTMLTag_strong, eHTMLTag_sub, eHTMLTag_sup,
eHTMLTag_td, eHTMLTag_text,
@@ -944,9 +975,9 @@ PRBool CNavDTD::CanContain(PRInt32 aParent,PRInt32 aChild) {
}
switch((eHTMLTags)aParent) {
case eHTMLTag_address:
result=PRBool(0!=strchr(gTagSet2,aChild));
break;
result=PRBool(0!=strchr(gTagSet2,aChild)); break;
case eHTMLTag_applet:
{
@@ -957,7 +988,8 @@ PRBool CNavDTD::CanContain(PRInt32 aParent,PRInt32 aChild) {
eHTMLTag_em, eHTMLTag_font, eHTMLTag_italic, eHTMLTag_iframe,
eHTMLTag_img, eHTMLTag_input, eHTMLTag_kbd, eHTMLTag_label,
eHTMLTag_map, eHTMLTag_object, eHTMLTag_param, eHTMLTag_quotation,
eHTMLTag_samp, eHTMLTag_script, eHTMLTag_select, eHTMLTag_small,
eHTMLTag_s, eHTMLTag_samp, eHTMLTag_script,
eHTMLTag_select, eHTMLTag_small,
eHTMLTag_span, eHTMLTag_strike, eHTMLTag_strong, eHTMLTag_sub,
eHTMLTag_sup, eHTMLTag_textarea, eHTMLTag_tt, eHTMLTag_u,
eHTMLTag_var,0};
@@ -1288,6 +1320,7 @@ PRBool CNavDTD::CanOmit(eHTMLTags aParent,eHTMLTags aChild) const {
case eHTMLTag_input: case eHTMLTag_isindex:
case eHTMLTag_label: case eHTMLTag_legend:
case eHTMLTag_select: case eHTMLTag_textarea:
case eHTMLTag_option:
if(PR_FALSE==HasOpenContainer(eHTMLTag_form))
result=PR_TRUE;
break;
@@ -1379,7 +1412,7 @@ PRBool CNavDTD::CanOmitEndTag(eHTMLTags aParent,eHTMLTags aChild) const {
break;
default:
if(IsCompatibleStyleTag(aChild,(eHTMLTags)GetTopNode()))
if(IsCompatibleStyleTag(aChild,GetTopNode()))
result=PR_FALSE;
else result=(!HasOpenContainer(aChild));
break;
@@ -1587,33 +1620,6 @@ PRBool CNavDTD::BackwardPropagate(nsString& aVector,eHTMLTags aParentTag,eHTMLTa
*/
PRInt32 CNavDTD::DidOpenContainer(eHTMLTags aTag,PRBool /*anExplicitOpen*/){
PRInt32 result=0;
switch (aTag) {
case eHTMLTag_a:
case eHTMLTag_bold:
case eHTMLTag_big:
case eHTMLTag_blink:
case eHTMLTag_cite:
case eHTMLTag_em:
case eHTMLTag_font:
case eHTMLTag_italic:
case eHTMLTag_kbd:
case eHTMLTag_small:
case eHTMLTag_spell:
case eHTMLTag_strike:
case eHTMLTag_strong:
case eHTMLTag_sub:
case eHTMLTag_sup:
case eHTMLTag_tt:
case eHTMLTag_u:
case eHTMLTag_var:
break;
default:
break;
}
return result;
}
@@ -1704,7 +1710,7 @@ PRInt32 CNavDTD::OpenTransientStyles(eHTMLTags aTag){
if(0==strchr(gWhitespaceTags,aTag)){
PRInt32 pos=0;
eHTMLTags parentTag=(eHTMLTags)GetTopNode();
eHTMLTags parentTag=GetTopNode();
if(CanContainStyles(parentTag)) {
for(pos=0;pos<mStyleStackPos;pos++) {
eHTMLTags theTag=(eHTMLTags)(int)mStyleStack[pos];
@@ -1832,7 +1838,7 @@ PRInt32 CNavDTD::OpenBody(const nsIParserNode& aNode){
NS_PRECONDITION(mContextStackPos >= 0, kInvalidTagStackPos);
PRInt32 result=kNoError;
eHTMLTags topTag=(eHTMLTags)CNavDTD::GetTopNode();
eHTMLTags topTag=GetTopNode();
if(eHTMLTag_html!=topTag) {
@@ -2126,7 +2132,7 @@ PRInt32 CNavDTD::CloseContainersTo(eHTMLTags aTag,PRBool aUpdateStyles){
return CloseContainersTo(pos,aTag,aUpdateStyles);
}
eHTMLTags theTopTag=(eHTMLTags)GetTopNode();
eHTMLTags theTopTag=GetTopNode();
if(IsCompatibleStyleTag(aTag,theTopTag)) {
//if you're here, it's because we're trying to close one style tag,
//but a different one is actually open. Because this is NAV4x
@@ -2263,13 +2269,13 @@ PRInt32 CNavDTD::CreateContextStackFor(eHTMLTags aChildTag){
*/
PRInt32 CNavDTD::ReduceContextStackFor(eHTMLTags aChildTag){
PRInt32 result=kNoError;
eHTMLTags topTag=(eHTMLTags)GetTopNode();
eHTMLTags topTag=GetTopNode();
while( (topTag!=kNotFound) &&
(PR_FALSE==CanContain(topTag,aChildTag)) &&
(PR_FALSE==CanContainIndirect(topTag,aChildTag))) {
CloseTopmostContainer();
topTag=(eHTMLTags)GetTopNode();
topTag=GetTopNode();
}
return result;
}
@@ -2297,6 +2303,7 @@ PRInt32 CNavDTD::UpdateStyleStackForOpenTag(eHTMLTags aTag,eHTMLTags anActualTag
case eHTMLTag_font:
case eHTMLTag_italic:
case eHTMLTag_kbd:
case eHTMLTag_s:
case eHTMLTag_small:
case eHTMLTag_spell:
case eHTMLTag_strike:
@@ -2346,6 +2353,7 @@ PRInt32 CNavDTD::UpdateStyleStackForCloseTag(eHTMLTags aTag,eHTMLTags anActualTa
case eHTMLTag_kbd:
case eHTMLTag_small:
case eHTMLTag_spell:
case eHTMLTag_s:
case eHTMLTag_strike:
case eHTMLTag_strong:
case eHTMLTag_sub:

31
mozilla/htmlparser/src/CNavDTD.h
View File

@@ -28,22 +28,22 @@
#define NS_NAVHTMLDTD__
#include "nsIDTD.h"
#include "nsISupports.h"
#include "nsHTMLTokens.h"
#include "nshtmlpars.h"
#include "nsParserNode.h"
#include "nsTokenHandler.h"
#include "nsVoidArray.h"
#include "nsDeque.h"
#define NS_INAVHTML_DTD_IID \
{0x5c5cce40, 0xcfd6, 0x11d1, \
{0xaa, 0xda, 0x00, 0x80, 0x5f, 0x8a, 0x3e, 0x14}}
class nsParser;
class nsIHTMLContentSink;
class nsIDTDDebug;
class nsIParserNode;
class CITokenHandler;
class nsParser;
class CNavDTD : public nsIDTD {
@@ -70,13 +70,21 @@ class CNavDTD : public nsIDTD {
*/
virtual ~CNavDTD();
/**
*
* @update gess7/1/98
* @param
* @return
*/
virtual PRBool IsCapableOf(eProcessType aProcessType, nsString& aString, PRInt32 aVersion);
/**
*
* @update jevering6/23/98
* @param
* @return
*/
virtual void SetDTDDebug(nsIDTDDebug * aDTDDebug);
virtual void SetDTDDebug(nsIDTDDebug * aDTDDebug);
/**
*
@@ -292,7 +300,7 @@ class CNavDTD : public nsIDTD {
* @param aNode is a node be updated with info from given token
* @return TRUE if the token was handled.
*/
PRInt32 HandleDefaultStartToken(CToken* aToken,eHTMLTags aChildTag,nsCParserNode& aNode);
PRInt32 HandleDefaultStartToken(CToken* aToken,eHTMLTags aChildTag,nsIParserNode& aNode);
/**
* This method gets called when an end token has been consumed and needs
@@ -359,7 +367,7 @@ class CNavDTD : public nsIDTD {
private:
protected:
/**
* Causes token handlers to be registered for this parser.
@@ -373,13 +381,13 @@ private:
* DEPRECATED
* @update gess5/11/98
*/
CTokenHandler* GetTokenHandler(eHTMLTokenTypes aType) const;
CITokenHandler* GetTokenHandler(eHTMLTokenTypes aType) const;
/**
* DEPRECATED
* @update gess5/11/98
*/
CTokenHandler* AddTokenHandler(CTokenHandler* aHandler);
CITokenHandler* AddTokenHandler(CITokenHandler* aHandler);
/**
* DEPRECATED
@@ -679,7 +687,7 @@ protected:
nsParser* mParser;
nsIHTMLContentSink* mSink;
CTokenHandler* mTokenHandlers[eToken_last];
CITokenHandler* mTokenHandlers[eToken_last];
nsVoidArray mLeafBits;
nsVoidArray mContextStack;
@@ -691,9 +699,10 @@ protected:
PRBool mHasOpenMap;
nsDeque mTokenDeque;
char* mFilename;
nsIDTDDebug* mDTDDebug;
nsIDTDDebug* mDTDDebug;
};
extern NS_HTMLPARS nsresult NS_NewNavHTMLDTD(nsIDTD** aInstancePtrResult);
#endif

1931
mozilla/htmlparser/src/COtherDTD.cpp
View File

File diff suppressed because it is too large Load Diff

112
mozilla/htmlparser/src/COtherDTD.h
View File

@@ -26,13 +26,7 @@
#ifndef NS_OTHERHTMLDTD__
#define NS_OTHERHTMLDTD__
#include "nsIDTD.h"
#include "nsHTMLTokens.h"
#include "nshtmlpars.h"
#include "nsParserNode.h"
#include "nsTokenHandler.h"
#include "nsDeque.h"
#include "nsVoidArray.h"
#include "CNavDTD.h"
#define NS_IOtherHTML_DTD_IID \
@@ -40,11 +34,10 @@
{0x80, 0x22, 0x00, 0x60, 0x8, 0x14, 0x98, 0x89}}
class nsParser;
class nsIHTMLContentSink;
class nsIDTDDebug;
//class nsParser;
//class nsIHTMLContentSink;
class COtherDTD : public nsIDTD {
class COtherDTD : public CNavDTD {
public:
@@ -69,14 +62,8 @@ class COtherDTD : public nsIDTD {
*/
virtual ~COtherDTD();
/**
*
* @update jevering6/23/98
* @param
* @return
*/
virtual void SetDTDDebug(nsIDTDDebug * aDTDDebug);
virtual PRBool IsCapableOf(eProcessType aProcessType, nsString& aString,PRInt32 aVersion);
/**
*
* @update gess5/18/98
@@ -101,15 +88,6 @@ class COtherDTD : public nsIDTD {
*/
virtual PRInt32 HandleToken(CToken* aToken);
/**
*
*
* @update gess 3/25/98
* @param
* @return
*/
virtual void SetParser(nsIParser* aParser);
/**
* Cause the tokenizer to consume the next token, and
* return an error result.
@@ -137,14 +115,6 @@ class COtherDTD : public nsIDTD {
*/
virtual void WillInterruptParse(void);
/**
* Select given content sink into parser for parser output
* @update gess5/11/98
* @param aSink is the new sink to be used by parser
* @return old sink, or NULL
*/
virtual nsIContentSink* SetContentSink(nsIContentSink* aSink);
/**
* This method is called to determine whether or not a tag
* of one type can contain a tag of another type.
@@ -239,30 +209,6 @@ class COtherDTD : public nsIDTD {
*/
virtual PRInt32 DidOpenContainer(eHTMLTags aTag,PRBool anExplicitOpen);
/**
* Ask parser if a given container is open ANYWHERE on stack
* @update gess5/11/98
* @param id of container you want to test for
* @return TRUE if the given container type is open -- otherwise FALSE
*/
virtual PRBool HasOpenContainer(eHTMLTags aContainer) const;
/**
* Retrieve the tag type of the topmost item on context vector stack
* @update gess5/11/98
* @return tag type (may be unknown)
*/
virtual eHTMLTags GetTopNode() const;
/**
* Finds the topmost occurance of given tag within context vector stack.
* @update gess5/11/98
* @param tag to be found
* @return index of topmost tag occurance -- may be -1 (kNotFound).
*/
virtual PRInt32 GetTopmostIndexOf(eHTMLTags aTag) const;
/**
*
* @update gess6/4/98
@@ -291,7 +237,7 @@ class COtherDTD : public nsIDTD {
* @param aNode is a node be updated with info from given token
* @return TRUE if the token was handled.
*/
PRInt32 HandleDefaultStartToken(CToken* aToken,eHTMLTags aChildTag,nsCParserNode& aNode);
PRInt32 HandleDefaultStartToken(CToken* aToken,eHTMLTags aChildTag,nsIParserNode& aNode);
/**
* This method gets called when an end token has been consumed and needs
@@ -358,33 +304,6 @@ class COtherDTD : public nsIDTD {
private:
/**
* Causes token handlers to be registered for this parser.
* DO NOT CALL THIS! IT'S DEPRECATED!
* @update gess5/11/98
*/
void InitializeDefaultTokenHandlers();
/**
* DEPRECATED
* @update gess5/11/98
*/
CTokenHandler* GetTokenHandler(eHTMLTokenTypes aType) const;
/**
* DEPRECATED
* @update gess5/11/98
*/
CTokenHandler* AddTokenHandler(CTokenHandler* aHandler);
/**
* DEPRECATED
* @update gess5/11/98
*/
void DeleteTokenHandlers(void);
//*************************************************
//these cover methods mimic the sink, and are used
@@ -673,23 +592,8 @@ protected:
PRBool CanContainFormElement(eHTMLTags aParent,eHTMLTags aChild) const;
nsParser* mParser;
nsIHTMLContentSink* mSink;
CTokenHandler* mTokenHandlers[eToken_last];
nsVoidArray mLeafBits;
nsVoidArray mContextStack;
PRInt32 mContextStackPos;
nsVoidArray mStyleStack;
PRInt32 mStyleStackPos;
PRBool mHasOpenForm;
PRBool mHasOpenMap;
nsDeque mTokenDeque;
char* mFilename;
nsIDTDDebug* mDTDDebug;
};
extern NS_HTMLPARS nsresult NS_NewOtherHTMLDTD(nsIDTD** aInstancePtrResult);
#endif

3
mozilla/htmlparser/src/Makefile
View File

@@ -42,8 +42,11 @@ EXPORTS = \
nsHTMLTokens.h \
nsIParserNode.h \
nsIParser.h \
nsIDTD.h \
nsIParserFilter.h \
nsToken.h \
CNavDTD.h \
COtherDTD.h \
$(NULL)
MODULE = raptor

3
mozilla/htmlparser/src/makefile.win
View File

@@ -32,7 +32,8 @@ CPPSRCS=nsHTMLContentSink.cpp \
EXPORTS=nshtmlpars.h nsIContentSink.h nsIHTMLContentSink.h \
nsHTMLTokens.h nsIParserNode.h nsIParser.h nsToken.h \
nsIDTDDebug.h nsIParserFilter.h
nsIDTDDebug.h nsIParserFilter.h \
CNavDTD.h COtherDTD.h nsIDTD.h
CPP_OBJS=.\$(OBJDIR)\nsHTMLContentSink.obj \
.\$(OBJDIR)\CNavDTD.obj \

244
mozilla/htmlparser/src/nsHTMLTokens.cpp
View File

@@ -45,8 +45,8 @@ static const char* kNullScanner = "Error: Scanner is null.";
const PRInt32 kMAXNAMELEN=10;
struct StrToUnicodeStruct
{
char fName[kMAXNAMELEN+1];
PRInt32 fValue;
char mName[kMAXNAMELEN+1];
PRInt32 mValue;
};
@@ -96,8 +96,9 @@ static StrToUnicodeStruct gStrToUnicodeTable[] =
struct HTMLTagEntry {
char fName[12];
eHTMLTags fTagID;
char mName[12];
eHTMLTags mTagID;
PRInt16 mUnused;
};
// KEEP THIS LIST SORTED!
@@ -106,103 +107,103 @@ struct HTMLTagEntry {
// the binary search code above will break!
HTMLTagEntry gHTMLTagTable[] =
{
{"!!UNKNOWN", eHTMLTag_unknown},
{"!DOCTYPE", eHTMLTag_doctype}, {"A", eHTMLTag_a},
{"ACRONYM", eHTMLTag_acronym}, {"ADDRESS", eHTMLTag_address},
{"APPLET", eHTMLTag_applet}, {"AREA", eHTMLTag_area},
{"!!UNKNOWN", eHTMLTag_unknown,0},
{"!DOCTYPE", eHTMLTag_doctype,0}, {"A", eHTMLTag_a,0},
{"ACRONYM", eHTMLTag_acronym,0}, {"ADDRESS", eHTMLTag_address,0},
{"APPLET", eHTMLTag_applet,0}, {"AREA", eHTMLTag_area,0},
{"B", eHTMLTag_bold}, {"BASE", eHTMLTag_base},
{"BASEFONT", eHTMLTag_basefont}, {"BDO", eHTMLTag_bdo},
{"BIG", eHTMLTag_big}, {"BLINK", eHTMLTag_blink},
{"BLOCKQUOTE", eHTMLTag_blockquote}, {"BODY", eHTMLTag_body},
{"BR", eHTMLTag_br}, {"BUTTON", eHTMLTag_button},
{"B", eHTMLTag_bold,0}, {"BASE", eHTMLTag_base,0},
{"BASEFONT", eHTMLTag_basefont,0}, {"BDO", eHTMLTag_bdo,0},
{"BIG", eHTMLTag_big,0}, {"BLINK", eHTMLTag_blink,0},
{"BLOCKQUOTE", eHTMLTag_blockquote,0}, {"BODY", eHTMLTag_body,0},
{"BR", eHTMLTag_br,0}, {"BUTTON", eHTMLTag_button,0},
{"CAPTION", eHTMLTag_caption}, {"CENTER", eHTMLTag_center},
{"CERTIFICATE", eHTMLTag_certificate},
{"CITE", eHTMLTag_cite}, {"CODE", eHTMLTag_code},
{"COL", eHTMLTag_col}, {"COLGROUP", eHTMLTag_colgroup},
{"COMMENT", eHTMLTag_comment},
{"CAPTION", eHTMLTag_caption,0}, {"CENTER", eHTMLTag_center,0},
{"CERTIFICATE", eHTMLTag_certificate,0},
{"CITE", eHTMLTag_cite,0}, {"CODE", eHTMLTag_code,0},
{"COL", eHTMLTag_col,0}, {"COLGROUP", eHTMLTag_colgroup,0},
{"COMMENT", eHTMLTag_comment,0},
{"DD", eHTMLTag_dd}, {"DEL", eHTMLTag_del},
{"DFN", eHTMLTag_dfn}, {"DIR", eHTMLTag_dir},
{"DIV", eHTMLTag_div}, {"DL", eHTMLTag_dl},
{"DT", eHTMLTag_dt},
{"DD", eHTMLTag_dd,0}, {"DEL", eHTMLTag_del,0},
{"DFN", eHTMLTag_dfn,0}, {"DIR", eHTMLTag_dir,0},
{"DIV", eHTMLTag_div,0}, {"DL", eHTMLTag_dl,0},
{"DT", eHTMLTag_dt,0},
{"EM", eHTMLTag_em}, {"EMBED", eHTMLTag_embed},
{"ENTITY", eHTMLTag_entity}, //a pseudo tag
{"EM", eHTMLTag_em,0}, {"EMBED", eHTMLTag_embed,0},
{"ENTITY", eHTMLTag_entity,0}, //a pseudo tag
{"FIELDSET", eHTMLTag_fieldset}, {"FONT", eHTMLTag_font},
{"FOOTER", eHTMLTag_footer}, {"FORM", eHTMLTag_form},
{"FRAME", eHTMLTag_frame}, {"FRAMESET", eHTMLTag_frameset},
{"FIELDSET", eHTMLTag_fieldset,0}, {"FONT", eHTMLTag_font,0},
{"FOOTER", eHTMLTag_footer,0}, {"FORM", eHTMLTag_form,0},
{"FRAME", eHTMLTag_frame,0}, {"FRAMESET", eHTMLTag_frameset,0},
{"H1", eHTMLTag_h1}, {"H2", eHTMLTag_h2},
{"H3", eHTMLTag_h3}, {"H4", eHTMLTag_h4},
{"H5", eHTMLTag_h5}, {"H6", eHTMLTag_h6},
{"HEAD", eHTMLTag_head}, {"HEADER", eHTMLTag_header},
{"HR", eHTMLTag_hr}, {"HTML", eHTMLTag_html},
{"H1", eHTMLTag_h1,0}, {"H2", eHTMLTag_h2,0},
{"H3", eHTMLTag_h3,0}, {"H4", eHTMLTag_h4,0},
{"H5", eHTMLTag_h5,0}, {"H6", eHTMLTag_h6,0},
{"HEAD", eHTMLTag_head,0}, {"HEADER", eHTMLTag_header,0},
{"HR", eHTMLTag_hr,0}, {"HTML", eHTMLTag_html,0},
{"I", eHTMLTag_italic}, {"IFRAME", eHTMLTag_iframe},
{"ILAYER", eHTMLTag_ilayer}, {"IMG", eHTMLTag_img},
{"INPUT", eHTMLTag_input}, {"INS", eHTMLTag_ins},
{"ISINDEX", eHTMLTag_isindex},
{"I", eHTMLTag_italic,0}, {"IFRAME", eHTMLTag_iframe,0},
{"ILAYER", eHTMLTag_ilayer,0}, {"IMG", eHTMLTag_img,0},
{"INPUT", eHTMLTag_input,0}, {"INS", eHTMLTag_ins,0},
{"ISINDEX", eHTMLTag_isindex,0},
{"KBD", eHTMLTag_kbd}, {"KEYGEN", eHTMLTag_keygen},
{"KBD", eHTMLTag_kbd,0}, {"KEYGEN", eHTMLTag_keygen,0},
{"LABEL", eHTMLTag_label}, {"LAYER", eHTMLTag_layer},
{"LEGEND", eHTMLTag_legend}, {"LI", eHTMLTag_listitem},
{"LINK", eHTMLTag_link}, {"LISTING", eHTMLTag_listing},
{"LABEL", eHTMLTag_label,0}, {"LAYER", eHTMLTag_layer,0},
{"LEGEND", eHTMLTag_legend,0}, {"LI", eHTMLTag_listitem,0},
{"LINK", eHTMLTag_link,0}, {"LISTING", eHTMLTag_listing,0},
{"MAP", eHTMLTag_map}, {"MARQUEE", eHTMLTag_marquee},
{"MATH", eHTMLTag_math},
{"MENU", eHTMLTag_menu}, {"META", eHTMLTag_meta},
{"MAP", eHTMLTag_map,0}, {"MARQUEE", eHTMLTag_marquee,0},
{"MATH", eHTMLTag_math,0},
{"MENU", eHTMLTag_menu,0}, {"META", eHTMLTag_meta,0},
{"NEWLINE", eHTMLTag_newline}, {"NOBR", eHTMLTag_nobr},
{"NEWLINE", eHTMLTag_newline,0}, {"NOBR", eHTMLTag_nobr,0},
{"NOEMBED", eHTMLTag_noembed}, {"NOFRAMES", eHTMLTag_noframes},
{"NOLAYER", eHTMLTag_nolayer}, {"NOSCRIPT", eHTMLTag_noscript},
{"NOTE", eHTMLTag_note},
{"NOEMBED", eHTMLTag_noembed,0}, {"NOFRAMES", eHTMLTag_noframes,0},
{"NOLAYER", eHTMLTag_nolayer,0}, {"NOSCRIPT", eHTMLTag_noscript,0},
{"NOTE", eHTMLTag_note,0},
{"OBJECT", eHTMLTag_object}, {"OL", eHTMLTag_ol},
{"OPTION", eHTMLTag_option},
{"OBJECT", eHTMLTag_object,0}, {"OL", eHTMLTag_ol,0},
{"OPTION", eHTMLTag_option,0},
{"P", eHTMLTag_paragraph}, {"PARAM", eHTMLTag_param},
{"PLAINTEXT", eHTMLTag_plaintext},
{"P", eHTMLTag_paragraph,0}, {"PARAM", eHTMLTag_param,0},
{"PLAINTEXT", eHTMLTag_plaintext,0},
{"PRE", eHTMLTag_pre},
{"PRE", eHTMLTag_pre,0},
{"Q", eHTMLTag_quotation},
{"Q", eHTMLTag_quotation,0},
{"S", eHTMLTag_strike}, {"SAMP", eHTMLTag_samp},
{"SCRIPT", eHTMLTag_script}, {"SELECT", eHTMLTag_select},
{"SERVER", eHTMLTag_server}, {"SMALL", eHTMLTag_small},
{"SPACER", eHTMLTag_spacer},
{"SPAN", eHTMLTag_span}, {"SPELL", eHTMLTag_spell},
{"STRIKE", eHTMLTag_strike},
{"STRONG", eHTMLTag_strong}, {"STYLE", eHTMLTag_style},
{"SUB", eHTMLTag_sub}, {"SUP", eHTMLTag_sup},
{"S", eHTMLTag_s,0}, {"SAMP", eHTMLTag_samp,0},
{"SCRIPT", eHTMLTag_script,0}, {"SELECT", eHTMLTag_select,0},
{"SERVER", eHTMLTag_server,0}, {"SMALL", eHTMLTag_small,0},
{"SPACER", eHTMLTag_spacer,0},
{"SPAN", eHTMLTag_span,0}, {"SPELL", eHTMLTag_spell,0},
{"STRIKE", eHTMLTag_strike,0},
{"STRONG", eHTMLTag_strong,0}, {"STYLE", eHTMLTag_style,0},
{"SUB", eHTMLTag_sub,0}, {"SUP", eHTMLTag_sup,0},
{"TABLE", eHTMLTag_table}, {"TBODY", eHTMLTag_tbody},
{"TD", eHTMLTag_td},
{"TABLE", eHTMLTag_table,0}, {"TBODY", eHTMLTag_tbody,0},
{"TD", eHTMLTag_td,0},
{"TEXT", eHTMLTag_text},
{"TEXT", eHTMLTag_text,0},
{"TEXTAREA", eHTMLTag_textarea},
{"TFOOT", eHTMLTag_tfoot}, {"TH", eHTMLTag_th},
{"THEAD", eHTMLTag_thead}, {"TITLE", eHTMLTag_title},
{"TR", eHTMLTag_tr}, {"TT", eHTMLTag_tt},
{"TEXTAREA", eHTMLTag_textarea,0},
{"TFOOT", eHTMLTag_tfoot,0}, {"TH", eHTMLTag_th,0},
{"THEAD", eHTMLTag_thead,0}, {"TITLE", eHTMLTag_title,0},
{"TR", eHTMLTag_tr,0}, {"TT", eHTMLTag_tt,0},
{"U", eHTMLTag_u}, {"UL", eHTMLTag_ul},
{"USERDEF", eHTMLTag_userdefined},
{"VAR", eHTMLTag_var}, {"WBR", eHTMLTag_wbr},
{"WS", eHTMLTag_whitespace},
{"U", eHTMLTag_u,0}, {"UL", eHTMLTag_ul,0},
{"VAR", eHTMLTag_var,0}, {"WBR", eHTMLTag_wbr,0},
{"WS", eHTMLTag_whitespace,0},
{"X-USERDEF", eHTMLTag_userdefined,0}, //make sure this is always last!
};
struct HTMLAttrEntry
{
char fName[11];
eHTMLAttributes fAttrID;
char mName[11];
eHTMLAttributes mAttrID;
};
HTMLAttrEntry gHTMLAttributeTable[] =
@@ -1073,21 +1074,20 @@ PRInt32 CEntityToken::TranslateToUnicodeStr(nsString& aString) {
aString.Append(PRUnichar(index));
}
else {
char* cp = mTextValue.ToNewCString();
index=FindEntityIndex(cp);
index=FindEntityIndex(mTextValue);
if(kNotFound!=index) {
PRUnichar ch=gStrToUnicodeTable[index].fValue;
PRUnichar ch=gStrToUnicodeTable[index].mValue;
aString=ch;
}
else {
#ifdef GESS_MACHINE
index=TranslateExtendedEntity(cp,aString);
index=TranslateExtendedEntity(mTextValue,aString);
#endif
}
delete cp;
}
return index;
}
/*
@@ -1102,7 +1102,7 @@ PRBool CEntityToken::VerifyEntityTable(){
PRInt32 i,j;
for(i=1;i<count-1;i++)
{
j=strcmp(gStrToUnicodeTable[i-1].fName,gStrToUnicodeTable[i].fName);
j=strcmp(gStrToUnicodeTable[i-1].mName,gStrToUnicodeTable[i].mName);
if(j>0)
return PR_FALSE;
}
@@ -1119,23 +1119,19 @@ PRBool CEntityToken::VerifyEntityTable(){
* @param aBuflen -- optional string length
* @return integer offset of string in table, or kNotFound
*/
PRInt32 CEntityToken::FindEntityIndex(const char* aBuffer,PRInt32 aBufLen) {
PRInt32 CEntityToken::FindEntityIndex(nsString& aString) {
PRInt32 result=kNotFound;
PRInt32 cnt=sizeof(gStrToUnicodeTable)/sizeof(StrToUnicodeStruct);
PRInt32 low=0;
PRInt32 high=cnt-1;
PRInt32 middle=kNotFound;
if(kNotFound==aBufLen) {
aBufLen=strlen(aBuffer);
}
if (aBuffer && aBufLen && cnt) {
if(cnt) {
while(low<=high)
{
middle=(PRInt32)(low+high)/2;
// result=strncmp(aBuffer,gStrToUnicodeTable[middle].fName,aBufLen);
result=strcmp(aBuffer,gStrToUnicodeTable[middle].fName);
result=aString.Compare(gStrToUnicodeTable[middle].mName);
// result=strcmp(aBuffer,gStrToUnicodeTable[middle].mName);
if (result==0) {
return middle;
}
@@ -1149,6 +1145,46 @@ PRInt32 CEntityToken::FindEntityIndex(const char* aBuffer,PRInt32 aBufLen) {
}
/*
* This method is used to convert from a given string (char*)
* into a entity index (offset within entity table).
*
* @update gess 3/25/98
* @param aBuffer -- string to be converted
* @param aBuflen -- optional string length
* @return integer offset of string in table, or kNotFound
*/
PRInt32 CEntityToken::FindEntityIndexMax(const char* aBuffer,PRInt32 aBufLen) {
PRInt32 result=kNotFound;
PRInt32 cnt=sizeof(gStrToUnicodeTable)/sizeof(StrToUnicodeStruct);
PRInt32 low=0;
PRInt32 high=cnt-1;
PRInt32 middle=kNotFound;
if(aBuffer) {
if(-1==aBufLen) {
aBufLen=strlen(aBuffer);
}
if(aBufLen && cnt) {
while(low<=high)
{
middle=(PRInt32)(low+high)/2;
result=strcmp(aBuffer,gStrToUnicodeTable[middle].mName);
if (result==0) {
return middle;
}
if (result<0) {
high=middle-1;
}
else low=middle+1;
}
}
}
return kNotFound;
}
/*
* This method reduces all text entities into their char
* representation.
@@ -1171,9 +1207,9 @@ PRInt32 CEntityToken::ReduceEntities(nsString& aString) {
if(kNotFound==endpos)
cnt=aString.Length()-1-amppos;
else cnt=endpos-amppos;
PRInt32 index=FindEntityIndex((const char*)&aString[amppos+1],cnt);
PRInt32 index=FindEntityIndexMax((const char*)&aString[amppos+1],cnt);
if(kNotFound!=index) {
aString[amppos]=gStrToUnicodeTable[index].fValue;
aString[amppos]=gStrToUnicodeTable[index].mValue;
aString.Cut(amppos+1,cnt+(endpos!=kNotFound));
}
else offset=amppos+1;
@@ -1316,19 +1352,19 @@ public:
PRInt32 i,j;
for(i=1;i<count-1;i++)
{
j=strcmp(gHTMLTagTable[i-1].fName,gHTMLTagTable[i].fName);
j=strcmp(gHTMLTagTable[i-1].mName,gHTMLTagTable[i].mName);
if(j>0) {
#ifdef VERBOSE_DEBUG
cout << "Tag Table is out of order at " << i << "!" << endl;
#endif
return;
cout << "Tag Table names are out of order at " << i << "!" << endl;
}
if(gHTMLTagTable[i-1].mTagID>=gHTMLTagTable[i].mTagID) {
cout << "Tag table ID's are out of order at " << i << "!" << endl;;
}
}
return;
}
};
/*
* This method accepts a string (and optionally, its length)
* and determines the eHTMLTag (id) value.
@@ -1347,9 +1383,9 @@ eHTMLTags DetermineHTMLTagType(const nsString& aString)
while(low<=high){
middle=(PRInt32)(low+high)/2;
result=aString.Compare(gHTMLTagTable[middle].fName, PR_TRUE);
result=aString.Compare(gHTMLTagTable[middle].mName, PR_TRUE);
if (result==0)
return gHTMLTagTable[middle].fTagID;
return gHTMLTagTable[middle].mTagID;
if (result<0)
high=middle-1;
else low=middle+1;
@@ -1373,9 +1409,9 @@ const char* GetTagName(PRInt32 aTag) {
while(low<=high) {
middle=(PRInt32)(low+high)/2;
if(aTag==gHTMLTagTable[middle].fTagID)
return gHTMLTagTable[middle].fName;
if(aTag<gHTMLTagTable[middle].fTagID)
if(aTag==gHTMLTagTable[middle].mTagID)
return gHTMLTagTable[middle].mName;
if(aTag<gHTMLTagTable[middle].mTagID)
high=middle-1;
else low=middle+1;
}
@@ -1399,7 +1435,7 @@ public:
PRInt32 i,j;
for(i=1;i<count-1;i++)
{
j=strcmp(gHTMLAttributeTable[i-1].fName,gHTMLAttributeTable[i].fName);
j=strcmp(gHTMLAttributeTable[i-1].mName,gHTMLAttributeTable[i].mName);
if(j>0) {
#ifdef VERBOSE_DEBUG
cout << "Attribute table is out of order at " << j << "!" << endl;

26
mozilla/htmlparser/src/nsHTMLTokens.h
View File

@@ -1,3 +1,4 @@
/* -*- Mode: C++; tab-width: 2; indent-tabs-mode: nil; c-basic-offset: 2 -*- */
/*
* The contents of this file are subject to the Netscape Public License
@@ -75,22 +76,24 @@ enum eHTMLTags
eHTMLTag_link, eHTMLTag_listing, eHTMLTag_map, eHTMLTag_marquee,
eHTMLTag_math, eHTMLTag_menu, eHTMLTag_meta, eHTMLTag_newline,
eHTMLTag_nobr,
eHTMLTag_noembed, eHTMLTag_noframes, eHTMLTag_nolayer, eHTMLTag_noscript, //74
eHTMLTag_noembed, eHTMLTag_noframes, eHTMLTag_nolayer, eHTMLTag_noscript, //
eHTMLTag_note, eHTMLTag_object, eHTMLTag_ol,
eHTMLTag_option, eHTMLTag_paragraph, eHTMLTag_param, eHTMLTag_plaintext,
eHTMLTag_pre, eHTMLTag_quotation, eHTMLTag_samp, eHTMLTag_script,
eHTMLTag_pre, eHTMLTag_quotation,
eHTMLTag_s, eHTMLTag_samp, eHTMLTag_script,
eHTMLTag_select, eHTMLTag_server, eHTMLTag_small,
eHTMLTag_spacer, eHTMLTag_span, eHTMLTag_spell,
eHTMLTag_strong, eHTMLTag_style, eHTMLTag_strike, eHTMLTag_sub,
eHTMLTag_sup,
eHTMLTag_table, eHTMLTag_tbody, eHTMLTag_td, //98
eHTMLTag_text, //used for plain text; this is not really a tag.
eHTMLTag_textarea, //100
eHTMLTag_strike, eHTMLTag_strong, eHTMLTag_style,
eHTMLTag_sub, eHTMLTag_sup,
eHTMLTag_table, eHTMLTag_tbody, eHTMLTag_td, //
eHTMLTag_text, //used for plain text; this is not really a tag.
eHTMLTag_textarea, //
eHTMLTag_tfoot,
eHTMLTag_th, eHTMLTag_thead, eHTMLTag_title, eHTMLTag_tr,
eHTMLTag_tt, eHTMLTag_monofont, eHTMLTag_u, eHTMLTag_ul,
eHTMLTag_tt, eHTMLTag_u, eHTMLTag_ul,
eHTMLTag_var, eHTMLTag_wbr, eHTMLTag_whitespace,
eHTMLTag_xmp,
@@ -121,9 +124,9 @@ enum eHTMLAttributes {
PRInt32 ConsumeQuotedString(PRUnichar aChar,nsString& aString,CScanner& aScanner);
PRInt32 ConsumeAttributeText(PRUnichar aChar,nsString& aString,CScanner& aScanner);
PRInt32 FindEntityIndex(const char* aBuffer,PRInt32 aBufLen=-1);
eHTMLTags DetermineHTMLTagType(const nsString& aString);
const char* GetTagName(PRInt32 aTag);
//PRInt32 FindEntityIndex(nsString& aString,PRInt32 aCount=-1);
/**
@@ -219,7 +222,8 @@ class CEntityToken : public CHTMLToken {
virtual PRInt32 Consume(PRUnichar aChar,CScanner& aScanner);
static PRInt32 ConsumeEntity(PRUnichar aChar,nsString& aString,CScanner& aScanner);
static PRInt32 TranslateToUnicodeStr(PRInt32 aValue,nsString& aString);
static PRInt32 FindEntityIndex(const char* aBuffer,PRInt32 aBufLen=-1);
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);

22
mozilla/htmlparser/src/nsIDTD.h
View File

@@ -39,11 +39,23 @@ class CToken;
class nsIContentSink;
class nsIDTDDebug;
class nsIURL;
class nsString;
enum eProcessType {eParsing, eConverting};
enum eAutoDetectResult {eUnknownDetect, eValidDetect, eInvalidDetect};
class nsIDTD : public nsISupports {
public:
/**
*
* @update gess6/24/98
* @param
* @return
*/
virtual ~nsIDTD() {};
/**
* This method is called to determine whether or not a tag
* of one type can contain a tag of another type.
@@ -63,6 +75,14 @@ class nsIDTD : public nsISupports {
*/
virtual nsIContentSink* SetContentSink(nsIContentSink* aSink)=0;
/**
*
* @update gess6/24/98
* @param
* @return
*/
virtual PRBool IsCapableOf(eProcessType aProcessType, nsString& aContentType, PRInt32 aVersion)=0;
/**
*
* @update gess5/18/98
@@ -130,7 +150,7 @@ class nsIDTD : public nsISupports {
* @param
* @return
*/
virtual void SetDTDDebug(nsIDTDDebug * aDTDDebug) = 0;
virtual void SetDTDDebug(nsIDTDDebug * aDTDDebug) = 0;
};

29
mozilla/htmlparser/src/nsIParser.h
View File

@@ -21,7 +21,8 @@
#include "nshtmlpars.h"
#include "nsISupports.h"
#include "nsIStreamListener.h"
#include "nsIDTD.h"
#define NS_IPARSER_IID \
{0x355cbba0, 0xbf7d, 0x11d1, \
@@ -33,7 +34,6 @@ class nsIStreamObserver;
class nsString;
class CToken;
class nsIURL;
class nsIDTD;
class nsIDTDDebug;
/**
@@ -45,11 +45,11 @@ class nsIDTDDebug;
class nsIParser : public nsISupports {
public:
virtual void RegisterDTD(nsIDTD* aDTD)=0;
virtual nsIContentSink* SetContentSink(nsIContentSink* aContentSink)=0;
virtual void SetDTD(nsIDTD* aDTD)=0;
virtual nsIDTD* GetDTD(void)=0;
virtual eAutoDetectResult AutoDetectContentType(nsString& aBuffer)=0;
/**
* Cause the tokenizer to consume the next token, and
@@ -65,14 +65,23 @@ class nsIParser : public nsISupports {
nsIStreamObserver* aListener = nsnull,
nsIDTDDebug * aDTDDebug = 0) = 0;
virtual PRInt32 Parse(nsIURL* aURL,
nsIStreamObserver* aListener = nsnull,
nsIDTDDebug * aDTDDebug = 0) = 0;
/******************************************************************************************
* Parse methods always begin with an input source, and perform conversions
* until you wind up with HTML in your actual content model.
******************************************************************************************/
virtual PRInt32 Parse(nsIURL* aURL,nsIStreamObserver* aListener = nsnull,nsIDTDDebug * aDTDDebug = 0) = 0;
virtual PRInt32 Parse(const char* aFilename)=0;
virtual PRInt32 Parse(nsString& anHTMLString,PRBool appendTokens)=0;
/******************************************************************************************
* Convert methods start input source (of known or unknown form), and perform conversions
* until you wind up with a <i>stream</i> in your target form.
* The internal content model is never effected.
******************************************************************************************/
virtual PRInt32 Convert(nsIURL* aURL,char* aSourceForm,char* aTargetForm,nsIStreamListener* aListener) = 0;
virtual PRInt32 Convert(const char* aFilename,char* aSourceForm,char* aTargetForm)=0;
virtual PRInt32 Convert(nsString& anHTMLString,char* aSourceForm,char* aTargetForm,PRBool appendTokens)=0;
virtual PRInt32 ResumeParse(void)=0;
/**

1
mozilla/htmlparser/src/nsITokenHandler.h
View File

@@ -35,6 +35,7 @@ class CITokenHandler {
public:
virtual PRInt32 operator()(CToken* aToken,nsIDTD* aDTD)=0;
virtual PRInt32 GetTokenType(void)=0;
};

386
mozilla/htmlparser/src/nsParser.cpp
View File

@@ -22,8 +22,6 @@
#include "nsIContentSink.h"
#include "nsString.h"
#include "nsCRT.h"
#include "COtherDTD.h"
#include "CNavDTD.h"
#include "nsScanner.h"
#include "prenv.h" //this is here for debug reasons...
#include "plstr.h"
@@ -31,6 +29,7 @@
#include "nsIInputStream.h"
#include "nsIParserFilter.h"
#include "nsIDTDDebug.h"
#include "nshtmlpars.h"
static NS_DEFINE_IID(kISupportsIID, NS_ISUPPORTS_IID);
static NS_DEFINE_IID(kClassIID, NS_PARSER_IID);
@@ -40,6 +39,7 @@ static NS_DEFINE_IID(kIStreamListenerIID, NS_ISTREAMLISTENER_IID);
static const char* kNullURL = "Error: Null URL given";
static const char* kNullFilename= "Error: Null filename given";
static const char* kNullTokenizer = "Error: Unable to construct tokenizer";
static const char* kHTMLTextContentType = "text/html";
static const int gTransferBufferSize=4096; //size of the buffer used in moving data from iistream
@@ -70,13 +70,62 @@ NS_HTMLPARS nsresult NS_NewParser(nsIParser** aInstancePtrResult)
class CTokenDeallocator: public nsDequeFunctor{
public:
virtual void operator()(void* anObject) {
virtual void* operator()(void* anObject) {
CToken* aToken = (CToken*)anObject;
delete aToken;
return 0;
}
};
CTokenDeallocator gTokenKiller;
CTokenDeallocator gTokenDeallocator;
class CDTDDeallocator: public nsDequeFunctor{
public:
virtual void* operator()(void* anObject) {
nsIDTD* aDTD =(nsIDTD*)anObject;
NS_RELEASE(aDTD);
return 0;
}
};
class CDTDFinder: public nsDequeFunctor{
public:
CDTDFinder(nsIDTD* aDTD) {
NS_IF_ADDREF(mTargetDTD=aDTD);
}
~CDTDFinder() {
NS_IF_RELEASE(mTargetDTD);
}
virtual void* operator()(void* anObject) {
return (anObject==(void*)mTargetDTD) ? anObject : 0;
}
nsIDTD* mTargetDTD;
};
class CSharedParserObjects {
public:
CSharedParserObjects() : mDeallocator(), mDTDDeque(mDeallocator) {
}
~CSharedParserObjects() {
}
void RegisterDTD(nsIDTD* aDTD){
CDTDFinder theFinder(aDTD);
if(!mDTDDeque.ForEach(theFinder))
mDTDDeque.Push(aDTD);
}
nsIDTD* FindDTD(nsIDTD* aDTD){
return 0;
}
CDTDDeallocator mDeallocator;
nsDeque mDTDDeque;
};
CSharedParserObjects gSharedParserObjects;
/**
* default constructor
@@ -85,7 +134,10 @@ CTokenDeallocator gTokenKiller;
* @param
* @return
*/
nsParser::nsParser() : mTokenDeque(gTokenKiller) {
nsParser::nsParser() :
mTokenDeque(gTokenDeallocator),
mContentType()
{
NS_INIT_REFCNT();
mDTDDebug = 0;
mParserFilter = 0;
@@ -97,6 +149,9 @@ nsParser::nsParser() : mTokenDeque(gTokenKiller) {
mParseMode=eParseMode_unknown;
mURL=0;
mDTD=0;
mScanner=0;
mTransferBuffer=new char[gTransferBufferSize+1];
mAutoDetectStatus=eUnknownDetect;
}
@@ -121,7 +176,6 @@ nsParser::~nsParser() {
delete mScanner;
mScanner=0;
NS_IF_RELEASE(mDTD);
NS_IF_RELEASE(mURL);
}
@@ -207,26 +261,16 @@ nsIContentSink* nsParser::SetContentSink(nsIContentSink* aSink) {
return old;
}
/**
*
* Call this static method when you want to
* register your dynamic DTD's with the parser.
*
* @update gess 6/9/98
* @param
* @return
* @param aDTD is the object to be registered.
* @return nothing.
*/
void nsParser::SetDTD(nsIDTD* aDTD) {
mDTD=aDTD;
}
/**
* Retrieve the DTD from the parser.
* @update gess6/18/98
* @param
* @return
*/
nsIDTD * nsParser::GetDTD(void) {
return mDTD;
void nsParser::RegisterDTD(nsIDTD* aDTD){
gSharedParserObjects.RegisterDTD(aDTD);
}
/**
@@ -240,39 +284,6 @@ CScanner* nsParser::GetScanner(void){
return mScanner;
}
/**
* This is where we loop over the tokens created in the
* tokenization phase, and try to make sense out of them.
*
* @update gess 3/25/98
* @param
* @return PR_TRUE if parse succeeded, PR_FALSE otherwise.
*/
PRInt32 nsParser::IterateTokens() {
nsDequeIterator e=mTokenDeque.End();
nsDequeIterator theMarkPos(e);
mMajorIteration++;
if(!mCurrentPos)
mCurrentPos=new nsDequeIterator(mTokenDeque.Begin());
PRInt32 result=kNoError;
while((kNoError==result) && ((*mCurrentPos<e))){
mMinorIteration++;
CToken* theToken=(CToken*)mCurrentPos->GetCurrent();
theMarkPos=*mCurrentPos;
result=mDTD->HandleToken(theToken);
++(*mCurrentPos);
}
if(kInterrupted==result)
*mCurrentPos=theMarkPos;
return result;
}
/**
*
*
@@ -291,7 +302,6 @@ eParseMode DetermineParseMode() {
return result;
}
/**
*
*
@@ -299,34 +309,59 @@ eParseMode DetermineParseMode() {
* @param
* @return
*/
nsIDTD* CreateDTD(eParseMode aMode,const char* aContentType) {
nsIDTD* aDTD=0;
switch(aMode) {
case eParseMode_navigator:
aDTD=new CNavDTD(); break;
case eParseMode_other:
aDTD=new COtherDTD(); break;
default:
break;
PRBool FindSuitableDTD(eProcessType aProcessType,eParseMode aMode,nsString& aContentType, nsIDTD*& aDefaultDTD) {
PRBool result=PR_FALSE;
if(aDefaultDTD) {
if(aDefaultDTD->IsCapableOf(aProcessType,aContentType,0)) {
result=PR_TRUE;
}
}
if (aDTD)
aDTD->AddRef();
return aDTD;
}
else {
nsDequeIterator b=gSharedParserObjects.mDTDDeque.Begin();
nsDequeIterator e=gSharedParserObjects.mDTDDeque.End();
/**
*
* @update gess6/22/98
* @param
* @return
*/
PRBool nsParser::DetermineContentType(const char* aContentType) {
PRBool result=PR_TRUE;
aContentType="application/x-unknown-content-type";
while(b<e){
nsIDTD* theDTD=(nsIDTD*)b.GetCurrent();
if(theDTD && theDTD->IsCapableOf(aProcessType,aContentType,0)) {
aDefaultDTD=theDTD;
result=PR_TRUE;
break;
}
++b;
}
}
return result;
}
/**
* Call this method if you want the known DTD's to try
* to detect the document type based through analysis
* of the underlying stream.
*
* @update gess6/22/98
* @param aBuffer -- nsString containing sample data to be analyzed.
* @return auto-detect result: eValid, eInvalid, eUnknown
*/
eAutoDetectResult nsParser::AutoDetectContentType(nsString& aBuffer) {
//The process:
// You should go out and ask each DTD if they
// recognize the content in the scanner.
// Somebody should say yes, or we can't continue.
//This method may change mContentType and mDTD.
//It absolutely changes mAutoDetectStatus
mAutoDetectStatus=eUnknownDetect;
if(PR_TRUE==mContentType.Equals(kHTMLTextContentType)) {
mAutoDetectStatus=eValidDetect;
}
return mAutoDetectStatus;
}
/**
* This gets called just prior to the model actually
* being constructed. It's important to make this the
@@ -338,14 +373,13 @@ PRBool nsParser::DetermineContentType(const char* aContentType) {
* @param
* @return
*/
PRInt32 nsParser::WillBuildModel(const char* aFilename, const char* aContentType){
PRInt32 nsParser::WillBuildModel(eProcessType aProcessType, const char* aFilename){
mMajorIteration=-1;
mMinorIteration=-1;
mParseMode=DetermineParseMode();
mDTD=(0==mDTD) ? CreateDTD(mParseMode,aContentType) : mDTD;
if(mDTD) {
mDTD->SetDTDDebug(mDTDDebug);
if(PR_TRUE==FindSuitableDTD(aProcessType,mParseMode,mContentType,mDTD)) {
mDTD->SetParser(this);
mDTD->SetContentSink(mSink);
mDTD->WillBuildModel(aFilename);
@@ -406,16 +440,66 @@ PRBool nsParser::Parse(const char* aFilename){
//ok, time to create our tokenizer and begin the process
mScanner=new CScanner(aFilename,mParseMode);
char theContentType[600];
DetermineContentType(theContentType);
WillBuildModel(aFilename,theContentType);
status=ResumeParse();
DidBuildModel(status);
if(mScanner) {
mScanner->Eof();
if(eValidDetect==AutoDetectContentType(mScanner->GetBuffer())) {
WillBuildModel(eParsing,aFilename);
status=ResumeParse();
DidBuildModel(status);
}
} //if
}
return status;
}
/**
* This is the main controlling routine in the parsing process.
* Note that it may get called multiple times for the same scanner,
* since this is a pushed based system, and all the tokens may
* not have been consumed by the scanner during a given invocation
* of this method.
*
* NOTE: We don't call willbuildmodel here, because it will happen
* as a result of calling OnStartBinding later on.
*
* @update gess 3/25/98
* @param aFilename -- const char* containing file to be parsed.
* @return PR_TRUE if parse succeeded, PR_FALSE otherwise.
*/
PRInt32 nsParser::Parse(nsIURL* aURL,nsIStreamObserver* aListener, nsIDTDDebug * aDTDDebug) {
PRInt32 status;
status = BeginParse(aURL, aListener, aDTDDebug);
if (NS_OK == status) {
status=mURL->Open(this);
}
return status;
}
/**
* Call this method if all you want to do is parse 1 string full of HTML text.
*
* @update gess5/11/98
* @param anHTMLString contains a string-full of real HTML
* @param appendTokens tells us whether we should insert tokens inline, or append them.
* @return TRUE if all went well -- FALSE otherwise
*/
PRInt32 nsParser::Parse(nsString& aSourceBuffer,PRBool appendTokens){
PRInt32 result=kNoError;
mScanner=new CScanner();
mScanner->Append(aSourceBuffer);
if(eValidDetect==AutoDetectContentType(aSourceBuffer)) {
WillBuildModel(eParsing,"from-string");
result=ResumeParse();
DidBuildModel(result);
}
return result;
}
/**
* This is the main controlling routine in the parsing process.
* Note that it may get called multiple times for the same scanner,
@@ -454,46 +538,35 @@ PRInt32 nsParser::BeginParse(nsIURL* aURL,nsIStreamObserver* aListener, nsIDTDDe
}
/**
* This is the main controlling routine in the parsing process.
* Note that it may get called multiple times for the same scanner,
* since this is a pushed based system, and all the tokens may
* not have been consumed by the scanner during a given invocation
* of this method.
*
* NOTE: We don't call willbuildmodel here, because it will happen
* as a result of calling OnStartBinding later on.
*
* @update gess 3/25/98
* @param aFilename -- const char* containing file to be parsed.
* @return PR_TRUE if parse succeeded, PR_FALSE otherwise.
*
* @update gess6/23/98
* @param
* @return
*/
PRInt32 nsParser::Parse(nsIURL* aURL,nsIStreamObserver* aListener, nsIDTDDebug * aDTDDebug) {
PRInt32 status;
status = BeginParse(aURL, aListener, aDTDDebug);
if (NS_OK == status) {
status=mURL->Open(this);
}
return status;
PRInt32 nsParser::Convert(nsIURL* aURL,char* aSourceForm,char* aTargetForm,nsIStreamListener* aListener) {
PRInt32 result=0;
return result;
}
/**
* Call this method if all you want to do is parse 1 string full of HTML text.
*
* @update gess5/11/98
* @param anHTMLString contains a string-full of real HTML
* @param appendTokens tells us whether we should insert tokens inline, or append them.
* @return TRUE if all went well -- FALSE otherwise
*
* @update gess6/23/98
* @param
* @return
*/
PRInt32 nsParser::Parse(nsString& aSourceBuffer,PRBool appendTokens){
PRInt32 result=kNoError;
WillBuildModel("nsString","text/html");
mScanner->Append(aSourceBuffer);
result=ResumeParse();
DidBuildModel(result);
PRInt32 nsParser::Convert(const char* aFilename,char* aSourceForm,char* aTargetForm) {
PRInt32 result=0;
return result;
}
/**
*
* @update gess6/23/98
* @param
* @return
*/
PRInt32 nsParser::Convert(nsString& anHTMLString,char* aSourceForm,char* aTargetForm,PRBool appendTokens){
PRInt32 result=0;
return result;
}
@@ -520,6 +593,37 @@ PRInt32 nsParser::ResumeParse() {
return result;
}
/**
* This is where we loop over the tokens created in the
* tokenization phase, and try to make sense out of them.
*
* @update gess 3/25/98
* @param
* @return PR_TRUE if parse succeeded, PR_FALSE otherwise.
*/
PRInt32 nsParser::IterateTokens() {
nsDequeIterator e=mTokenDeque.End();
nsDequeIterator theMarkPos(e);
mMajorIteration++;
if(!mCurrentPos)
mCurrentPos=new nsDequeIterator(mTokenDeque.Begin());
PRInt32 result=kNoError;
while((kNoError==result) && ((*mCurrentPos<e))){
mMinorIteration++;
CToken* theToken=(CToken*)mCurrentPos->GetCurrent();
theMarkPos=*mCurrentPos;
result=mDTD->HandleToken(theToken);
++(*mCurrentPos);
}
if(kInterrupted==result)
*mCurrentPos=theMarkPos;
return result;
}
/**
* Retrieve the attributes for this node, and add then into
@@ -577,6 +681,10 @@ PRInt32 nsParser::CollectSkippedContent(nsCParserNode& aNode,PRInt32& aCount) {
return result;
}
/*******************************************************************
These methods are used to talk to the netlib system...
*******************************************************************/
/**
*
*
@@ -618,11 +726,11 @@ nsresult nsParser::OnStartBinding(const char *aContentType){
if (nsnull != mObserver) {
mObserver->OnStartBinding(aContentType);
}
nsresult result=WillBuildModel(mURL->GetSpec(),aContentType);
if(!mTransferBuffer) {
mTransferBuffer=new char[gTransferBufferSize+1];
}
return result;
mAutoDetectStatus=eUnknownDetect;
mDTD=0;
mContentType=aContentType;
return kNoError;
}
/**
@@ -634,9 +742,21 @@ nsresult nsParser::OnStartBinding(const char *aContentType){
* @return error code (usually 0)
*/
nsresult nsParser::OnDataAvailable(nsIInputStream *pIStream, PRInt32 length){
/* if (nsnull != mListener) {
//Rick potts removed this.
//Does it need to be here?
mListener->OnDataAvailable(pIStream, length);
}
*/
int len=0;
int offset=0;
if(eInvalidDetect==mAutoDetectStatus) {
if(mScanner) {
mScanner->GetBuffer().Truncate();
}
}
do {
PRInt32 err;
len = pIStream->Read(&err, mTransferBuffer, 0, gTransferBufferSize);
@@ -650,9 +770,13 @@ nsresult nsParser::OnDataAvailable(nsIInputStream *pIStream, PRInt32 length){
if (mParserFilter)
mParserFilter->RawBuffer(mTransferBuffer, &len);
mScanner->Append(mTransferBuffer,len);
mScanner->Append(&mTransferBuffer[offset],len);
if(eUnknownDetect==mAutoDetectStatus) {
if(eValidDetect==AutoDetectContentType(mScanner->GetBuffer())) {
nsresult result=WillBuildModel(eParsing,mURL->GetSpec());
} //if
}
} //if
} while (len > 0);

24
mozilla/htmlparser/src/nsParser.h
View File

@@ -60,7 +60,6 @@
#include "nsParserNode.h"
#include "nsParserTypes.h"
#include "nsIURL.h"
#include "nsIStreamListener.h"
#define NS_PARSER_IID \
@@ -107,11 +106,9 @@ friend class CTokenHandler;
virtual nsIContentSink* SetContentSink(nsIContentSink* aSink);
virtual nsIParserFilter* SetParserFilter(nsIParserFilter* aFilter);
virtual void RegisterDTD(nsIDTD* aDTD);
virtual void SetDTD(nsIDTD* aDTD);
virtual nsIDTD* GetDTD(void);
/**
*
*
@@ -152,6 +149,15 @@ friend class CTokenHandler;
*/
virtual PRInt32 Parse(nsString& anHTMLString,PRBool appendTokens);
/******************************************************************************************
* Convert methods start input source (of known or unknown form), and perform conversions
* until you wind up with a <i>stream</i> in your target form.
* The internal content model is never effected.
******************************************************************************************/
virtual PRInt32 Convert(nsIURL* aURL,char* aSourceForm, char* aTargetForm,nsIStreamListener* aListener);
virtual PRInt32 Convert(const char* aFilename,char* aSourceForm,char* aTargetForm);
virtual PRInt32 Convert(nsString& anHTMLString,char* aSourceForm,char* aTargetForm,PRBool appendTokens);
/**
* This method gets called (automatically) during incremental parsing
* @update gess5/11/98
@@ -208,7 +214,7 @@ protected:
* @param
* @return
*/
PRInt32 WillBuildModel(const char* aFilename=0,const char* aContentType=0);
PRInt32 WillBuildModel(eProcessType theProcessType=eParsing,const char* aFilename=0);
/**
*
@@ -303,7 +309,7 @@ private:
* @param
* @return TRUE if we figured it out.
*/
PRBool DetermineContentType(const char* aContentType);
eAutoDetectResult AutoDetectContentType(nsString& aBuffer);
protected:
@@ -328,7 +334,9 @@ protected:
nsDeque mTokenDeque;
CScanner* mScanner;
nsIURL* mURL;
nsIDTDDebug* mDTDDebug;
nsIDTDDebug* mDTDDebug;
nsString mContentType;
eAutoDetectResult mAutoDetectStatus;
};

8
mozilla/htmlparser/src/nsParserNode.cpp
View File

@@ -19,7 +19,7 @@
#include "nsParserNode.h"
#include "string.h"
#include "nsHTMLTokens.h"
/**
* Default constructor
@@ -32,7 +32,7 @@ nsCParserNode::nsCParserNode(CToken* aToken): nsIParserNode(),
mName(), mEmptyString() {
NS_PRECONDITION(0!=aToken, "Null Token");
mAttributeCount=0;
mToken=(CHTMLToken*)aToken;
mToken=aToken;
memset(mAttributes,0,sizeof(mAttributes));
}
@@ -60,7 +60,7 @@ void nsCParserNode::AddAttribute(CToken* aToken) {
NS_PRECONDITION(mAttributeCount<sizeof(mAttributes), "Buffer overrun!");
NS_PRECONDITION(0!=aToken, "Error: Token shouldn't be null!");
if(aToken) {
mAttributes[mAttributeCount++]=(CHTMLToken*)aToken;
mAttributes[mAttributeCount++]=aToken;
}
}
@@ -79,7 +79,7 @@ void nsCParserNode::SetSkippedContent(CToken* aToken){
NS_PRECONDITION(mAttributeCount<sizeof(mAttributes)-1, "Buffer overrun!");
NS_PRECONDITION(0!=aToken, "Error: Token shouldn't be null!");
if(aToken) {
mAttributes[mAttributeCount++]=(CHTMLToken*)aToken;
mAttributes[mAttributeCount++]=aToken;
}
}

12
mozilla/htmlparser/src/nsParserNode.h
View File

@@ -37,7 +37,7 @@
#define NS_PARSERNODE__
#include "nsIParserNode.h"
#include "nsHTMLTokens.h"
#include "nsToken.h"
#include "nsString.h"
//class nsParser;
@@ -148,11 +148,11 @@ class nsCParserNode : public nsIParserNode {
protected:
PRInt32 mAttributeCount;
CHTMLToken* mToken;
CHTMLToken* mAttributes[20]; // XXX Ack! This needs to be dynamic!
nsString mName;
nsString mEmptyString;
PRInt32 mAttributeCount;
CToken* mToken;
CToken* mAttributes[20]; // XXX Ack! This needs to be dynamic!
nsString mName;
nsString mEmptyString;
};

9
mozilla/htmlparser/src/nsScanner.cpp
View File

@@ -503,6 +503,15 @@ PRInt32 CScanner::ReadUntil(nsString& aString,PRUnichar aTerminalChar,PRBool add
return result;
}
/**
*
* @update gess 3/25/98
* @param
* @return
*/
nsString& CScanner::GetBuffer(void) {
return mBuffer;
}
/**
* Conduct self test. Actually, selftesting for this class

9
mozilla/htmlparser/src/nsScanner.h
View File

@@ -233,6 +233,15 @@ class CScanner {
*/
PRInt32 IncrementalAppend(const char* aBuffer,PRInt32 aSize);
/**
*
*
* @update gess 5/12/98
* @param
* @return
*/
nsString& GetBuffer(void);
static void SelfTest();
protected:

12
mozilla/layout/html/document/src/nsHTMLDocument.cpp
View File

@@ -34,6 +34,7 @@
#include "nsIStreamListener.h"
#include "nsIURL.h"
#include "nsIWebWidget.h"
#include "CNavDTD.h"
static NS_DEFINE_IID(kIDocumentIID, NS_IDOCUMENT_IID);
static NS_DEFINE_IID(kIDOMElementIID, NS_IDOMELEMENT_IID);
@@ -119,6 +120,17 @@ nsHTMLDocument::StartDocumentLoad(nsIURL *aURL,
rv = parser->QueryInterface(kIStreamListenerIID, (void**)aDocListener);
if (NS_OK == rv) {
//The following lines were added by Rick.
//These perform "dynamic" DTD registration, allowing
//the caller total control over process, and decoupling
//parser from any given grammar.
nsIDTD* theDTD=0;
NS_NewNavHTMLDTD(&theDTD);
parser->RegisterDTD(theDTD);
parser->RegisterDTD(theDTD);
parser->SetContentSink(sink);
parser->BeginParse(aURL);
}

168
mozilla/parser/htmlparser/src/CNavDTD.cpp
View File

@@ -40,6 +40,7 @@
#include "nsScanner.h"
#include "nsParserTypes.h"
#include "nsVoidArray.h"
#include "nsTokenHandler.h"
#include "prenv.h" //this is here for debug reasons...
#include "prtypes.h" //this is here for debug reasons...
@@ -61,6 +62,7 @@ static const char* kNullFilename= "Error: Null filename given";
static const char* kNullTokenizer = "Error: Unable to construct tokenizer";
static const char* kNullToken = "Error: Null token given";
static const char* kInvalidTagStackPos = "Error: invalid tag stack position";
static const char* kHTMLTextContentType = "text/html";
static nsAutoString gEmpty;
@@ -78,7 +80,8 @@ static char gStyleTags[]={
eHTMLTag_a, eHTMLTag_bold, eHTMLTag_big,
eHTMLTag_blink, eHTMLTag_cite, eHTMLTag_em,
eHTMLTag_font, eHTMLTag_italic, eHTMLTag_kbd,
eHTMLTag_small, eHTMLTag_spell, eHTMLTag_strike,
eHTMLTag_s, eHTMLTag_small,
eHTMLTag_spell, eHTMLTag_strike,
eHTMLTag_strong, eHTMLTag_sub, eHTMLTag_sup,
eHTMLTag_tt, eHTMLTag_u, eHTMLTag_var,
0};
@@ -215,9 +218,10 @@ void CNavDTD::InitializeDefaultTokenHandlers() {
class CNavTokenDeallocator: public nsDequeFunctor{
public:
virtual void operator()(void* anObject) {
virtual void* operator()(void* anObject) {
CToken* aToken = (CToken*)anObject;
delete aToken;
return 0;
}
};
@@ -277,6 +281,20 @@ void CNavDTD::SetDTDDebug(nsIDTDDebug * aDTDDebug)
NS_ADDREF(mDTDDebug);
}
/**
* This method gets called by the parser to determine if this DTD can handle
* the requested process for the requested content type.
*
* @update gess6/24/98
* @param
* @return TRUE if the DTD can handle the process for this type; FALSE otherwise.
*/
PRBool CNavDTD::IsCapableOf(eProcessType aProcessType, nsString& aContentType,PRInt32 aVersion){
PRBool result=aContentType.Equals(kHTMLTextContentType);
return result;
}
/**
*
* @update gess5/18/98
@@ -337,10 +355,10 @@ PRInt32 CNavDTD::HandleToken(CToken* aToken){
if(aToken) {
CHTMLToken* theToken= (CHTMLToken*)(aToken);
eHTMLTokenTypes theType=eHTMLTokenTypes(theToken->GetTokenType());
CTokenHandler* aHandler=GetTokenHandler(theType);
CITokenHandler* theHandler=GetTokenHandler(theType);
if(aHandler) {
result=(*aHandler)(theToken,this);
if(theHandler) {
result=(*theHandler)(theToken,this);
if (mDTDDebug)
mDTDDebug->Verify(this, mParser, mContextStackPos, mContextStack, mFilename);
}
@@ -365,10 +383,10 @@ PRInt32 CNavDTD::HandleToken(CToken* aToken){
* @param aNode -- CParserNode representing this start token
* @return PR_TRUE if all went well; PR_FALSE if error occured
*/
PRInt32 CNavDTD::HandleDefaultStartToken(CToken* aToken,eHTMLTags aChildTag,nsCParserNode& aNode) {
PRInt32 CNavDTD::HandleDefaultStartToken(CToken* aToken,eHTMLTags aChildTag,nsIParserNode& aNode) {
NS_PRECONDITION(0!=aToken,kNullToken);
eHTMLTags parentTag=(eHTMLTags)GetTopNode();
eHTMLTags parentTag=GetTopNode();
PRInt32 result=kNoError;
PRBool contains=CanContain(parentTag,aChildTag);
@@ -487,30 +505,7 @@ PRInt32 CNavDTD::HandleStartToken(CToken* aToken) {
break;
case eHTMLTag_script:
{
PRInt32 pos=GetTopmostIndexOf(eHTMLTag_body);
nsCParserNode theNode((CHTMLToken*)aToken);
if (kNotFound == pos) {
// We're in the HEAD
result=OpenHead(theNode);
if(kNoError==result) {
mParser->CollectSkippedContent(attrNode,theCount);
if(kNoError==result) {
result=AddLeaf(attrNode);
if(kNoError==result)
result=CloseHead(theNode);
}
}
}
else {
// We're in the BODY
mParser->CollectSkippedContent(attrNode,theCount);
if(kNoError==result) {
result=AddLeaf(attrNode);
}
}
break;
}
result=HandleScriptToken(st); break;
case eHTMLTag_head:
break; //ignore head tags...
@@ -570,7 +565,7 @@ PRInt32 CNavDTD::HandleEndToken(CToken* aToken) {
/*
if(0!=strchr(gStyleTags,tokenTagType)){
eHTMLTags topTag=(eHTMLTags)GetTopNode();
eHTMLTags topTag=GetTopNode();
if(0!=strchr(gStyleTags,topTag)){
tokenTagType=topTag;
}
@@ -704,10 +699,32 @@ PRInt32 CNavDTD::HandleAttributeToken(CToken* aToken) {
*/
PRInt32 CNavDTD::HandleScriptToken(CToken* aToken) {
NS_PRECONDITION(0!=aToken,kNullToken);
PRInt32 result=kNoError;
PRInt32 pos=GetTopmostIndexOf(eHTMLTag_body);
nsCParserNode theNode((CHTMLToken*)aToken);
PRInt32 theCount=0;
PRInt32 result=mParser->CollectSkippedContent(theNode,theCount);
nsCParserNode attrNode((CHTMLToken*)aToken);
PRInt32 attrCount=aToken->GetAttributeCount();
if (kNotFound == pos) {
// We're in the HEAD
result=OpenHead(theNode);
if(kNoError==result) {
mParser->CollectSkippedContent(attrNode,attrCount);
if(kNoError==result) {
result=AddLeaf(attrNode);
if(kNoError==result)
result=CloseHead(theNode);
}
}
}
else {
// We're in the BODY
mParser->CollectSkippedContent(attrNode,attrCount);
if(kNoError==result) {
result=AddLeaf(attrNode);
}
}
return result;
}
@@ -753,8 +770,8 @@ void CNavDTD::DeleteTokenHandlers(void) {
* @param aTagType type of tag to be handled
* @return valid tag handler (if found) or null
*/
CTokenHandler* CNavDTD::GetTokenHandler(eHTMLTokenTypes aType) const {
CTokenHandler* result=0;
CITokenHandler* CNavDTD::GetTokenHandler(eHTMLTokenTypes aType) const {
CITokenHandler* result=0;
if((aType>0) && (aType<eToken_last)) {
result=mTokenHandlers[aType];
}
@@ -771,13 +788,13 @@ CTokenHandler* CNavDTD::GetTokenHandler(eHTMLTokenTypes aType) const {
* @param
* @return
*/
CTokenHandler* CNavDTD::AddTokenHandler(CTokenHandler* aHandler) {
CITokenHandler* CNavDTD::AddTokenHandler(CITokenHandler* aHandler) {
NS_ASSERTION(0!=aHandler,"Error: Null handler");
if(aHandler) {
eHTMLTokenTypes type=(eHTMLTokenTypes)aHandler->GetTokenType();
if(type<eToken_last) {
CTokenHandler* old=mTokenHandlers[type];
CITokenHandler* old=mTokenHandlers[type];
mTokenHandlers[type]=aHandler;
}
else {
@@ -836,8 +853,8 @@ PRBool CNavDTD::CanContainStyles(eHTMLTags aParent) const {
}
/**
* This method is called to determine whether or not a tag
* of one type can contain a tag of another type.
* This method is called to determine whether or not a
* form-type tag can contain a tag of another form-type tag.
*
* @update gess 4/8/98
* @param aParent -- tag enum of parent container
@@ -845,7 +862,18 @@ PRBool CNavDTD::CanContainStyles(eHTMLTags aParent) const {
* @return PR_TRUE if parent can contain child
*/
PRBool CNavDTD::CanContainFormElement(eHTMLTags aParent,eHTMLTags aChild) const {
PRBool result=(mParser) ? HasOpenContainer(eHTMLTag_form) : PR_FALSE;
PRBool result=PR_FALSE;
if(mParser && HasOpenContainer(eHTMLTag_form)) {
eHTMLTags topTag=GetTopNode();
switch(aChild) {
case eHTMLTag_option:
result=PRBool(eHTMLTag_select==topTag); break;
default:
result=PR_TRUE;
break;
} //switch
}//if
return result;
}
@@ -880,7 +908,8 @@ PRBool CNavDTD::CanContain(PRInt32 aParent,PRInt32 aChild) {
eHTMLTag_map, eHTMLTag_menu, eHTMLTag_newline, eHTMLTag_nobr,
eHTMLTag_noframes, eHTMLTag_noscript,
eHTMLTag_object, eHTMLTag_ol, eHTMLTag_paragraph, eHTMLTag_pre,
eHTMLTag_quotation, eHTMLTag_strike, eHTMLTag_samp, eHTMLTag_script,
eHTMLTag_quotation, eHTMLTag_s, eHTMLTag_strike,
eHTMLTag_samp, eHTMLTag_script,
eHTMLTag_select, eHTMLTag_small, eHTMLTag_span, eHTMLTag_strong,
eHTMLTag_sub, eHTMLTag_sup, eHTMLTag_table, eHTMLTag_text,
@@ -901,7 +930,8 @@ PRBool CNavDTD::CanContain(PRInt32 aParent,PRInt32 aChild) {
eHTMLTag_label, eHTMLTag_map, eHTMLTag_newline, eHTMLTag_nobr,
eHTMLTag_object, eHTMLTag_paragraph,
eHTMLTag_quotation, eHTMLTag_strike, eHTMLTag_samp, eHTMLTag_script,
eHTMLTag_quotation, eHTMLTag_s, eHTMLTag_strike,
eHTMLTag_samp, eHTMLTag_script,
eHTMLTag_select, eHTMLTag_small, eHTMLTag_span, eHTMLTag_strong,
eHTMLTag_sub, eHTMLTag_sup, eHTMLTag_text, eHTMLTag_textarea,
eHTMLTag_tt, eHTMLTag_u, eHTMLTag_userdefined, eHTMLTag_var,
@@ -925,7 +955,8 @@ PRBool CNavDTD::CanContain(PRInt32 aParent,PRInt32 aChild) {
eHTMLTag_noframes,
eHTMLTag_noscript, eHTMLTag_object, eHTMLTag_paragraph, eHTMLTag_pre,
eHTMLTag_quotation, eHTMLTag_strike, eHTMLTag_samp, eHTMLTag_small,
eHTMLTag_quotation, eHTMLTag_s, eHTMLTag_strike,
eHTMLTag_samp, eHTMLTag_small,
eHTMLTag_span, eHTMLTag_strong, eHTMLTag_sub, eHTMLTag_sup,
eHTMLTag_td, eHTMLTag_text,
@@ -944,9 +975,9 @@ PRBool CNavDTD::CanContain(PRInt32 aParent,PRInt32 aChild) {
}
switch((eHTMLTags)aParent) {
case eHTMLTag_address:
result=PRBool(0!=strchr(gTagSet2,aChild));
break;
result=PRBool(0!=strchr(gTagSet2,aChild)); break;
case eHTMLTag_applet:
{
@@ -957,7 +988,8 @@ PRBool CNavDTD::CanContain(PRInt32 aParent,PRInt32 aChild) {
eHTMLTag_em, eHTMLTag_font, eHTMLTag_italic, eHTMLTag_iframe,
eHTMLTag_img, eHTMLTag_input, eHTMLTag_kbd, eHTMLTag_label,
eHTMLTag_map, eHTMLTag_object, eHTMLTag_param, eHTMLTag_quotation,
eHTMLTag_samp, eHTMLTag_script, eHTMLTag_select, eHTMLTag_small,
eHTMLTag_s, eHTMLTag_samp, eHTMLTag_script,
eHTMLTag_select, eHTMLTag_small,
eHTMLTag_span, eHTMLTag_strike, eHTMLTag_strong, eHTMLTag_sub,
eHTMLTag_sup, eHTMLTag_textarea, eHTMLTag_tt, eHTMLTag_u,
eHTMLTag_var,0};
@@ -1288,6 +1320,7 @@ PRBool CNavDTD::CanOmit(eHTMLTags aParent,eHTMLTags aChild) const {
case eHTMLTag_input: case eHTMLTag_isindex:
case eHTMLTag_label: case eHTMLTag_legend:
case eHTMLTag_select: case eHTMLTag_textarea:
case eHTMLTag_option:
if(PR_FALSE==HasOpenContainer(eHTMLTag_form))
result=PR_TRUE;
break;
@@ -1379,7 +1412,7 @@ PRBool CNavDTD::CanOmitEndTag(eHTMLTags aParent,eHTMLTags aChild) const {
break;
default:
if(IsCompatibleStyleTag(aChild,(eHTMLTags)GetTopNode()))
if(IsCompatibleStyleTag(aChild,GetTopNode()))
result=PR_FALSE;
else result=(!HasOpenContainer(aChild));
break;
@@ -1587,33 +1620,6 @@ PRBool CNavDTD::BackwardPropagate(nsString& aVector,eHTMLTags aParentTag,eHTMLTa
*/
PRInt32 CNavDTD::DidOpenContainer(eHTMLTags aTag,PRBool /*anExplicitOpen*/){
PRInt32 result=0;
switch (aTag) {
case eHTMLTag_a:
case eHTMLTag_bold:
case eHTMLTag_big:
case eHTMLTag_blink:
case eHTMLTag_cite:
case eHTMLTag_em:
case eHTMLTag_font:
case eHTMLTag_italic:
case eHTMLTag_kbd:
case eHTMLTag_small:
case eHTMLTag_spell:
case eHTMLTag_strike:
case eHTMLTag_strong:
case eHTMLTag_sub:
case eHTMLTag_sup:
case eHTMLTag_tt:
case eHTMLTag_u:
case eHTMLTag_var:
break;
default:
break;
}
return result;
}
@@ -1704,7 +1710,7 @@ PRInt32 CNavDTD::OpenTransientStyles(eHTMLTags aTag){
if(0==strchr(gWhitespaceTags,aTag)){
PRInt32 pos=0;
eHTMLTags parentTag=(eHTMLTags)GetTopNode();
eHTMLTags parentTag=GetTopNode();
if(CanContainStyles(parentTag)) {
for(pos=0;pos<mStyleStackPos;pos++) {
eHTMLTags theTag=(eHTMLTags)(int)mStyleStack[pos];
@@ -1832,7 +1838,7 @@ PRInt32 CNavDTD::OpenBody(const nsIParserNode& aNode){
NS_PRECONDITION(mContextStackPos >= 0, kInvalidTagStackPos);
PRInt32 result=kNoError;
eHTMLTags topTag=(eHTMLTags)CNavDTD::GetTopNode();
eHTMLTags topTag=GetTopNode();
if(eHTMLTag_html!=topTag) {
@@ -2126,7 +2132,7 @@ PRInt32 CNavDTD::CloseContainersTo(eHTMLTags aTag,PRBool aUpdateStyles){
return CloseContainersTo(pos,aTag,aUpdateStyles);
}
eHTMLTags theTopTag=(eHTMLTags)GetTopNode();
eHTMLTags theTopTag=GetTopNode();
if(IsCompatibleStyleTag(aTag,theTopTag)) {
//if you're here, it's because we're trying to close one style tag,
//but a different one is actually open. Because this is NAV4x
@@ -2263,13 +2269,13 @@ PRInt32 CNavDTD::CreateContextStackFor(eHTMLTags aChildTag){
*/
PRInt32 CNavDTD::ReduceContextStackFor(eHTMLTags aChildTag){
PRInt32 result=kNoError;
eHTMLTags topTag=(eHTMLTags)GetTopNode();
eHTMLTags topTag=GetTopNode();
while( (topTag!=kNotFound) &&
(PR_FALSE==CanContain(topTag,aChildTag)) &&
(PR_FALSE==CanContainIndirect(topTag,aChildTag))) {
CloseTopmostContainer();
topTag=(eHTMLTags)GetTopNode();
topTag=GetTopNode();
}
return result;
}
@@ -2297,6 +2303,7 @@ PRInt32 CNavDTD::UpdateStyleStackForOpenTag(eHTMLTags aTag,eHTMLTags anActualTag
case eHTMLTag_font:
case eHTMLTag_italic:
case eHTMLTag_kbd:
case eHTMLTag_s:
case eHTMLTag_small:
case eHTMLTag_spell:
case eHTMLTag_strike:
@@ -2346,6 +2353,7 @@ PRInt32 CNavDTD::UpdateStyleStackForCloseTag(eHTMLTags aTag,eHTMLTags anActualTa
case eHTMLTag_kbd:
case eHTMLTag_small:
case eHTMLTag_spell:
case eHTMLTag_s:
case eHTMLTag_strike:
case eHTMLTag_strong:
case eHTMLTag_sub:

31
mozilla/parser/htmlparser/src/CNavDTD.h
View File

@@ -28,22 +28,22 @@
#define NS_NAVHTMLDTD__
#include "nsIDTD.h"
#include "nsISupports.h"
#include "nsHTMLTokens.h"
#include "nshtmlpars.h"
#include "nsParserNode.h"
#include "nsTokenHandler.h"
#include "nsVoidArray.h"
#include "nsDeque.h"
#define NS_INAVHTML_DTD_IID \
{0x5c5cce40, 0xcfd6, 0x11d1, \
{0xaa, 0xda, 0x00, 0x80, 0x5f, 0x8a, 0x3e, 0x14}}
class nsParser;
class nsIHTMLContentSink;
class nsIDTDDebug;
class nsIParserNode;
class CITokenHandler;
class nsParser;
class CNavDTD : public nsIDTD {
@@ -70,13 +70,21 @@ class CNavDTD : public nsIDTD {
*/
virtual ~CNavDTD();
/**
*
* @update gess7/1/98
* @param
* @return
*/
virtual PRBool IsCapableOf(eProcessType aProcessType, nsString& aString, PRInt32 aVersion);
/**
*
* @update jevering6/23/98
* @param
* @return
*/
virtual void SetDTDDebug(nsIDTDDebug * aDTDDebug);
virtual void SetDTDDebug(nsIDTDDebug * aDTDDebug);
/**
*
@@ -292,7 +300,7 @@ class CNavDTD : public nsIDTD {
* @param aNode is a node be updated with info from given token
* @return TRUE if the token was handled.
*/
PRInt32 HandleDefaultStartToken(CToken* aToken,eHTMLTags aChildTag,nsCParserNode& aNode);
PRInt32 HandleDefaultStartToken(CToken* aToken,eHTMLTags aChildTag,nsIParserNode& aNode);
/**
* This method gets called when an end token has been consumed and needs
@@ -359,7 +367,7 @@ class CNavDTD : public nsIDTD {
private:
protected:
/**
* Causes token handlers to be registered for this parser.
@@ -373,13 +381,13 @@ private:
* DEPRECATED
* @update gess5/11/98
*/
CTokenHandler* GetTokenHandler(eHTMLTokenTypes aType) const;
CITokenHandler* GetTokenHandler(eHTMLTokenTypes aType) const;
/**
* DEPRECATED
* @update gess5/11/98
*/
CTokenHandler* AddTokenHandler(CTokenHandler* aHandler);
CITokenHandler* AddTokenHandler(CITokenHandler* aHandler);
/**
* DEPRECATED
@@ -679,7 +687,7 @@ protected:
nsParser* mParser;
nsIHTMLContentSink* mSink;
CTokenHandler* mTokenHandlers[eToken_last];
CITokenHandler* mTokenHandlers[eToken_last];
nsVoidArray mLeafBits;
nsVoidArray mContextStack;
@@ -691,9 +699,10 @@ protected:
PRBool mHasOpenMap;
nsDeque mTokenDeque;
char* mFilename;
nsIDTDDebug* mDTDDebug;
nsIDTDDebug* mDTDDebug;
};
extern NS_HTMLPARS nsresult NS_NewNavHTMLDTD(nsIDTD** aInstancePtrResult);
#endif

1931
mozilla/parser/htmlparser/src/COtherDTD.cpp
View File

File diff suppressed because it is too large Load Diff

112
mozilla/parser/htmlparser/src/COtherDTD.h
View File

@@ -26,13 +26,7 @@
#ifndef NS_OTHERHTMLDTD__
#define NS_OTHERHTMLDTD__
#include "nsIDTD.h"
#include "nsHTMLTokens.h"
#include "nshtmlpars.h"
#include "nsParserNode.h"
#include "nsTokenHandler.h"
#include "nsDeque.h"
#include "nsVoidArray.h"
#include "CNavDTD.h"
#define NS_IOtherHTML_DTD_IID \
@@ -40,11 +34,10 @@
{0x80, 0x22, 0x00, 0x60, 0x8, 0x14, 0x98, 0x89}}
class nsParser;
class nsIHTMLContentSink;
class nsIDTDDebug;
//class nsParser;
//class nsIHTMLContentSink;
class COtherDTD : public nsIDTD {
class COtherDTD : public CNavDTD {
public:
@@ -69,14 +62,8 @@ class COtherDTD : public nsIDTD {
*/
virtual ~COtherDTD();
/**
*
* @update jevering6/23/98
* @param
* @return
*/
virtual void SetDTDDebug(nsIDTDDebug * aDTDDebug);
virtual PRBool IsCapableOf(eProcessType aProcessType, nsString& aString,PRInt32 aVersion);
/**
*
* @update gess5/18/98
@@ -101,15 +88,6 @@ class COtherDTD : public nsIDTD {
*/
virtual PRInt32 HandleToken(CToken* aToken);
/**
*
*
* @update gess 3/25/98
* @param
* @return
*/
virtual void SetParser(nsIParser* aParser);
/**
* Cause the tokenizer to consume the next token, and
* return an error result.
@@ -137,14 +115,6 @@ class COtherDTD : public nsIDTD {
*/
virtual void WillInterruptParse(void);
/**
* Select given content sink into parser for parser output
* @update gess5/11/98
* @param aSink is the new sink to be used by parser
* @return old sink, or NULL
*/
virtual nsIContentSink* SetContentSink(nsIContentSink* aSink);
/**
* This method is called to determine whether or not a tag
* of one type can contain a tag of another type.
@@ -239,30 +209,6 @@ class COtherDTD : public nsIDTD {
*/
virtual PRInt32 DidOpenContainer(eHTMLTags aTag,PRBool anExplicitOpen);
/**
* Ask parser if a given container is open ANYWHERE on stack
* @update gess5/11/98
* @param id of container you want to test for
* @return TRUE if the given container type is open -- otherwise FALSE
*/
virtual PRBool HasOpenContainer(eHTMLTags aContainer) const;
/**
* Retrieve the tag type of the topmost item on context vector stack
* @update gess5/11/98
* @return tag type (may be unknown)
*/
virtual eHTMLTags GetTopNode() const;
/**
* Finds the topmost occurance of given tag within context vector stack.
* @update gess5/11/98
* @param tag to be found
* @return index of topmost tag occurance -- may be -1 (kNotFound).
*/
virtual PRInt32 GetTopmostIndexOf(eHTMLTags aTag) const;
/**
*
* @update gess6/4/98
@@ -291,7 +237,7 @@ class COtherDTD : public nsIDTD {
* @param aNode is a node be updated with info from given token
* @return TRUE if the token was handled.
*/
PRInt32 HandleDefaultStartToken(CToken* aToken,eHTMLTags aChildTag,nsCParserNode& aNode);
PRInt32 HandleDefaultStartToken(CToken* aToken,eHTMLTags aChildTag,nsIParserNode& aNode);
/**
* This method gets called when an end token has been consumed and needs
@@ -358,33 +304,6 @@ class COtherDTD : public nsIDTD {
private:
/**
* Causes token handlers to be registered for this parser.
* DO NOT CALL THIS! IT'S DEPRECATED!
* @update gess5/11/98
*/
void InitializeDefaultTokenHandlers();
/**
* DEPRECATED
* @update gess5/11/98
*/
CTokenHandler* GetTokenHandler(eHTMLTokenTypes aType) const;
/**
* DEPRECATED
* @update gess5/11/98
*/
CTokenHandler* AddTokenHandler(CTokenHandler* aHandler);
/**
* DEPRECATED
* @update gess5/11/98
*/
void DeleteTokenHandlers(void);
//*************************************************
//these cover methods mimic the sink, and are used
@@ -673,23 +592,8 @@ protected:
PRBool CanContainFormElement(eHTMLTags aParent,eHTMLTags aChild) const;
nsParser* mParser;
nsIHTMLContentSink* mSink;
CTokenHandler* mTokenHandlers[eToken_last];
nsVoidArray mLeafBits;
nsVoidArray mContextStack;
PRInt32 mContextStackPos;
nsVoidArray mStyleStack;
PRInt32 mStyleStackPos;
PRBool mHasOpenForm;
PRBool mHasOpenMap;
nsDeque mTokenDeque;
char* mFilename;
nsIDTDDebug* mDTDDebug;
};
extern NS_HTMLPARS nsresult NS_NewOtherHTMLDTD(nsIDTD** aInstancePtrResult);
#endif

3
mozilla/parser/htmlparser/src/Makefile
View File

@@ -42,8 +42,11 @@ EXPORTS = \
nsHTMLTokens.h \
nsIParserNode.h \
nsIParser.h \
nsIDTD.h \
nsIParserFilter.h \
nsToken.h \
CNavDTD.h \
COtherDTD.h \
$(NULL)
MODULE = raptor

3
mozilla/parser/htmlparser/src/makefile.win
View File

@@ -32,7 +32,8 @@ CPPSRCS=nsHTMLContentSink.cpp \
EXPORTS=nshtmlpars.h nsIContentSink.h nsIHTMLContentSink.h \
nsHTMLTokens.h nsIParserNode.h nsIParser.h nsToken.h \
nsIDTDDebug.h nsIParserFilter.h
nsIDTDDebug.h nsIParserFilter.h \
CNavDTD.h COtherDTD.h nsIDTD.h
CPP_OBJS=.\$(OBJDIR)\nsHTMLContentSink.obj \
.\$(OBJDIR)\CNavDTD.obj \

244
mozilla/parser/htmlparser/src/nsHTMLTokens.cpp
View File

@@ -45,8 +45,8 @@ static const char* kNullScanner = "Error: Scanner is null.";
const PRInt32 kMAXNAMELEN=10;
struct StrToUnicodeStruct
{
char fName[kMAXNAMELEN+1];
PRInt32 fValue;
char mName[kMAXNAMELEN+1];
PRInt32 mValue;
};
@@ -96,8 +96,9 @@ static StrToUnicodeStruct gStrToUnicodeTable[] =
struct HTMLTagEntry {
char fName[12];
eHTMLTags fTagID;
char mName[12];
eHTMLTags mTagID;
PRInt16 mUnused;
};
// KEEP THIS LIST SORTED!
@@ -106,103 +107,103 @@ struct HTMLTagEntry {
// the binary search code above will break!
HTMLTagEntry gHTMLTagTable[] =
{
{"!!UNKNOWN", eHTMLTag_unknown},
{"!DOCTYPE", eHTMLTag_doctype}, {"A", eHTMLTag_a},
{"ACRONYM", eHTMLTag_acronym}, {"ADDRESS", eHTMLTag_address},
{"APPLET", eHTMLTag_applet}, {"AREA", eHTMLTag_area},
{"!!UNKNOWN", eHTMLTag_unknown,0},
{"!DOCTYPE", eHTMLTag_doctype,0}, {"A", eHTMLTag_a,0},
{"ACRONYM", eHTMLTag_acronym,0}, {"ADDRESS", eHTMLTag_address,0},
{"APPLET", eHTMLTag_applet,0}, {"AREA", eHTMLTag_area,0},
{"B", eHTMLTag_bold}, {"BASE", eHTMLTag_base},
{"BASEFONT", eHTMLTag_basefont}, {"BDO", eHTMLTag_bdo},
{"BIG", eHTMLTag_big}, {"BLINK", eHTMLTag_blink},
{"BLOCKQUOTE", eHTMLTag_blockquote}, {"BODY", eHTMLTag_body},
{"BR", eHTMLTag_br}, {"BUTTON", eHTMLTag_button},
{"B", eHTMLTag_bold,0}, {"BASE", eHTMLTag_base,0},
{"BASEFONT", eHTMLTag_basefont,0}, {"BDO", eHTMLTag_bdo,0},
{"BIG", eHTMLTag_big,0}, {"BLINK", eHTMLTag_blink,0},
{"BLOCKQUOTE", eHTMLTag_blockquote,0}, {"BODY", eHTMLTag_body,0},
{"BR", eHTMLTag_br,0}, {"BUTTON", eHTMLTag_button,0},
{"CAPTION", eHTMLTag_caption}, {"CENTER", eHTMLTag_center},
{"CERTIFICATE", eHTMLTag_certificate},
{"CITE", eHTMLTag_cite}, {"CODE", eHTMLTag_code},
{"COL", eHTMLTag_col}, {"COLGROUP", eHTMLTag_colgroup},
{"COMMENT", eHTMLTag_comment},
{"CAPTION", eHTMLTag_caption,0}, {"CENTER", eHTMLTag_center,0},
{"CERTIFICATE", eHTMLTag_certificate,0},
{"CITE", eHTMLTag_cite,0}, {"CODE", eHTMLTag_code,0},
{"COL", eHTMLTag_col,0}, {"COLGROUP", eHTMLTag_colgroup,0},
{"COMMENT", eHTMLTag_comment,0},
{"DD", eHTMLTag_dd}, {"DEL", eHTMLTag_del},
{"DFN", eHTMLTag_dfn}, {"DIR", eHTMLTag_dir},
{"DIV", eHTMLTag_div}, {"DL", eHTMLTag_dl},
{"DT", eHTMLTag_dt},
{"DD", eHTMLTag_dd,0}, {"DEL", eHTMLTag_del,0},
{"DFN", eHTMLTag_dfn,0}, {"DIR", eHTMLTag_dir,0},
{"DIV", eHTMLTag_div,0}, {"DL", eHTMLTag_dl,0},
{"DT", eHTMLTag_dt,0},
{"EM", eHTMLTag_em}, {"EMBED", eHTMLTag_embed},
{"ENTITY", eHTMLTag_entity}, //a pseudo tag
{"EM", eHTMLTag_em,0}, {"EMBED", eHTMLTag_embed,0},
{"ENTITY", eHTMLTag_entity,0}, //a pseudo tag
{"FIELDSET", eHTMLTag_fieldset}, {"FONT", eHTMLTag_font},
{"FOOTER", eHTMLTag_footer}, {"FORM", eHTMLTag_form},
{"FRAME", eHTMLTag_frame}, {"FRAMESET", eHTMLTag_frameset},
{"FIELDSET", eHTMLTag_fieldset,0}, {"FONT", eHTMLTag_font,0},
{"FOOTER", eHTMLTag_footer,0}, {"FORM", eHTMLTag_form,0},
{"FRAME", eHTMLTag_frame,0}, {"FRAMESET", eHTMLTag_frameset,0},
{"H1", eHTMLTag_h1}, {"H2", eHTMLTag_h2},
{"H3", eHTMLTag_h3}, {"H4", eHTMLTag_h4},
{"H5", eHTMLTag_h5}, {"H6", eHTMLTag_h6},
{"HEAD", eHTMLTag_head}, {"HEADER", eHTMLTag_header},
{"HR", eHTMLTag_hr}, {"HTML", eHTMLTag_html},
{"H1", eHTMLTag_h1,0}, {"H2", eHTMLTag_h2,0},
{"H3", eHTMLTag_h3,0}, {"H4", eHTMLTag_h4,0},
{"H5", eHTMLTag_h5,0}, {"H6", eHTMLTag_h6,0},
{"HEAD", eHTMLTag_head,0}, {"HEADER", eHTMLTag_header,0},
{"HR", eHTMLTag_hr,0}, {"HTML", eHTMLTag_html,0},
{"I", eHTMLTag_italic}, {"IFRAME", eHTMLTag_iframe},
{"ILAYER", eHTMLTag_ilayer}, {"IMG", eHTMLTag_img},
{"INPUT", eHTMLTag_input}, {"INS", eHTMLTag_ins},
{"ISINDEX", eHTMLTag_isindex},
{"I", eHTMLTag_italic,0}, {"IFRAME", eHTMLTag_iframe,0},
{"ILAYER", eHTMLTag_ilayer,0}, {"IMG", eHTMLTag_img,0},
{"INPUT", eHTMLTag_input,0}, {"INS", eHTMLTag_ins,0},
{"ISINDEX", eHTMLTag_isindex,0},
{"KBD", eHTMLTag_kbd}, {"KEYGEN", eHTMLTag_keygen},
{"KBD", eHTMLTag_kbd,0}, {"KEYGEN", eHTMLTag_keygen,0},
{"LABEL", eHTMLTag_label}, {"LAYER", eHTMLTag_layer},
{"LEGEND", eHTMLTag_legend}, {"LI", eHTMLTag_listitem},
{"LINK", eHTMLTag_link}, {"LISTING", eHTMLTag_listing},
{"LABEL", eHTMLTag_label,0}, {"LAYER", eHTMLTag_layer,0},
{"LEGEND", eHTMLTag_legend,0}, {"LI", eHTMLTag_listitem,0},
{"LINK", eHTMLTag_link,0}, {"LISTING", eHTMLTag_listing,0},
{"MAP", eHTMLTag_map}, {"MARQUEE", eHTMLTag_marquee},
{"MATH", eHTMLTag_math},
{"MENU", eHTMLTag_menu}, {"META", eHTMLTag_meta},
{"MAP", eHTMLTag_map,0}, {"MARQUEE", eHTMLTag_marquee,0},
{"MATH", eHTMLTag_math,0},
{"MENU", eHTMLTag_menu,0}, {"META", eHTMLTag_meta,0},
{"NEWLINE", eHTMLTag_newline}, {"NOBR", eHTMLTag_nobr},
{"NEWLINE", eHTMLTag_newline,0}, {"NOBR", eHTMLTag_nobr,0},
{"NOEMBED", eHTMLTag_noembed}, {"NOFRAMES", eHTMLTag_noframes},
{"NOLAYER", eHTMLTag_nolayer}, {"NOSCRIPT", eHTMLTag_noscript},
{"NOTE", eHTMLTag_note},
{"NOEMBED", eHTMLTag_noembed,0}, {"NOFRAMES", eHTMLTag_noframes,0},
{"NOLAYER", eHTMLTag_nolayer,0}, {"NOSCRIPT", eHTMLTag_noscript,0},
{"NOTE", eHTMLTag_note,0},
{"OBJECT", eHTMLTag_object}, {"OL", eHTMLTag_ol},
{"OPTION", eHTMLTag_option},
{"OBJECT", eHTMLTag_object,0}, {"OL", eHTMLTag_ol,0},
{"OPTION", eHTMLTag_option,0},
{"P", eHTMLTag_paragraph}, {"PARAM", eHTMLTag_param},
{"PLAINTEXT", eHTMLTag_plaintext},
{"P", eHTMLTag_paragraph,0}, {"PARAM", eHTMLTag_param,0},
{"PLAINTEXT", eHTMLTag_plaintext,0},
{"PRE", eHTMLTag_pre},
{"PRE", eHTMLTag_pre,0},
{"Q", eHTMLTag_quotation},
{"Q", eHTMLTag_quotation,0},
{"S", eHTMLTag_strike}, {"SAMP", eHTMLTag_samp},
{"SCRIPT", eHTMLTag_script}, {"SELECT", eHTMLTag_select},
{"SERVER", eHTMLTag_server}, {"SMALL", eHTMLTag_small},
{"SPACER", eHTMLTag_spacer},
{"SPAN", eHTMLTag_span}, {"SPELL", eHTMLTag_spell},
{"STRIKE", eHTMLTag_strike},
{"STRONG", eHTMLTag_strong}, {"STYLE", eHTMLTag_style},
{"SUB", eHTMLTag_sub}, {"SUP", eHTMLTag_sup},
{"S", eHTMLTag_s,0}, {"SAMP", eHTMLTag_samp,0},
{"SCRIPT", eHTMLTag_script,0}, {"SELECT", eHTMLTag_select,0},
{"SERVER", eHTMLTag_server,0}, {"SMALL", eHTMLTag_small,0},
{"SPACER", eHTMLTag_spacer,0},
{"SPAN", eHTMLTag_span,0}, {"SPELL", eHTMLTag_spell,0},
{"STRIKE", eHTMLTag_strike,0},
{"STRONG", eHTMLTag_strong,0}, {"STYLE", eHTMLTag_style,0},
{"SUB", eHTMLTag_sub,0}, {"SUP", eHTMLTag_sup,0},
{"TABLE", eHTMLTag_table}, {"TBODY", eHTMLTag_tbody},
{"TD", eHTMLTag_td},
{"TABLE", eHTMLTag_table,0}, {"TBODY", eHTMLTag_tbody,0},
{"TD", eHTMLTag_td,0},
{"TEXT", eHTMLTag_text},
{"TEXT", eHTMLTag_text,0},
{"TEXTAREA", eHTMLTag_textarea},
{"TFOOT", eHTMLTag_tfoot}, {"TH", eHTMLTag_th},
{"THEAD", eHTMLTag_thead}, {"TITLE", eHTMLTag_title},
{"TR", eHTMLTag_tr}, {"TT", eHTMLTag_tt},
{"TEXTAREA", eHTMLTag_textarea,0},
{"TFOOT", eHTMLTag_tfoot,0}, {"TH", eHTMLTag_th,0},
{"THEAD", eHTMLTag_thead,0}, {"TITLE", eHTMLTag_title,0},
{"TR", eHTMLTag_tr,0}, {"TT", eHTMLTag_tt,0},
{"U", eHTMLTag_u}, {"UL", eHTMLTag_ul},
{"USERDEF", eHTMLTag_userdefined},
{"VAR", eHTMLTag_var}, {"WBR", eHTMLTag_wbr},
{"WS", eHTMLTag_whitespace},
{"U", eHTMLTag_u,0}, {"UL", eHTMLTag_ul,0},
{"VAR", eHTMLTag_var,0}, {"WBR", eHTMLTag_wbr,0},
{"WS", eHTMLTag_whitespace,0},
{"X-USERDEF", eHTMLTag_userdefined,0}, //make sure this is always last!
};
struct HTMLAttrEntry
{
char fName[11];
eHTMLAttributes fAttrID;
char mName[11];
eHTMLAttributes mAttrID;
};
HTMLAttrEntry gHTMLAttributeTable[] =
@@ -1073,21 +1074,20 @@ PRInt32 CEntityToken::TranslateToUnicodeStr(nsString& aString) {
aString.Append(PRUnichar(index));
}
else {
char* cp = mTextValue.ToNewCString();
index=FindEntityIndex(cp);
index=FindEntityIndex(mTextValue);
if(kNotFound!=index) {
PRUnichar ch=gStrToUnicodeTable[index].fValue;
PRUnichar ch=gStrToUnicodeTable[index].mValue;
aString=ch;
}
else {
#ifdef GESS_MACHINE
index=TranslateExtendedEntity(cp,aString);
index=TranslateExtendedEntity(mTextValue,aString);
#endif
}
delete cp;
}
return index;
}
/*
@@ -1102,7 +1102,7 @@ PRBool CEntityToken::VerifyEntityTable(){
PRInt32 i,j;
for(i=1;i<count-1;i++)
{
j=strcmp(gStrToUnicodeTable[i-1].fName,gStrToUnicodeTable[i].fName);
j=strcmp(gStrToUnicodeTable[i-1].mName,gStrToUnicodeTable[i].mName);
if(j>0)
return PR_FALSE;
}
@@ -1119,23 +1119,19 @@ PRBool CEntityToken::VerifyEntityTable(){
* @param aBuflen -- optional string length
* @return integer offset of string in table, or kNotFound
*/
PRInt32 CEntityToken::FindEntityIndex(const char* aBuffer,PRInt32 aBufLen) {
PRInt32 CEntityToken::FindEntityIndex(nsString& aString) {
PRInt32 result=kNotFound;
PRInt32 cnt=sizeof(gStrToUnicodeTable)/sizeof(StrToUnicodeStruct);
PRInt32 low=0;
PRInt32 high=cnt-1;
PRInt32 middle=kNotFound;
if(kNotFound==aBufLen) {
aBufLen=strlen(aBuffer);
}
if (aBuffer && aBufLen && cnt) {
if(cnt) {
while(low<=high)
{
middle=(PRInt32)(low+high)/2;
// result=strncmp(aBuffer,gStrToUnicodeTable[middle].fName,aBufLen);
result=strcmp(aBuffer,gStrToUnicodeTable[middle].fName);
result=aString.Compare(gStrToUnicodeTable[middle].mName);
// result=strcmp(aBuffer,gStrToUnicodeTable[middle].mName);
if (result==0) {
return middle;
}
@@ -1149,6 +1145,46 @@ PRInt32 CEntityToken::FindEntityIndex(const char* aBuffer,PRInt32 aBufLen) {
}
/*
* This method is used to convert from a given string (char*)
* into a entity index (offset within entity table).
*
* @update gess 3/25/98
* @param aBuffer -- string to be converted
* @param aBuflen -- optional string length
* @return integer offset of string in table, or kNotFound
*/
PRInt32 CEntityToken::FindEntityIndexMax(const char* aBuffer,PRInt32 aBufLen) {
PRInt32 result=kNotFound;
PRInt32 cnt=sizeof(gStrToUnicodeTable)/sizeof(StrToUnicodeStruct);
PRInt32 low=0;
PRInt32 high=cnt-1;
PRInt32 middle=kNotFound;
if(aBuffer) {
if(-1==aBufLen) {
aBufLen=strlen(aBuffer);
}
if(aBufLen && cnt) {
while(low<=high)
{
middle=(PRInt32)(low+high)/2;
result=strcmp(aBuffer,gStrToUnicodeTable[middle].mName);
if (result==0) {
return middle;
}
if (result<0) {
high=middle-1;
}
else low=middle+1;
}
}
}
return kNotFound;
}
/*
* This method reduces all text entities into their char
* representation.
@@ -1171,9 +1207,9 @@ PRInt32 CEntityToken::ReduceEntities(nsString& aString) {
if(kNotFound==endpos)
cnt=aString.Length()-1-amppos;
else cnt=endpos-amppos;
PRInt32 index=FindEntityIndex((const char*)&aString[amppos+1],cnt);
PRInt32 index=FindEntityIndexMax((const char*)&aString[amppos+1],cnt);
if(kNotFound!=index) {
aString[amppos]=gStrToUnicodeTable[index].fValue;
aString[amppos]=gStrToUnicodeTable[index].mValue;
aString.Cut(amppos+1,cnt+(endpos!=kNotFound));
}
else offset=amppos+1;
@@ -1316,19 +1352,19 @@ public:
PRInt32 i,j;
for(i=1;i<count-1;i++)
{
j=strcmp(gHTMLTagTable[i-1].fName,gHTMLTagTable[i].fName);
j=strcmp(gHTMLTagTable[i-1].mName,gHTMLTagTable[i].mName);
if(j>0) {
#ifdef VERBOSE_DEBUG
cout << "Tag Table is out of order at " << i << "!" << endl;
#endif
return;
cout << "Tag Table names are out of order at " << i << "!" << endl;
}
if(gHTMLTagTable[i-1].mTagID>=gHTMLTagTable[i].mTagID) {
cout << "Tag table ID's are out of order at " << i << "!" << endl;;
}
}
return;
}
};
/*
* This method accepts a string (and optionally, its length)
* and determines the eHTMLTag (id) value.
@@ -1347,9 +1383,9 @@ eHTMLTags DetermineHTMLTagType(const nsString& aString)
while(low<=high){
middle=(PRInt32)(low+high)/2;
result=aString.Compare(gHTMLTagTable[middle].fName, PR_TRUE);
result=aString.Compare(gHTMLTagTable[middle].mName, PR_TRUE);
if (result==0)
return gHTMLTagTable[middle].fTagID;
return gHTMLTagTable[middle].mTagID;
if (result<0)
high=middle-1;
else low=middle+1;
@@ -1373,9 +1409,9 @@ const char* GetTagName(PRInt32 aTag) {
while(low<=high) {
middle=(PRInt32)(low+high)/2;
if(aTag==gHTMLTagTable[middle].fTagID)
return gHTMLTagTable[middle].fName;
if(aTag<gHTMLTagTable[middle].fTagID)
if(aTag==gHTMLTagTable[middle].mTagID)
return gHTMLTagTable[middle].mName;
if(aTag<gHTMLTagTable[middle].mTagID)
high=middle-1;
else low=middle+1;
}
@@ -1399,7 +1435,7 @@ public:
PRInt32 i,j;
for(i=1;i<count-1;i++)
{
j=strcmp(gHTMLAttributeTable[i-1].fName,gHTMLAttributeTable[i].fName);
j=strcmp(gHTMLAttributeTable[i-1].mName,gHTMLAttributeTable[i].mName);
if(j>0) {
#ifdef VERBOSE_DEBUG
cout << "Attribute table is out of order at " << j << "!" << endl;

26
mozilla/parser/htmlparser/src/nsHTMLTokens.h
View File

@@ -1,3 +1,4 @@
/* -*- Mode: C++; tab-width: 2; indent-tabs-mode: nil; c-basic-offset: 2 -*- */
/*
* The contents of this file are subject to the Netscape Public License
@@ -75,22 +76,24 @@ enum eHTMLTags
eHTMLTag_link, eHTMLTag_listing, eHTMLTag_map, eHTMLTag_marquee,
eHTMLTag_math, eHTMLTag_menu, eHTMLTag_meta, eHTMLTag_newline,
eHTMLTag_nobr,
eHTMLTag_noembed, eHTMLTag_noframes, eHTMLTag_nolayer, eHTMLTag_noscript, //74
eHTMLTag_noembed, eHTMLTag_noframes, eHTMLTag_nolayer, eHTMLTag_noscript, //
eHTMLTag_note, eHTMLTag_object, eHTMLTag_ol,
eHTMLTag_option, eHTMLTag_paragraph, eHTMLTag_param, eHTMLTag_plaintext,
eHTMLTag_pre, eHTMLTag_quotation, eHTMLTag_samp, eHTMLTag_script,
eHTMLTag_pre, eHTMLTag_quotation,
eHTMLTag_s, eHTMLTag_samp, eHTMLTag_script,
eHTMLTag_select, eHTMLTag_server, eHTMLTag_small,
eHTMLTag_spacer, eHTMLTag_span, eHTMLTag_spell,
eHTMLTag_strong, eHTMLTag_style, eHTMLTag_strike, eHTMLTag_sub,
eHTMLTag_sup,
eHTMLTag_table, eHTMLTag_tbody, eHTMLTag_td, //98
eHTMLTag_text, //used for plain text; this is not really a tag.
eHTMLTag_textarea, //100
eHTMLTag_strike, eHTMLTag_strong, eHTMLTag_style,
eHTMLTag_sub, eHTMLTag_sup,
eHTMLTag_table, eHTMLTag_tbody, eHTMLTag_td, //
eHTMLTag_text, //used for plain text; this is not really a tag.
eHTMLTag_textarea, //
eHTMLTag_tfoot,
eHTMLTag_th, eHTMLTag_thead, eHTMLTag_title, eHTMLTag_tr,
eHTMLTag_tt, eHTMLTag_monofont, eHTMLTag_u, eHTMLTag_ul,
eHTMLTag_tt, eHTMLTag_u, eHTMLTag_ul,
eHTMLTag_var, eHTMLTag_wbr, eHTMLTag_whitespace,
eHTMLTag_xmp,
@@ -121,9 +124,9 @@ enum eHTMLAttributes {
PRInt32 ConsumeQuotedString(PRUnichar aChar,nsString& aString,CScanner& aScanner);
PRInt32 ConsumeAttributeText(PRUnichar aChar,nsString& aString,CScanner& aScanner);
PRInt32 FindEntityIndex(const char* aBuffer,PRInt32 aBufLen=-1);
eHTMLTags DetermineHTMLTagType(const nsString& aString);
const char* GetTagName(PRInt32 aTag);
//PRInt32 FindEntityIndex(nsString& aString,PRInt32 aCount=-1);
/**
@@ -219,7 +222,8 @@ class CEntityToken : public CHTMLToken {
virtual PRInt32 Consume(PRUnichar aChar,CScanner& aScanner);
static PRInt32 ConsumeEntity(PRUnichar aChar,nsString& aString,CScanner& aScanner);
static PRInt32 TranslateToUnicodeStr(PRInt32 aValue,nsString& aString);
static PRInt32 FindEntityIndex(const char* aBuffer,PRInt32 aBufLen=-1);
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);

22
mozilla/parser/htmlparser/src/nsIDTD.h
View File

@@ -39,11 +39,23 @@ class CToken;
class nsIContentSink;
class nsIDTDDebug;
class nsIURL;
class nsString;
enum eProcessType {eParsing, eConverting};
enum eAutoDetectResult {eUnknownDetect, eValidDetect, eInvalidDetect};
class nsIDTD : public nsISupports {
public:
/**
*
* @update gess6/24/98
* @param
* @return
*/
virtual ~nsIDTD() {};
/**
* This method is called to determine whether or not a tag
* of one type can contain a tag of another type.
@@ -63,6 +75,14 @@ class nsIDTD : public nsISupports {
*/
virtual nsIContentSink* SetContentSink(nsIContentSink* aSink)=0;
/**
*
* @update gess6/24/98
* @param
* @return
*/
virtual PRBool IsCapableOf(eProcessType aProcessType, nsString& aContentType, PRInt32 aVersion)=0;
/**
*
* @update gess5/18/98
@@ -130,7 +150,7 @@ class nsIDTD : public nsISupports {
* @param
* @return
*/
virtual void SetDTDDebug(nsIDTDDebug * aDTDDebug) = 0;
virtual void SetDTDDebug(nsIDTDDebug * aDTDDebug) = 0;
};

29
mozilla/parser/htmlparser/src/nsIParser.h
View File

@@ -21,7 +21,8 @@
#include "nshtmlpars.h"
#include "nsISupports.h"
#include "nsIStreamListener.h"
#include "nsIDTD.h"
#define NS_IPARSER_IID \
{0x355cbba0, 0xbf7d, 0x11d1, \
@@ -33,7 +34,6 @@ class nsIStreamObserver;
class nsString;
class CToken;
class nsIURL;
class nsIDTD;
class nsIDTDDebug;
/**
@@ -45,11 +45,11 @@ class nsIDTDDebug;
class nsIParser : public nsISupports {
public:
virtual void RegisterDTD(nsIDTD* aDTD)=0;
virtual nsIContentSink* SetContentSink(nsIContentSink* aContentSink)=0;
virtual void SetDTD(nsIDTD* aDTD)=0;
virtual nsIDTD* GetDTD(void)=0;
virtual eAutoDetectResult AutoDetectContentType(nsString& aBuffer)=0;
/**
* Cause the tokenizer to consume the next token, and
@@ -65,14 +65,23 @@ class nsIParser : public nsISupports {
nsIStreamObserver* aListener = nsnull,
nsIDTDDebug * aDTDDebug = 0) = 0;
virtual PRInt32 Parse(nsIURL* aURL,
nsIStreamObserver* aListener = nsnull,
nsIDTDDebug * aDTDDebug = 0) = 0;
/******************************************************************************************
* Parse methods always begin with an input source, and perform conversions
* until you wind up with HTML in your actual content model.
******************************************************************************************/
virtual PRInt32 Parse(nsIURL* aURL,nsIStreamObserver* aListener = nsnull,nsIDTDDebug * aDTDDebug = 0) = 0;
virtual PRInt32 Parse(const char* aFilename)=0;
virtual PRInt32 Parse(nsString& anHTMLString,PRBool appendTokens)=0;
/******************************************************************************************
* Convert methods start input source (of known or unknown form), and perform conversions
* until you wind up with a <i>stream</i> in your target form.
* The internal content model is never effected.
******************************************************************************************/
virtual PRInt32 Convert(nsIURL* aURL,char* aSourceForm,char* aTargetForm,nsIStreamListener* aListener) = 0;
virtual PRInt32 Convert(const char* aFilename,char* aSourceForm,char* aTargetForm)=0;
virtual PRInt32 Convert(nsString& anHTMLString,char* aSourceForm,char* aTargetForm,PRBool appendTokens)=0;
virtual PRInt32 ResumeParse(void)=0;
/**

1
mozilla/parser/htmlparser/src/nsITokenHandler.h
View File

@@ -35,6 +35,7 @@ class CITokenHandler {
public:
virtual PRInt32 operator()(CToken* aToken,nsIDTD* aDTD)=0;
virtual PRInt32 GetTokenType(void)=0;
};

386
mozilla/parser/htmlparser/src/nsParser.cpp
View File

@@ -22,8 +22,6 @@
#include "nsIContentSink.h"
#include "nsString.h"
#include "nsCRT.h"
#include "COtherDTD.h"
#include "CNavDTD.h"
#include "nsScanner.h"
#include "prenv.h" //this is here for debug reasons...
#include "plstr.h"
@@ -31,6 +29,7 @@
#include "nsIInputStream.h"
#include "nsIParserFilter.h"
#include "nsIDTDDebug.h"
#include "nshtmlpars.h"
static NS_DEFINE_IID(kISupportsIID, NS_ISUPPORTS_IID);
static NS_DEFINE_IID(kClassIID, NS_PARSER_IID);
@@ -40,6 +39,7 @@ static NS_DEFINE_IID(kIStreamListenerIID, NS_ISTREAMLISTENER_IID);
static const char* kNullURL = "Error: Null URL given";
static const char* kNullFilename= "Error: Null filename given";
static const char* kNullTokenizer = "Error: Unable to construct tokenizer";
static const char* kHTMLTextContentType = "text/html";
static const int gTransferBufferSize=4096; //size of the buffer used in moving data from iistream
@@ -70,13 +70,62 @@ NS_HTMLPARS nsresult NS_NewParser(nsIParser** aInstancePtrResult)
class CTokenDeallocator: public nsDequeFunctor{
public:
virtual void operator()(void* anObject) {
virtual void* operator()(void* anObject) {
CToken* aToken = (CToken*)anObject;
delete aToken;
return 0;
}
};
CTokenDeallocator gTokenKiller;
CTokenDeallocator gTokenDeallocator;
class CDTDDeallocator: public nsDequeFunctor{
public:
virtual void* operator()(void* anObject) {
nsIDTD* aDTD =(nsIDTD*)anObject;
NS_RELEASE(aDTD);
return 0;
}
};
class CDTDFinder: public nsDequeFunctor{
public:
CDTDFinder(nsIDTD* aDTD) {
NS_IF_ADDREF(mTargetDTD=aDTD);
}
~CDTDFinder() {
NS_IF_RELEASE(mTargetDTD);
}
virtual void* operator()(void* anObject) {
return (anObject==(void*)mTargetDTD) ? anObject : 0;
}
nsIDTD* mTargetDTD;
};
class CSharedParserObjects {
public:
CSharedParserObjects() : mDeallocator(), mDTDDeque(mDeallocator) {
}
~CSharedParserObjects() {
}
void RegisterDTD(nsIDTD* aDTD){
CDTDFinder theFinder(aDTD);
if(!mDTDDeque.ForEach(theFinder))
mDTDDeque.Push(aDTD);
}
nsIDTD* FindDTD(nsIDTD* aDTD){
return 0;
}
CDTDDeallocator mDeallocator;
nsDeque mDTDDeque;
};
CSharedParserObjects gSharedParserObjects;
/**
* default constructor
@@ -85,7 +134,10 @@ CTokenDeallocator gTokenKiller;
* @param
* @return
*/
nsParser::nsParser() : mTokenDeque(gTokenKiller) {
nsParser::nsParser() :
mTokenDeque(gTokenDeallocator),
mContentType()
{
NS_INIT_REFCNT();
mDTDDebug = 0;
mParserFilter = 0;
@@ -97,6 +149,9 @@ nsParser::nsParser() : mTokenDeque(gTokenKiller) {
mParseMode=eParseMode_unknown;
mURL=0;
mDTD=0;
mScanner=0;
mTransferBuffer=new char[gTransferBufferSize+1];
mAutoDetectStatus=eUnknownDetect;
}
@@ -121,7 +176,6 @@ nsParser::~nsParser() {
delete mScanner;
mScanner=0;
NS_IF_RELEASE(mDTD);
NS_IF_RELEASE(mURL);
}
@@ -207,26 +261,16 @@ nsIContentSink* nsParser::SetContentSink(nsIContentSink* aSink) {
return old;
}
/**
*
* Call this static method when you want to
* register your dynamic DTD's with the parser.
*
* @update gess 6/9/98
* @param
* @return
* @param aDTD is the object to be registered.
* @return nothing.
*/
void nsParser::SetDTD(nsIDTD* aDTD) {
mDTD=aDTD;
}
/**
* Retrieve the DTD from the parser.
* @update gess6/18/98
* @param
* @return
*/
nsIDTD * nsParser::GetDTD(void) {
return mDTD;
void nsParser::RegisterDTD(nsIDTD* aDTD){
gSharedParserObjects.RegisterDTD(aDTD);
}
/**
@@ -240,39 +284,6 @@ CScanner* nsParser::GetScanner(void){
return mScanner;
}
/**
* This is where we loop over the tokens created in the
* tokenization phase, and try to make sense out of them.
*
* @update gess 3/25/98
* @param
* @return PR_TRUE if parse succeeded, PR_FALSE otherwise.
*/
PRInt32 nsParser::IterateTokens() {
nsDequeIterator e=mTokenDeque.End();
nsDequeIterator theMarkPos(e);
mMajorIteration++;
if(!mCurrentPos)
mCurrentPos=new nsDequeIterator(mTokenDeque.Begin());
PRInt32 result=kNoError;
while((kNoError==result) && ((*mCurrentPos<e))){
mMinorIteration++;
CToken* theToken=(CToken*)mCurrentPos->GetCurrent();
theMarkPos=*mCurrentPos;
result=mDTD->HandleToken(theToken);
++(*mCurrentPos);
}
if(kInterrupted==result)
*mCurrentPos=theMarkPos;
return result;
}
/**
*
*
@@ -291,7 +302,6 @@ eParseMode DetermineParseMode() {
return result;
}
/**
*
*
@@ -299,34 +309,59 @@ eParseMode DetermineParseMode() {
* @param
* @return
*/
nsIDTD* CreateDTD(eParseMode aMode,const char* aContentType) {
nsIDTD* aDTD=0;
switch(aMode) {
case eParseMode_navigator:
aDTD=new CNavDTD(); break;
case eParseMode_other:
aDTD=new COtherDTD(); break;
default:
break;
PRBool FindSuitableDTD(eProcessType aProcessType,eParseMode aMode,nsString& aContentType, nsIDTD*& aDefaultDTD) {
PRBool result=PR_FALSE;
if(aDefaultDTD) {
if(aDefaultDTD->IsCapableOf(aProcessType,aContentType,0)) {
result=PR_TRUE;
}
}
if (aDTD)
aDTD->AddRef();
return aDTD;
}
else {
nsDequeIterator b=gSharedParserObjects.mDTDDeque.Begin();
nsDequeIterator e=gSharedParserObjects.mDTDDeque.End();
/**
*
* @update gess6/22/98
* @param
* @return
*/
PRBool nsParser::DetermineContentType(const char* aContentType) {
PRBool result=PR_TRUE;
aContentType="application/x-unknown-content-type";
while(b<e){
nsIDTD* theDTD=(nsIDTD*)b.GetCurrent();
if(theDTD && theDTD->IsCapableOf(aProcessType,aContentType,0)) {
aDefaultDTD=theDTD;
result=PR_TRUE;
break;
}
++b;
}
}
return result;
}
/**
* Call this method if you want the known DTD's to try
* to detect the document type based through analysis
* of the underlying stream.
*
* @update gess6/22/98
* @param aBuffer -- nsString containing sample data to be analyzed.
* @return auto-detect result: eValid, eInvalid, eUnknown
*/
eAutoDetectResult nsParser::AutoDetectContentType(nsString& aBuffer) {
//The process:
// You should go out and ask each DTD if they
// recognize the content in the scanner.
// Somebody should say yes, or we can't continue.
//This method may change mContentType and mDTD.
//It absolutely changes mAutoDetectStatus
mAutoDetectStatus=eUnknownDetect;
if(PR_TRUE==mContentType.Equals(kHTMLTextContentType)) {
mAutoDetectStatus=eValidDetect;
}
return mAutoDetectStatus;
}
/**
* This gets called just prior to the model actually
* being constructed. It's important to make this the
@@ -338,14 +373,13 @@ PRBool nsParser::DetermineContentType(const char* aContentType) {
* @param
* @return
*/
PRInt32 nsParser::WillBuildModel(const char* aFilename, const char* aContentType){
PRInt32 nsParser::WillBuildModel(eProcessType aProcessType, const char* aFilename){
mMajorIteration=-1;
mMinorIteration=-1;
mParseMode=DetermineParseMode();
mDTD=(0==mDTD) ? CreateDTD(mParseMode,aContentType) : mDTD;
if(mDTD) {
mDTD->SetDTDDebug(mDTDDebug);
if(PR_TRUE==FindSuitableDTD(aProcessType,mParseMode,mContentType,mDTD)) {
mDTD->SetParser(this);
mDTD->SetContentSink(mSink);
mDTD->WillBuildModel(aFilename);
@@ -406,16 +440,66 @@ PRBool nsParser::Parse(const char* aFilename){
//ok, time to create our tokenizer and begin the process
mScanner=new CScanner(aFilename,mParseMode);
char theContentType[600];
DetermineContentType(theContentType);
WillBuildModel(aFilename,theContentType);
status=ResumeParse();
DidBuildModel(status);
if(mScanner) {
mScanner->Eof();
if(eValidDetect==AutoDetectContentType(mScanner->GetBuffer())) {
WillBuildModel(eParsing,aFilename);
status=ResumeParse();
DidBuildModel(status);
}
} //if
}
return status;
}
/**
* This is the main controlling routine in the parsing process.
* Note that it may get called multiple times for the same scanner,
* since this is a pushed based system, and all the tokens may
* not have been consumed by the scanner during a given invocation
* of this method.
*
* NOTE: We don't call willbuildmodel here, because it will happen
* as a result of calling OnStartBinding later on.
*
* @update gess 3/25/98
* @param aFilename -- const char* containing file to be parsed.
* @return PR_TRUE if parse succeeded, PR_FALSE otherwise.
*/
PRInt32 nsParser::Parse(nsIURL* aURL,nsIStreamObserver* aListener, nsIDTDDebug * aDTDDebug) {
PRInt32 status;
status = BeginParse(aURL, aListener, aDTDDebug);
if (NS_OK == status) {
status=mURL->Open(this);
}
return status;
}
/**
* Call this method if all you want to do is parse 1 string full of HTML text.
*
* @update gess5/11/98
* @param anHTMLString contains a string-full of real HTML
* @param appendTokens tells us whether we should insert tokens inline, or append them.
* @return TRUE if all went well -- FALSE otherwise
*/
PRInt32 nsParser::Parse(nsString& aSourceBuffer,PRBool appendTokens){
PRInt32 result=kNoError;
mScanner=new CScanner();
mScanner->Append(aSourceBuffer);
if(eValidDetect==AutoDetectContentType(aSourceBuffer)) {
WillBuildModel(eParsing,"from-string");
result=ResumeParse();
DidBuildModel(result);
}
return result;
}
/**
* This is the main controlling routine in the parsing process.
* Note that it may get called multiple times for the same scanner,
@@ -454,46 +538,35 @@ PRInt32 nsParser::BeginParse(nsIURL* aURL,nsIStreamObserver* aListener, nsIDTDDe
}
/**
* This is the main controlling routine in the parsing process.
* Note that it may get called multiple times for the same scanner,
* since this is a pushed based system, and all the tokens may
* not have been consumed by the scanner during a given invocation
* of this method.
*
* NOTE: We don't call willbuildmodel here, because it will happen
* as a result of calling OnStartBinding later on.
*
* @update gess 3/25/98
* @param aFilename -- const char* containing file to be parsed.
* @return PR_TRUE if parse succeeded, PR_FALSE otherwise.
*
* @update gess6/23/98
* @param
* @return
*/
PRInt32 nsParser::Parse(nsIURL* aURL,nsIStreamObserver* aListener, nsIDTDDebug * aDTDDebug) {
PRInt32 status;
status = BeginParse(aURL, aListener, aDTDDebug);
if (NS_OK == status) {
status=mURL->Open(this);
}
return status;
PRInt32 nsParser::Convert(nsIURL* aURL,char* aSourceForm,char* aTargetForm,nsIStreamListener* aListener) {
PRInt32 result=0;
return result;
}
/**
* Call this method if all you want to do is parse 1 string full of HTML text.
*
* @update gess5/11/98
* @param anHTMLString contains a string-full of real HTML
* @param appendTokens tells us whether we should insert tokens inline, or append them.
* @return TRUE if all went well -- FALSE otherwise
*
* @update gess6/23/98
* @param
* @return
*/
PRInt32 nsParser::Parse(nsString& aSourceBuffer,PRBool appendTokens){
PRInt32 result=kNoError;
WillBuildModel("nsString","text/html");
mScanner->Append(aSourceBuffer);
result=ResumeParse();
DidBuildModel(result);
PRInt32 nsParser::Convert(const char* aFilename,char* aSourceForm,char* aTargetForm) {
PRInt32 result=0;
return result;
}
/**
*
* @update gess6/23/98
* @param
* @return
*/
PRInt32 nsParser::Convert(nsString& anHTMLString,char* aSourceForm,char* aTargetForm,PRBool appendTokens){
PRInt32 result=0;
return result;
}
@@ -520,6 +593,37 @@ PRInt32 nsParser::ResumeParse() {
return result;
}
/**
* This is where we loop over the tokens created in the
* tokenization phase, and try to make sense out of them.
*
* @update gess 3/25/98
* @param
* @return PR_TRUE if parse succeeded, PR_FALSE otherwise.
*/
PRInt32 nsParser::IterateTokens() {
nsDequeIterator e=mTokenDeque.End();
nsDequeIterator theMarkPos(e);
mMajorIteration++;
if(!mCurrentPos)
mCurrentPos=new nsDequeIterator(mTokenDeque.Begin());
PRInt32 result=kNoError;
while((kNoError==result) && ((*mCurrentPos<e))){
mMinorIteration++;
CToken* theToken=(CToken*)mCurrentPos->GetCurrent();
theMarkPos=*mCurrentPos;
result=mDTD->HandleToken(theToken);
++(*mCurrentPos);
}
if(kInterrupted==result)
*mCurrentPos=theMarkPos;
return result;
}
/**
* Retrieve the attributes for this node, and add then into
@@ -577,6 +681,10 @@ PRInt32 nsParser::CollectSkippedContent(nsCParserNode& aNode,PRInt32& aCount) {
return result;
}
/*******************************************************************
These methods are used to talk to the netlib system...
*******************************************************************/
/**
*
*
@@ -618,11 +726,11 @@ nsresult nsParser::OnStartBinding(const char *aContentType){
if (nsnull != mObserver) {
mObserver->OnStartBinding(aContentType);
}
nsresult result=WillBuildModel(mURL->GetSpec(),aContentType);
if(!mTransferBuffer) {
mTransferBuffer=new char[gTransferBufferSize+1];
}
return result;
mAutoDetectStatus=eUnknownDetect;
mDTD=0;
mContentType=aContentType;
return kNoError;
}
/**
@@ -634,9 +742,21 @@ nsresult nsParser::OnStartBinding(const char *aContentType){
* @return error code (usually 0)
*/
nsresult nsParser::OnDataAvailable(nsIInputStream *pIStream, PRInt32 length){
/* if (nsnull != mListener) {
//Rick potts removed this.
//Does it need to be here?
mListener->OnDataAvailable(pIStream, length);
}
*/
int len=0;
int offset=0;
if(eInvalidDetect==mAutoDetectStatus) {
if(mScanner) {
mScanner->GetBuffer().Truncate();
}
}
do {
PRInt32 err;
len = pIStream->Read(&err, mTransferBuffer, 0, gTransferBufferSize);
@@ -650,9 +770,13 @@ nsresult nsParser::OnDataAvailable(nsIInputStream *pIStream, PRInt32 length){
if (mParserFilter)
mParserFilter->RawBuffer(mTransferBuffer, &len);
mScanner->Append(mTransferBuffer,len);
mScanner->Append(&mTransferBuffer[offset],len);
if(eUnknownDetect==mAutoDetectStatus) {
if(eValidDetect==AutoDetectContentType(mScanner->GetBuffer())) {
nsresult result=WillBuildModel(eParsing,mURL->GetSpec());
} //if
}
} //if
} while (len > 0);

24
mozilla/parser/htmlparser/src/nsParser.h
View File

@@ -60,7 +60,6 @@
#include "nsParserNode.h"
#include "nsParserTypes.h"
#include "nsIURL.h"
#include "nsIStreamListener.h"
#define NS_PARSER_IID \
@@ -107,11 +106,9 @@ friend class CTokenHandler;
virtual nsIContentSink* SetContentSink(nsIContentSink* aSink);
virtual nsIParserFilter* SetParserFilter(nsIParserFilter* aFilter);
virtual void RegisterDTD(nsIDTD* aDTD);
virtual void SetDTD(nsIDTD* aDTD);
virtual nsIDTD* GetDTD(void);
/**
*
*
@@ -152,6 +149,15 @@ friend class CTokenHandler;
*/
virtual PRInt32 Parse(nsString& anHTMLString,PRBool appendTokens);
/******************************************************************************************
* Convert methods start input source (of known or unknown form), and perform conversions
* until you wind up with a <i>stream</i> in your target form.
* The internal content model is never effected.
******************************************************************************************/
virtual PRInt32 Convert(nsIURL* aURL,char* aSourceForm, char* aTargetForm,nsIStreamListener* aListener);
virtual PRInt32 Convert(const char* aFilename,char* aSourceForm,char* aTargetForm);
virtual PRInt32 Convert(nsString& anHTMLString,char* aSourceForm,char* aTargetForm,PRBool appendTokens);
/**
* This method gets called (automatically) during incremental parsing
* @update gess5/11/98
@@ -208,7 +214,7 @@ protected:
* @param
* @return
*/
PRInt32 WillBuildModel(const char* aFilename=0,const char* aContentType=0);
PRInt32 WillBuildModel(eProcessType theProcessType=eParsing,const char* aFilename=0);
/**
*
@@ -303,7 +309,7 @@ private:
* @param
* @return TRUE if we figured it out.
*/
PRBool DetermineContentType(const char* aContentType);
eAutoDetectResult AutoDetectContentType(nsString& aBuffer);
protected:
@@ -328,7 +334,9 @@ protected:
nsDeque mTokenDeque;
CScanner* mScanner;
nsIURL* mURL;
nsIDTDDebug* mDTDDebug;
nsIDTDDebug* mDTDDebug;
nsString mContentType;
eAutoDetectResult mAutoDetectStatus;
};

8
mozilla/parser/htmlparser/src/nsParserNode.cpp
View File

@@ -19,7 +19,7 @@
#include "nsParserNode.h"
#include "string.h"
#include "nsHTMLTokens.h"
/**
* Default constructor
@@ -32,7 +32,7 @@ nsCParserNode::nsCParserNode(CToken* aToken): nsIParserNode(),
mName(), mEmptyString() {
NS_PRECONDITION(0!=aToken, "Null Token");
mAttributeCount=0;
mToken=(CHTMLToken*)aToken;
mToken=aToken;
memset(mAttributes,0,sizeof(mAttributes));
}
@@ -60,7 +60,7 @@ void nsCParserNode::AddAttribute(CToken* aToken) {
NS_PRECONDITION(mAttributeCount<sizeof(mAttributes), "Buffer overrun!");
NS_PRECONDITION(0!=aToken, "Error: Token shouldn't be null!");
if(aToken) {
mAttributes[mAttributeCount++]=(CHTMLToken*)aToken;
mAttributes[mAttributeCount++]=aToken;
}
}
@@ -79,7 +79,7 @@ void nsCParserNode::SetSkippedContent(CToken* aToken){
NS_PRECONDITION(mAttributeCount<sizeof(mAttributes)-1, "Buffer overrun!");
NS_PRECONDITION(0!=aToken, "Error: Token shouldn't be null!");
if(aToken) {
mAttributes[mAttributeCount++]=(CHTMLToken*)aToken;
mAttributes[mAttributeCount++]=aToken;
}
}

12
mozilla/parser/htmlparser/src/nsParserNode.h
View File

@@ -37,7 +37,7 @@
#define NS_PARSERNODE__
#include "nsIParserNode.h"
#include "nsHTMLTokens.h"
#include "nsToken.h"
#include "nsString.h"
//class nsParser;
@@ -148,11 +148,11 @@ class nsCParserNode : public nsIParserNode {
protected:
PRInt32 mAttributeCount;
CHTMLToken* mToken;
CHTMLToken* mAttributes[20]; // XXX Ack! This needs to be dynamic!
nsString mName;
nsString mEmptyString;
PRInt32 mAttributeCount;
CToken* mToken;
CToken* mAttributes[20]; // XXX Ack! This needs to be dynamic!
nsString mName;
nsString mEmptyString;
};

9
mozilla/parser/htmlparser/src/nsScanner.cpp
View File

@@ -503,6 +503,15 @@ PRInt32 CScanner::ReadUntil(nsString& aString,PRUnichar aTerminalChar,PRBool add
return result;
}
/**
*
* @update gess 3/25/98
* @param
* @return
*/
nsString& CScanner::GetBuffer(void) {
return mBuffer;
}
/**
* Conduct self test. Actually, selftesting for this class

9
mozilla/parser/htmlparser/src/nsScanner.h
View File

@@ -233,6 +233,15 @@ class CScanner {
*/
PRInt32 IncrementalAppend(const char* aBuffer,PRInt32 aSize);
/**
*
*
* @update gess 5/12/98
* @param
* @return
*/
nsString& GetBuffer(void);
static void SelfTest();
protected:

13
mozilla/xpcom/ds/nsDeque.cpp
View File

@@ -219,13 +219,15 @@ nsDequeIterator nsDeque::End(void) const{
* @param aFunctor object to call for each member
* @return *this
*/
const nsDeque& nsDeque::ForEach(nsDequeFunctor& aFunctor) const{
const void* nsDeque::ForEach(nsDequeFunctor& aFunctor) const{
int i=0;
for(i=0;i<mSize;i++){
void* obj=ObjectAt(i);
aFunctor(obj);
obj=aFunctor(obj);
if(obj)
return obj;
}
return *this;
return 0;
}
/******************************************************
@@ -379,9 +381,8 @@ void* nsDequeIterator::GetCurrent(void) {
* @param aFunctor object to call for each member
* @return *this
*/
const nsDequeIterator& nsDequeIterator::ForEach(nsDequeFunctor& aFunctor) const{
mDeque.ForEach(aFunctor);
return *this;
const void* nsDequeIterator::ForEach(nsDequeFunctor& aFunctor) const{
return mDeque.ForEach(aFunctor);
}

6
mozilla/xpcom/ds/nsDeque.h
View File

@@ -50,7 +50,7 @@
*/
class NS_BASE nsDequeFunctor{
public:
virtual void operator()(void* anObject)=0;
virtual void* operator()(void* anObject)=0;
};
@@ -157,7 +157,7 @@ friend class nsDequeIterator;
* @param aFunctor object to call for each member
* @return *this
*/
const nsDeque& ForEach(nsDequeFunctor& aFunctor) const;
const void* ForEach(nsDequeFunctor& aFunctor) const;
/**
* Perform automated selftest on the deque
@@ -346,7 +346,7 @@ public:
* @param aFunctor object to call for each member
* @return *this
*/
const nsDequeIterator& ForEach(nsDequeFunctor& aFunctor) const;
const void* ForEach(nsDequeFunctor& aFunctor) const;
protected:
PRInt32 mIndex;