mirror of
https://github.com/mozilla/gecko-dev.git
synced 2025-02-11 01:57:00 +00:00
34458 - Added static APIs to determine DTD for a given
DOCTYPE,MIMETYPE,COMMAND. r=rickg,pollmann 33397 - Fixed the NOSCRIPT interference on residual style handling. r=rickg. 35456 - Emulating Nav. for documents with unclosed STYLE and SCRIPT. 37052 - Handling XMP content. r=pollmann
This commit is contained in:
harishd%netscape.com
parent
bc9cc0db50
commit
9bc8378e24
@ -1332,7 +1332,7 @@ nsresult CNavDTD::HandleStartToken(CToken* aToken) {
|
||||
break;
|
||||
|
||||
case eHTMLTag_noscript:
|
||||
mHasOpenNoXXX++;
|
||||
//mHasOpenNoXXX++; // Fix for 33397 - Enable this when we handle NOSCRIPTS.
|
||||
isTokenHandled=PR_TRUE; // XXX - Throwing NOSCRIPT to the floor...yet another time..
|
||||
break;
|
||||
|
||||
|
||||
@ -142,186 +142,4 @@ void CParserContext::SetMimeType(const nsString& aMimeType){
|
||||
mDocType=eXMLText;
|
||||
}
|
||||
|
||||
/*************************************************************************************************
|
||||
First, let's define our modalities:
|
||||
|
||||
1. compatibility-mode: behave as much like nav4 as possible (unless it's too broken to bother)
|
||||
2. standard-mode: do html as well as you can per spec, and throw out navigator quirks
|
||||
3. strict-mode: adhere to the strict DTD specificiation to the highest degree possible
|
||||
|
||||
Assume the doctype is in the following form:
|
||||
<!DOCTYPE [Top Level Element] [Availability] "[Registration]// [Owner-ID] // [Type] [desc-text] // [Language]" "URI|text-identifier">
|
||||
[HTML] [PUBLIC|...] [+|-] [W3C|IETF|...] [DTD] "..." [EN]|...] "..."
|
||||
|
||||
|
||||
Here are the new rules for DTD handling; comments welcome:
|
||||
|
||||
XHTML and XML documents are always strict-mode:
|
||||
example: <!DOCTYPE \"-//W3C//DTD XHTML 1.0 Strict//EN\">
|
||||
|
||||
HTML strict dtd's enable strict-mode:
|
||||
example: <!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0//EN">
|
||||
example: <!DOCTYPE \"ISO/IEC 15445:1999//DTD HTML//EN\">
|
||||
|
||||
HTML 4.0 (or greater) transitional, frameset, (etc), without URI enables compatibility-mode:
|
||||
example: <!DOCTYPE \"-//W3C//DTD HTML 4.01 Transitional//EN\">
|
||||
|
||||
HTML 4.0 (or greater) transitional, frameset, (etc), with a URI that points to the strict.dtd will become strict:
|
||||
example: <!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN"
|
||||
"http://www.w3.org/TR/REC-html40/strict.dtd">
|
||||
|
||||
doctypes with systemID's or internal subset are handled in strict-mode:
|
||||
example: <!DOCTYPE HTML PUBLIC PublicID SystemID>
|
||||
example: <!DOCTYPE HTML SYSTEM SystemID>
|
||||
example: <!DOCTYPE HTML (PUBLIC PublicID SystemID? | SYSTEM SystemID) [ Internal-SS ]>
|
||||
|
||||
All other doctypes (<4.0), and documents without a doctype are handled in compatibility-mode.
|
||||
|
||||
*****************************************************************************************************/
|
||||
|
||||
/**
|
||||
* This is called when it's time to find out
|
||||
* what mode the parser/DTD should run for this document.
|
||||
* (Each parsercontext can have it's own mode).
|
||||
*
|
||||
* @update gess 02/17/00
|
||||
* @return parsermode (define in nsIParser.h)
|
||||
*/
|
||||
eParseMode CParserContext::DetermineParseMode(const nsString& theBuffer) {
|
||||
const char* theModeStr= PR_GetEnv("PARSE_MODE");
|
||||
|
||||
mParseMode = eParseMode_unknown;
|
||||
|
||||
PRInt32 theIndex=theBuffer.Find("DOCTYPE",PR_TRUE,0,10);
|
||||
if(kNotFound<theIndex) {
|
||||
|
||||
//good, we found "DOCTYPE" -- now go find it's end delimiter '>'
|
||||
PRInt32 theGTPos=theBuffer.FindChar(kGreaterThan,theIndex+1);
|
||||
PRInt32 theEnd=(kNotFound==theGTPos) ? 512 : MinInt(512,theGTPos);
|
||||
PRInt32 theSubIndex=theBuffer.Find("//DTD",PR_TRUE,theIndex+8,theEnd-(theIndex+8)); //skip to the type and desc-text...
|
||||
PRInt32 theErr=0;
|
||||
PRInt32 theMajorVersion=3;
|
||||
|
||||
//note that if we don't find '>', then we just scan the first 512 bytes.
|
||||
|
||||
if(0<=theSubIndex) {
|
||||
PRInt32 theStartPos=theSubIndex+5;
|
||||
PRInt32 theCount=theEnd-theStartPos;
|
||||
|
||||
if(kNotFound<theSubIndex) {
|
||||
|
||||
theSubIndex=theBuffer.Find("XHTML",PR_TRUE,theStartPos,theCount);
|
||||
if(0<=theSubIndex) {
|
||||
mDocType=eXHTMLText;
|
||||
mParseMode=eParseMode_strict;
|
||||
return mParseMode;
|
||||
}
|
||||
else {
|
||||
theSubIndex=theBuffer.Find("ISO/IEC 15445:",PR_TRUE,theIndex+8,theEnd-(theIndex+8));
|
||||
if(0<=theSubIndex) {
|
||||
mDocType=eHTML4Text;
|
||||
mParseMode=eParseMode_strict;
|
||||
theMajorVersion=4;
|
||||
theSubIndex+=15;
|
||||
}
|
||||
else {
|
||||
theSubIndex=theBuffer.Find("HTML",PR_TRUE,theStartPos,theCount);
|
||||
if(0<=theSubIndex) {
|
||||
mDocType=eHTML4Text;
|
||||
mParseMode=eParseMode_strict;
|
||||
theMajorVersion=3;
|
||||
}
|
||||
else {
|
||||
theSubIndex=theBuffer.Find("HYPERTEXT MARKUP",PR_TRUE,theStartPos,theCount);
|
||||
if(0<=theSubIndex) {
|
||||
mDocType=eHTML3Text;
|
||||
mParseMode=eParseMode_quirks;
|
||||
theSubIndex+=20;
|
||||
}
|
||||
}
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
theStartPos=theSubIndex+5;
|
||||
theCount=theEnd-theStartPos;
|
||||
nsAutoString theNum;
|
||||
|
||||
//get the next substring from the buffer, which should be a number.
|
||||
//now see what the version number is...
|
||||
|
||||
theStartPos=theBuffer.FindCharInSet("123456789",theStartPos);
|
||||
if(0<=theStartPos) {
|
||||
PRInt32 theTerminal=theBuffer.FindCharInSet(" />",theStartPos+1);
|
||||
if(theTerminal) {
|
||||
theBuffer.Mid(theNum,theStartPos,theTerminal-theStartPos);
|
||||
}
|
||||
else theBuffer.Mid(theNum,theStartPos,3);
|
||||
theMajorVersion=theNum.ToInteger(&theErr);
|
||||
}
|
||||
|
||||
//now see what the
|
||||
theStartPos+=theNum.Length();
|
||||
theCount=theEnd-theStartPos;
|
||||
if((theBuffer.Find("TRANSITIONAL",PR_TRUE,theStartPos,theCount)>kNotFound)||
|
||||
(theBuffer.Find("LOOSE",PR_TRUE,theStartPos,theCount)>kNotFound) ||
|
||||
(theBuffer.Find("FRAMESET",PR_TRUE,theStartPos,theCount)>kNotFound) ||
|
||||
(theBuffer.Find("LATIN1", PR_TRUE,theStartPos,theCount) >kNotFound) ||
|
||||
(theBuffer.Find("SYMBOLS",PR_TRUE,theStartPos,theCount) >kNotFound) ||
|
||||
(theBuffer.Find("SPECIAL",PR_TRUE,theStartPos,theCount) >kNotFound)) {
|
||||
mParseMode=eParseMode_quirks;
|
||||
}
|
||||
|
||||
//one last thing: look for a URI that specifies the strict.dtd
|
||||
theStartPos+=6;
|
||||
theCount=theEnd-theStartPos;
|
||||
theSubIndex=theBuffer.Find("STRICT.DTD",PR_TRUE,theStartPos,theCount);
|
||||
if(0<theSubIndex) {
|
||||
//Since we found it, regardless of what's in the descr-text, kick into strict mode.
|
||||
mParseMode=eParseMode_strict;
|
||||
mDocType=eHTML4Text;
|
||||
}
|
||||
|
||||
if (0==theErr){
|
||||
switch(theMajorVersion) {
|
||||
case 0: case 1: case 2: case 3:
|
||||
if(mDocType!=eXHTMLText){
|
||||
mParseMode=eParseMode_quirks; //be as backward compatible as possible
|
||||
mDocType=eHTML3Text;
|
||||
}
|
||||
break;
|
||||
|
||||
default:
|
||||
//XXX hack -- someday, the next line of code will be criticized
|
||||
//for it's lack of vision...
|
||||
if(theMajorVersion>20) {
|
||||
mParseMode=eParseMode_noquirks;
|
||||
}
|
||||
break;
|
||||
} //switch
|
||||
}
|
||||
|
||||
} //if
|
||||
else {
|
||||
PRInt32 thePos=theBuffer.Find("HTML",PR_TRUE,1,50);
|
||||
if(kNotFound!=thePos) {
|
||||
mDocType=eHTML4Text;
|
||||
PRInt32 theIDPos=theBuffer.Find("PublicID",thePos);
|
||||
if(kNotFound==theIDPos)
|
||||
theIDPos=theBuffer.Find("SystemID",thePos);
|
||||
mParseMode=(kNotFound==theIDPos) ? eParseMode_quirks : eParseMode_strict;
|
||||
}
|
||||
}
|
||||
}
|
||||
else if(kNotFound<(theIndex=theBuffer.Find("?XML",PR_TRUE,0,128))) {
|
||||
mParseMode=eParseMode_strict;
|
||||
}
|
||||
|
||||
if(theModeStr) {
|
||||
if(0==nsCRT::strcasecmp(theModeStr,"strict"))
|
||||
mParseMode=eParseMode_strict;
|
||||
}
|
||||
else mParseMode = (eParseMode_unknown==mParseMode)? eParseMode_quirks : mParseMode;
|
||||
|
||||
return mParseMode;
|
||||
}
|
||||
|
||||
@ -58,8 +58,6 @@ public:
|
||||
PRBool aCopyUnused=PR_FALSE);
|
||||
|
||||
CParserContext( const CParserContext& aContext);
|
||||
eParseMode DetermineParseMode(const nsString& theBuffer);
|
||||
|
||||
|
||||
~CParserContext();
|
||||
|
||||
|
||||
@ -992,7 +992,7 @@ nsHTMLContentSinkStream::AddDocTypeDecl(const nsIParserNode& aNode, PRInt32 aMod
|
||||
DebugDump("<",aNode.GetText(),(mNodeStackPos)*2);
|
||||
#endif
|
||||
|
||||
Write("<!");
|
||||
// Write("<!");
|
||||
|
||||
return NS_OK;
|
||||
}
|
||||
|
||||
@ -540,7 +540,7 @@ nsresult nsHTMLTokenizer::ConsumeStartTag(PRUnichar aChar,CToken*& aToken,nsScan
|
||||
if(NS_SUCCEEDED(result)) {
|
||||
|
||||
//XXX - Find a better soution to record content
|
||||
if(theTag==eHTMLTag_textarea && !mRecordTrailingContent) {
|
||||
if((theTag==eHTMLTag_textarea || theTag==eHTMLTag_xmp) && !mRecordTrailingContent) {
|
||||
mRecordTrailingContent=PR_TRUE;
|
||||
}
|
||||
|
||||
@ -589,7 +589,8 @@ nsresult nsHTMLTokenizer::ConsumeEndTag(PRUnichar aChar,CToken*& aToken,nsScanne
|
||||
nsresult result=NS_OK;
|
||||
|
||||
if(aToken) {
|
||||
if(aToken->GetTypeID()==eHTMLTag_textarea && mRecordTrailingContent) {
|
||||
eHTMLTags theTag=(eHTMLTags)aToken->GetTypeID();
|
||||
if((theTag==eHTMLTag_textarea || theTag==eHTMLTag_xmp) && mRecordTrailingContent) {
|
||||
mRecordTrailingContent=PR_FALSE;
|
||||
}
|
||||
result= aToken->Consume(aChar,aScanner,mParseMode); //tell new token to finish consuming text...
|
||||
|
||||
@ -282,16 +282,17 @@ void CStartToken::DebugDumpSource(nsOutputStream& out) {
|
||||
void CStartToken::GetSource(nsString& anOutputString){
|
||||
anOutputString.AssignWithConversion("<");
|
||||
/*
|
||||
* mTextValue used to contain the name of the tag.
|
||||
* But for the sake of performance we now rely on the tagID
|
||||
* rather than tag name. This however, caused bug 15204
|
||||
* to reincarnate. Since, mTextvalue is not being used here..
|
||||
* I'm just going to comment it out.
|
||||
*
|
||||
* Watch out for Bug 15204
|
||||
*/
|
||||
// anOutputString+=mTextValue;
|
||||
if(mTrailingContent.Length()>0)
|
||||
anOutputString+=mTrailingContent;
|
||||
anOutputString=mTrailingContent;
|
||||
else {
|
||||
if(mTextValue.Length()>0)
|
||||
anOutputString=mTextValue;
|
||||
else
|
||||
anOutputString.AssignWithConversion(GetTagName(mTypeID));
|
||||
anOutputString+='>';
|
||||
}
|
||||
}
|
||||
|
||||
/*
|
||||
@ -304,16 +305,17 @@ void CStartToken::GetSource(nsString& anOutputString){
|
||||
void CStartToken::AppendSource(nsString& anOutputString){
|
||||
anOutputString.AppendWithConversion("<");
|
||||
/*
|
||||
* mTextValue used to contain the name of the tag.
|
||||
* But for the sake of performance we now rely on the tagID
|
||||
* rather than tag name. This however, caused bug 15204
|
||||
* to reincarnate. Since, mTextvalue is not being used here..
|
||||
* I'm just going to comment it out.
|
||||
*
|
||||
* Watch out for Bug 15204
|
||||
*/
|
||||
// anOutputString+=mTextValue;
|
||||
if(mTrailingContent.Length()>0)
|
||||
anOutputString+=mTrailingContent;
|
||||
else {
|
||||
if(mTextValue.Length()>0)
|
||||
anOutputString+=mTextValue;
|
||||
else
|
||||
anOutputString.AppendWithConversion(GetTagName(mTypeID));
|
||||
anOutputString+='>';
|
||||
}
|
||||
}
|
||||
|
||||
/*
|
||||
@ -655,10 +657,16 @@ nsresult CTextToken::ConsumeUntil(PRUnichar aChar,PRBool aIgnoreComments,nsScann
|
||||
else disaster=PR_TRUE;
|
||||
|
||||
if(disaster) {
|
||||
if((!aScanner.IsIncremental()) && (theAltTermStrPos>kNotFound)) {
|
||||
// If you're here it means..we hit the rock bottom and therefore switch to plan B.
|
||||
theCurrOffset=theAltTermStrPos;
|
||||
theLastIteration=PR_TRUE;
|
||||
if(!aScanner.IsIncremental()) {
|
||||
if(theAltTermStrPos>kNotFound) {
|
||||
// If you're here it means..we hit the rock bottom and therefore switch to plan B.
|
||||
theCurrOffset=theAltTermStrPos;
|
||||
theLastIteration=PR_TRUE;
|
||||
}
|
||||
else {
|
||||
aTerminalString.Cut(0,2); // Do this to fix Bug. 35456
|
||||
done=PR_TRUE;
|
||||
}
|
||||
}
|
||||
else
|
||||
result=kEOF;
|
||||
|
||||
@ -42,15 +42,24 @@
|
||||
#include "CRtfDTD.h"
|
||||
#include "CNavDTD.h"
|
||||
#include "COtherDTD.h"
|
||||
|
||||
//#define rickgdebug
|
||||
|
||||
#include "prenv.h"
|
||||
#include "nsParserCIID.h"
|
||||
|
||||
//#define rickgdebug
|
||||
|
||||
#define TEST_DOCTYPES 1
|
||||
|
||||
static NS_DEFINE_IID(kISupportsIID, NS_ISUPPORTS_IID);
|
||||
static NS_DEFINE_IID(kClassIID, NS_PARSER_IID);
|
||||
static NS_DEFINE_IID(kIParserIID, NS_IPARSER_IID);
|
||||
static NS_DEFINE_IID(kIStreamListenerIID, NS_ISTREAMLISTENER_IID);
|
||||
|
||||
static NS_DEFINE_CID(kWellFormedDTDCID, NS_WELLFORMEDDTD_CID);
|
||||
static NS_DEFINE_CID(kNavDTDCID, NS_CNAVDTD_CID);
|
||||
static NS_DEFINE_CID(kCOtherDTDCID, NS_COTHER_DTD_CID);
|
||||
static NS_DEFINE_CID(kViewSourceDTDCID, NS_VIEWSOURCE_DTD_CID);
|
||||
static NS_DEFINE_CID(kRtfDTDCID, NS_CRTF_DTD_CID);
|
||||
|
||||
static const char* kNullURL = "Error: Null URL given";
|
||||
static const char* kOnStartNotCalled = "Error: OnStartRequest() must be called before OnDataAvailable()";
|
||||
static const char* kBadListenerInit = "Error: Parser's IStreamListener API was not setup correctly in constructor.";
|
||||
@ -427,6 +436,194 @@ eParseMode nsParser::GetParseMode(void){
|
||||
}
|
||||
|
||||
|
||||
/*************************************************************************************************
|
||||
First, let's define our modalities:
|
||||
|
||||
1. compatibility-mode: behave as much like nav4 as possible (unless it's too broken to bother)
|
||||
2. standard-mode: do html as well as you can per spec, and throw out navigator quirks
|
||||
3. strict-mode: adhere to the strict DTD specificiation to the highest degree possible
|
||||
|
||||
Assume the doctype is in the following form:
|
||||
<!DOCTYPE [Top Level Element] [Availability] "[Registration]// [Owner-ID] // [Type] [desc-text] // [Language]" "URI|text-identifier">
|
||||
[HTML] [PUBLIC|...] [+|-] [W3C|IETF|...] [DTD] "..." [EN]|...] "..."
|
||||
|
||||
|
||||
Here are the new rules for DTD handling; comments welcome:
|
||||
|
||||
XHTML and XML documents are always strict-mode:
|
||||
example: <!DOCTYPE \"-//W3C//DTD XHTML 1.0 Strict//EN\">
|
||||
|
||||
HTML strict dtd's enable strict-mode:
|
||||
example: <!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0//EN">
|
||||
example: <!DOCTYPE \"ISO/IEC 15445:1999//DTD HTML//EN\">
|
||||
|
||||
HTML 4.0 (or greater) transitional, frameset, (etc), without URI enables compatibility-mode:
|
||||
example: <!DOCTYPE \"-//W3C//DTD HTML 4.01 Transitional//EN\">
|
||||
|
||||
HTML 4.0 (or greater) transitional, frameset, (etc), with a URI that points to the strict.dtd will become strict:
|
||||
example: <!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN"
|
||||
"http://www.w3.org/TR/REC-html40/strict.dtd">
|
||||
|
||||
doctypes with systemID's or internal subset are handled in strict-mode:
|
||||
example: <!DOCTYPE HTML PUBLIC PublicID SystemID>
|
||||
example: <!DOCTYPE HTML SYSTEM SystemID>
|
||||
example: <!DOCTYPE HTML (PUBLIC PublicID SystemID? | SYSTEM SystemID) [ Internal-SS ]>
|
||||
|
||||
All other doctypes (<4.0), and documents without a doctype are handled in compatibility-mode.
|
||||
|
||||
*****************************************************************************************************/
|
||||
|
||||
/**
|
||||
* This is called when it's time to find out
|
||||
* what mode the parser/DTD should run for this document.
|
||||
* (Each parsercontext can have it's own mode).
|
||||
*
|
||||
* @update gess 02/17/00
|
||||
* @return parsermode (define in nsIParser.h)
|
||||
*/
|
||||
static
|
||||
void DetermineParseMode(nsString& aBuffer,eParseMode& aParseMode,eParserDocType aDocType) {
|
||||
const char* theModeStr= PR_GetEnv("PARSE_MODE");
|
||||
|
||||
aParseMode = eParseMode_unknown;
|
||||
|
||||
PRInt32 theIndex=aBuffer.Find("DOCTYPE",PR_TRUE,0,10);
|
||||
if(kNotFound<theIndex) {
|
||||
|
||||
//good, we found "DOCTYPE" -- now go find it's end delimiter '>'
|
||||
PRInt32 theGTPos=aBuffer.FindChar(kGreaterThan,theIndex+1);
|
||||
PRInt32 theEnd=(kNotFound==theGTPos) ? 512 : MinInt(512,theGTPos);
|
||||
PRInt32 theSubIndex=aBuffer.Find("//DTD",PR_TRUE,theIndex+8,theEnd-(theIndex+8)); //skip to the type and desc-text...
|
||||
PRInt32 theErr=0;
|
||||
PRInt32 theMajorVersion=3;
|
||||
|
||||
//note that if we don't find '>', then we just scan the first 512 bytes.
|
||||
|
||||
if(0<=theSubIndex) {
|
||||
PRInt32 theStartPos=theSubIndex+5;
|
||||
PRInt32 theCount=theEnd-theStartPos;
|
||||
|
||||
if(kNotFound<theSubIndex) {
|
||||
|
||||
theSubIndex=aBuffer.Find("XHTML",PR_TRUE,theStartPos,theCount);
|
||||
if(0<=theSubIndex) {
|
||||
aDocType=eXHTMLText;
|
||||
aParseMode=eParseMode_strict;
|
||||
}
|
||||
else {
|
||||
theSubIndex=aBuffer.Find("ISO/IEC 15445:",PR_TRUE,theIndex+8,theEnd-(theIndex+8));
|
||||
if(0<=theSubIndex) {
|
||||
aDocType=eHTML4Text;
|
||||
aParseMode=eParseMode_strict;
|
||||
theMajorVersion=4;
|
||||
theSubIndex+=15;
|
||||
}
|
||||
else {
|
||||
theSubIndex=aBuffer.Find("HTML",PR_TRUE,theStartPos,theCount);
|
||||
if(0<=theSubIndex) {
|
||||
aDocType=eHTML4Text;
|
||||
aParseMode=eParseMode_strict;
|
||||
theMajorVersion=3;
|
||||
}
|
||||
else {
|
||||
theSubIndex=aBuffer.Find("HYPERTEXT MARKUP",PR_TRUE,theStartPos,theCount);
|
||||
if(0<=theSubIndex) {
|
||||
aDocType=eHTML3Text;
|
||||
aParseMode=eParseMode_quirks;
|
||||
theSubIndex+=20;
|
||||
}
|
||||
}
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
theStartPos=theSubIndex+5;
|
||||
theCount=theEnd-theStartPos;
|
||||
nsAutoString theNum;
|
||||
|
||||
//get the next substring from the buffer, which should be a number.
|
||||
//now see what the version number is...
|
||||
|
||||
theStartPos=aBuffer.FindCharInSet("123456789",theStartPos);
|
||||
if(0<=theStartPos) {
|
||||
PRInt32 theTerminal=aBuffer.FindCharInSet(" />",theStartPos+1);
|
||||
if(theTerminal) {
|
||||
aBuffer.Mid(theNum,theStartPos,theTerminal-theStartPos);
|
||||
}
|
||||
else aBuffer.Mid(theNum,theStartPos,3);
|
||||
theMajorVersion=theNum.ToInteger(&theErr);
|
||||
}
|
||||
|
||||
//now see what the
|
||||
theStartPos+=theNum.Length();
|
||||
theCount=theEnd-theStartPos;
|
||||
if((aBuffer.Find("TRANSITIONAL",PR_TRUE,theStartPos,theCount)>kNotFound)||
|
||||
(aBuffer.Find("LOOSE",PR_TRUE,theStartPos,theCount)>kNotFound) ||
|
||||
(aBuffer.Find("FRAMESET",PR_TRUE,theStartPos,theCount)>kNotFound) ||
|
||||
(aBuffer.Find("LATIN1", PR_TRUE,theStartPos,theCount) >kNotFound) ||
|
||||
(aBuffer.Find("SYMBOLS",PR_TRUE,theStartPos,theCount) >kNotFound) ||
|
||||
(aBuffer.Find("SPECIAL",PR_TRUE,theStartPos,theCount) >kNotFound)) {
|
||||
aParseMode=eParseMode_quirks;
|
||||
}
|
||||
|
||||
//one last thing: look for a URI that specifies the strict.dtd
|
||||
theStartPos+=6;
|
||||
theCount=theEnd-theStartPos;
|
||||
theSubIndex=aBuffer.Find("STRICT.DTD",PR_TRUE,theStartPos,theCount);
|
||||
if(0<theSubIndex) {
|
||||
//Since we found it, regardless of what's in the descr-text, kick into strict mode.
|
||||
aParseMode=eParseMode_strict;
|
||||
aDocType=eHTML4Text;
|
||||
}
|
||||
|
||||
if (0==theErr){
|
||||
switch(theMajorVersion) {
|
||||
case 0: case 1: case 2: case 3:
|
||||
if(aDocType!=eXHTMLText){
|
||||
aParseMode=eParseMode_quirks; //be as backward compatible as possible
|
||||
aDocType=eHTML3Text;
|
||||
}
|
||||
break;
|
||||
|
||||
default:
|
||||
//XXX hack -- someday, the next line of code will be criticized
|
||||
//for it's lack of vision...
|
||||
if(theMajorVersion>20) {
|
||||
aParseMode=eParseMode_noquirks;
|
||||
}
|
||||
break;
|
||||
} //switch
|
||||
}
|
||||
|
||||
} //if
|
||||
else {
|
||||
PRInt32 thePos=aBuffer.Find("HTML",PR_TRUE,1,50);
|
||||
if(kNotFound!=thePos) {
|
||||
aDocType=eHTML4Text;
|
||||
PRInt32 theIDPos=aBuffer.Find("PublicID",thePos);
|
||||
if(kNotFound==theIDPos)
|
||||
theIDPos=aBuffer.Find("SystemID",thePos);
|
||||
aParseMode=(kNotFound==theIDPos) ? eParseMode_quirks : eParseMode_strict;
|
||||
}
|
||||
}
|
||||
}
|
||||
else if(kNotFound<(theIndex=aBuffer.Find("?XML",PR_TRUE,0,128))) {
|
||||
aDocType=eXMLText;
|
||||
aParseMode=eParseMode_strict;
|
||||
}
|
||||
|
||||
if(theModeStr) {
|
||||
if(0==nsCRT::strcasecmp(theModeStr,"strict"))
|
||||
aParseMode=eParseMode_strict;
|
||||
}
|
||||
else {
|
||||
if(eParseMode_unknown==aParseMode) {
|
||||
aBuffer.InsertWithConversion("<!DOCTYPE HTML PUBLIC \"-//W3C//DTD HTML 3.2 Final//EN\">\n",0);
|
||||
aDocType=eHTML3Text;
|
||||
aParseMode=eParseMode_quirks;
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
/**
|
||||
*
|
||||
@ -496,6 +693,131 @@ PRBool FindSuitableDTD( CParserContext& aParserContext,nsString& aBuffer) {
|
||||
return PR_FALSE;
|
||||
}
|
||||
|
||||
/**
|
||||
* Call this method to determine a DTD for a DOCTYPE
|
||||
*
|
||||
* @update harishd 05/01/00
|
||||
* @param aDTD -- Carries the deduced ( from DOCTYPE ) DTD.
|
||||
* @param aDocTypeStr -- A doctype for which a DTD is to be selected.
|
||||
* @param aMimeType -- A mimetype for which a DTD is to be selected.
|
||||
* Note: aParseMode might be required.
|
||||
* @param aCommand -- A command for which a DTD is to be selected.
|
||||
* @param aParseMode -- Used with aMimeType to choose the correct DTD.
|
||||
* @return NS_OK if succeeded else ERROR.
|
||||
*/
|
||||
nsresult nsParser::CreateCompatibleDTDForDocType(nsIDTD** aDTD, nsString* aDocTypeStr)
|
||||
{
|
||||
nsresult result=NS_OK;
|
||||
const nsCID* theDTDClassID=0;
|
||||
|
||||
if(aDocTypeStr) {
|
||||
eParseMode theParseMode=eParseMode_unknown;
|
||||
eParserDocType theDocType=ePlainText;
|
||||
|
||||
DetermineParseMode(*aDocTypeStr,theParseMode,theDocType);
|
||||
|
||||
switch(theDocType) {
|
||||
case eHTML4Text:
|
||||
if(theParseMode==eParseMode_strict) {
|
||||
theDTDClassID=&kCOtherDTDCID;
|
||||
break;
|
||||
}
|
||||
case eHTML3Text:
|
||||
theDTDClassID=&kNavDTDCID;
|
||||
break;
|
||||
case eXHTMLText:
|
||||
case eXMLText:
|
||||
theDTDClassID=&kWellFormedDTDCID;
|
||||
break;
|
||||
default:
|
||||
theDTDClassID=&kNavDTDCID;
|
||||
break;
|
||||
}
|
||||
}
|
||||
|
||||
result=(theDTDClassID)? nsComponentManager::CreateInstance(*theDTDClassID, nsnull, NS_GET_IID(nsIDTD),(void**)aDTD):NS_OK;
|
||||
|
||||
return result;
|
||||
|
||||
}
|
||||
|
||||
/**
|
||||
* Call this method to determine a DTD for a given mime type.
|
||||
*
|
||||
* @update harishd 05/01/00
|
||||
* @param aDTD -- Carries the deduced ( from DOCTYPE ) DTD.
|
||||
* @param aMimeType -- A mimetype for which a DTD is to be selected.
|
||||
* Note: aParseMode might be required.
|
||||
* @param aParseMode -- Used with aMimeType to choose the correct DTD.
|
||||
* @return NS_OK if succeeded else ERROR.
|
||||
*/
|
||||
nsresult CreateCompatibleDTDForMimeType(nsIDTD** aDTD, const nsString* aMimeType=nsnull,
|
||||
eParseMode aParseMode=eParseMode_unknown)
|
||||
{
|
||||
nsresult result=NS_OK;
|
||||
const nsCID* theDTDClassID=0;
|
||||
|
||||
if(aMimeType) {
|
||||
|
||||
NS_ASSERTION(aParseMode!=eParseMode_unknown,"DTD selection might require a parsemode");
|
||||
|
||||
if(aMimeType->EqualsWithConversion(kHTMLTextContentType)) {
|
||||
if(aParseMode==eParseMode_strict) {
|
||||
theDTDClassID=&kCOtherDTDCID;
|
||||
}
|
||||
else {
|
||||
theDTDClassID=&kNavDTDCID;
|
||||
}
|
||||
}
|
||||
else if(aMimeType->EqualsWithConversion(kPlainTextContentType)) {
|
||||
theDTDClassID=&kNavDTDCID;
|
||||
}
|
||||
else if(aMimeType->EqualsWithConversion(kXMLTextContentType) ||
|
||||
aMimeType->EqualsWithConversion(kXULTextContentType) ||
|
||||
aMimeType->EqualsWithConversion(kRDFTextContentType)) {
|
||||
theDTDClassID=&kWellFormedDTDCID;
|
||||
}
|
||||
else if(aMimeType->EqualsWithConversion(kXIFTextContentType)) {
|
||||
theDTDClassID=&kRtfDTDCID;
|
||||
}
|
||||
else {
|
||||
theDTDClassID=&kNavDTDCID;
|
||||
}
|
||||
result=(theDTDClassID)? nsComponentManager::CreateInstance(*theDTDClassID, nsnull, NS_GET_IID(nsIDTD),(void**)aDTD):NS_OK;
|
||||
}
|
||||
|
||||
return result;
|
||||
}
|
||||
|
||||
/**
|
||||
* Call this method to determine a DTD for a given command
|
||||
*
|
||||
* @update harishd 05/01/00
|
||||
* @param aDTD -- Carries the deduced ( from DOCTYPE ) DTD.
|
||||
* @param aCommand -- A command for which a DTD is to be selected.
|
||||
* @return NS_OK if succeeded else ERROR.
|
||||
*/
|
||||
nsresult CreateCompatibleDTDForCommand(nsIDTD** aDTD, eParserCommands aCommand=eViewNormal)
|
||||
{
|
||||
nsresult result=NS_OK;
|
||||
const nsCID* theDTDClassID=0;
|
||||
|
||||
switch(aCommand) {
|
||||
case eViewSource:
|
||||
theDTDClassID=&kViewSourceDTDCID;
|
||||
break;
|
||||
case eViewNormal:
|
||||
theDTDClassID=&kNavDTDCID;
|
||||
break;
|
||||
case eViewErrors:
|
||||
default:
|
||||
break;
|
||||
}
|
||||
result=(theDTDClassID)? nsComponentManager::CreateInstance(*theDTDClassID, nsnull, NS_GET_IID(nsIDTD),(void**)aDTD):NS_OK;
|
||||
|
||||
return result;
|
||||
}
|
||||
|
||||
#ifdef TEST_DOCTYPES
|
||||
static const char* doctypes[] = {
|
||||
|
||||
@ -608,7 +930,7 @@ static const char* doctypes[] = {
|
||||
*/
|
||||
nsresult nsParser::WillBuildModel(nsString& aFilename){
|
||||
|
||||
nsresult result=NS_OK;
|
||||
nsresult result=NS_OK;
|
||||
|
||||
#if TEST_DOCTYPES
|
||||
|
||||
@ -617,9 +939,13 @@ nsresult nsParser::WillBuildModel(nsString& aFilename){
|
||||
|
||||
if(!tested) {
|
||||
tested=PR_TRUE;
|
||||
eParseMode theParseMode=eParseMode_unknown;
|
||||
eParserDocType theDocumentType=ePlainText;
|
||||
|
||||
while(*theDocType) {
|
||||
nsAutoString theType(*theDocType);
|
||||
eParseMode result=mParserContext->DetermineParseMode(theType);
|
||||
nsAutoString theType;
|
||||
theType.AssignWithConversion(*theDocType);
|
||||
DetermineParseMode(theType,theParseMode,theDocumentType);
|
||||
theDocType++;
|
||||
}
|
||||
}
|
||||
@ -631,7 +957,7 @@ nsresult nsParser::WillBuildModel(nsString& aFilename){
|
||||
mMinorIteration=-1;
|
||||
|
||||
nsString& theBuffer=mParserContext->mScanner->GetBuffer();
|
||||
mParserContext->DetermineParseMode(theBuffer);
|
||||
DetermineParseMode(theBuffer,mParserContext->mParseMode,mParserContext->mDocType);
|
||||
|
||||
if(PR_TRUE==FindSuitableDTD(*mParserContext,theBuffer)) {
|
||||
mParserContext->mDTD->WillBuildModel( *mParserContext,mSink);
|
||||
|
||||
@ -360,7 +360,40 @@ private:
|
||||
*/
|
||||
PRBool DidTokenize(PRBool aIsFinalChunk = PR_FALSE);
|
||||
|
||||
|
||||
/**
|
||||
* Call this method to determine a DTD for a DOCTYPE
|
||||
*
|
||||
* @update harishd 05/01/00
|
||||
* @param aDTD -- Carries the deduced ( from DOCTYPE ) DTD.
|
||||
* @param aDocTypeStr -- A doctype for which a DTD is to be selected.
|
||||
* @param aMimeType -- A mimetype for which a DTD is to be selected.
|
||||
Note: aParseMode might be required.
|
||||
* @param aCommand -- A command for which a DTD is to be selected.
|
||||
* @param aParseMode -- Used with aMimeType to choose the correct DTD.
|
||||
* @return NS_OK if succeeded else ERROR.
|
||||
*/
|
||||
static nsresult CreateCompatibleDTDForDocType(nsIDTD** aDTD, nsString* aDocTypeStr);
|
||||
/**
|
||||
* Call this method to determine a DTD for a given mime type.
|
||||
*
|
||||
* @update harishd 05/01/00
|
||||
* @param aDTD -- Carries the deduced ( from DOCTYPE ) DTD.
|
||||
* @param aMimeType -- A mimetype for which a DTD is to be selected.
|
||||
Note: aParseMode might be required.
|
||||
* @param aParseMode -- Used with aMimeType to choose the correct DTD.
|
||||
* @return NS_OK if succeeded else ERROR.
|
||||
*/
|
||||
static nsresult CreateCompatibleDTDForMimeType(nsIDTD** aDTD, const nsString* aMimeType=nsnull,
|
||||
eParseMode aParseMode=eParseMode_unknown);
|
||||
/**
|
||||
* Call this method to determine a DTD for a given command
|
||||
*
|
||||
* @update harishd 05/01/00
|
||||
* @param aDTD -- Carries the deduced ( from DOCTYPE ) DTD.
|
||||
* @param aCommand -- A command for which a DTD is to be selected.
|
||||
* @return NS_OK if succeeded else ERROR.
|
||||
*/
|
||||
static nsresult CreateCompatibleDTDForCommand(nsIDTD** aDTD, eParserCommands aCommand=eViewNormal);
|
||||
protected:
|
||||
//*********************************************
|
||||
// And now, some data members...
|
||||
|
||||
@ -54,6 +54,19 @@
|
||||
#define NS_XIF_DTD_CID \
|
||||
{ 0xa6cf910e, 0x15b3, 0x11d2, { 0x93, 0x2e, 0x0, 0x80, 0x5f, 0x8a, 0xdd, 0x32 } }
|
||||
|
||||
// {CCF5BED0-1AF8-11d4-812B-0010A4E0C706}
|
||||
#define NS_COTHER_DTD_CID \
|
||||
{ 0xccf5bed0, 0x1af8, 0x11d4, { 0x81, 0x2b, 0x0, 0x10, 0xa4, 0xe0, 0xc7, 0x6 } };
|
||||
|
||||
// {8323FAD0-2102-11d4-8142-000064657374}
|
||||
#define NS_VIEWSOURCE_DTD_CID \
|
||||
{ 0x8323fad0, 0x2102, 0x11d4, { 0x81, 0x42, 0x0, 0x0, 0x64, 0x65, 0x73, 0x74 } };
|
||||
|
||||
// {8323FAD1-2102-11d4-8142-000064657374}
|
||||
#define NS_CRTF_DTD_CID \
|
||||
{ 0x8323fad1, 0x2102, 0x11d4, { 0x81, 0x42, 0x0, 0x0, 0x64, 0x65, 0x73, 0x74 } };
|
||||
|
||||
|
||||
// {a6cf910f-15b3-11d2-932e-00805f8add32}
|
||||
#define NS_HTMLCONTENTSINKSTREAM_CID \
|
||||
{ 0xa6cf910f, 0x15b3, 0x11d2, { 0x93, 0x2e, 0x0, 0x80, 0x5f, 0x8a, 0xdd, 0x32 } }
|
||||
|
||||
@ -31,6 +31,9 @@
|
||||
#include "nsWellFormedDTD.h"
|
||||
#include "CNavDTD.h"
|
||||
#include "nsXIFDTD.h"
|
||||
#include "COtherDTD.h"
|
||||
#include "CRtfDTD.h"
|
||||
#include "nsViewSourceHTML.h"
|
||||
#include "nsHTMLContentSinkStream.h"
|
||||
#include "nsHTMLToTXTSinkStream.h"
|
||||
#include "nsHTMLEntities.h"
|
||||
@ -98,6 +101,9 @@ static NS_DEFINE_CID(kLoggingSinkCID, NS_LOGGING_SINK_CID);
|
||||
static NS_DEFINE_CID(kWellFormedDTDCID, NS_WELLFORMEDDTD_CID);
|
||||
static NS_DEFINE_CID(kNavDTDCID, NS_CNAVDTD_CID);
|
||||
static NS_DEFINE_CID(kXIFDTDCID, NS_XIF_DTD_CID);
|
||||
static NS_DEFINE_CID(kCOtherDTDCID, NS_COTHER_DTD_CID);
|
||||
static NS_DEFINE_CID(kViewSourceDTDCID, NS_VIEWSOURCE_DTD_CID);
|
||||
static NS_DEFINE_CID(kRtfDTDCID, NS_CRTF_DTD_CID);
|
||||
static NS_DEFINE_CID(kHTMLContentSinkStreamCID, NS_HTMLCONTENTSINKSTREAM_CID);
|
||||
static NS_DEFINE_CID(kHTMLToTXTSinkStreamCID, NS_HTMLTOTXTSINKSTREAM_CID);
|
||||
static NS_DEFINE_CID(kParserServiceCID, NS_PARSERSERVICE_CID);
|
||||
@ -114,6 +120,9 @@ static Components gComponents[] = {
|
||||
{ "Well formed DTD", &kWellFormedDTDCID },
|
||||
{ "Navigator HTML DTD", &kNavDTDCID },
|
||||
{ "XIF DTD", &kXIFDTDCID },
|
||||
{ "OTHER DTD", &kCOtherDTDCID },
|
||||
{ "ViewSource DTD", &kViewSourceDTDCID },
|
||||
{ "Rtf DTD", &kRtfDTDCID },
|
||||
{ "HTML Content Sink Stream", &kHTMLContentSinkStreamCID },
|
||||
{ "HTML To Text Sink Stream", &kHTMLToTXTSinkStreamCID },
|
||||
{ "ParserService", &kParserServiceCID },
|
||||
@ -127,6 +136,9 @@ NS_GENERIC_FACTORY_CONSTRUCTOR(nsLoggingSink)
|
||||
NS_GENERIC_FACTORY_CONSTRUCTOR(CWellFormedDTD)
|
||||
NS_GENERIC_FACTORY_CONSTRUCTOR(CNavDTD)
|
||||
NS_GENERIC_FACTORY_CONSTRUCTOR(nsXIFDTD)
|
||||
NS_GENERIC_FACTORY_CONSTRUCTOR(COtherDTD)
|
||||
NS_GENERIC_FACTORY_CONSTRUCTOR(CViewSourceHTML)
|
||||
NS_GENERIC_FACTORY_CONSTRUCTOR(CRtfDTD)
|
||||
NS_GENERIC_FACTORY_CONSTRUCTOR(nsHTMLContentSinkStream)
|
||||
NS_GENERIC_FACTORY_CONSTRUCTOR(nsHTMLToTXTSinkStream)
|
||||
NS_GENERIC_FACTORY_CONSTRUCTOR(nsParserService)
|
||||
@ -154,6 +166,9 @@ protected:
|
||||
nsCOMPtr<nsIGenericFactory> mWellFormedDTDFactory;
|
||||
nsCOMPtr<nsIGenericFactory> mNavHTMLDTDFactory;
|
||||
nsCOMPtr<nsIGenericFactory> mXIFDTDFactory;
|
||||
nsCOMPtr<nsIGenericFactory> mOtherHTMLDTDFactory;
|
||||
nsCOMPtr<nsIGenericFactory> mViewSourceHTMLDTDFactory;
|
||||
nsCOMPtr<nsIGenericFactory> mRtfHTMLDTDFactory;
|
||||
nsCOMPtr<nsIGenericFactory> mHTMLContentSinkStreamFactory;
|
||||
nsCOMPtr<nsIGenericFactory> mHTMLToTXTSinkStreamFactory;
|
||||
nsCOMPtr<nsIGenericFactory> mParserServiceFactory;
|
||||
@ -260,6 +275,27 @@ nsParserModule::GetClassObject(nsIComponentManager *aCompMgr,
|
||||
&nsXIFDTDConstructor);
|
||||
}
|
||||
fact = mXIFDTDFactory;
|
||||
}
|
||||
else if (aClass.Equals(kCOtherDTDCID)) {
|
||||
if (!mOtherHTMLDTDFactory) {
|
||||
rv = NS_NewGenericFactory(getter_AddRefs(mOtherHTMLDTDFactory),
|
||||
&COtherDTDConstructor);
|
||||
}
|
||||
fact = mOtherHTMLDTDFactory;
|
||||
}
|
||||
else if (aClass.Equals(kViewSourceDTDCID)) {
|
||||
if (!mViewSourceHTMLDTDFactory) {
|
||||
rv = NS_NewGenericFactory(getter_AddRefs(mViewSourceHTMLDTDFactory),
|
||||
&CViewSourceHTMLConstructor);
|
||||
}
|
||||
fact = mViewSourceHTMLDTDFactory;
|
||||
}
|
||||
else if (aClass.Equals(kRtfDTDCID)) {
|
||||
if (!mRtfHTMLDTDFactory) {
|
||||
rv = NS_NewGenericFactory(getter_AddRefs(mRtfHTMLDTDFactory),
|
||||
&CRtfDTDConstructor);
|
||||
}
|
||||
fact = mRtfHTMLDTDFactory;
|
||||
}
|
||||
else if (aClass.Equals(kHTMLContentSinkStreamCID)) {
|
||||
if (!mHTMLContentSinkStreamFactory) {
|
||||
|
||||
@ -1332,7 +1332,7 @@ nsresult CNavDTD::HandleStartToken(CToken* aToken) {
|
||||
break;
|
||||
|
||||
case eHTMLTag_noscript:
|
||||
mHasOpenNoXXX++;
|
||||
//mHasOpenNoXXX++; // Fix for 33397 - Enable this when we handle NOSCRIPTS.
|
||||
isTokenHandled=PR_TRUE; // XXX - Throwing NOSCRIPT to the floor...yet another time..
|
||||
break;
|
||||
|
||||
|
||||
@ -142,186 +142,4 @@ void CParserContext::SetMimeType(const nsString& aMimeType){
|
||||
mDocType=eXMLText;
|
||||
}
|
||||
|
||||
/*************************************************************************************************
|
||||
First, let's define our modalities:
|
||||
|
||||
1. compatibility-mode: behave as much like nav4 as possible (unless it's too broken to bother)
|
||||
2. standard-mode: do html as well as you can per spec, and throw out navigator quirks
|
||||
3. strict-mode: adhere to the strict DTD specificiation to the highest degree possible
|
||||
|
||||
Assume the doctype is in the following form:
|
||||
<!DOCTYPE [Top Level Element] [Availability] "[Registration]// [Owner-ID] // [Type] [desc-text] // [Language]" "URI|text-identifier">
|
||||
[HTML] [PUBLIC|...] [+|-] [W3C|IETF|...] [DTD] "..." [EN]|...] "..."
|
||||
|
||||
|
||||
Here are the new rules for DTD handling; comments welcome:
|
||||
|
||||
XHTML and XML documents are always strict-mode:
|
||||
example: <!DOCTYPE \"-//W3C//DTD XHTML 1.0 Strict//EN\">
|
||||
|
||||
HTML strict dtd's enable strict-mode:
|
||||
example: <!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0//EN">
|
||||
example: <!DOCTYPE \"ISO/IEC 15445:1999//DTD HTML//EN\">
|
||||
|
||||
HTML 4.0 (or greater) transitional, frameset, (etc), without URI enables compatibility-mode:
|
||||
example: <!DOCTYPE \"-//W3C//DTD HTML 4.01 Transitional//EN\">
|
||||
|
||||
HTML 4.0 (or greater) transitional, frameset, (etc), with a URI that points to the strict.dtd will become strict:
|
||||
example: <!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN"
|
||||
"http://www.w3.org/TR/REC-html40/strict.dtd">
|
||||
|
||||
doctypes with systemID's or internal subset are handled in strict-mode:
|
||||
example: <!DOCTYPE HTML PUBLIC PublicID SystemID>
|
||||
example: <!DOCTYPE HTML SYSTEM SystemID>
|
||||
example: <!DOCTYPE HTML (PUBLIC PublicID SystemID? | SYSTEM SystemID) [ Internal-SS ]>
|
||||
|
||||
All other doctypes (<4.0), and documents without a doctype are handled in compatibility-mode.
|
||||
|
||||
*****************************************************************************************************/
|
||||
|
||||
/**
|
||||
* This is called when it's time to find out
|
||||
* what mode the parser/DTD should run for this document.
|
||||
* (Each parsercontext can have it's own mode).
|
||||
*
|
||||
* @update gess 02/17/00
|
||||
* @return parsermode (define in nsIParser.h)
|
||||
*/
|
||||
eParseMode CParserContext::DetermineParseMode(const nsString& theBuffer) {
|
||||
const char* theModeStr= PR_GetEnv("PARSE_MODE");
|
||||
|
||||
mParseMode = eParseMode_unknown;
|
||||
|
||||
PRInt32 theIndex=theBuffer.Find("DOCTYPE",PR_TRUE,0,10);
|
||||
if(kNotFound<theIndex) {
|
||||
|
||||
//good, we found "DOCTYPE" -- now go find it's end delimiter '>'
|
||||
PRInt32 theGTPos=theBuffer.FindChar(kGreaterThan,theIndex+1);
|
||||
PRInt32 theEnd=(kNotFound==theGTPos) ? 512 : MinInt(512,theGTPos);
|
||||
PRInt32 theSubIndex=theBuffer.Find("//DTD",PR_TRUE,theIndex+8,theEnd-(theIndex+8)); //skip to the type and desc-text...
|
||||
PRInt32 theErr=0;
|
||||
PRInt32 theMajorVersion=3;
|
||||
|
||||
//note that if we don't find '>', then we just scan the first 512 bytes.
|
||||
|
||||
if(0<=theSubIndex) {
|
||||
PRInt32 theStartPos=theSubIndex+5;
|
||||
PRInt32 theCount=theEnd-theStartPos;
|
||||
|
||||
if(kNotFound<theSubIndex) {
|
||||
|
||||
theSubIndex=theBuffer.Find("XHTML",PR_TRUE,theStartPos,theCount);
|
||||
if(0<=theSubIndex) {
|
||||
mDocType=eXHTMLText;
|
||||
mParseMode=eParseMode_strict;
|
||||
return mParseMode;
|
||||
}
|
||||
else {
|
||||
theSubIndex=theBuffer.Find("ISO/IEC 15445:",PR_TRUE,theIndex+8,theEnd-(theIndex+8));
|
||||
if(0<=theSubIndex) {
|
||||
mDocType=eHTML4Text;
|
||||
mParseMode=eParseMode_strict;
|
||||
theMajorVersion=4;
|
||||
theSubIndex+=15;
|
||||
}
|
||||
else {
|
||||
theSubIndex=theBuffer.Find("HTML",PR_TRUE,theStartPos,theCount);
|
||||
if(0<=theSubIndex) {
|
||||
mDocType=eHTML4Text;
|
||||
mParseMode=eParseMode_strict;
|
||||
theMajorVersion=3;
|
||||
}
|
||||
else {
|
||||
theSubIndex=theBuffer.Find("HYPERTEXT MARKUP",PR_TRUE,theStartPos,theCount);
|
||||
if(0<=theSubIndex) {
|
||||
mDocType=eHTML3Text;
|
||||
mParseMode=eParseMode_quirks;
|
||||
theSubIndex+=20;
|
||||
}
|
||||
}
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
theStartPos=theSubIndex+5;
|
||||
theCount=theEnd-theStartPos;
|
||||
nsAutoString theNum;
|
||||
|
||||
//get the next substring from the buffer, which should be a number.
|
||||
//now see what the version number is...
|
||||
|
||||
theStartPos=theBuffer.FindCharInSet("123456789",theStartPos);
|
||||
if(0<=theStartPos) {
|
||||
PRInt32 theTerminal=theBuffer.FindCharInSet(" />",theStartPos+1);
|
||||
if(theTerminal) {
|
||||
theBuffer.Mid(theNum,theStartPos,theTerminal-theStartPos);
|
||||
}
|
||||
else theBuffer.Mid(theNum,theStartPos,3);
|
||||
theMajorVersion=theNum.ToInteger(&theErr);
|
||||
}
|
||||
|
||||
//now see what the
|
||||
theStartPos+=theNum.Length();
|
||||
theCount=theEnd-theStartPos;
|
||||
if((theBuffer.Find("TRANSITIONAL",PR_TRUE,theStartPos,theCount)>kNotFound)||
|
||||
(theBuffer.Find("LOOSE",PR_TRUE,theStartPos,theCount)>kNotFound) ||
|
||||
(theBuffer.Find("FRAMESET",PR_TRUE,theStartPos,theCount)>kNotFound) ||
|
||||
(theBuffer.Find("LATIN1", PR_TRUE,theStartPos,theCount) >kNotFound) ||
|
||||
(theBuffer.Find("SYMBOLS",PR_TRUE,theStartPos,theCount) >kNotFound) ||
|
||||
(theBuffer.Find("SPECIAL",PR_TRUE,theStartPos,theCount) >kNotFound)) {
|
||||
mParseMode=eParseMode_quirks;
|
||||
}
|
||||
|
||||
//one last thing: look for a URI that specifies the strict.dtd
|
||||
theStartPos+=6;
|
||||
theCount=theEnd-theStartPos;
|
||||
theSubIndex=theBuffer.Find("STRICT.DTD",PR_TRUE,theStartPos,theCount);
|
||||
if(0<theSubIndex) {
|
||||
//Since we found it, regardless of what's in the descr-text, kick into strict mode.
|
||||
mParseMode=eParseMode_strict;
|
||||
mDocType=eHTML4Text;
|
||||
}
|
||||
|
||||
if (0==theErr){
|
||||
switch(theMajorVersion) {
|
||||
case 0: case 1: case 2: case 3:
|
||||
if(mDocType!=eXHTMLText){
|
||||
mParseMode=eParseMode_quirks; //be as backward compatible as possible
|
||||
mDocType=eHTML3Text;
|
||||
}
|
||||
break;
|
||||
|
||||
default:
|
||||
//XXX hack -- someday, the next line of code will be criticized
|
||||
//for it's lack of vision...
|
||||
if(theMajorVersion>20) {
|
||||
mParseMode=eParseMode_noquirks;
|
||||
}
|
||||
break;
|
||||
} //switch
|
||||
}
|
||||
|
||||
} //if
|
||||
else {
|
||||
PRInt32 thePos=theBuffer.Find("HTML",PR_TRUE,1,50);
|
||||
if(kNotFound!=thePos) {
|
||||
mDocType=eHTML4Text;
|
||||
PRInt32 theIDPos=theBuffer.Find("PublicID",thePos);
|
||||
if(kNotFound==theIDPos)
|
||||
theIDPos=theBuffer.Find("SystemID",thePos);
|
||||
mParseMode=(kNotFound==theIDPos) ? eParseMode_quirks : eParseMode_strict;
|
||||
}
|
||||
}
|
||||
}
|
||||
else if(kNotFound<(theIndex=theBuffer.Find("?XML",PR_TRUE,0,128))) {
|
||||
mParseMode=eParseMode_strict;
|
||||
}
|
||||
|
||||
if(theModeStr) {
|
||||
if(0==nsCRT::strcasecmp(theModeStr,"strict"))
|
||||
mParseMode=eParseMode_strict;
|
||||
}
|
||||
else mParseMode = (eParseMode_unknown==mParseMode)? eParseMode_quirks : mParseMode;
|
||||
|
||||
return mParseMode;
|
||||
}
|
||||
|
||||
@ -58,8 +58,6 @@ public:
|
||||
PRBool aCopyUnused=PR_FALSE);
|
||||
|
||||
CParserContext( const CParserContext& aContext);
|
||||
eParseMode DetermineParseMode(const nsString& theBuffer);
|
||||
|
||||
|
||||
~CParserContext();
|
||||
|
||||
|
||||
@ -992,7 +992,7 @@ nsHTMLContentSinkStream::AddDocTypeDecl(const nsIParserNode& aNode, PRInt32 aMod
|
||||
DebugDump("<",aNode.GetText(),(mNodeStackPos)*2);
|
||||
#endif
|
||||
|
||||
Write("<!");
|
||||
// Write("<!");
|
||||
|
||||
return NS_OK;
|
||||
}
|
||||
|
||||
@ -540,7 +540,7 @@ nsresult nsHTMLTokenizer::ConsumeStartTag(PRUnichar aChar,CToken*& aToken,nsScan
|
||||
if(NS_SUCCEEDED(result)) {
|
||||
|
||||
//XXX - Find a better soution to record content
|
||||
if(theTag==eHTMLTag_textarea && !mRecordTrailingContent) {
|
||||
if((theTag==eHTMLTag_textarea || theTag==eHTMLTag_xmp) && !mRecordTrailingContent) {
|
||||
mRecordTrailingContent=PR_TRUE;
|
||||
}
|
||||
|
||||
@ -589,7 +589,8 @@ nsresult nsHTMLTokenizer::ConsumeEndTag(PRUnichar aChar,CToken*& aToken,nsScanne
|
||||
nsresult result=NS_OK;
|
||||
|
||||
if(aToken) {
|
||||
if(aToken->GetTypeID()==eHTMLTag_textarea && mRecordTrailingContent) {
|
||||
eHTMLTags theTag=(eHTMLTags)aToken->GetTypeID();
|
||||
if((theTag==eHTMLTag_textarea || theTag==eHTMLTag_xmp) && mRecordTrailingContent) {
|
||||
mRecordTrailingContent=PR_FALSE;
|
||||
}
|
||||
result= aToken->Consume(aChar,aScanner,mParseMode); //tell new token to finish consuming text...
|
||||
|
||||
@ -282,16 +282,17 @@ void CStartToken::DebugDumpSource(nsOutputStream& out) {
|
||||
void CStartToken::GetSource(nsString& anOutputString){
|
||||
anOutputString.AssignWithConversion("<");
|
||||
/*
|
||||
* mTextValue used to contain the name of the tag.
|
||||
* But for the sake of performance we now rely on the tagID
|
||||
* rather than tag name. This however, caused bug 15204
|
||||
* to reincarnate. Since, mTextvalue is not being used here..
|
||||
* I'm just going to comment it out.
|
||||
*
|
||||
* Watch out for Bug 15204
|
||||
*/
|
||||
// anOutputString+=mTextValue;
|
||||
if(mTrailingContent.Length()>0)
|
||||
anOutputString+=mTrailingContent;
|
||||
anOutputString=mTrailingContent;
|
||||
else {
|
||||
if(mTextValue.Length()>0)
|
||||
anOutputString=mTextValue;
|
||||
else
|
||||
anOutputString.AssignWithConversion(GetTagName(mTypeID));
|
||||
anOutputString+='>';
|
||||
}
|
||||
}
|
||||
|
||||
/*
|
||||
@ -304,16 +305,17 @@ void CStartToken::GetSource(nsString& anOutputString){
|
||||
void CStartToken::AppendSource(nsString& anOutputString){
|
||||
anOutputString.AppendWithConversion("<");
|
||||
/*
|
||||
* mTextValue used to contain the name of the tag.
|
||||
* But for the sake of performance we now rely on the tagID
|
||||
* rather than tag name. This however, caused bug 15204
|
||||
* to reincarnate. Since, mTextvalue is not being used here..
|
||||
* I'm just going to comment it out.
|
||||
*
|
||||
* Watch out for Bug 15204
|
||||
*/
|
||||
// anOutputString+=mTextValue;
|
||||
if(mTrailingContent.Length()>0)
|
||||
anOutputString+=mTrailingContent;
|
||||
else {
|
||||
if(mTextValue.Length()>0)
|
||||
anOutputString+=mTextValue;
|
||||
else
|
||||
anOutputString.AppendWithConversion(GetTagName(mTypeID));
|
||||
anOutputString+='>';
|
||||
}
|
||||
}
|
||||
|
||||
/*
|
||||
@ -655,10 +657,16 @@ nsresult CTextToken::ConsumeUntil(PRUnichar aChar,PRBool aIgnoreComments,nsScann
|
||||
else disaster=PR_TRUE;
|
||||
|
||||
if(disaster) {
|
||||
if((!aScanner.IsIncremental()) && (theAltTermStrPos>kNotFound)) {
|
||||
// If you're here it means..we hit the rock bottom and therefore switch to plan B.
|
||||
theCurrOffset=theAltTermStrPos;
|
||||
theLastIteration=PR_TRUE;
|
||||
if(!aScanner.IsIncremental()) {
|
||||
if(theAltTermStrPos>kNotFound) {
|
||||
// If you're here it means..we hit the rock bottom and therefore switch to plan B.
|
||||
theCurrOffset=theAltTermStrPos;
|
||||
theLastIteration=PR_TRUE;
|
||||
}
|
||||
else {
|
||||
aTerminalString.Cut(0,2); // Do this to fix Bug. 35456
|
||||
done=PR_TRUE;
|
||||
}
|
||||
}
|
||||
else
|
||||
result=kEOF;
|
||||
|
||||
@ -42,15 +42,24 @@
|
||||
#include "CRtfDTD.h"
|
||||
#include "CNavDTD.h"
|
||||
#include "COtherDTD.h"
|
||||
|
||||
//#define rickgdebug
|
||||
|
||||
#include "prenv.h"
|
||||
#include "nsParserCIID.h"
|
||||
|
||||
//#define rickgdebug
|
||||
|
||||
#define TEST_DOCTYPES 1
|
||||
|
||||
static NS_DEFINE_IID(kISupportsIID, NS_ISUPPORTS_IID);
|
||||
static NS_DEFINE_IID(kClassIID, NS_PARSER_IID);
|
||||
static NS_DEFINE_IID(kIParserIID, NS_IPARSER_IID);
|
||||
static NS_DEFINE_IID(kIStreamListenerIID, NS_ISTREAMLISTENER_IID);
|
||||
|
||||
static NS_DEFINE_CID(kWellFormedDTDCID, NS_WELLFORMEDDTD_CID);
|
||||
static NS_DEFINE_CID(kNavDTDCID, NS_CNAVDTD_CID);
|
||||
static NS_DEFINE_CID(kCOtherDTDCID, NS_COTHER_DTD_CID);
|
||||
static NS_DEFINE_CID(kViewSourceDTDCID, NS_VIEWSOURCE_DTD_CID);
|
||||
static NS_DEFINE_CID(kRtfDTDCID, NS_CRTF_DTD_CID);
|
||||
|
||||
static const char* kNullURL = "Error: Null URL given";
|
||||
static const char* kOnStartNotCalled = "Error: OnStartRequest() must be called before OnDataAvailable()";
|
||||
static const char* kBadListenerInit = "Error: Parser's IStreamListener API was not setup correctly in constructor.";
|
||||
@ -427,6 +436,194 @@ eParseMode nsParser::GetParseMode(void){
|
||||
}
|
||||
|
||||
|
||||
/*************************************************************************************************
|
||||
First, let's define our modalities:
|
||||
|
||||
1. compatibility-mode: behave as much like nav4 as possible (unless it's too broken to bother)
|
||||
2. standard-mode: do html as well as you can per spec, and throw out navigator quirks
|
||||
3. strict-mode: adhere to the strict DTD specificiation to the highest degree possible
|
||||
|
||||
Assume the doctype is in the following form:
|
||||
<!DOCTYPE [Top Level Element] [Availability] "[Registration]// [Owner-ID] // [Type] [desc-text] // [Language]" "URI|text-identifier">
|
||||
[HTML] [PUBLIC|...] [+|-] [W3C|IETF|...] [DTD] "..." [EN]|...] "..."
|
||||
|
||||
|
||||
Here are the new rules for DTD handling; comments welcome:
|
||||
|
||||
XHTML and XML documents are always strict-mode:
|
||||
example: <!DOCTYPE \"-//W3C//DTD XHTML 1.0 Strict//EN\">
|
||||
|
||||
HTML strict dtd's enable strict-mode:
|
||||
example: <!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0//EN">
|
||||
example: <!DOCTYPE \"ISO/IEC 15445:1999//DTD HTML//EN\">
|
||||
|
||||
HTML 4.0 (or greater) transitional, frameset, (etc), without URI enables compatibility-mode:
|
||||
example: <!DOCTYPE \"-//W3C//DTD HTML 4.01 Transitional//EN\">
|
||||
|
||||
HTML 4.0 (or greater) transitional, frameset, (etc), with a URI that points to the strict.dtd will become strict:
|
||||
example: <!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN"
|
||||
"http://www.w3.org/TR/REC-html40/strict.dtd">
|
||||
|
||||
doctypes with systemID's or internal subset are handled in strict-mode:
|
||||
example: <!DOCTYPE HTML PUBLIC PublicID SystemID>
|
||||
example: <!DOCTYPE HTML SYSTEM SystemID>
|
||||
example: <!DOCTYPE HTML (PUBLIC PublicID SystemID? | SYSTEM SystemID) [ Internal-SS ]>
|
||||
|
||||
All other doctypes (<4.0), and documents without a doctype are handled in compatibility-mode.
|
||||
|
||||
*****************************************************************************************************/
|
||||
|
||||
/**
|
||||
* This is called when it's time to find out
|
||||
* what mode the parser/DTD should run for this document.
|
||||
* (Each parsercontext can have it's own mode).
|
||||
*
|
||||
* @update gess 02/17/00
|
||||
* @return parsermode (define in nsIParser.h)
|
||||
*/
|
||||
static
|
||||
void DetermineParseMode(nsString& aBuffer,eParseMode& aParseMode,eParserDocType aDocType) {
|
||||
const char* theModeStr= PR_GetEnv("PARSE_MODE");
|
||||
|
||||
aParseMode = eParseMode_unknown;
|
||||
|
||||
PRInt32 theIndex=aBuffer.Find("DOCTYPE",PR_TRUE,0,10);
|
||||
if(kNotFound<theIndex) {
|
||||
|
||||
//good, we found "DOCTYPE" -- now go find it's end delimiter '>'
|
||||
PRInt32 theGTPos=aBuffer.FindChar(kGreaterThan,theIndex+1);
|
||||
PRInt32 theEnd=(kNotFound==theGTPos) ? 512 : MinInt(512,theGTPos);
|
||||
PRInt32 theSubIndex=aBuffer.Find("//DTD",PR_TRUE,theIndex+8,theEnd-(theIndex+8)); //skip to the type and desc-text...
|
||||
PRInt32 theErr=0;
|
||||
PRInt32 theMajorVersion=3;
|
||||
|
||||
//note that if we don't find '>', then we just scan the first 512 bytes.
|
||||
|
||||
if(0<=theSubIndex) {
|
||||
PRInt32 theStartPos=theSubIndex+5;
|
||||
PRInt32 theCount=theEnd-theStartPos;
|
||||
|
||||
if(kNotFound<theSubIndex) {
|
||||
|
||||
theSubIndex=aBuffer.Find("XHTML",PR_TRUE,theStartPos,theCount);
|
||||
if(0<=theSubIndex) {
|
||||
aDocType=eXHTMLText;
|
||||
aParseMode=eParseMode_strict;
|
||||
}
|
||||
else {
|
||||
theSubIndex=aBuffer.Find("ISO/IEC 15445:",PR_TRUE,theIndex+8,theEnd-(theIndex+8));
|
||||
if(0<=theSubIndex) {
|
||||
aDocType=eHTML4Text;
|
||||
aParseMode=eParseMode_strict;
|
||||
theMajorVersion=4;
|
||||
theSubIndex+=15;
|
||||
}
|
||||
else {
|
||||
theSubIndex=aBuffer.Find("HTML",PR_TRUE,theStartPos,theCount);
|
||||
if(0<=theSubIndex) {
|
||||
aDocType=eHTML4Text;
|
||||
aParseMode=eParseMode_strict;
|
||||
theMajorVersion=3;
|
||||
}
|
||||
else {
|
||||
theSubIndex=aBuffer.Find("HYPERTEXT MARKUP",PR_TRUE,theStartPos,theCount);
|
||||
if(0<=theSubIndex) {
|
||||
aDocType=eHTML3Text;
|
||||
aParseMode=eParseMode_quirks;
|
||||
theSubIndex+=20;
|
||||
}
|
||||
}
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
theStartPos=theSubIndex+5;
|
||||
theCount=theEnd-theStartPos;
|
||||
nsAutoString theNum;
|
||||
|
||||
//get the next substring from the buffer, which should be a number.
|
||||
//now see what the version number is...
|
||||
|
||||
theStartPos=aBuffer.FindCharInSet("123456789",theStartPos);
|
||||
if(0<=theStartPos) {
|
||||
PRInt32 theTerminal=aBuffer.FindCharInSet(" />",theStartPos+1);
|
||||
if(theTerminal) {
|
||||
aBuffer.Mid(theNum,theStartPos,theTerminal-theStartPos);
|
||||
}
|
||||
else aBuffer.Mid(theNum,theStartPos,3);
|
||||
theMajorVersion=theNum.ToInteger(&theErr);
|
||||
}
|
||||
|
||||
//now see what the
|
||||
theStartPos+=theNum.Length();
|
||||
theCount=theEnd-theStartPos;
|
||||
if((aBuffer.Find("TRANSITIONAL",PR_TRUE,theStartPos,theCount)>kNotFound)||
|
||||
(aBuffer.Find("LOOSE",PR_TRUE,theStartPos,theCount)>kNotFound) ||
|
||||
(aBuffer.Find("FRAMESET",PR_TRUE,theStartPos,theCount)>kNotFound) ||
|
||||
(aBuffer.Find("LATIN1", PR_TRUE,theStartPos,theCount) >kNotFound) ||
|
||||
(aBuffer.Find("SYMBOLS",PR_TRUE,theStartPos,theCount) >kNotFound) ||
|
||||
(aBuffer.Find("SPECIAL",PR_TRUE,theStartPos,theCount) >kNotFound)) {
|
||||
aParseMode=eParseMode_quirks;
|
||||
}
|
||||
|
||||
//one last thing: look for a URI that specifies the strict.dtd
|
||||
theStartPos+=6;
|
||||
theCount=theEnd-theStartPos;
|
||||
theSubIndex=aBuffer.Find("STRICT.DTD",PR_TRUE,theStartPos,theCount);
|
||||
if(0<theSubIndex) {
|
||||
//Since we found it, regardless of what's in the descr-text, kick into strict mode.
|
||||
aParseMode=eParseMode_strict;
|
||||
aDocType=eHTML4Text;
|
||||
}
|
||||
|
||||
if (0==theErr){
|
||||
switch(theMajorVersion) {
|
||||
case 0: case 1: case 2: case 3:
|
||||
if(aDocType!=eXHTMLText){
|
||||
aParseMode=eParseMode_quirks; //be as backward compatible as possible
|
||||
aDocType=eHTML3Text;
|
||||
}
|
||||
break;
|
||||
|
||||
default:
|
||||
//XXX hack -- someday, the next line of code will be criticized
|
||||
//for it's lack of vision...
|
||||
if(theMajorVersion>20) {
|
||||
aParseMode=eParseMode_noquirks;
|
||||
}
|
||||
break;
|
||||
} //switch
|
||||
}
|
||||
|
||||
} //if
|
||||
else {
|
||||
PRInt32 thePos=aBuffer.Find("HTML",PR_TRUE,1,50);
|
||||
if(kNotFound!=thePos) {
|
||||
aDocType=eHTML4Text;
|
||||
PRInt32 theIDPos=aBuffer.Find("PublicID",thePos);
|
||||
if(kNotFound==theIDPos)
|
||||
theIDPos=aBuffer.Find("SystemID",thePos);
|
||||
aParseMode=(kNotFound==theIDPos) ? eParseMode_quirks : eParseMode_strict;
|
||||
}
|
||||
}
|
||||
}
|
||||
else if(kNotFound<(theIndex=aBuffer.Find("?XML",PR_TRUE,0,128))) {
|
||||
aDocType=eXMLText;
|
||||
aParseMode=eParseMode_strict;
|
||||
}
|
||||
|
||||
if(theModeStr) {
|
||||
if(0==nsCRT::strcasecmp(theModeStr,"strict"))
|
||||
aParseMode=eParseMode_strict;
|
||||
}
|
||||
else {
|
||||
if(eParseMode_unknown==aParseMode) {
|
||||
aBuffer.InsertWithConversion("<!DOCTYPE HTML PUBLIC \"-//W3C//DTD HTML 3.2 Final//EN\">\n",0);
|
||||
aDocType=eHTML3Text;
|
||||
aParseMode=eParseMode_quirks;
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
/**
|
||||
*
|
||||
@ -496,6 +693,131 @@ PRBool FindSuitableDTD( CParserContext& aParserContext,nsString& aBuffer) {
|
||||
return PR_FALSE;
|
||||
}
|
||||
|
||||
/**
|
||||
* Call this method to determine a DTD for a DOCTYPE
|
||||
*
|
||||
* @update harishd 05/01/00
|
||||
* @param aDTD -- Carries the deduced ( from DOCTYPE ) DTD.
|
||||
* @param aDocTypeStr -- A doctype for which a DTD is to be selected.
|
||||
* @param aMimeType -- A mimetype for which a DTD is to be selected.
|
||||
* Note: aParseMode might be required.
|
||||
* @param aCommand -- A command for which a DTD is to be selected.
|
||||
* @param aParseMode -- Used with aMimeType to choose the correct DTD.
|
||||
* @return NS_OK if succeeded else ERROR.
|
||||
*/
|
||||
nsresult nsParser::CreateCompatibleDTDForDocType(nsIDTD** aDTD, nsString* aDocTypeStr)
|
||||
{
|
||||
nsresult result=NS_OK;
|
||||
const nsCID* theDTDClassID=0;
|
||||
|
||||
if(aDocTypeStr) {
|
||||
eParseMode theParseMode=eParseMode_unknown;
|
||||
eParserDocType theDocType=ePlainText;
|
||||
|
||||
DetermineParseMode(*aDocTypeStr,theParseMode,theDocType);
|
||||
|
||||
switch(theDocType) {
|
||||
case eHTML4Text:
|
||||
if(theParseMode==eParseMode_strict) {
|
||||
theDTDClassID=&kCOtherDTDCID;
|
||||
break;
|
||||
}
|
||||
case eHTML3Text:
|
||||
theDTDClassID=&kNavDTDCID;
|
||||
break;
|
||||
case eXHTMLText:
|
||||
case eXMLText:
|
||||
theDTDClassID=&kWellFormedDTDCID;
|
||||
break;
|
||||
default:
|
||||
theDTDClassID=&kNavDTDCID;
|
||||
break;
|
||||
}
|
||||
}
|
||||
|
||||
result=(theDTDClassID)? nsComponentManager::CreateInstance(*theDTDClassID, nsnull, NS_GET_IID(nsIDTD),(void**)aDTD):NS_OK;
|
||||
|
||||
return result;
|
||||
|
||||
}
|
||||
|
||||
/**
|
||||
* Call this method to determine a DTD for a given mime type.
|
||||
*
|
||||
* @update harishd 05/01/00
|
||||
* @param aDTD -- Carries the deduced ( from DOCTYPE ) DTD.
|
||||
* @param aMimeType -- A mimetype for which a DTD is to be selected.
|
||||
* Note: aParseMode might be required.
|
||||
* @param aParseMode -- Used with aMimeType to choose the correct DTD.
|
||||
* @return NS_OK if succeeded else ERROR.
|
||||
*/
|
||||
nsresult CreateCompatibleDTDForMimeType(nsIDTD** aDTD, const nsString* aMimeType=nsnull,
|
||||
eParseMode aParseMode=eParseMode_unknown)
|
||||
{
|
||||
nsresult result=NS_OK;
|
||||
const nsCID* theDTDClassID=0;
|
||||
|
||||
if(aMimeType) {
|
||||
|
||||
NS_ASSERTION(aParseMode!=eParseMode_unknown,"DTD selection might require a parsemode");
|
||||
|
||||
if(aMimeType->EqualsWithConversion(kHTMLTextContentType)) {
|
||||
if(aParseMode==eParseMode_strict) {
|
||||
theDTDClassID=&kCOtherDTDCID;
|
||||
}
|
||||
else {
|
||||
theDTDClassID=&kNavDTDCID;
|
||||
}
|
||||
}
|
||||
else if(aMimeType->EqualsWithConversion(kPlainTextContentType)) {
|
||||
theDTDClassID=&kNavDTDCID;
|
||||
}
|
||||
else if(aMimeType->EqualsWithConversion(kXMLTextContentType) ||
|
||||
aMimeType->EqualsWithConversion(kXULTextContentType) ||
|
||||
aMimeType->EqualsWithConversion(kRDFTextContentType)) {
|
||||
theDTDClassID=&kWellFormedDTDCID;
|
||||
}
|
||||
else if(aMimeType->EqualsWithConversion(kXIFTextContentType)) {
|
||||
theDTDClassID=&kRtfDTDCID;
|
||||
}
|
||||
else {
|
||||
theDTDClassID=&kNavDTDCID;
|
||||
}
|
||||
result=(theDTDClassID)? nsComponentManager::CreateInstance(*theDTDClassID, nsnull, NS_GET_IID(nsIDTD),(void**)aDTD):NS_OK;
|
||||
}
|
||||
|
||||
return result;
|
||||
}
|
||||
|
||||
/**
|
||||
* Call this method to determine a DTD for a given command
|
||||
*
|
||||
* @update harishd 05/01/00
|
||||
* @param aDTD -- Carries the deduced ( from DOCTYPE ) DTD.
|
||||
* @param aCommand -- A command for which a DTD is to be selected.
|
||||
* @return NS_OK if succeeded else ERROR.
|
||||
*/
|
||||
nsresult CreateCompatibleDTDForCommand(nsIDTD** aDTD, eParserCommands aCommand=eViewNormal)
|
||||
{
|
||||
nsresult result=NS_OK;
|
||||
const nsCID* theDTDClassID=0;
|
||||
|
||||
switch(aCommand) {
|
||||
case eViewSource:
|
||||
theDTDClassID=&kViewSourceDTDCID;
|
||||
break;
|
||||
case eViewNormal:
|
||||
theDTDClassID=&kNavDTDCID;
|
||||
break;
|
||||
case eViewErrors:
|
||||
default:
|
||||
break;
|
||||
}
|
||||
result=(theDTDClassID)? nsComponentManager::CreateInstance(*theDTDClassID, nsnull, NS_GET_IID(nsIDTD),(void**)aDTD):NS_OK;
|
||||
|
||||
return result;
|
||||
}
|
||||
|
||||
#ifdef TEST_DOCTYPES
|
||||
static const char* doctypes[] = {
|
||||
|
||||
@ -608,7 +930,7 @@ static const char* doctypes[] = {
|
||||
*/
|
||||
nsresult nsParser::WillBuildModel(nsString& aFilename){
|
||||
|
||||
nsresult result=NS_OK;
|
||||
nsresult result=NS_OK;
|
||||
|
||||
#if TEST_DOCTYPES
|
||||
|
||||
@ -617,9 +939,13 @@ nsresult nsParser::WillBuildModel(nsString& aFilename){
|
||||
|
||||
if(!tested) {
|
||||
tested=PR_TRUE;
|
||||
eParseMode theParseMode=eParseMode_unknown;
|
||||
eParserDocType theDocumentType=ePlainText;
|
||||
|
||||
while(*theDocType) {
|
||||
nsAutoString theType(*theDocType);
|
||||
eParseMode result=mParserContext->DetermineParseMode(theType);
|
||||
nsAutoString theType;
|
||||
theType.AssignWithConversion(*theDocType);
|
||||
DetermineParseMode(theType,theParseMode,theDocumentType);
|
||||
theDocType++;
|
||||
}
|
||||
}
|
||||
@ -631,7 +957,7 @@ nsresult nsParser::WillBuildModel(nsString& aFilename){
|
||||
mMinorIteration=-1;
|
||||
|
||||
nsString& theBuffer=mParserContext->mScanner->GetBuffer();
|
||||
mParserContext->DetermineParseMode(theBuffer);
|
||||
DetermineParseMode(theBuffer,mParserContext->mParseMode,mParserContext->mDocType);
|
||||
|
||||
if(PR_TRUE==FindSuitableDTD(*mParserContext,theBuffer)) {
|
||||
mParserContext->mDTD->WillBuildModel( *mParserContext,mSink);
|
||||
|
||||
@ -360,7 +360,40 @@ private:
|
||||
*/
|
||||
PRBool DidTokenize(PRBool aIsFinalChunk = PR_FALSE);
|
||||
|
||||
|
||||
/**
|
||||
* Call this method to determine a DTD for a DOCTYPE
|
||||
*
|
||||
* @update harishd 05/01/00
|
||||
* @param aDTD -- Carries the deduced ( from DOCTYPE ) DTD.
|
||||
* @param aDocTypeStr -- A doctype for which a DTD is to be selected.
|
||||
* @param aMimeType -- A mimetype for which a DTD is to be selected.
|
||||
Note: aParseMode might be required.
|
||||
* @param aCommand -- A command for which a DTD is to be selected.
|
||||
* @param aParseMode -- Used with aMimeType to choose the correct DTD.
|
||||
* @return NS_OK if succeeded else ERROR.
|
||||
*/
|
||||
static nsresult CreateCompatibleDTDForDocType(nsIDTD** aDTD, nsString* aDocTypeStr);
|
||||
/**
|
||||
* Call this method to determine a DTD for a given mime type.
|
||||
*
|
||||
* @update harishd 05/01/00
|
||||
* @param aDTD -- Carries the deduced ( from DOCTYPE ) DTD.
|
||||
* @param aMimeType -- A mimetype for which a DTD is to be selected.
|
||||
Note: aParseMode might be required.
|
||||
* @param aParseMode -- Used with aMimeType to choose the correct DTD.
|
||||
* @return NS_OK if succeeded else ERROR.
|
||||
*/
|
||||
static nsresult CreateCompatibleDTDForMimeType(nsIDTD** aDTD, const nsString* aMimeType=nsnull,
|
||||
eParseMode aParseMode=eParseMode_unknown);
|
||||
/**
|
||||
* Call this method to determine a DTD for a given command
|
||||
*
|
||||
* @update harishd 05/01/00
|
||||
* @param aDTD -- Carries the deduced ( from DOCTYPE ) DTD.
|
||||
* @param aCommand -- A command for which a DTD is to be selected.
|
||||
* @return NS_OK if succeeded else ERROR.
|
||||
*/
|
||||
static nsresult CreateCompatibleDTDForCommand(nsIDTD** aDTD, eParserCommands aCommand=eViewNormal);
|
||||
protected:
|
||||
//*********************************************
|
||||
// And now, some data members...
|
||||
|
||||
@ -54,6 +54,19 @@
|
||||
#define NS_XIF_DTD_CID \
|
||||
{ 0xa6cf910e, 0x15b3, 0x11d2, { 0x93, 0x2e, 0x0, 0x80, 0x5f, 0x8a, 0xdd, 0x32 } }
|
||||
|
||||
// {CCF5BED0-1AF8-11d4-812B-0010A4E0C706}
|
||||
#define NS_COTHER_DTD_CID \
|
||||
{ 0xccf5bed0, 0x1af8, 0x11d4, { 0x81, 0x2b, 0x0, 0x10, 0xa4, 0xe0, 0xc7, 0x6 } };
|
||||
|
||||
// {8323FAD0-2102-11d4-8142-000064657374}
|
||||
#define NS_VIEWSOURCE_DTD_CID \
|
||||
{ 0x8323fad0, 0x2102, 0x11d4, { 0x81, 0x42, 0x0, 0x0, 0x64, 0x65, 0x73, 0x74 } };
|
||||
|
||||
// {8323FAD1-2102-11d4-8142-000064657374}
|
||||
#define NS_CRTF_DTD_CID \
|
||||
{ 0x8323fad1, 0x2102, 0x11d4, { 0x81, 0x42, 0x0, 0x0, 0x64, 0x65, 0x73, 0x74 } };
|
||||
|
||||
|
||||
// {a6cf910f-15b3-11d2-932e-00805f8add32}
|
||||
#define NS_HTMLCONTENTSINKSTREAM_CID \
|
||||
{ 0xa6cf910f, 0x15b3, 0x11d2, { 0x93, 0x2e, 0x0, 0x80, 0x5f, 0x8a, 0xdd, 0x32 } }
|
||||
|
||||
@ -31,6 +31,9 @@
|
||||
#include "nsWellFormedDTD.h"
|
||||
#include "CNavDTD.h"
|
||||
#include "nsXIFDTD.h"
|
||||
#include "COtherDTD.h"
|
||||
#include "CRtfDTD.h"
|
||||
#include "nsViewSourceHTML.h"
|
||||
#include "nsHTMLContentSinkStream.h"
|
||||
#include "nsHTMLToTXTSinkStream.h"
|
||||
#include "nsHTMLEntities.h"
|
||||
@ -98,6 +101,9 @@ static NS_DEFINE_CID(kLoggingSinkCID, NS_LOGGING_SINK_CID);
|
||||
static NS_DEFINE_CID(kWellFormedDTDCID, NS_WELLFORMEDDTD_CID);
|
||||
static NS_DEFINE_CID(kNavDTDCID, NS_CNAVDTD_CID);
|
||||
static NS_DEFINE_CID(kXIFDTDCID, NS_XIF_DTD_CID);
|
||||
static NS_DEFINE_CID(kCOtherDTDCID, NS_COTHER_DTD_CID);
|
||||
static NS_DEFINE_CID(kViewSourceDTDCID, NS_VIEWSOURCE_DTD_CID);
|
||||
static NS_DEFINE_CID(kRtfDTDCID, NS_CRTF_DTD_CID);
|
||||
static NS_DEFINE_CID(kHTMLContentSinkStreamCID, NS_HTMLCONTENTSINKSTREAM_CID);
|
||||
static NS_DEFINE_CID(kHTMLToTXTSinkStreamCID, NS_HTMLTOTXTSINKSTREAM_CID);
|
||||
static NS_DEFINE_CID(kParserServiceCID, NS_PARSERSERVICE_CID);
|
||||
@ -114,6 +120,9 @@ static Components gComponents[] = {
|
||||
{ "Well formed DTD", &kWellFormedDTDCID },
|
||||
{ "Navigator HTML DTD", &kNavDTDCID },
|
||||
{ "XIF DTD", &kXIFDTDCID },
|
||||
{ "OTHER DTD", &kCOtherDTDCID },
|
||||
{ "ViewSource DTD", &kViewSourceDTDCID },
|
||||
{ "Rtf DTD", &kRtfDTDCID },
|
||||
{ "HTML Content Sink Stream", &kHTMLContentSinkStreamCID },
|
||||
{ "HTML To Text Sink Stream", &kHTMLToTXTSinkStreamCID },
|
||||
{ "ParserService", &kParserServiceCID },
|
||||
@ -127,6 +136,9 @@ NS_GENERIC_FACTORY_CONSTRUCTOR(nsLoggingSink)
|
||||
NS_GENERIC_FACTORY_CONSTRUCTOR(CWellFormedDTD)
|
||||
NS_GENERIC_FACTORY_CONSTRUCTOR(CNavDTD)
|
||||
NS_GENERIC_FACTORY_CONSTRUCTOR(nsXIFDTD)
|
||||
NS_GENERIC_FACTORY_CONSTRUCTOR(COtherDTD)
|
||||
NS_GENERIC_FACTORY_CONSTRUCTOR(CViewSourceHTML)
|
||||
NS_GENERIC_FACTORY_CONSTRUCTOR(CRtfDTD)
|
||||
NS_GENERIC_FACTORY_CONSTRUCTOR(nsHTMLContentSinkStream)
|
||||
NS_GENERIC_FACTORY_CONSTRUCTOR(nsHTMLToTXTSinkStream)
|
||||
NS_GENERIC_FACTORY_CONSTRUCTOR(nsParserService)
|
||||
@ -154,6 +166,9 @@ protected:
|
||||
nsCOMPtr<nsIGenericFactory> mWellFormedDTDFactory;
|
||||
nsCOMPtr<nsIGenericFactory> mNavHTMLDTDFactory;
|
||||
nsCOMPtr<nsIGenericFactory> mXIFDTDFactory;
|
||||
nsCOMPtr<nsIGenericFactory> mOtherHTMLDTDFactory;
|
||||
nsCOMPtr<nsIGenericFactory> mViewSourceHTMLDTDFactory;
|
||||
nsCOMPtr<nsIGenericFactory> mRtfHTMLDTDFactory;
|
||||
nsCOMPtr<nsIGenericFactory> mHTMLContentSinkStreamFactory;
|
||||
nsCOMPtr<nsIGenericFactory> mHTMLToTXTSinkStreamFactory;
|
||||
nsCOMPtr<nsIGenericFactory> mParserServiceFactory;
|
||||
@ -260,6 +275,27 @@ nsParserModule::GetClassObject(nsIComponentManager *aCompMgr,
|
||||
&nsXIFDTDConstructor);
|
||||
}
|
||||
fact = mXIFDTDFactory;
|
||||
}
|
||||
else if (aClass.Equals(kCOtherDTDCID)) {
|
||||
if (!mOtherHTMLDTDFactory) {
|
||||
rv = NS_NewGenericFactory(getter_AddRefs(mOtherHTMLDTDFactory),
|
||||
&COtherDTDConstructor);
|
||||
}
|
||||
fact = mOtherHTMLDTDFactory;
|
||||
}
|
||||
else if (aClass.Equals(kViewSourceDTDCID)) {
|
||||
if (!mViewSourceHTMLDTDFactory) {
|
||||
rv = NS_NewGenericFactory(getter_AddRefs(mViewSourceHTMLDTDFactory),
|
||||
&CViewSourceHTMLConstructor);
|
||||
}
|
||||
fact = mViewSourceHTMLDTDFactory;
|
||||
}
|
||||
else if (aClass.Equals(kRtfDTDCID)) {
|
||||
if (!mRtfHTMLDTDFactory) {
|
||||
rv = NS_NewGenericFactory(getter_AddRefs(mRtfHTMLDTDFactory),
|
||||
&CRtfDTDConstructor);
|
||||
}
|
||||
fact = mRtfHTMLDTDFactory;
|
||||
}
|
||||
else if (aClass.Equals(kHTMLContentSinkStreamCID)) {
|
||||
if (!mHTMLContentSinkStreamFactory) {
|
||||
|
||||
Loading…
x
Reference in New Issue
Block a user