diff --git a/mozilla/htmlparser/src/CNavDTD.cpp b/mozilla/htmlparser/src/CNavDTD.cpp
index 77cf497ec02..3d8929884a5 100644
--- a/mozilla/htmlparser/src/CNavDTD.cpp
+++ b/mozilla/htmlparser/src/CNavDTD.cpp
@@ -942,10 +942,19 @@ nsresult CNavDTD::WillHandleStartTag(CToken* aToken,eHTMLTags aTag,nsCParserNode
if(theKey.EqualsIgnoreCase("HTTP-EQUIV")) {
const nsString& theKey2=aNode.GetKeyAt(1);
if(theKey2.EqualsIgnoreCase("CONTENT")) {
- nsScanner* theScanner=mParser->GetScanner();
- if(theScanner) {
- const nsString& theValue=aNode.GetValueAt(1);
- theScanner->SetDocumentCharset(theValue);
+ nsScanner* theScanner=mParser->GetScanner();
+ if(theScanner) {
+ const nsString& theValue=aNode.GetValueAt(1);
+ PRInt32 charsetValueStart = theValue.RFind("charset=", PR_TRUE ) ;
+ if(kNotFound != charsetValueStart) {
+ charsetValueStart += 8; // 8 = "charset=".length
+ PRInt32 charsetValueEnd = theValue.FindCharInSet("\'\";", charsetValueStart );
+ if(kNotFound == charsetValueEnd )
+ charsetValueEnd = theValue.Length();
+ nsAutoString theCharset;
+ theValue.Mid(theCharset, charsetValueStart, charsetValueEnd - charsetValueStart);
+ theScanner->SetDocumentCharset(theCharset, kCharsetFromMetaTag);
+ } //if
} //if
}
} //if
diff --git a/mozilla/htmlparser/src/nsParser.cpp b/mozilla/htmlparser/src/nsParser.cpp
index cab5e31ed91..c0d95d0c591 100644
--- a/mozilla/htmlparser/src/nsParser.cpp
+++ b/mozilla/htmlparser/src/nsParser.cpp
@@ -225,6 +225,7 @@ nsIParserFilter * nsParser::SetParserFilter(nsIParserFilter * aFilter)
return old;
}
+
/**
* Call this method once you've created a parser, and want to instruct it
* about the command which caused the parser to be constructed. For example,
@@ -526,6 +527,24 @@ PRBool nsParser::EnableParser(PRBool aState){
nsresult nsParser::Parse(nsIURL* aURL,nsIStreamObserver* aListener,PRBool aVerifyEnabled) {
NS_PRECONDITION(0!=aURL,kNullURL);
+ nsAutoString charset;
+ nsCharsetSource charsetSource;
+
+ // XXXX get HTTP charset here
+ // charset =
+ // charsetSource = kCharsetFromHTTPHeader;
+
+ // XXXX get User Prefernce charset here
+ // charset =
+ // charsetSource = kCharsetFromUserDefault;
+
+ // XXXX get Doc Type Default (e.g. UTF-8 for XML)
+
+ // XXX We should really put if doc == html for the following line
+ charset = "ISO-8859-1";
+ charsetSource = kCharsetFromDocTypeDefault;
+
+
nsresult result=kBadURL;
mDTDVerification=aVerifyEnabled;
if(aURL) {
@@ -533,7 +552,7 @@ nsresult nsParser::Parse(nsIURL* aURL,nsIStreamObserver* aListener,PRBool aVerif
nsresult rv = aURL->GetSpec(&spec);
if (rv != NS_OK) return rv;
nsAutoString theName(spec);
- CParserContext* pc=new CParserContext(new nsScanner(theName,PR_FALSE),aURL,aListener);
+ CParserContext* pc=new CParserContext(new nsScanner(theName,PR_FALSE, charset, charsetSource),aURL,aListener);
if(pc) {
pc->mMultipart=PR_TRUE;
pc->mContextType=CParserContext::eCTURL;
@@ -555,10 +574,27 @@ nsresult nsParser::Parse(fstream& aStream,PRBool aVerifyEnabled){
mDTDVerification=aVerifyEnabled;
nsresult result=NS_ERROR_OUT_OF_MEMORY;
+
+ nsAutoString charset;
+ nsCharsetSource charsetSource;
+
+ // XXXX get HTTP charset here
+ // charset =
+ // charsetSource = kCharsetFromHTTPHeader;
+
+ // XXXX get User Prefernce charset here
+ // charset =
+ // charsetSource = kCharsetFromUserDefault;
+
+ // XXXX get Doc Type Default (e.g. UTF-8 for XML)
+
+ // XXX We should really put if doc == html for the following line
+ charset = "ISO-8859-1";
+ charsetSource = kCharsetFromDocTypeDefault;
//ok, time to create our tokenizer and begin the process
nsAutoString theUnknownFilename("unknown");
- CParserContext* pc=new CParserContext(new nsScanner(theUnknownFilename,aStream,PR_FALSE),&aStream,0);
+ CParserContext* pc=new CParserContext(new nsScanner(theUnknownFilename,aStream, charset, charsetSource,PR_FALSE),&aStream,0);
if(pc) {
PushContext(*pc);
pc->mSourceType=kHTMLTextContentType;
@@ -593,6 +629,22 @@ nsresult nsParser::Parse(nsString& aSourceBuffer,void* aKey,const nsString& aCon
}
#endif
+ nsAutoString charset;
+ nsCharsetSource charsetSource;
+
+ // XXXX get HTTP charset here
+ // charset =
+ // charsetSource = kCharsetFromHTTPHeader;
+
+ // XXXX get User Prefernce charset here
+ // charset =
+ // charsetSource = kCharsetFromUserDefault;
+
+ // XXXX get Doc Type Default (e.g. UTF-8 for XML)
+
+ // XXX We should really put if doc == html for the following line
+ charset = "ISO-8859-1";
+ charsetSource = kCharsetFromDocTypeDefault;
//NOTE: Make sure that updates to this method don't cause
// bug #2361 to break again!
@@ -607,7 +659,7 @@ nsresult nsParser::Parse(nsString& aSourceBuffer,void* aKey,const nsString& aCon
if((!mParserContext) || (mParserContext->mKey!=aKey)) {
//only make a new context if we dont have one, OR if we do, but has a different context key...
- pc=new CParserContext(new nsScanner(mUnusedInput),aKey,0);
+ pc=new CParserContext(new nsScanner(mUnusedInput, charset, charsetSource),aKey, 0);
if(pc) {
PushContext(*pc);
pc->mStreamListenerState=eOnStart;
diff --git a/mozilla/htmlparser/src/nsScanner.cpp b/mozilla/htmlparser/src/nsScanner.cpp
index 070d0174afe..efe74c83906 100644
--- a/mozilla/htmlparser/src/nsScanner.cpp
+++ b/mozilla/htmlparser/src/nsScanner.cpp
@@ -23,6 +23,7 @@
#include "nsDebug.h"
#include "nsIServiceManager.h"
#include "nsICharsetConverterManager.h"
+#include "nsICharsetAlias.h"
const char* kBadHTMLText="Oops...
You just tried to read a non-existent document:
";
@@ -34,8 +35,6 @@ const int kBufsize=1;
const int kBufsize=64;
#endif
-// #define DEFAULTCHARSET "Shift_JIS"
-#define DEFAULTCHARSET "ISO-8859-1"
/**
* Use this constructor if you want i/o to be based on
@@ -46,8 +45,8 @@ const int kBufsize=64;
* @param aMode represents the parser mode (nav, other)
* @return
*/
-nsScanner::nsScanner(nsString& anHTMLString) :
- mBuffer(anHTMLString), mFilename("") , mCharset("")
+nsScanner::nsScanner(nsString& anHTMLString, const nsString& aCharset, nsCharsetSource aSource) :
+ mBuffer(anHTMLString), mFilename("")
{
mTotalRead=mBuffer.Length();
mIncremental=PR_TRUE;
@@ -55,9 +54,10 @@ nsScanner::nsScanner(nsString& anHTMLString) :
mOffset=0;
mMarkPos=-1;
mFileStream=0;
- nsAutoString defaultCharset(DEFAULTCHARSET);
mUnicodeDecoder = 0;
- SetDocumentCharset(defaultCharset);
+ mCharset = "";
+ mCharsetSource = kCharsetUninitialized;
+ SetDocumentCharset(aCharset, aSource);
}
/**
@@ -69,8 +69,8 @@ nsScanner::nsScanner(nsString& anHTMLString) :
* @param aFilename --
* @return
*/
-nsScanner::nsScanner(nsString& aFilename,PRBool aCreateStream) :
- mBuffer(""), mFilename(aFilename) , mCharset("")
+nsScanner::nsScanner(nsString& aFilename,PRBool aCreateStream, const nsString& aCharset, nsCharsetSource aSource) :
+ mBuffer(""), mFilename(aFilename)
{
mIncremental=PR_TRUE;
mOffset=0;
@@ -91,8 +91,9 @@ nsScanner::nsScanner(nsString& aFilename,PRBool aCreateStream) :
#endif
} //if
mUnicodeDecoder = 0;
- nsAutoString defaultCharset(DEFAULTCHARSET);
- SetDocumentCharset(defaultCharset);
+ mCharset = "";
+ mCharsetSource = kCharsetUninitialized;
+ SetDocumentCharset(aCharset, aSource);
}
@@ -105,8 +106,8 @@ nsScanner::nsScanner(nsString& aFilename,PRBool aCreateStream) :
* @param aFilename --
* @return
*/
-nsScanner::nsScanner(nsString& aFilename,fstream& aStream,PRBool assumeOwnership) :
- mBuffer(""), mFilename(aFilename) , mCharset("")
+nsScanner::nsScanner(nsString& aFilename,fstream& aStream,const nsString& aCharset, nsCharsetSource aSource, PRBool assumeOwnership) :
+ mBuffer(""), mFilename(aFilename)
{
mIncremental=PR_TRUE;
mOffset=0;
@@ -115,15 +116,46 @@ nsScanner::nsScanner(nsString& aFilename,fstream& aStream,PRBool assumeOwnership
mOwnsStream=assumeOwnership;
mFileStream=&aStream;
mUnicodeDecoder = 0;
- nsAutoString defaultCharset(DEFAULTCHARSET);
- SetDocumentCharset(defaultCharset);
+ mCharset = "";
+ mCharsetSource = kCharsetUninitialized;
+ SetDocumentCharset(aCharset, aSource);
}
-nsresult nsScanner::SetDocumentCharset(const nsString& aCharset )
+nsresult nsScanner::SetDocumentCharset(const nsString& aCharset , nsCharsetSource aSource)
{
+
nsresult res = NS_OK;
- if(! mCharset.EqualsIgnoreCase(aCharset)) // see do we need to change a converter.
+
+ if( aSource < mCharsetSource) // priority is lower the the current one , just
+ return res;
+
+ nsICharsetAlias* calias = nsnull;
+ res = nsServiceManager::GetService(kCharsetAliasCID,
+ kICharsetAliasIID,
+ (nsISupports**)&calias);
+
+ NS_ASSERTION( nsnull != calias, "cannot find charet alias");
+ nsAutoString charsetName = aCharset;
+ if( NS_SUCCEEDED(res) && (nsnull != calias))
{
+ PRBool same = PR_FALSE;
+ res = calias->Equals(aCharset, mCharset, &same);
+ if(NS_SUCCEEDED(res) && same)
+ {
+ return NS_OK; // no difference, don't change it
+ }
+ // different, need to change it
+ res = calias->GetPreferred(aCharset, charsetName);
+ nsServiceManager::ReleaseService(kCharsetAliasCID, calias);
+
+ if(NS_FAILED(res) && (kCharsetUninitialized == mCharsetSource) )
+ {
+ // failed - unknown alias , fallback to ISO-8859-1
+ charsetName = "ISO-8859-1";
+ }
+ mCharset = charsetName;
+ mCharsetSource = aSource;
+
nsICharsetConverterManager * ccm = nsnull;
res = nsServiceManager::GetService(kCharsetConverterManagerCID,
kICharsetConverterManagerIID,
@@ -131,13 +163,12 @@ nsresult nsScanner::SetDocumentCharset(const nsString& aCharset )
if(NS_SUCCEEDED(res) && (nsnull != ccm))
{
nsIUnicodeDecoder * decoder = nsnull;
- res = ccm->GetUnicodeDecoder(&aCharset, &decoder);
+ res = ccm->GetUnicodeDecoder(&mCharset, &decoder);
if(NS_SUCCEEDED(res) && (nsnull != decoder))
{
NS_IF_RELEASE(mUnicodeDecoder);
mUnicodeDecoder = decoder;
- mCharset = aCharset;
}
nsServiceManager::ReleaseService(kCharsetConverterManagerCID, ccm);
}
diff --git a/mozilla/htmlparser/src/nsScanner.h b/mozilla/htmlparser/src/nsScanner.h
index acbb3773c57..7fb72c89db8 100644
--- a/mozilla/htmlparser/src/nsScanner.h
+++ b/mozilla/htmlparser/src/nsScanner.h
@@ -39,6 +39,16 @@
#include
+typedef enum {
+ kCharsetUninitialized = 0,
+ kCharsetFromUserDefault ,
+ kCharsetFromDocTypeDefault,
+ kCharsetFromParentFrame,
+ kCharsetFromAutoDetection,
+ kCharsetFromMetaTag,
+ kCharsetFromHTTPHeader
+} nsCharsetSource;
+
class nsScanner {
public:
@@ -47,30 +57,36 @@ class nsScanner {
* a single string you hand in during construction.
* This short cut was added for Javascript.
*
- * @update gess 5/12/98
+ * @update ftang 3/02/99
+ * @param aCharset charset
+ * @param aCharsetSource - where the charset info came from
* @param aMode represents the parser mode (nav, other)
* @return
*/
- nsScanner(nsString& anHTMLString);
+ nsScanner(nsString& anHTMLString, const nsString& aCharset, nsCharsetSource aSource);
/**
* Use this constructor if you want i/o to be based on
* a file (therefore a stream) or just data you provide via Append().
*
- * @update gess 5/12/98
+ * @update ftang 3/02/99
+ * @param aCharset charset
+ * @param aCharsetSource - where the charset info came from
* @param aMode represents the parser mode (nav, other)
* @return
*/
- nsScanner(nsString& aFilename,PRBool aCreateStream);
+ nsScanner(nsString& aFilename,PRBool aCreateStream, const nsString& aCharset, nsCharsetSource aSource);
/**
* Use this constructor if you want i/o to be stream based.
*
- * @update gess 5/12/98
+ * @update ftang 3/02/99
+ * @param aCharset charset
+ * @param aCharsetSource - where the charset info came from
* @param aMode represents the parser mode (nav, other)
* @return
*/
- nsScanner(nsString& aFilename,fstream& aStream,PRBool assumeOwnership=PR_TRUE);
+ nsScanner(nsString& aFilename,fstream& aStream, const nsString& aCharset, nsCharsetSource aSource,PRBool assumeOwnership=PR_TRUE);
~nsScanner();
@@ -269,11 +285,12 @@ class nsScanner {
/**
* Use this setter to change the scanner's unicode decoder
*
- * @update ftang 2/12/99
+ * @update ftang 3/02/99
* @param aCharset a normalized (alias resolved) charset name
+ * @param aCharsetSource- where the charset info came from
* @return
*/
- nsresult SetDocumentCharset(const nsString& aCharset);
+ nsresult SetDocumentCharset(const nsString& aCharset, nsCharsetSource aSource);
protected:
@@ -295,6 +312,7 @@ class nsScanner {
PRUint32 mTotalRead;
PRBool mOwnsStream;
PRBool mIncremental;
+ nsCharsetSource mCharsetSource;
nsString mCharset;
nsIUnicodeDecoder *mUnicodeDecoder;
};
diff --git a/mozilla/parser/htmlparser/src/CNavDTD.cpp b/mozilla/parser/htmlparser/src/CNavDTD.cpp
index 77cf497ec02..3d8929884a5 100644
--- a/mozilla/parser/htmlparser/src/CNavDTD.cpp
+++ b/mozilla/parser/htmlparser/src/CNavDTD.cpp
@@ -942,10 +942,19 @@ nsresult CNavDTD::WillHandleStartTag(CToken* aToken,eHTMLTags aTag,nsCParserNode
if(theKey.EqualsIgnoreCase("HTTP-EQUIV")) {
const nsString& theKey2=aNode.GetKeyAt(1);
if(theKey2.EqualsIgnoreCase("CONTENT")) {
- nsScanner* theScanner=mParser->GetScanner();
- if(theScanner) {
- const nsString& theValue=aNode.GetValueAt(1);
- theScanner->SetDocumentCharset(theValue);
+ nsScanner* theScanner=mParser->GetScanner();
+ if(theScanner) {
+ const nsString& theValue=aNode.GetValueAt(1);
+ PRInt32 charsetValueStart = theValue.RFind("charset=", PR_TRUE ) ;
+ if(kNotFound != charsetValueStart) {
+ charsetValueStart += 8; // 8 = "charset=".length
+ PRInt32 charsetValueEnd = theValue.FindCharInSet("\'\";", charsetValueStart );
+ if(kNotFound == charsetValueEnd )
+ charsetValueEnd = theValue.Length();
+ nsAutoString theCharset;
+ theValue.Mid(theCharset, charsetValueStart, charsetValueEnd - charsetValueStart);
+ theScanner->SetDocumentCharset(theCharset, kCharsetFromMetaTag);
+ } //if
} //if
}
} //if
diff --git a/mozilla/parser/htmlparser/src/nsParser.cpp b/mozilla/parser/htmlparser/src/nsParser.cpp
index cab5e31ed91..c0d95d0c591 100644
--- a/mozilla/parser/htmlparser/src/nsParser.cpp
+++ b/mozilla/parser/htmlparser/src/nsParser.cpp
@@ -225,6 +225,7 @@ nsIParserFilter * nsParser::SetParserFilter(nsIParserFilter * aFilter)
return old;
}
+
/**
* Call this method once you've created a parser, and want to instruct it
* about the command which caused the parser to be constructed. For example,
@@ -526,6 +527,24 @@ PRBool nsParser::EnableParser(PRBool aState){
nsresult nsParser::Parse(nsIURL* aURL,nsIStreamObserver* aListener,PRBool aVerifyEnabled) {
NS_PRECONDITION(0!=aURL,kNullURL);
+ nsAutoString charset;
+ nsCharsetSource charsetSource;
+
+ // XXXX get HTTP charset here
+ // charset =
+ // charsetSource = kCharsetFromHTTPHeader;
+
+ // XXXX get User Prefernce charset here
+ // charset =
+ // charsetSource = kCharsetFromUserDefault;
+
+ // XXXX get Doc Type Default (e.g. UTF-8 for XML)
+
+ // XXX We should really put if doc == html for the following line
+ charset = "ISO-8859-1";
+ charsetSource = kCharsetFromDocTypeDefault;
+
+
nsresult result=kBadURL;
mDTDVerification=aVerifyEnabled;
if(aURL) {
@@ -533,7 +552,7 @@ nsresult nsParser::Parse(nsIURL* aURL,nsIStreamObserver* aListener,PRBool aVerif
nsresult rv = aURL->GetSpec(&spec);
if (rv != NS_OK) return rv;
nsAutoString theName(spec);
- CParserContext* pc=new CParserContext(new nsScanner(theName,PR_FALSE),aURL,aListener);
+ CParserContext* pc=new CParserContext(new nsScanner(theName,PR_FALSE, charset, charsetSource),aURL,aListener);
if(pc) {
pc->mMultipart=PR_TRUE;
pc->mContextType=CParserContext::eCTURL;
@@ -555,10 +574,27 @@ nsresult nsParser::Parse(fstream& aStream,PRBool aVerifyEnabled){
mDTDVerification=aVerifyEnabled;
nsresult result=NS_ERROR_OUT_OF_MEMORY;
+
+ nsAutoString charset;
+ nsCharsetSource charsetSource;
+
+ // XXXX get HTTP charset here
+ // charset =
+ // charsetSource = kCharsetFromHTTPHeader;
+
+ // XXXX get User Prefernce charset here
+ // charset =
+ // charsetSource = kCharsetFromUserDefault;
+
+ // XXXX get Doc Type Default (e.g. UTF-8 for XML)
+
+ // XXX We should really put if doc == html for the following line
+ charset = "ISO-8859-1";
+ charsetSource = kCharsetFromDocTypeDefault;
//ok, time to create our tokenizer and begin the process
nsAutoString theUnknownFilename("unknown");
- CParserContext* pc=new CParserContext(new nsScanner(theUnknownFilename,aStream,PR_FALSE),&aStream,0);
+ CParserContext* pc=new CParserContext(new nsScanner(theUnknownFilename,aStream, charset, charsetSource,PR_FALSE),&aStream,0);
if(pc) {
PushContext(*pc);
pc->mSourceType=kHTMLTextContentType;
@@ -593,6 +629,22 @@ nsresult nsParser::Parse(nsString& aSourceBuffer,void* aKey,const nsString& aCon
}
#endif
+ nsAutoString charset;
+ nsCharsetSource charsetSource;
+
+ // XXXX get HTTP charset here
+ // charset =
+ // charsetSource = kCharsetFromHTTPHeader;
+
+ // XXXX get User Prefernce charset here
+ // charset =
+ // charsetSource = kCharsetFromUserDefault;
+
+ // XXXX get Doc Type Default (e.g. UTF-8 for XML)
+
+ // XXX We should really put if doc == html for the following line
+ charset = "ISO-8859-1";
+ charsetSource = kCharsetFromDocTypeDefault;
//NOTE: Make sure that updates to this method don't cause
// bug #2361 to break again!
@@ -607,7 +659,7 @@ nsresult nsParser::Parse(nsString& aSourceBuffer,void* aKey,const nsString& aCon
if((!mParserContext) || (mParserContext->mKey!=aKey)) {
//only make a new context if we dont have one, OR if we do, but has a different context key...
- pc=new CParserContext(new nsScanner(mUnusedInput),aKey,0);
+ pc=new CParserContext(new nsScanner(mUnusedInput, charset, charsetSource),aKey, 0);
if(pc) {
PushContext(*pc);
pc->mStreamListenerState=eOnStart;
diff --git a/mozilla/parser/htmlparser/src/nsScanner.cpp b/mozilla/parser/htmlparser/src/nsScanner.cpp
index 070d0174afe..efe74c83906 100644
--- a/mozilla/parser/htmlparser/src/nsScanner.cpp
+++ b/mozilla/parser/htmlparser/src/nsScanner.cpp
@@ -23,6 +23,7 @@
#include "nsDebug.h"
#include "nsIServiceManager.h"
#include "nsICharsetConverterManager.h"
+#include "nsICharsetAlias.h"
const char* kBadHTMLText="Oops...
You just tried to read a non-existent document:
";
@@ -34,8 +35,6 @@ const int kBufsize=1;
const int kBufsize=64;
#endif
-// #define DEFAULTCHARSET "Shift_JIS"
-#define DEFAULTCHARSET "ISO-8859-1"
/**
* Use this constructor if you want i/o to be based on
@@ -46,8 +45,8 @@ const int kBufsize=64;
* @param aMode represents the parser mode (nav, other)
* @return
*/
-nsScanner::nsScanner(nsString& anHTMLString) :
- mBuffer(anHTMLString), mFilename("") , mCharset("")
+nsScanner::nsScanner(nsString& anHTMLString, const nsString& aCharset, nsCharsetSource aSource) :
+ mBuffer(anHTMLString), mFilename("")
{
mTotalRead=mBuffer.Length();
mIncremental=PR_TRUE;
@@ -55,9 +54,10 @@ nsScanner::nsScanner(nsString& anHTMLString) :
mOffset=0;
mMarkPos=-1;
mFileStream=0;
- nsAutoString defaultCharset(DEFAULTCHARSET);
mUnicodeDecoder = 0;
- SetDocumentCharset(defaultCharset);
+ mCharset = "";
+ mCharsetSource = kCharsetUninitialized;
+ SetDocumentCharset(aCharset, aSource);
}
/**
@@ -69,8 +69,8 @@ nsScanner::nsScanner(nsString& anHTMLString) :
* @param aFilename --
* @return
*/
-nsScanner::nsScanner(nsString& aFilename,PRBool aCreateStream) :
- mBuffer(""), mFilename(aFilename) , mCharset("")
+nsScanner::nsScanner(nsString& aFilename,PRBool aCreateStream, const nsString& aCharset, nsCharsetSource aSource) :
+ mBuffer(""), mFilename(aFilename)
{
mIncremental=PR_TRUE;
mOffset=0;
@@ -91,8 +91,9 @@ nsScanner::nsScanner(nsString& aFilename,PRBool aCreateStream) :
#endif
} //if
mUnicodeDecoder = 0;
- nsAutoString defaultCharset(DEFAULTCHARSET);
- SetDocumentCharset(defaultCharset);
+ mCharset = "";
+ mCharsetSource = kCharsetUninitialized;
+ SetDocumentCharset(aCharset, aSource);
}
@@ -105,8 +106,8 @@ nsScanner::nsScanner(nsString& aFilename,PRBool aCreateStream) :
* @param aFilename --
* @return
*/
-nsScanner::nsScanner(nsString& aFilename,fstream& aStream,PRBool assumeOwnership) :
- mBuffer(""), mFilename(aFilename) , mCharset("")
+nsScanner::nsScanner(nsString& aFilename,fstream& aStream,const nsString& aCharset, nsCharsetSource aSource, PRBool assumeOwnership) :
+ mBuffer(""), mFilename(aFilename)
{
mIncremental=PR_TRUE;
mOffset=0;
@@ -115,15 +116,46 @@ nsScanner::nsScanner(nsString& aFilename,fstream& aStream,PRBool assumeOwnership
mOwnsStream=assumeOwnership;
mFileStream=&aStream;
mUnicodeDecoder = 0;
- nsAutoString defaultCharset(DEFAULTCHARSET);
- SetDocumentCharset(defaultCharset);
+ mCharset = "";
+ mCharsetSource = kCharsetUninitialized;
+ SetDocumentCharset(aCharset, aSource);
}
-nsresult nsScanner::SetDocumentCharset(const nsString& aCharset )
+nsresult nsScanner::SetDocumentCharset(const nsString& aCharset , nsCharsetSource aSource)
{
+
nsresult res = NS_OK;
- if(! mCharset.EqualsIgnoreCase(aCharset)) // see do we need to change a converter.
+
+ if( aSource < mCharsetSource) // priority is lower the the current one , just
+ return res;
+
+ nsICharsetAlias* calias = nsnull;
+ res = nsServiceManager::GetService(kCharsetAliasCID,
+ kICharsetAliasIID,
+ (nsISupports**)&calias);
+
+ NS_ASSERTION( nsnull != calias, "cannot find charet alias");
+ nsAutoString charsetName = aCharset;
+ if( NS_SUCCEEDED(res) && (nsnull != calias))
{
+ PRBool same = PR_FALSE;
+ res = calias->Equals(aCharset, mCharset, &same);
+ if(NS_SUCCEEDED(res) && same)
+ {
+ return NS_OK; // no difference, don't change it
+ }
+ // different, need to change it
+ res = calias->GetPreferred(aCharset, charsetName);
+ nsServiceManager::ReleaseService(kCharsetAliasCID, calias);
+
+ if(NS_FAILED(res) && (kCharsetUninitialized == mCharsetSource) )
+ {
+ // failed - unknown alias , fallback to ISO-8859-1
+ charsetName = "ISO-8859-1";
+ }
+ mCharset = charsetName;
+ mCharsetSource = aSource;
+
nsICharsetConverterManager * ccm = nsnull;
res = nsServiceManager::GetService(kCharsetConverterManagerCID,
kICharsetConverterManagerIID,
@@ -131,13 +163,12 @@ nsresult nsScanner::SetDocumentCharset(const nsString& aCharset )
if(NS_SUCCEEDED(res) && (nsnull != ccm))
{
nsIUnicodeDecoder * decoder = nsnull;
- res = ccm->GetUnicodeDecoder(&aCharset, &decoder);
+ res = ccm->GetUnicodeDecoder(&mCharset, &decoder);
if(NS_SUCCEEDED(res) && (nsnull != decoder))
{
NS_IF_RELEASE(mUnicodeDecoder);
mUnicodeDecoder = decoder;
- mCharset = aCharset;
}
nsServiceManager::ReleaseService(kCharsetConverterManagerCID, ccm);
}
diff --git a/mozilla/parser/htmlparser/src/nsScanner.h b/mozilla/parser/htmlparser/src/nsScanner.h
index acbb3773c57..7fb72c89db8 100644
--- a/mozilla/parser/htmlparser/src/nsScanner.h
+++ b/mozilla/parser/htmlparser/src/nsScanner.h
@@ -39,6 +39,16 @@
#include
+typedef enum {
+ kCharsetUninitialized = 0,
+ kCharsetFromUserDefault ,
+ kCharsetFromDocTypeDefault,
+ kCharsetFromParentFrame,
+ kCharsetFromAutoDetection,
+ kCharsetFromMetaTag,
+ kCharsetFromHTTPHeader
+} nsCharsetSource;
+
class nsScanner {
public:
@@ -47,30 +57,36 @@ class nsScanner {
* a single string you hand in during construction.
* This short cut was added for Javascript.
*
- * @update gess 5/12/98
+ * @update ftang 3/02/99
+ * @param aCharset charset
+ * @param aCharsetSource - where the charset info came from
* @param aMode represents the parser mode (nav, other)
* @return
*/
- nsScanner(nsString& anHTMLString);
+ nsScanner(nsString& anHTMLString, const nsString& aCharset, nsCharsetSource aSource);
/**
* Use this constructor if you want i/o to be based on
* a file (therefore a stream) or just data you provide via Append().
*
- * @update gess 5/12/98
+ * @update ftang 3/02/99
+ * @param aCharset charset
+ * @param aCharsetSource - where the charset info came from
* @param aMode represents the parser mode (nav, other)
* @return
*/
- nsScanner(nsString& aFilename,PRBool aCreateStream);
+ nsScanner(nsString& aFilename,PRBool aCreateStream, const nsString& aCharset, nsCharsetSource aSource);
/**
* Use this constructor if you want i/o to be stream based.
*
- * @update gess 5/12/98
+ * @update ftang 3/02/99
+ * @param aCharset charset
+ * @param aCharsetSource - where the charset info came from
* @param aMode represents the parser mode (nav, other)
* @return
*/
- nsScanner(nsString& aFilename,fstream& aStream,PRBool assumeOwnership=PR_TRUE);
+ nsScanner(nsString& aFilename,fstream& aStream, const nsString& aCharset, nsCharsetSource aSource,PRBool assumeOwnership=PR_TRUE);
~nsScanner();
@@ -269,11 +285,12 @@ class nsScanner {
/**
* Use this setter to change the scanner's unicode decoder
*
- * @update ftang 2/12/99
+ * @update ftang 3/02/99
* @param aCharset a normalized (alias resolved) charset name
+ * @param aCharsetSource- where the charset info came from
* @return
*/
- nsresult SetDocumentCharset(const nsString& aCharset);
+ nsresult SetDocumentCharset(const nsString& aCharset, nsCharsetSource aSource);
protected:
@@ -295,6 +312,7 @@ class nsScanner {
PRUint32 mTotalRead;
PRBool mOwnsStream;
PRBool mIncremental;
+ nsCharsetSource mCharsetSource;
nsString mCharset;
nsIUnicodeDecoder *mUnicodeDecoder;
};