📄 html_documentimpl.cpp
字号:
int oldCount = (int)namedImageAndFormCounts.find(name);
namedImageAndFormCounts.insert(name, (char *)(oldCount + 1));
}
void HTMLDocumentImpl::removeNamedImageOrForm(const QString &name)
{
if (name.length() == 0) {
return;
}
int oldVal = (int)(namedImageAndFormCounts.find(name));
if (oldVal != 0) {
int newVal = oldVal - 1;
if (newVal == 0) {
namedImageAndFormCounts.remove(name);
} else {
namedImageAndFormCounts.insert(name, (char *)newVal);
}
}
}
bool HTMLDocumentImpl::haveNamedImageOrForm(const QString &name)
{
return namedImageAndFormCounts.find(name) != NULL;
}
const int PARSEMODE_HAVE_DOCTYPE = (1<<0);
const int PARSEMODE_HAVE_PUBLIC_ID = (1<<1);
const int PARSEMODE_HAVE_SYSTEM_ID = (1<<2);
const int PARSEMODE_HAVE_INTERNAL = (1<<3);
static int parseDocTypePart(const QString& buffer, int index)
{
while (true) {
QChar ch = buffer[index];
if (ch == ' ' || ch == '\t' || ch == '\n' || ch == '\r')
++index;
else if (ch == '-') {
int tmpIndex=index;
if (buffer[index+1] == '-' &&
((tmpIndex=buffer.find("--", index+2)) != -1))
index = tmpIndex+2;
else
return index;
}
else
return index;
}
}
static bool containsString(const char* str, const QString& buffer, int offset)
{
QString startString(str);
if (offset + startString.length() > buffer.length())
return false;
QString bufferString = buffer.mid(offset, startString.length()).lower();
QString lowerStart = startString.lower();
return bufferString.startsWith(lowerStart);
}
static bool parseDocTypeDeclaration(const QString& buffer,
int* resultFlags,
QString& publicID,
QString& systemID)
{
bool haveDocType = false;
*resultFlags = 0;
// Skip through any comments and processing instructions.
int index = 0;
do {
index = buffer.find('<', index);
if (index == -1) break;
QChar nextChar = buffer[index+1];
if (nextChar == '!') {
if (containsString("doctype", buffer, index+2)) {
haveDocType = true;
index += 9; // Skip "<!DOCTYPE"
break;
}
index = parseDocTypePart(buffer,index);
index = buffer.find('>', index);
}
else if (nextChar == '?')
index = buffer.find('>', index);
else
break;
} while (index != -1);
if (!haveDocType)
return true;
*resultFlags |= PARSEMODE_HAVE_DOCTYPE;
index = parseDocTypePart(buffer, index);
if (!containsString("html", buffer, index))
return false;
index = parseDocTypePart(buffer, index+4);
bool hasPublic = containsString("public", buffer, index);
if (hasPublic) {
index = parseDocTypePart(buffer, index+6);
// We've read <!DOCTYPE HTML PUBLIC (not case sensitive).
// Now we find the beginning and end of the public identifers
// and system identifiers (assuming they're even present).
QChar theChar = buffer[index];
if (theChar != '\"' && theChar != '\'')
return false;
// |start| is the first character (after the quote) and |end|
// is the final quote, so there are |end|-|start| characters.
int publicIDStart = index+1;
int publicIDEnd = buffer.find(theChar, publicIDStart);
if (publicIDEnd == -1)
return false;
index = parseDocTypePart(buffer, publicIDEnd+1);
QChar next = buffer[index];
if (next == '>') {
// Public identifier present, but no system identifier.
// Do nothing. Note that this is the most common
// case.
}
else if (next == '\"' || next == '\'') {
// We have a system identifier.
*resultFlags |= PARSEMODE_HAVE_SYSTEM_ID;
int systemIDStart = index+1;
int systemIDEnd = buffer.find(next, systemIDStart);
if (systemIDEnd == -1)
return false;
systemID = buffer.mid(systemIDStart, systemIDEnd - systemIDStart);
}
else if (next == '[') {
// We found an internal subset.
*resultFlags |= PARSEMODE_HAVE_INTERNAL;
}
else
return false; // Something's wrong.
// We need to trim whitespace off the public identifier.
publicID = buffer.mid(publicIDStart, publicIDEnd - publicIDStart);
publicID = publicID.stripWhiteSpace();
*resultFlags |= PARSEMODE_HAVE_PUBLIC_ID;
} else {
if (containsString("system", buffer, index)) {
// Doctype has a system ID but no public ID
*resultFlags |= PARSEMODE_HAVE_SYSTEM_ID;
index = parseDocTypePart(buffer, index+6);
QChar next = buffer[index];
if (next != '\"' && next != '\'')
return false;
int systemIDStart = index+1;
int systemIDEnd = buffer.find(next, systemIDStart);
if (systemIDEnd == -1)
return false;
systemID = buffer.mid(systemIDStart, systemIDEnd - systemIDStart);
index = parseDocTypePart(buffer, systemIDEnd+1);
}
QChar nextChar = buffer[index];
if (nextChar == '[')
*resultFlags |= PARSEMODE_HAVE_INTERNAL;
else if (nextChar != '>')
return false;
}
return true;
}
void HTMLDocumentImpl::determineParseMode( const QString &str )
{
//kdDebug() << "DocumentImpl::determineParseMode str=" << str<< endl;
// This code more or less mimics Mozilla's implementation (specifically the
// doctype parsing implemented by David Baron in Mozilla's nsParser.cpp).
//
// There are three possible parse modes:
// COMPAT - quirks mode emulates WinIE
// and NS4. CSS parsing is also relaxed in this mode, e.g., unit types can
// be omitted from numbers.
// ALMOST STRICT - This mode is identical to strict mode
// except for its treatment of line-height in the inline box model. For
// now (until the inline box model is re-written), this mode is identical
// to STANDARDS mode.
// STRICT - no quirks apply. Web pages will obey the specifications to
// the letter.
QString systemID, publicID;
int resultFlags = 0;
if (parseDocTypeDeclaration(str, &resultFlags, publicID, systemID)) {
if (resultFlags & PARSEMODE_HAVE_DOCTYPE) {
m_doctype->setName("HTML");
m_doctype->setPublicId(publicID);
m_doctype->setSystemId(systemID);
}
if (!(resultFlags & PARSEMODE_HAVE_DOCTYPE)) {
// No doctype found at all. Default to quirks mode and Html4.
pMode = Compat;
hMode = Html4;
}
else if ((resultFlags & PARSEMODE_HAVE_INTERNAL) ||
!(resultFlags & PARSEMODE_HAVE_PUBLIC_ID)) {
// Internal subsets always denote full standards, as does
// a doctype without a public ID.
pMode = Strict;
hMode = Html4;
}
else {
// We have to check a list of public IDs to see what we
// should do.
QString lowerPubID = publicID.lower();
const char* pubIDStr = lowerPubID.latin1();
// Look up the entry in our gperf-generated table.
const PubIDInfo* doctypeEntry = findDoctypeEntry(pubIDStr, publicID.length());
if (!doctypeEntry) {
// The DOCTYPE is not in the list. Assume strict mode.
pMode = Strict;
hMode = Html4;
#if NOKIA_CHANGES
if (lowerPubID.contains("xhtml mobile"))
m_xHtmlMobile = true;
#endif
return;
}
switch ((resultFlags & PARSEMODE_HAVE_SYSTEM_ID) ?
doctypeEntry->mode_if_sysid :
doctypeEntry->mode_if_no_sysid)
{
case PubIDInfo::eQuirks3:
pMode = Compat;
hMode = Html3;
break;
case PubIDInfo::eQuirks:
pMode = Compat;
hMode = Html4;
break;
case PubIDInfo::eAlmostStandards:
pMode = AlmostStrict;
hMode = Html4;
break;
default:
assert(false);
}
}
}
else {
// Malformed doctype implies quirks mode.
pMode = Compat;
hMode = Html3;
}
m_styleSelector->strictParsing = !inCompatMode();
// kdDebug() << "DocumentImpl::determineParseMode: publicId =" << publicId << " systemId = " << systemId << endl;
// kdDebug() << "DocumentImpl::determineParseMode: htmlMode = " << hMode<< endl;
// if( pMode == Strict )
// kdDebug(6020) << " using strict parseMode" << endl;
// else if (pMode == Compat )
// kdDebug(6020) << " using compatibility parseMode" << endl;
// else
// kdDebug(6020) << " using almost strict parseMode" << endl;
}
#include "html_documentimpl.moc"
⌨️ 快捷键说明
复制代码
Ctrl + C
搜索代码
Ctrl + F
全屏模式
F11
切换主题
Ctrl + Shift + D
显示快捷键
?
增大字号
Ctrl + =
减小字号
Ctrl + -