summaryrefslogtreecommitdiffstats
path: root/qtools/qxml.cpp
diff options
context:
space:
mode:
authordimitri <dimitri@afe2bf4a-e733-0410-8a33-86f594647bc7>2000-12-03 19:13:07 (GMT)
committerdimitri <dimitri@afe2bf4a-e733-0410-8a33-86f594647bc7>2000-12-03 19:13:07 (GMT)
commitfe67b8eb68129713327965c201f2d7226b83202f (patch)
tree485fb83c5a301dd4b0edb3c534b1f31eeb08ab1f /qtools/qxml.cpp
parenta30c2e3c5ea41ae1947e9893c82ed8c8b6d7c5a2 (diff)
downloadDoxygen-fe67b8eb68129713327965c201f2d7226b83202f.zip
Doxygen-fe67b8eb68129713327965c201f2d7226b83202f.tar.gz
Doxygen-fe67b8eb68129713327965c201f2d7226b83202f.tar.bz2
Release-1.2.3-20001203
Diffstat (limited to 'qtools/qxml.cpp')
-rw-r--r--qtools/qxml.cpp6007
1 files changed, 6007 insertions, 0 deletions
diff --git a/qtools/qxml.cpp b/qtools/qxml.cpp
new file mode 100644
index 0000000..d171828
--- /dev/null
+++ b/qtools/qxml.cpp
@@ -0,0 +1,6007 @@
+/****************************************************************************
+**
+**
+** Implementation of QXmlSimpleReader and related classes.
+**
+** Created : 000518
+**
+** Copyright (C) 1992-2000 Trolltech AS. All rights reserved.
+**
+** This file is part of the XML module of the Qt GUI Toolkit.
+**
+** This file may be distributed under the terms of the Q Public License
+** as defined by Trolltech AS of Norway and appearing in the file
+** LICENSE.QPL included in the packaging of this file.
+**
+** This file may be distributed and/or modified under the terms of the
+** GNU General Public License version 2 as published by the Free Software
+** Foundation and appearing in the file LICENSE.GPL included in the
+** packaging of this file.
+**
+** Licensees holding valid Qt Enterprise Edition licenses may use this
+** file in accordance with the Qt Commercial License Agreement provided
+** with the Software.
+**
+** This file is provided AS IS with NO WARRANTY OF ANY KIND, INCLUDING THE
+** WARRANTY OF DESIGN, MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE.
+**
+** See http://www.trolltech.com/pricing.html or email sales@trolltech.com for
+** information about Qt Commercial License Agreements.
+** See http://www.trolltech.com/qpl/ for QPL licensing information.
+** See http://www.trolltech.com/gpl/ for GPL licensing information.
+**
+** Contact info@trolltech.com if any conditions of this licensing are
+** not clear to you.
+**
+**********************************************************************/
+
+#include "qxml.h"
+#include "qtextcodec.h"
+#include "qbuffer.h"
+
+#ifndef QT_NO_XML
+// NOT REVISED
+
+// Error strings for the XML reader
+#define XMLERR_OK "no error occured"
+#define XMLERR_TAGMISMATCH "tag mismatch"
+#define XMLERR_UNEXPECTEDEOF "unexpected end of file"
+#define XMLERR_FINISHEDPARSINGWHILENOTEOF "parsing is finished but end of file is not reached"
+#define XMLERR_LETTEREXPECTED "letter is expected"
+#define XMLERR_ERRORPARSINGELEMENT "error while parsing element"
+#define XMLERR_ERRORPARSINGPROLOG "error while parsing prolog"
+#define XMLERR_ERRORPARSINGMAINELEMENT "error while parsing main element"
+#define XMLERR_ERRORPARSINGCONTENT "error while parsing content"
+#define XMLERR_ERRORPARSINGNAME "error while parsing name"
+#define XMLERR_ERRORPARSINGNMTOKEN "error while parsing Nmtoken"
+#define XMLERR_ERRORPARSINGATTRIBUTE "error while parsing attribute"
+#define XMLERR_ERRORPARSINGMISC "error while parsing misc"
+#define XMLERR_ERRORPARSINGCHOICE "error while parsing choice or seq"
+#define XMLERR_ERRORBYCONSUMER "error triggered by consumer"
+#define XMLERR_UNEXPECTEDCHARACTER "unexpected character"
+#define XMLERR_EQUALSIGNEXPECTED "expected '=' but not found"
+#define XMLERR_QUOTATIONEXPECTED "expected \" or ' but not found"
+#define XMLERR_ERRORPARSINGREFERENCE "error while parsing reference"
+#define XMLERR_ERRORPARSINGPI "error while parsing processing instruction"
+#define XMLERR_ERRORPARSINGATTLISTDECL "error while parsing attribute list declaration"
+#define XMLERR_ERRORPARSINGATTTYPE "error while parsing attribute type declaration"
+#define XMLERR_ERRORPARSINGATTVALUE "error while parsing attribute value declaration"
+#define XMLERR_ERRORPARSINGELEMENTDECL "error while parsing element declaration"
+#define XMLERR_ERRORPARSINGENTITYDECL "error while parsing entity declaration"
+#define XMLERR_ERRORPARSINGNOTATIONDECL "error while parsing notation declaration"
+#define XMLERR_ERRORPARSINGEXTERNALID "error while parsing external id"
+#define XMLERR_ERRORPARSINGCOMMENT "error while parsing comment"
+#define XMLERR_ERRORPARSINGENTITYVALUE "error while parsing entity value declaration"
+#define XMLERR_CDSECTHEADEREXPECTED "expected the header for a cdata section"
+#define XMLERR_MORETHANONEDOCTYPE "more than one document type definition"
+#define XMLERR_ERRORPARSINGDOCTYPE "error while parsing document type definition"
+#define XMLERR_INVALIDNAMEFORPI "invalid name for processing instruction"
+#define XMLERR_VERSIONEXPECTED "version expected while reading the XML declaration"
+#define XMLERR_EDECLORSDDECLEXPECTED "EDecl or SDDecl expected while reading the XML declaration"
+#define XMLERR_SDDECLEXPECTED "SDDecl expected while reading the XML declaration"
+#define XMLERR_WRONGVALUEFORSDECL "wrong value for standalone declaration"
+#define XMLERR_UNPARSEDENTITYREFERENCE "unparsed entity reference"
+#define XMLERR_INTERNALGENERALENTITYINDTD "internal general entity reference not allowed in DTD"
+#define XMLERR_EXTERNALGENERALENTITYINDTD "external parsed general entity reference not allowed in DTD"
+#define XMLERR_EXTERNALGENERALENTITYINAV "external parsed general entity reference not allowed in attribute value"
+
+
+// the constants for the lookup table
+static const signed char cltWS = 0; // white space
+static const signed char cltPer = 1; // %
+static const signed char cltAmp = 2; // &
+static const signed char cltGt = 3; // >
+static const signed char cltLt = 4; // <
+static const signed char cltSlash = 5; // /
+static const signed char cltQm = 6; // ?
+static const signed char cltEm = 7; // !
+static const signed char cltDash = 8; // -
+static const signed char cltCB = 9; // ]
+static const signed char cltOB = 10; // [
+static const signed char cltEq = 11; // =
+static const signed char cltDq = 12; // "
+static const signed char cltSq = 13; // '
+static const signed char cltUnknown = 14;
+
+// character lookup table
+static const signed char charLookupTable[256]={
+ cltUnknown, cltUnknown, cltUnknown, cltUnknown, cltUnknown, cltUnknown, cltUnknown, cltUnknown, // 0x00 - 0x07
+ cltUnknown, // 0x08
+ cltWS, // 0x09 \t
+ cltWS, // 0x0A \n
+ cltUnknown, // 0x0B
+ cltUnknown, // 0x0C
+ cltWS, // 0x0D \r
+ cltUnknown, // 0x0E
+ cltUnknown, // 0x0F
+ cltUnknown, cltUnknown, cltUnknown, cltUnknown, cltUnknown, cltUnknown, cltUnknown, cltUnknown, // 0x17 - 0x16
+ cltUnknown, cltUnknown, cltUnknown, cltUnknown, cltUnknown, cltUnknown, cltUnknown, cltUnknown, // 0x18 - 0x1F
+ cltWS, // 0x20 Space
+ cltEm, // 0x21 !
+ cltDq, // 0x22 "
+ cltUnknown, // 0x23
+ cltUnknown, // 0x24
+ cltPer, // 0x25 %
+ cltAmp, // 0x26 &
+ cltSq, // 0x27 '
+ cltUnknown, // 0x28
+ cltUnknown, // 0x29
+ cltUnknown, // 0x2A
+ cltUnknown, // 0x2B
+ cltUnknown, // 0x2C
+ cltDash, // 0x2D -
+ cltUnknown, // 0x2E
+ cltSlash, // 0x2F /
+ cltUnknown, cltUnknown, cltUnknown, cltUnknown, cltUnknown, cltUnknown, cltUnknown, cltUnknown, // 0x30 - 0x37
+ cltUnknown, // 0x38
+ cltUnknown, // 0x39
+ cltUnknown, // 0x3A
+ cltUnknown, // 0x3B
+ cltLt, // 0x3C <
+ cltEq, // 0x3D =
+ cltGt, // 0x3E >
+ cltQm, // 0x3F ?
+ cltUnknown, cltUnknown, cltUnknown, cltUnknown, cltUnknown, cltUnknown, cltUnknown, cltUnknown, // 0x40 - 0x47
+ cltUnknown, cltUnknown, cltUnknown, cltUnknown, cltUnknown, cltUnknown, cltUnknown, cltUnknown, // 0x48 - 0x4F
+ cltUnknown, cltUnknown, cltUnknown, cltUnknown, cltUnknown, cltUnknown, cltUnknown, cltUnknown, // 0x50 - 0x57
+ cltUnknown, // 0x58
+ cltUnknown, // 0x59
+ cltUnknown, // 0x5A
+ cltOB, // 0x5B [
+ cltUnknown, // 0x5C
+ cltCB, // 0x5D ]
+ cltUnknown, // 0x5E
+ cltUnknown, // 0x5F
+ cltUnknown, cltUnknown, cltUnknown, cltUnknown, cltUnknown, cltUnknown, cltUnknown, cltUnknown, // 0x60 - 0x67
+ cltUnknown, cltUnknown, cltUnknown, cltUnknown, cltUnknown, cltUnknown, cltUnknown, cltUnknown, // 0x68 - 0x6F
+ cltUnknown, cltUnknown, cltUnknown, cltUnknown, cltUnknown, cltUnknown, cltUnknown, cltUnknown, // 0x70 - 0x77
+ cltUnknown, cltUnknown, cltUnknown, cltUnknown, cltUnknown, cltUnknown, cltUnknown, cltUnknown, // 0x78 - 0x7F
+ cltUnknown, cltUnknown, cltUnknown, cltUnknown, cltUnknown, cltUnknown, cltUnknown, cltUnknown, // 0x80 - 0x87
+ cltUnknown, cltUnknown, cltUnknown, cltUnknown, cltUnknown, cltUnknown, cltUnknown, cltUnknown, // 0x88 - 0x8F
+ cltUnknown, cltUnknown, cltUnknown, cltUnknown, cltUnknown, cltUnknown, cltUnknown, cltUnknown, // 0x90 - 0x97
+ cltUnknown, cltUnknown, cltUnknown, cltUnknown, cltUnknown, cltUnknown, cltUnknown, cltUnknown, // 0x98 - 0x9F
+ cltUnknown, cltUnknown, cltUnknown, cltUnknown, cltUnknown, cltUnknown, cltUnknown, cltUnknown, // 0xA0 - 0xA7
+ cltUnknown, cltUnknown, cltUnknown, cltUnknown, cltUnknown, cltUnknown, cltUnknown, cltUnknown, // 0xA8 - 0xAF
+ cltUnknown, cltUnknown, cltUnknown, cltUnknown, cltUnknown, cltUnknown, cltUnknown, cltUnknown, // 0xB0 - 0xB7
+ cltUnknown, cltUnknown, cltUnknown, cltUnknown, cltUnknown, cltUnknown, cltUnknown, cltUnknown, // 0xB8 - 0xBF
+ cltUnknown, cltUnknown, cltUnknown, cltUnknown, cltUnknown, cltUnknown, cltUnknown, cltUnknown, // 0xC0 - 0xC7
+ cltUnknown, cltUnknown, cltUnknown, cltUnknown, cltUnknown, cltUnknown, cltUnknown, cltUnknown, // 0xC8 - 0xCF
+ cltUnknown, cltUnknown, cltUnknown, cltUnknown, cltUnknown, cltUnknown, cltUnknown, cltUnknown, // 0xD0 - 0xD7
+ cltUnknown, cltUnknown, cltUnknown, cltUnknown, cltUnknown, cltUnknown, cltUnknown, cltUnknown, // 0xD8 - 0xDF
+ cltUnknown, cltUnknown, cltUnknown, cltUnknown, cltUnknown, cltUnknown, cltUnknown, cltUnknown, // 0xE0 - 0xE7
+ cltUnknown, cltUnknown, cltUnknown, cltUnknown, cltUnknown, cltUnknown, cltUnknown, cltUnknown, // 0xE8 - 0xEF
+ cltUnknown, cltUnknown, cltUnknown, cltUnknown, cltUnknown, cltUnknown, cltUnknown, cltUnknown, // 0xF0 - 0xF7
+ cltUnknown, cltUnknown, cltUnknown, cltUnknown, cltUnknown, cltUnknown, cltUnknown, cltUnknown // 0xF8 - 0xFF
+};
+
+
+class QXmlNamespaceSupportPrivate
+{
+};
+class QXmlAttributesPrivate
+{
+};
+class QXmlInputSourcePrivate
+{
+};
+class QXmlParseExceptionPrivate
+{
+};
+class QXmlLocatorPrivate
+{
+};
+class QXmlDefaultHandlerPrivate
+{
+};
+
+#if defined(Q_FULL_TEMPLATE_INSTANTIATION)
+bool operator==( const QMap<QString, QString>, const QMap<QString, QString> )
+{
+ return FALSE;
+}
+#endif
+
+/*!
+ \class QXmlParseException qxml.h
+ \brief The QXmlParseException class is used to report errors with the
+ QXmlErrorHandler interface.
+
+ \module XML
+
+ \sa QXmlErrorHandler
+*/
+/*!
+ \fn QXmlParseException::QXmlParseException( const QString& name, int c, int l, const QString& p, const QString& s )
+
+ Constructs a parse exception with the error string \a name in the column
+ \a c and line \a l for the public identifier \a p and the system identifier
+ \a s.
+*/
+/*!
+ Returns the error message.
+*/
+QString QXmlParseException::message() const
+{
+ return msg;
+}
+/*!
+ Returns the column number the error occured.
+*/
+int QXmlParseException::columnNumber() const
+{
+ return column;
+}
+/*!
+ Returns the line number the error occured.
+*/
+int QXmlParseException::lineNumber() const
+{
+ return line;
+}
+/*!
+ Returns the public identifier the error occured.
+*/
+QString QXmlParseException::publicId() const
+{
+ return pub;
+}
+/*!
+ Returns the system identifier the error occured.
+*/
+QString QXmlParseException::systemId() const
+{
+ return sys;
+}
+
+
+/*!
+ \class QXmlLocator qxml.h
+ \brief The QXmlLocator class provides the XML handler classes with
+ information about the actual parsing position.
+
+ \module XML
+
+ The reader reports a QXmlLocator to the content handler before he starts to
+ parse the document. This is done with the
+ QXmlContentHandler::setDocumentLocator() function. The handler classes can
+ now use this locator to get the actual position the reader is at.
+*/
+/*!
+ \fn QXmlLocator::QXmlLocator( QXmlSimpleReader* parent )
+
+ Constructor.
+*/
+/*!
+ \fn QXmlLocator::~QXmlLocator()
+
+ Destructor.
+*/
+/*!
+ Gets the column number (starting with 1) or -1 if there is no column number
+ available.
+*/
+int QXmlLocator::columnNumber()
+{
+ return ( reader->columnNr == -1 ? -1 : reader->columnNr + 1 );
+}
+/*!
+ Gets the line number (starting with 1) or -1 if there is no line number
+ available.
+*/
+int QXmlLocator::lineNumber()
+{
+ return ( reader->lineNr == -1 ? -1 : reader->lineNr + 1 );
+}
+
+
+/*********************************************
+ *
+ * QXmlNamespaceSupport
+ *
+ *********************************************/
+
+/*!
+ \class QXmlNamespaceSupport qxml.h
+ \brief The QXmlNamespaceSupport class is a helper class for XML readers which
+ want to include namespace support.
+
+ \module XML
+
+ It provides some functions that makes it easy to handle namespaces. Its main
+ use is for subclasses of QXmlReader which want to provide namespace
+ support.
+
+ See also the <a href="xml-sax.html#namespaces">namespace description</a>.
+*/
+
+/*!
+ Constructs a QXmlNamespaceSupport.
+*/
+QXmlNamespaceSupport::QXmlNamespaceSupport()
+{
+ reset();
+}
+
+/*!
+ Destructs a QXmlNamespaceSupport.
+*/
+QXmlNamespaceSupport::~QXmlNamespaceSupport()
+{
+}
+
+/*!
+ This function declares a prefix in the current namespace context; the prefix
+ will remain in force until this context is popped, unless it is shadowed in a
+ descendant context.
+
+ Note that there is an asymmetry in this library: while prefix() will not
+ return the default "" prefix, even if you have declared one; to check for a
+ default prefix, you have to look it up explicitly using uri(). This
+ asymmetry exists to make it easier to look up prefixes for attribute names,
+ where the default prefix is not allowed.
+*/
+void QXmlNamespaceSupport::setPrefix( const QString& pre, const QString& uri )
+{
+ if( pre.isNull() ) {
+ ns.insert( "", uri );
+ } else {
+ ns.insert( pre, uri );
+ }
+}
+
+/*!
+ Returns one of the prefixes mapped to a namespace URI.
+
+ If more than one prefix is currently mapped to the same URI, this function
+ will make an arbitrary selection; if you want all of the prefixes, use the
+ prefixes() function instead.
+
+ Note: this will never return the empty (default) prefix; to check for a
+ default prefix, use the uri() function with an argument of "".
+*/
+QString QXmlNamespaceSupport::prefix( const QString& uri ) const
+{
+ QMap<QString, QString>::ConstIterator itc, it = ns.begin();
+ while ( (itc=it) != ns.end() ) {
+ ++it;
+ if ( itc.data() == uri && !itc.key().isEmpty() )
+ return itc.key();
+ }
+ return "";
+}
+
+/*!
+ Looks up a prefix in the current context and returns the currently-mapped
+ namespace URI. Use the empty string ("") for the default namespace.
+*/
+QString QXmlNamespaceSupport::uri( const QString& prefix ) const
+{
+ const QString& returi = ns[ prefix ];
+ return returi;
+}
+
+/*!
+ Splits the name at the ':' and returns the prefix and the local name.
+*/
+void QXmlNamespaceSupport::splitName( const QString& qname,
+ QString& prefix, QString& localname ) const
+{
+ uint pos;
+ // search the ':'
+ for( pos=0; pos<qname.length(); pos++ ) {
+ if ( qname.at(pos) == ':' )
+ break;
+ }
+ // and split
+ prefix = qname.left( pos );
+ localname = qname.mid( pos+1 );
+}
+
+/*!
+ Processes a raw XML 1.0 name in the current context by removing the prefix
+ and looking it up among the prefixes currently declared.
+
+ First parameter is the raw XML 1.0 name to be processed. The second parameter
+ is a flag wheter the name is the name of an attribute (TRUE) or not (FALSE).
+
+ The return values will be stored in the last two parameters as follows:
+ <ul>
+ <li> The namespace URI, or an empty string if none is in use.
+ <li> The local name (without prefix).
+ </ul>
+
+ If the raw name has a prefix that has not been declared, then the return
+ value will be empty.
+
+ Note that attribute names are processed differently than element names: an
+ unprefixed element name will received the default namespace (if any), while
+ an unprefixed element name will not
+*/
+void QXmlNamespaceSupport::processName( const QString& qname,
+ bool isAttribute,
+ QString& nsuri, QString& localname ) const
+{
+ uint pos;
+ // search the ':'
+ for( pos=0; pos<qname.length(); pos++ ) {
+ if ( qname.at(pos) == ':' )
+ break;
+ }
+ if ( pos < qname.length() ) {
+ // there was a ':'
+ nsuri = uri( qname.left( pos ) );
+ localname = qname.mid( pos+1 );
+ } else {
+ // there was no ':'
+ if ( isAttribute ) {
+ nsuri = ""; // attributes don't take default namespace
+ } else {
+ nsuri = uri( "" ); // get default namespace
+ }
+ localname = qname;
+ }
+}
+
+/*!
+ Returns an enumeration of all prefixes currently declared.
+
+ Note: if there is a default prefix, it will not be returned in this
+ enumeration; check for the default prefix using uri() with an argument
+ of "".
+*/
+QStringList QXmlNamespaceSupport::prefixes() const
+{
+ QStringList list;
+
+ QMap<QString, QString>::ConstIterator itc, it = ns.begin();
+ while ( (itc=it) != ns.end() ) {
+ ++it;
+ if ( !itc.key().isEmpty() )
+ list.append( itc.key() );
+ }
+ return list;
+}
+
+/*!
+ Returns a list of all prefixes currently declared for a URI.
+
+ The xml: prefix will be included. If you want only one prefix that's
+ mapped to the namespace URI, and you don't care which one you get, use the
+ prefix() function instead.
+
+ Note: the empty (default) prefix is never included in this enumeration; to
+ check for the presence of a default namespace, use uri() with an
+ argument of "".
+*/
+QStringList QXmlNamespaceSupport::prefixes( const QString& uri ) const
+{
+ QStringList list;
+
+ QMap<QString, QString>::ConstIterator itc, it = ns.begin();
+ while ( (itc=it) != ns.end() ) {
+ ++it;
+ if ( itc.data() == uri && !itc.key().isEmpty() )
+ list.append( itc.key() );
+ }
+ return list;
+}
+
+/*!
+ Starts a new namespace context.
+
+ Normally, you should push a new context at the beginning of each XML element:
+ the new context will automatically inherit the declarations of its parent
+ context, but it will also keep track of which declarations were made within
+ this context.
+*/
+void QXmlNamespaceSupport::pushContext()
+{
+ nsStack.push( ns );
+}
+
+/*!
+ Reverts to the previous namespace context.
+
+ Normally, you should pop the context at the end of each XML element. After
+ popping the context, all namespace prefix mappings that were previously in
+ force are restored.
+*/
+void QXmlNamespaceSupport::popContext()
+{
+ if( !nsStack.isEmpty() )
+ ns = nsStack.pop();
+}
+
+/*!
+ Resets this namespace support object for reuse.
+*/
+void QXmlNamespaceSupport::reset()
+{
+ nsStack.clear();
+ ns.clear();
+ ns.insert( "xml", "http://www.w3.org/XML/1998/namespace" ); // the XML namespace
+}
+
+
+
+/*********************************************
+ *
+ * QXmlAttributes
+ *
+ *********************************************/
+
+/*!
+ \class QXmlAttributes qxml.h
+ \brief The QXmlAttributes class provides XML attributes.
+
+ \module XML
+
+ If attributes are reported by QXmlContentHandler::startElement() this
+ class is used to pass the attribute values. It provides you with different
+ functions to access the attribute names and values.
+*/
+/*!
+ \fn QXmlAttributes::QXmlAttributes()
+
+ Constructs an empty attribute list.
+*/
+/*!
+ \fn QXmlAttributes::~QXmlAttributes()
+
+ Destructs attributes.
+*/
+
+/*!
+ Look up the index of an attribute by an XML 1.0 qualified name.
+
+ Returns the index of the attribute (starting with 0) or -1 if it wasn't
+ found.
+
+ See also the <a href="xml-sax.html#namespaces">namespace description</a>.
+*/
+int QXmlAttributes::index( const QString& qName ) const
+{
+ return qnameList.findIndex( qName );
+}
+
+/*!
+ Looks up the index of an attribute by a namespace name.
+
+ \a uri specifies the namespace URI, or the empty string if the name has no
+ namespace URI. \a localPart specifies the attribute's local name.
+
+ Returns the index of the attribute (starting with 0) or -1 if it wasn't
+ found.
+
+ See also the <a href="xml-sax.html#namespaces">namespace description</a>.
+*/
+int QXmlAttributes::index( const QString& uri, const QString& localPart ) const
+{
+ uint count = uriList.count();
+ for ( uint i=0; i<count; i++ ) {
+ if ( uriList[i] == uri && localnameList[i] == localPart )
+ return i;
+ }
+ return -1;
+}
+
+/*!
+ Returns the number of attributes in the list.
+*/
+int QXmlAttributes::length() const
+{
+ return valueList.count();
+}
+
+/*!
+ Looks up an attribute's local name by index (starting with 0).
+
+ See also the <a href="xml-sax.html#namespaces">namespace description</a>.
+*/
+QString QXmlAttributes::localName( int index ) const
+{
+ return localnameList[index];
+}
+
+/*!
+ Looks up an attribute's XML 1.0 qualified name by index (starting with 0).
+
+ See also the <a href="xml-sax.html#namespaces">namespace description</a>.
+*/
+QString QXmlAttributes::qName( int index ) const
+{
+ return qnameList[index];
+}
+
+/*!
+ Looks up an attribute's namespace URI by index (starting with 0).
+
+ See also the <a href="xml-sax.html#namespaces">namespace description</a>.
+*/
+QString QXmlAttributes::uri( int index ) const
+{
+ return uriList[index];
+}
+
+/*!
+ Looks up an attribute's type by index (starting with 0).
+
+ At the moment only 'CDATA' is returned.
+*/
+QString QXmlAttributes::type( int ) const
+{
+ return "CDATA";
+}
+
+/*!
+ Looks up an attribute's type by XML 1.0 qualified name.
+
+ At the moment only 'CDATA' is returned.
+*/
+QString QXmlAttributes::type( const QString& ) const
+{
+ return "CDATA";
+}
+
+/*!
+ Looks up an attribute's type by namespace name.
+
+ The first parameter specifies the namespace URI, or the empty string if
+ the name has no namespace URI. The second parameter specifies the
+ attribute's local name.
+
+ At the moment only 'CDATA' is returned.
+*/
+QString QXmlAttributes::type( const QString&, const QString& ) const
+{
+ return "CDATA";
+}
+
+/*!
+ Looks up an attribute's value by index (starting with 0).
+*/
+QString QXmlAttributes::value( int index ) const
+{
+ return valueList[index];
+}
+
+/*!
+ Looks up an attribute's value by XML 1.0 qualified name.
+
+ See also the <a href="xml-sax.html#namespaces">namespace description</a>.
+*/
+QString QXmlAttributes::value( const QString& qName ) const
+{
+ int i = index( qName );
+ if ( i == -1 )
+ return QString::null;
+ return valueList[ i ];
+}
+
+/*!
+ Looks up an attribute's value by namespace name.
+
+ \a uri specifies the namespace URI, or the empty string if the name has no
+ namespace URI. \a localName specifies the attribute's local name.
+
+ See also the <a href="xml-sax.html#namespaces">namespace description</a>.
+*/
+QString QXmlAttributes::value( const QString& uri, const QString& localName ) const
+{
+ int i = index( uri, localName );
+ if ( i == -1 )
+ return QString::null;
+ return valueList[ i ];
+}
+
+
+/*********************************************
+ *
+ * QXmlInputSource
+ *
+ *********************************************/
+
+/*!
+ \class QXmlInputSource qxml.h
+ \brief The QXmlInputSource class is the source where XML data is read from.
+
+ \module XML
+
+ All subclasses of QXmlReader read the input from this class.
+*/
+
+/*!
+ Returns all the data this input source contains.
+*/
+const QString& QXmlInputSource::data() const
+{
+ return input;
+}
+
+/*!
+ Constructs a input source which contains no data.
+*/
+QXmlInputSource::QXmlInputSource( )
+{
+ input = "";
+}
+
+/*!
+ Constructs a input source and get the data from the text stream.
+*/
+QXmlInputSource::QXmlInputSource( QTextStream& stream )
+{
+ QByteArray rawData;
+ if ( stream.device()->isDirectAccess() ) {
+ rawData = stream.device()->readAll();
+ } else {
+ int nread = 0;
+ const int bufsize = 512;
+ while ( !stream.device()->atEnd() ) {
+ rawData.resize( nread + bufsize );
+ nread += stream.device()->readBlock( rawData.data()+nread, bufsize );
+ }
+ rawData.resize( nread );
+ }
+ readInput( rawData );
+}
+
+/*!
+ Constructs a input source and get the data from a file. If the file cannot be
+ read the input source is empty.
+*/
+QXmlInputSource::QXmlInputSource( QFile& file )
+{
+ if ( !file.open(IO_ReadOnly) ) {
+ input = "";
+ return;
+ }
+ QByteArray rawData = file.readAll();
+ readInput( rawData );
+ file.close();
+}
+
+/*!
+ Destructor.
+*/
+QXmlInputSource::~QXmlInputSource()
+{
+}
+
+/*!
+ Sets the data of the input source to \a dat.
+*/
+void QXmlInputSource::setData( const QString& dat )
+{
+ input = dat;
+}
+
+/*!
+ Read the XML file from the byte array; try to recoginize the encoding.
+*/
+// ### The input source should not do the encoding detection!
+void QXmlInputSource::readInput( QByteArray& rawData )
+{
+ QBuffer buf( rawData );
+ buf.open( IO_ReadOnly );
+ QTextStream *stream = new QTextStream( &buf );
+ QChar tmp;
+ // assume UTF8 or UTF16 at first
+ stream->setEncoding( QTextStream::UnicodeUTF8 );
+ input = "";
+ // read the first 5 characters
+ for ( int i=0; i<5; i++ ) {
+ *stream >> tmp;
+ input += tmp;
+ }
+ // starts the document with an XML declaration?
+ if ( input == "<?xml" ) {
+ // read the whole XML declaration
+ do {
+ *stream >> tmp;
+ input += tmp;
+ } while( tmp != '>' );
+ // and try to find out if there is an encoding
+ int pos = input.find( "encoding" );
+ if ( pos != -1 ) {
+ QString encoding;
+ do {
+ pos++;
+ if ( pos > (int)input.length() )
+ goto finished;
+ } while( input[pos] != '"' && input[pos] != '\'' );
+ pos++;
+ while( input[pos] != '"' && input[pos] != '\'' ) {
+ encoding += input[pos];
+ pos++;
+ if ( pos > (int)input.length() )
+ goto finished;
+ }
+ delete stream;
+ stream = new QTextStream( &buf );
+ stream->setCodec( QTextCodec::codecForName( encoding ) );
+ buf.reset();
+ input = "";
+ }
+ }
+finished:
+ input += stream->read();
+ delete stream;
+ buf.close();
+}
+
+
+/*********************************************
+ *
+ * QXmlDefaultHandler
+ *
+ *********************************************/
+
+/*!
+ \class QXmlContentHandler qxml.h
+ \brief The QXmlContentHandler class provides an interface to report logical
+ content of XML data.
+
+ \module XML
+
+ If the application needs to be informed of basic parsing events, it
+ implements this interface and sets it with QXmlReader::setContentHandler().
+ The reader reports basic document-related events like the start and end of
+ elements and character data through this interface.
+
+ The order of events in this interface is very important, and mirrors the
+ order of information in the document itself. For example, all of an element's
+ content (character data, processing instructions, and/or subelements) will
+ appear, in order, between the startElement() event and the corresponding
+ endElement() event.
+
+ The class QXmlDefaultHandler gives a default implementation for this
+ interface; subclassing from this class is very convenient if you want only be
+ informed of some parsing events.
+
+ See also the <a href="xml.html#introSAX2">Introduction to SAX2</a>.
+
+ \sa QXmlDTDHandler QXmlDeclHandler QXmlEntityResolver QXmlErrorHandler
+ QXmlLexicalHandler
+*/
+/*!
+ \fn void QXmlContentHandler::setDocumentLocator( QXmlLocator* locator )
+
+ The reader calls this function before he starts parsing the document. The
+ argument \a locator is a pointer to a QXmlLocator which allows the
+ application to get the actual position of the parsing in the document.
+
+ Do not destroy the \a locator; it is destroyed when the reader is destroyed
+ (do not use the \a locator after the reader got destroyed).
+*/
+/*!
+ \fn bool QXmlContentHandler::startDocument()
+
+ The reader calls this function when he starts parsing the document.
+ The reader will call this function only once before any other functions in
+ this class or in the QXmlDTDHandler class are called (except
+ QXmlContentHandler::setDocumentLocator()).
+
+ If this function returns FALSE the reader will stop parsing and will report
+ an error. The reader will use the function errorString() to get the error
+ message that will be used for reporting the error.
+
+ \sa endDocument()
+*/
+/*!
+ \fn bool QXmlContentHandler::endDocument()
+
+ The reader calls this function after he has finished the parsing. It
+ is only called once. It is the last function of all handler functions that is
+ called. It is called after the reader has read all input or has abandoned
+ parsing because of a fatal error.
+
+ If this function returns FALSE the reader will stop parsing and will report
+ an error. The reader will use the function errorString() to get the error
+ message that will be used for reporting the error.
+
+ \sa startDocument()
+*/
+/*!
+ \fn bool QXmlContentHandler::startPrefixMapping( const QString& prefix, const QString& uri )
+
+ The reader calls this function to signal the begin of a prefix-URI
+ namespace mapping scope. This information is not necessary for normal
+ namespace processing since the reader automatically replaces prefixes for
+ element and attribute names.
+
+ Note that startPrefixMapping and endPrefixMapping calls are not guaranteed to
+ be properly nested relative to each-other: all startPrefixMapping events will
+ occur before the corresponding startElement event, and all endPrefixMapping
+ events will occur after the corresponding endElement event, but their order
+ is not otherwise guaranteed.
+
+ The argument \a prefix is the namespace prefix being declared and the
+ argument \a uri is the namespace URI the prefix is mapped to.
+
+ If this function returns FALSE the reader will stop parsing and will report
+ an error. The reader will use the function errorString() to get the error
+ message that will be used for reporting the error.
+
+ See also the <a href="xml-sax.html#namespaces">namespace description</a>.
+
+ \sa endPrefixMapping()
+*/
+/*!
+ \fn bool QXmlContentHandler::endPrefixMapping( const QString& prefix )
+
+ The reader calls this function to signal the end of a prefix mapping.
+
+ If this function returns FALSE the reader will stop parsing and will report
+ an error. The reader will use the function errorString() to get the error
+ message that will be used for reporting the error.
+
+ See also the <a href="xml-sax.html#namespaces">namespace description</a>.
+
+ \sa startPrefixMapping()
+*/
+/*!
+ \fn bool QXmlContentHandler::startElement( const QString& namespaceURI, const QString& localName, const QString& qName, const QXmlAttributes& atts )
+
+ The reader calls this function when he has parsed a start element tag.
+
+ There will be a corresponding endElement() call when the corresponding end
+ element tag was read. The startElement() and endElement() calls are always
+ nested correctly. Empty element tags (e.g. &lt;a/&gt;) are reported by
+ startElement() directly followed by a call to endElement().
+
+ The attribute list provided will contain only attributes with explicit
+ values. The attribute list will contain attributes used for namespace
+ declaration (i.e. attributes starting with xmlns) only if the
+ namespace-prefix property of the reader is TRUE.
+
+ The argument \a uri is the namespace URI, or the empty string if the element
+ has no namespace URI or if namespace processing is not being performed, \a
+ localName is the local name (without prefix), or the empty string if
+ namespace processing is not being performed, \a qName is the qualified name
+ (with prefix), or the empty string if qualified names are not available and
+ \a atts are the attributes attached to the element. If there are no
+ attributes, \a atts is an empty attributes object
+
+ If this function returns FALSE the reader will stop parsing and will report
+ an error. The reader will use the function errorString() to get the error
+ message that will be used for reporting the error.
+
+ See also the <a href="xml-sax.html#namespaces">namespace description</a>.
+
+ \sa endElement()
+*/
+/*!
+ \fn bool QXmlContentHandler::endElement( const QString& namespaceURI, const QString& localName, const QString& qName )
+
+ The reader calls this function when he has parsed an end element tag.
+
+ If this function returns FALSE the reader will stop parsing and will report
+ an error. The reader will use the function errorString() to get the error
+ message that will be used for reporting the error.
+
+ See also the <a href="xml-sax.html#namespaces">namespace description</a>.
+
+ \sa startElement()
+*/
+/*!
+ \fn bool QXmlContentHandler::characters( const QString& ch )
+
+ The reader calls this function when he has parsed a chunk of character
+ data (either normal character data or character data inside a CDATA section;
+ if you have to distinguish between those two types you have to use
+ QXmlLexicalHandler::startCDATA() and QXmlLexicalHandler::endCDATA() in
+ addition).
+
+ Some readers will report whitespace in element content using the
+ ignorableWhitespace() function rather than this one (QXmlSimpleReader will
+ do it not though).
+
+ A reader is allowed to report the character data of an element in more than
+ one chunk; e.g. a reader might want to report "a &amp;lt; b" in three
+ characters() events ("a ", "<" and " b").
+
+ If this function returns FALSE the reader will stop parsing and will report
+ an error. The reader will use the function errorString() to get the error
+ message that will be used for reporting the error.
+*/
+/*!
+ \fn bool QXmlContentHandler::ignorableWhitespace( const QString& ch )
+
+ Some readers may use this function to report each chunk of whitespace in
+ element content (QXmlSimpleReader does not though).
+
+ If this function returns FALSE the reader will stop parsing and will report
+ an error. The reader will use the function errorString() to get the error
+ message that will be used for reporting the error.
+*/
+/*!
+ \fn bool QXmlContentHandler::processingInstruction( const QString& target, const QString& data )
+
+ The reader calls this function when he has parsed a processing
+ instruction.
+
+ \a target is the target name of the processing instruction and \a data is the
+ data of the processing instruction.
+
+ If this function returns FALSE the reader will stop parsing and will report
+ an error. The reader will use the function errorString() to get the error
+ message that will be used for reporting the error.
+*/
+/*!
+ \fn bool QXmlContentHandler::skippedEntity( const QString& name )
+
+ Some readers may skip entities if they have not seen the declarations (e.g.
+ because they are in an external DTD). If they do so they will report it by
+ calling this function.
+
+ If this function returns FALSE the reader will stop parsing and will report
+ an error. The reader will use the function errorString() to get the error
+ message that will be used for reporting the error.
+*/
+/*!
+ \fn QString QXmlContentHandler::errorString()
+
+ The reader calls this function to get an error string if any of the handler
+ functions returns FALSE to him.
+*/
+
+
+/*!
+ \class QXmlErrorHandler qxml.h
+ \brief The QXmlErrorHandler class provides an interface to report errors in
+ XML data.
+
+ \module XML
+
+ If the application is interested in reporting errors to the user or any other
+ customized error handling, you should subclass this class.
+
+ You can set the error handler with QXmlReader::setErrorHandler().
+
+ See also the <a href="xml.html#introSAX2">Introduction to SAX2</a>.
+
+ \sa QXmlDTDHandler QXmlDeclHandler QXmlContentHandler QXmlEntityResolver
+ QXmlLexicalHandler
+*/
+/*!
+ \fn bool QXmlErrorHandler::warning( const QXmlParseException& exception )
+
+ A reader might use this function to report a warning. Warnings are conditions
+ that are not errors or fatal errors as defined by the XML 1.0 specification.
+
+ If this function returns FALSE the reader will stop parsing and will report
+ an error. The reader will use the function errorString() to get the error
+ message that will be used for reporting the error.
+*/
+/*!
+ \fn bool QXmlErrorHandler::error( const QXmlParseException& exception )
+
+ A reader might use this function to report a recoverable error. A recoverable
+ error corresponds to the definiton of "error" in section 1.2 of the XML 1.0
+ specification.
+
+ The reader must continue to provide normal parsing events after invoking this
+ function.
+
+ If this function returns FALSE the reader will stop parsing and will report
+ an error. The reader will use the function errorString() to get the error
+ message that will be used for reporting the error.
+*/
+/*!
+ \fn bool QXmlErrorHandler::fatalError( const QXmlParseException& exception )
+
+ A reader must use this function to report a non-recoverable error.
+
+ If this function returns TRUE the reader might try to go on parsing and
+ reporting further errors; but no regular parsing events are reported.
+*/
+/*!
+ \fn QString QXmlErrorHandler::errorString()
+
+ The reader calls this function to get an error string if any of the handler
+ functions returns FALSE to him.
+*/
+
+
+/*!
+ \class QXmlDTDHandler qxml.h
+ \brief The QXmlDTDHandler class provides an interface to report DTD content
+ of XML data.
+
+ \module XML
+
+ If an application needs information about notations and unparsed entities,
+ then the application implements this interface and registers an instance with
+ QXmlReader::setDTDHandler().
+
+ Note that this interface includes only those DTD events that the XML
+ recommendation requires processors to report: notation and unparsed entity
+ declarations.
+
+ See also the <a href="xml.html#introSAX2">Introduction to SAX2</a>.
+
+ \sa QXmlDeclHandler QXmlContentHandler QXmlEntityResolver QXmlErrorHandler
+ QXmlLexicalHandler
+*/
+/*!
+ \fn bool QXmlDTDHandler::notationDecl( const QString& name, const QString& publicId, const QString& systemId )
+
+ The reader calls this function when he has parsed a notation
+ declaration.
+
+ The argument \a name is the notation name, \a publicId is the notations's
+ public identifier and \a systemId is the notations's system identifier.
+
+ If this function returns FALSE the reader will stop parsing and will report
+ an error. The reader will use the function errorString() to get the error
+ message that will be used for reporting the error.
+*/
+/*!
+ \fn bool QXmlDTDHandler::unparsedEntityDecl( const QString& name, const QString& publicId, const QString& systemId, const QString& notationName )
+
+ The reader calls this function when he finds an unparsed entity declaration.
+
+ The argument \a name is the unparsed entity's name, \a publicId is the
+ entity's public identifier, \a systemId is the entity's system identifier and
+ \a notation is the name of the associated notation.
+
+ If this function returns FALSE the reader will stop parsing and will report
+ an error. The reader will use the function errorString() to get the error
+ message that will be used for reporting the error.
+*/
+/*!
+ \fn QString QXmlDTDHandler::errorString()
+
+ The reader calls this function to get an error string if any of the handler
+ functions returns FALSE to him.
+*/
+
+
+/*!
+ \class QXmlEntityResolver qxml.h
+ \brief The QXmlEntityResolver class provides an interface to resolve extern
+ entities contained in XML data.
+
+ \module XML
+
+ If an application needs to implement customized handling for external
+ entities, it must implement this interface and register it with
+ QXmlReader::setEntityResolver().
+
+ See also the <a href="xml.html#introSAX2">Introduction to SAX2</a>.
+
+ \sa QXmlDTDHandler QXmlDeclHandler QXmlContentHandler QXmlErrorHandler
+ QXmlLexicalHandler
+*/
+/*!
+ \fn bool QXmlEntityResolver::resolveEntity( const QString& publicId, const QString& systemId, QXmlInputSource* ret )
+
+ The reader will call this function before he opens any external entity,
+ except the top-level document entity. The application may request the reader
+ to resolve the entity itself (\a ret is 0) or to use an entirely different
+ input source (\a ret points to the input source).
+
+ The reader will delete the input source \a ret when he no longer needs it. So
+ you should allocate it on the heap with \c new.
+
+ The argument \a publicId is the public identifier of the external entity, \a
+ systemId is the system identifier of the external entity and \a ret is the
+ return value of this function: if it is 0 the reader should resolve the
+ entity itself, if it is non-zero it must point to an input source which the
+ reader will use instead.
+
+ If this function returns FALSE the reader will stop parsing and will report
+ an error. The reader will use the function errorString() to get the error
+ message that will be used for reporting the error.
+*/
+/*!
+ \fn QString QXmlEntityResolver::errorString()
+
+ The reader calls this function to get an error string if any of the handler
+ functions returns FALSE to him.
+*/
+
+
+/*!
+ \class QXmlLexicalHandler qxml.h
+ \brief The QXmlLexicalHandler class provides an interface to report lexical
+ content of XML data.
+
+ \module XML
+
+ The events in the lexical handler apply to the entire document, not just to
+ the document element, and all lexical handler events appear between the
+ content handler's startDocument and endDocument events.
+
+ You can set the lexical handler with QXmlReader::setLexicalHandler().
+
+ This interface is designed after the SAX2 extension LexicalHandler. The
+ functions startEntity() and endEntity() are not included though.
+
+ See also the <a href="xml.html#introSAX2">Introduction to SAX2</a>.
+
+ \sa QXmlDTDHandler QXmlDeclHandler QXmlContentHandler QXmlEntityResolver
+ QXmlErrorHandler
+*/
+/*!
+ \fn bool QXmlLexicalHandler::startDTD( const QString& name, const QString& publicId, const QString& systemId )
+
+ The reader calls this function to report the start of a DTD declaration, if
+ any.
+
+ All declarations reported through QXmlDTDHandler or QXmlDeclHandler appear
+ between the startDTD() and endDTD() calls.
+
+ If this function returns FALSE the reader will stop parsing and will report
+ an error. The reader will use the function errorString() to get the error
+ message that will be used for reporting the error.
+
+ \sa endDTD()
+*/
+/*!
+ \fn bool QXmlLexicalHandler::endDTD()
+
+ The reader calls this function to report the end of a DTD declaration, if
+ any.
+
+ If this function returns FALSE the reader will stop parsing and will report
+ an error. The reader will use the function errorString() to get the error
+ message that will be used for reporting the error.
+
+ \sa startDTD()
+*/
+/*!
+ \fn bool QXmlLexicalHandler::startCDATA()
+
+ The reader calls this function to report the start of a CDATA section. The
+ content of the CDATA section will be reported through the regular
+ QXmlContentHandler::characters(). This function is intended only to report
+ the boundary.
+
+ If this function returns FALSE the reader will stop parsing and will report
+ an error. The reader will use the function errorString() to get the error
+ message that will be used for reporting the error.
+
+ \sa endCDATA()
+*/
+/*!
+ \fn bool QXmlLexicalHandler::endCDATA()
+
+ The reader calls this function to report the end of a CDATA section.
+
+ If this function returns FALSE the reader will stop parsing and will report
+ an error. The reader will use the function errorString() to get the error
+ message that will be used for reporting the error.
+
+ \sa startCDATA()
+*/
+/*!
+ \fn bool QXmlLexicalHandler::comment( const QString& ch )
+
+ The reader calls this function to report an XML comment anywhere in the
+ document.
+
+ If this function returns FALSE the reader will stop parsing and will report
+ an error. The reader will use the function errorString() to get the error
+ message that will be used for reporting the error.
+*/
+/*!
+ \fn QString QXmlLexicalHandler::errorString()
+
+ The reader calls this function to get an error string if any of the handler
+ functions returns FALSE to him.
+*/
+
+
+/*!
+ \class QXmlDeclHandler qxml.h
+ \brief The QXmlDeclHandler class provides an interface to report declaration
+ content of XML data.
+
+ \module XML
+
+ You can set the declaration handler with QXmlReader::setDeclHandler().
+
+ This interface is designed after the SAX2 extension DeclHandler.
+
+ See also the <a href="xml.html#introSAX2">Introduction to SAX2</a>.
+
+ \sa QXmlDTDHandler QXmlContentHandler QXmlEntityResolver QXmlErrorHandler
+ QXmlLexicalHandler
+*/
+/*!
+ \fn bool QXmlDeclHandler::attributeDecl( const QString& eName, const QString& aName, const QString& type, const QString& valueDefault, const QString& value )
+
+ The reader calls this function to report an attribute type declaration. Only
+ the effective (first) declaration for an attribute will be reported.
+
+ If this function returns FALSE the reader will stop parsing and will report
+ an error. The reader will use the function errorString() to get the error
+ message that will be used for reporting the error.
+*/
+/*!
+ \fn bool QXmlDeclHandler::internalEntityDecl( const QString& name, const QString& value )
+
+ The reader calls this function to report an internal entity declaration. Only
+ the effective (first) declaration will be reported.
+
+ If this function returns FALSE the reader will stop parsing and will report
+ an error. The reader will use the function errorString() to get the error
+ message that will be used for reporting the error.
+*/
+/*!
+ \fn bool QXmlDeclHandler::externalEntityDecl( const QString& name, const QString& publicId, const QString& systemId )
+
+ The reader calls this function to report a parsed external entity
+ declaration. Only the effective (first) declaration for each entity will be
+ reported.
+
+ If this function returns FALSE the reader will stop parsing and will report
+ an error. The reader will use the function errorString() to get the error
+ message that will be used for reporting the error.
+*/
+/*!
+ \fn QString QXmlDeclHandler::errorString()
+
+ The reader calls this function to get an error string if any of the handler
+ functions returns FALSE to him.
+*/
+
+
+/*!
+ \class QXmlDefaultHandler qxml.h
+ \brief The QXmlDefaultHandler class provides a default implementation of all
+ XML handler classes.
+
+ \module XML
+
+ Very often you are only interested in parts of the things that that the
+ reader reports to you. This class simply implements a default behaviour of
+ the handler classes (most of the time: do nothing). Normally this is the
+ class you subclass for implementing your customized handler.
+
+ See also the <a href="xml.html#introSAX2">Introduction to SAX2</a>.
+
+ \sa QXmlDTDHandler QXmlDeclHandler QXmlContentHandler QXmlEntityResolver
+ QXmlErrorHandler QXmlLexicalHandler
+*/
+/*!
+ \fn QXmlDefaultHandler::QXmlDefaultHandler()
+
+ Constructor.
+*/
+/*!
+ \fn QXmlDefaultHandler::~QXmlDefaultHandler()
+
+ Destructor.
+*/
+
+/*!
+ Does nothing.
+*/
+void QXmlDefaultHandler::setDocumentLocator( QXmlLocator* )
+{
+}
+
+/*!
+ Does nothing.
+*/
+bool QXmlDefaultHandler::startDocument()
+{
+ return TRUE;
+}
+
+/*!
+ Does nothing.
+*/
+bool QXmlDefaultHandler::endDocument()
+{
+ return TRUE;
+}
+
+/*!
+ Does nothing.
+*/
+bool QXmlDefaultHandler::startPrefixMapping( const QString&, const QString& )
+{
+ return TRUE;
+}
+
+/*!
+ Does nothing.
+*/
+bool QXmlDefaultHandler::endPrefixMapping( const QString& )
+{
+ return TRUE;
+}
+
+/*!
+ Does nothing.
+*/
+bool QXmlDefaultHandler::startElement( const QString&, const QString&,
+ const QString&, const QXmlAttributes& )
+{
+ return TRUE;
+}
+
+/*!
+ Does nothing.
+*/
+bool QXmlDefaultHandler::endElement( const QString&, const QString&,
+ const QString& )
+{
+ return TRUE;
+}
+
+/*!
+ Does nothing.
+*/
+bool QXmlDefaultHandler::characters( const QString& )
+{
+ return TRUE;
+}
+
+/*!
+ Does nothing.
+*/
+bool QXmlDefaultHandler::ignorableWhitespace( const QString& )
+{
+ return TRUE;
+}
+
+/*!
+ Does nothing.
+*/
+bool QXmlDefaultHandler::processingInstruction( const QString&,
+ const QString& )
+{
+ return TRUE;
+}
+
+/*!
+ Does nothing.
+*/
+bool QXmlDefaultHandler::skippedEntity( const QString& )
+{
+ return TRUE;
+}
+
+/*!
+ Does nothing.
+*/
+bool QXmlDefaultHandler::warning( const QXmlParseException& )
+{
+ return TRUE;
+}
+
+/*!
+ Does nothing.
+*/
+bool QXmlDefaultHandler::error( const QXmlParseException& )
+{
+ return TRUE;
+}
+
+/*!
+ Does nothing.
+*/
+bool QXmlDefaultHandler::fatalError( const QXmlParseException& )
+{
+ return TRUE;
+}
+
+/*!
+ Does nothing.
+*/
+bool QXmlDefaultHandler::notationDecl( const QString&, const QString&,
+ const QString& )
+{
+ return TRUE;
+}
+
+/*!
+ Does nothing.
+*/
+bool QXmlDefaultHandler::unparsedEntityDecl( const QString&, const QString&,
+ const QString&, const QString& )
+{
+ return TRUE;
+}
+
+/*!
+ Always sets \a ret to 0, so that the reader will use the system identifier
+ provided in the XML document.
+*/
+bool QXmlDefaultHandler::resolveEntity( const QString&, const QString&,
+ QXmlInputSource* ret )
+{
+ ret = 0;
+ return TRUE;
+}
+
+/*!
+ Returns the default error string.
+*/
+QString QXmlDefaultHandler::errorString()
+{
+ return QString( XMLERR_ERRORBYCONSUMER );
+}
+
+/*!
+ Does nothing.
+*/
+bool QXmlDefaultHandler::startDTD( const QString&, const QString&, const QString& )
+{
+ return TRUE;
+}
+
+/*!
+ Does nothing.
+*/
+bool QXmlDefaultHandler::endDTD()
+{
+ return TRUE;
+}
+
+#if 0
+/*!
+ Does nothing.
+*/
+bool QXmlDefaultHandler::startEntity( const QString& )
+{
+ return TRUE;
+}
+
+/*!
+ Does nothing.
+*/
+bool QXmlDefaultHandler::endEntity( const QString& )
+{
+ return TRUE;
+}
+#endif
+
+/*!
+ Does nothing.
+*/
+bool QXmlDefaultHandler::startCDATA()
+{
+ return TRUE;
+}
+
+/*!
+ Does nothing.
+*/
+bool QXmlDefaultHandler::endCDATA()
+{
+ return TRUE;
+}
+
+/*!
+ Does nothing.
+*/
+bool QXmlDefaultHandler::comment( const QString& )
+{
+ return TRUE;
+}
+
+/*!
+ Does nothing.
+*/
+bool QXmlDefaultHandler::attributeDecl( const QString&, const QString&, const QString&, const QString&, const QString& )
+{
+ return TRUE;
+}
+
+/*!
+ Does nothing.
+*/
+bool QXmlDefaultHandler::internalEntityDecl( const QString&, const QString& )
+{
+ return TRUE;
+}
+
+/*!
+ Does nothing.
+*/
+bool QXmlDefaultHandler::externalEntityDecl( const QString&, const QString&, const QString& )
+{
+ return TRUE;
+}
+
+
+/*********************************************
+ *
+ * QXmlSimpleReaderPrivate
+ *
+ *********************************************/
+
+class QXmlSimpleReaderPrivate
+{
+private:
+ // constructor
+ QXmlSimpleReaderPrivate()
+ { }
+
+
+ // used for entity declarations
+ struct ExternParameterEntity
+ {
+ ExternParameterEntity( ) {}
+ ExternParameterEntity( const QString &p, const QString &s )
+ : publicId(p), systemId(s) {}
+ QString publicId;
+ QString systemId;
+ };
+ struct ExternEntity
+ {
+ ExternEntity( ) {}
+ ExternEntity( const QString &p, const QString &s, const QString &n )
+ : publicId(p), systemId(s), notation(n) {}
+ QString publicId;
+ QString systemId;
+ QString notation;
+ };
+ QMap<QString,ExternParameterEntity> externParameterEntities;
+ QMap<QString,QString> parameterEntities;
+ QMap<QString,ExternEntity> externEntities;
+ QMap<QString,QString> entities;
+
+ // used for standalone declaration
+ enum Standalone { Yes, No, Unknown };
+
+ QString doctype; // only used for the doctype
+ QString xmlVersion; // only used to store the version information
+ QString encoding; // only used to store the encoding
+ Standalone standalone; // used to store the value of the standalone declaration
+
+ QString publicId; // used by parseExternalID() to store the public ID
+ QString systemId; // used by parseExternalID() to store the system ID
+ QString attDeclEName; // use by parseAttlistDecl()
+ QString attDeclAName; // use by parseAttlistDecl()
+
+ // flags for some features support
+ bool useNamespaces;
+ bool useNamespacePrefixes;
+ bool reportWhitespaceCharData;
+
+ // used to build the attribute list
+ QXmlAttributes attList;
+
+ // helper classes
+ QXmlLocator *locator;
+ QXmlNamespaceSupport namespaceSupport;
+
+ // error string
+ QString error;
+
+ // friend declarations
+ friend class QXmlSimpleReader;
+};
+
+
+/*********************************************
+ *
+ * QXmlSimpleReader
+ *
+ *********************************************/
+
+/*!
+ \class QXmlReader qxml.h
+ \brief The QXmlReader class provides an interface for XML readers (i.e.
+ parsers).
+
+ \module XML
+
+ This abstract class describes an interface for all XML readers in Qt. At the
+ moment there is only one implementation of a reader included in the XML
+ module of Qt (QXmlSimpleReader). In future releases there might be more
+ readers with different properties available (e.g. a validating parser).
+
+ The design of the XML classes follow the
+ <a href="http://www.megginson.com/SAX/">SAX2 java interface</a>.
+ It was adopted to fit into the Qt naming conventions; so it should be very
+ easy for anybody who has worked with SAX2 to get started with the Qt XML
+ classes.
+
+ All readers use the class QXmlInputSource to read the input document from.
+ Since you are normally interested in certain contents of the XML document,
+ the reader reports those contents through special handler classes
+ (QXmlDTDHandler, QXmlDeclHandler, QXmlContentHandler, QXmlEntityResolver,
+ QXmlErrorHandler and QXmlLexicalHandler).
+
+ You have to subclass these classes. Since the handler classes describe only
+ interfaces you must implement all functions; there is a class
+ (QXmlDefaultHandler) to make this easier; it implements a default behaviour
+ (do nothing) for all functions.
+
+ For getting started see also the
+ <a href="xml-sax.html#quickStart">Quick start</a>.
+
+ \sa QXmlSimpleReader
+*/
+/*!
+ \fn bool QXmlReader::feature( const QString& name, bool *ok ) const
+
+ If the reader has the feature \a name, this function returns the value of the
+ feature.
+
+ If the reader has not the feature \a name, the return value may be anything.
+
+ If \a ok is not 0, then \a ok is set to TRUE if the reader has the feature
+ \a name, otherwise \a ok is set to FALSE.
+
+ \sa setFeature() hasFeature()
+*/
+/*!
+ \fn void QXmlReader::setFeature( const QString& name, bool value )
+
+ Sets the feature \a name to \a value. If the reader has not the feature \a
+ name, this value is ignored.
+
+ \sa feature() hasFeature()
+*/
+/*!
+ \fn bool QXmlReader::hasFeature( const QString& name ) const
+
+ Returns \c TRUE if the reader has the feature \a name, otherwise FALSE.
+
+ \sa feature() setFeature()
+*/
+/*!
+ \fn void* QXmlReader::property( const QString& name, bool *ok ) const
+
+ If the reader has the property \a name, this function returns the value of
+ the property.
+
+ If the reader has not the property \a name, the return value is 0.
+
+ If \a ok is not 0, then \a ok is set to TRUE if the reader has the property
+ \a name, otherwise \a ok is set to FALSE.
+
+ \sa setProperty() hasProperty()
+*/
+/*!
+ \fn void QXmlReader::setProperty( const QString& name, void* value )
+
+ Sets the property \a name to \a value. If the reader has not the property \a
+ name, this value is ignored.
+
+ \sa property() hasProperty()
+*/
+/*!
+ \fn bool QXmlReader::hasProperty( const QString& name ) const
+
+ Returns TRUE if the reader has the property \a name, otherwise FALSE.
+
+ \sa property() setProperty()
+*/
+/*!
+ \fn void QXmlReader::setEntityResolver( QXmlEntityResolver* handler )
+
+ Sets the entity resolver to \a handler.
+
+ \sa entityResolver()
+*/
+/*!
+ \fn QXmlEntityResolver* QXmlReader::entityResolver() const
+
+ Returns the entity resolver or 0 if none was set.
+
+ \sa setEntityResolver()
+*/
+/*!
+ \fn void QXmlReader::setDTDHandler( QXmlDTDHandler* handler )
+
+ Sets the DTD handler to \a handler.
+
+ \sa DTDHandler()
+*/
+/*!
+ \fn QXmlDTDHandler* QXmlReader::DTDHandler() const
+
+ Returns the DTD handler or 0 if none was set.
+
+ \sa setDTDHandler()
+*/
+/*!
+ \fn void QXmlReader::setContentHandler( QXmlContentHandler* handler )
+
+ Sets the content handler to \a handler.
+
+ \sa contentHandler()
+*/
+/*!
+ \fn QXmlContentHandler* QXmlReader::contentHandler() const
+
+ Returns the content handler or 0 if none was set.
+
+ \sa setContentHandler()
+*/
+/*!
+ \fn void QXmlReader::setErrorHandler( QXmlErrorHandler* handler )
+
+ Sets the error handler to \a handler.
+
+ \sa errorHandler()
+*/
+/*!
+ \fn QXmlErrorHandler* QXmlReader::errorHandler() const
+
+ Returns the error handler or 0 if none was set
+
+ \sa setErrorHandler()
+*/
+/*!
+ \fn void QXmlReader::setLexicalHandler( QXmlLexicalHandler* handler )
+
+ Sets the lexical handler to \a handler.
+
+ \sa lexicalHandler()
+*/
+/*!
+ \fn QXmlLexicalHandler* QXmlReader::lexicalHandler() const
+
+ Returns the lexical handler or 0 if none was set.
+
+ \sa setLexicalHandler()
+*/
+/*!
+ \fn void QXmlReader::setDeclHandler( QXmlDeclHandler* handler )
+
+ Sets the declaration handler to \a handler.
+
+ \sa declHandler()
+*/
+/*!
+ \fn QXmlDeclHandler* QXmlReader::declHandler() const
+
+ Returns the declaration handler or 0 if none was set.
+
+ \sa setDeclHandler()
+*/
+/*!
+ \fn bool QXmlReader::parse( const QXmlInputSource& input )
+
+ Parses the XML document \a input. Returns TRUE if the parsing was successful,
+ otherwise FALSE.
+*/
+/*!
+ \fn bool QXmlReader::parse( const QString& systemId )
+
+ Parses the XML document at the location \a systemId. Returns TRUE if the
+ parsing was successful, otherwise FALSE.
+*/
+
+
+/*!
+ \class QXmlSimpleReader qxml.h
+ \brief The QXmlSimpleReader class provides an implementation of a simple XML
+ reader (i.e. parser).
+
+ \module XML
+
+ This XML reader is sufficient for simple parsing tasks. Here is a short list
+ of the properties of this reader:
+ <ul>
+ <li> well-formed parser
+ <li> does not parse any external entities
+ <li> can do namespace processing
+ </ul>
+
+ For getting started see also the
+ <a href="xml-sax.html#quickStart">Quick start</a>.
+*/
+
+//guaranteed not to be a characater
+const QChar QXmlSimpleReader::QEOF = QChar((ushort)0xffff);
+
+/*!
+ Constructs a simple XML reader.
+*/
+QXmlSimpleReader::QXmlSimpleReader()
+{
+ d = new QXmlSimpleReaderPrivate();
+ d->locator = new QXmlLocator( this );
+
+ entityRes = 0;
+ dtdHnd = 0;
+ contentHnd = 0;
+ errorHnd = 0;
+ lexicalHnd = 0;
+ declHnd = 0;
+
+ // default feature settings
+ d->useNamespaces = TRUE;
+ d->useNamespacePrefixes = FALSE;
+ d->reportWhitespaceCharData = TRUE;
+}
+
+/*!
+ Destroys a simple XML reader.
+*/
+QXmlSimpleReader::~QXmlSimpleReader()
+{
+ delete d->locator;
+ delete d;
+}
+
+/*!
+ Gets the state of a feature.
+
+ \sa setFeature() hasFeature()
+*/
+bool QXmlSimpleReader::feature( const QString& name, bool *ok ) const
+{
+ if ( ok != 0 )
+ *ok = TRUE;
+ if ( name == "http://xml.org/sax/features/namespaces" ) {
+ return d->useNamespaces;
+ } else if ( name == "http://xml.org/sax/features/namespace-prefixes" ) {
+ return d->useNamespacePrefixes;
+ } else if ( name == "http://trolltech.com/xml/features/report-whitespace-only-CharData" ) {
+ return d->reportWhitespaceCharData;
+ } else {
+ qWarning( "Unknown feature " + name );
+ if ( ok != 0 )
+ *ok = FALSE;
+ }
+ return FALSE;
+}
+
+/*!
+ Sets the state of a feature.
+
+ Supported features are:
+ <ul>
+ <li> http://xml.org/sax/features/namespaces:
+ if this feature is TRUE, namespace processing is performed
+ <li> http://xml.org/sax/features/namespace-prefixes:
+ if this feature is TRUE, the the original prefixed names and attributes
+ used for namespace declarations are reported
+ <li> http://trolltech.com/xml/features/report-whitespace-only-CharData:
+ if this feature is TRUE, CharData that consists only of whitespace (and
+ no other characters) is not reported via
+ QXmlContentHandler::characters()
+ </ul>
+
+ \sa feature() hasFeature()
+*/
+void QXmlSimpleReader::setFeature( const QString& name, bool value )
+{
+ if ( name == "http://xml.org/sax/features/namespaces" ) {
+ d->useNamespaces = value;
+ } else if ( name == "http://xml.org/sax/features/namespace-prefixes" ) {
+ d->useNamespacePrefixes = value;
+ } else if ( name == "http://trolltech.com/xml/features/report-whitespace-only-CharData" ) {
+ d->reportWhitespaceCharData = value;
+ } else {
+ qWarning( "Unknown feature " + name );
+ }
+}
+
+/*!
+ Returns TRUE if the class has a feature named \a feature, otherwise FALSE.
+
+ \sa setFeature() feature()
+*/
+bool QXmlSimpleReader::hasFeature( const QString& name ) const
+{
+ if ( name == "http://xml.org/sax/features/namespaces" ||
+ name == "http://xml.org/sax/features/namespace-prefixes" ||
+ name == "http://trolltech.com/xml/features/report-whitespace-only-CharData" ) {
+ return TRUE;
+ } else {
+ return FALSE;
+ }
+}
+
+/*!
+ Returns 0 since this class does not support any properties.
+*/
+void* QXmlSimpleReader::property( const QString&, bool *ok ) const
+{
+ if ( ok != 0 )
+ *ok = FALSE;
+ return 0;
+}
+
+/*!
+ Does nothing since this class does not support any properties.
+*/
+void QXmlSimpleReader::setProperty( const QString&, void* )
+{
+}
+
+/*!
+ Returns FALSE since this class does not support any properties.
+*/
+bool QXmlSimpleReader::hasProperty( const QString& ) const
+{
+ return FALSE;
+}
+
+/*! \reimp */
+void QXmlSimpleReader::setEntityResolver( QXmlEntityResolver* handler )
+{ entityRes = handler; }
+
+/*! \reimp */
+QXmlEntityResolver* QXmlSimpleReader::entityResolver() const
+{ return entityRes; }
+
+/*! \reimp */
+void QXmlSimpleReader::setDTDHandler( QXmlDTDHandler* handler )
+{ dtdHnd = handler; }
+
+/*! \reimp */
+QXmlDTDHandler* QXmlSimpleReader::DTDHandler() const
+{ return dtdHnd; }
+
+/*! \reimp */
+void QXmlSimpleReader::setContentHandler( QXmlContentHandler* handler )
+{ contentHnd = handler; }
+
+/*! \reimp */
+QXmlContentHandler* QXmlSimpleReader::contentHandler() const
+{ return contentHnd; }
+
+/*! \reimp */
+void QXmlSimpleReader::setErrorHandler( QXmlErrorHandler* handler )
+{ errorHnd = handler; }
+
+/*! \reimp */
+QXmlErrorHandler* QXmlSimpleReader::errorHandler() const
+{ return errorHnd; }
+
+/*! \reimp */
+void QXmlSimpleReader::setLexicalHandler( QXmlLexicalHandler* handler )
+{ lexicalHnd = handler; }
+
+/*! \reimp */
+QXmlLexicalHandler* QXmlSimpleReader::lexicalHandler() const
+{ return lexicalHnd; }
+
+/*! \reimp */
+void QXmlSimpleReader::setDeclHandler( QXmlDeclHandler* handler )
+{ declHnd = handler; }
+
+/*! \reimp */
+QXmlDeclHandler* QXmlSimpleReader::declHandler() const
+{ return declHnd; }
+
+
+
+/*! \reimp */
+bool QXmlSimpleReader::parse( const QXmlInputSource& input )
+{
+ init( input );
+ // call the handler
+ if ( contentHnd ) {
+ contentHnd->setDocumentLocator( d->locator );
+ if ( !contentHnd->startDocument() ) {
+ d->error = contentHnd->errorString();
+ goto parseError;
+ }
+ }
+ // parse prolog
+ if ( !parseProlog() ) {
+ d->error = XMLERR_ERRORPARSINGPROLOG;
+ goto parseError;
+ }
+ // parse element
+ if ( !parseElement() ) {
+ d->error = XMLERR_ERRORPARSINGMAINELEMENT;
+ goto parseError;
+ }
+ // parse Misc*
+ while ( !atEnd() ) {
+ if ( !parseMisc() ) {
+ d->error = XMLERR_ERRORPARSINGMISC;
+ goto parseError;
+ }
+ }
+ // is stack empty?
+ if ( !tags.isEmpty() ) {
+ d->error = XMLERR_UNEXPECTEDEOF;
+ goto parseError;
+ }
+ // call the handler
+ if ( contentHnd ) {
+ if ( !contentHnd->endDocument() ) {
+ d->error = contentHnd->errorString();
+ goto parseError;
+ }
+ }
+
+ return TRUE;
+
+ // error handling
+
+parseError:
+ reportParseError();
+ tags.clear();
+ return FALSE;
+}
+
+/*!
+ Parses the prolog [22].
+*/
+bool QXmlSimpleReader::parseProlog()
+{
+ bool xmldecl_possible = TRUE;
+ bool doctype_read = FALSE;
+
+ const signed char Init = 0;
+ const signed char EatWS = 1; // eat white spaces
+ const signed char Lt = 2; // '<' read
+ const signed char Em = 3; // '!' read
+ const signed char DocType = 4; // read doctype
+ const signed char Comment = 5; // read comment
+ const signed char PI = 6; // read PI
+ const signed char Done = 7;
+
+ const signed char InpWs = 0;
+ const signed char InpLt = 1; // <
+ const signed char InpQm = 2; // ?
+ const signed char InpEm = 3; // !
+ const signed char InpD = 4; // D
+ const signed char InpDash = 5; // -
+ const signed char InpUnknown = 6;
+
+ // use some kind of state machine for parsing
+ static signed char table[7][7] = {
+ /* InpWs InpLt InpQm InpEm InpD InpDash InpUnknown */
+ { EatWS, Lt, -1, -1, -1, -1, -1 }, // Init
+ { -1, Lt, -1, -1, -1, -1, -1 }, // EatWS
+ { -1, -1, PI, Em, Done, -1, Done }, // Lt
+ { -1, -1, -1, -1, DocType, Comment, -1 }, // Em
+ { EatWS, Lt, -1, -1, -1, -1, -1 }, // DocType
+ { EatWS, Lt, -1, -1, -1, -1, -1 }, // Comment
+ { EatWS, Lt, -1, -1, -1, -1, -1 } // PI
+ };
+ signed char state = Init;
+ signed char input;
+ bool parseOk = TRUE;
+
+ while ( TRUE ) {
+
+ // read input
+ if ( atEnd() ) {
+ d->error = XMLERR_UNEXPECTEDEOF;
+ goto parseError;
+ }
+ if ( is_S(c) ) {
+ input = InpWs;
+ } else if ( c == '<' ) {
+ input = InpLt;
+ } else if ( c == '?' ) {
+ input = InpQm;
+ } else if ( c == '!' ) {
+ input = InpEm;
+ } else if ( c == 'D' ) {
+ input = InpD;
+ } else if ( c == '-' ) {
+ input = InpDash;
+ } else {
+ input = InpUnknown;
+ }
+ // get new state
+ state = table[state][input];
+
+ // in some cases do special actions depending on state
+ switch ( state ) {
+ case EatWS:
+ // XML declaration only on first position possible
+ xmldecl_possible = FALSE;
+ // eat white spaces
+ eat_ws();
+ break;
+ case Lt:
+ // next character
+ next();
+ break;
+ case Em:
+ // XML declaration only on first position possible
+ xmldecl_possible = FALSE;
+ // next character
+ next();
+ break;
+ case DocType:
+ parseOk = parseDoctype();
+ break;
+ case Comment:
+ parseOk = parseComment();
+ break;
+ case PI:
+ parseOk = parsePI( xmldecl_possible );
+ break;
+ }
+ // no input is read after this
+ switch ( state ) {
+ case DocType:
+ if ( !parseOk ) {
+ d->error = XMLERR_ERRORPARSINGPROLOG;
+ goto parseError;
+ }
+ if ( doctype_read ) {
+ d->error = XMLERR_MORETHANONEDOCTYPE;
+ goto parseError;
+ } else {
+ doctype_read = FALSE;
+ }
+ break;
+ case Comment:
+ if ( !parseOk ) {
+ d->error = XMLERR_ERRORPARSINGPROLOG;
+ goto parseError;
+ }
+ if ( lexicalHnd ) {
+ if ( !lexicalHnd->comment( string() ) ) {
+ d->error = lexicalHnd->errorString();
+ goto parseError;
+ }
+ }
+ break;
+ case PI:
+ if ( !parseOk ) {
+ d->error = XMLERR_ERRORPARSINGPROLOG;
+ goto parseError;
+ }
+ // call the handler
+ if ( contentHnd ) {
+ if ( xmldecl_possible && !d->xmlVersion.isEmpty() ) {
+ QString value( "version = '" );
+ value += d->xmlVersion;
+ value += "'";
+ if ( !d->encoding.isEmpty() ) {
+ value += " encoding = '";
+ value += d->encoding;
+ value += "'";
+ }
+ if ( d->standalone == QXmlSimpleReaderPrivate::Yes ) {
+ value += " standalone = 'yes'";
+ } else if ( d->standalone == QXmlSimpleReaderPrivate::No ) {
+ value += " standalone = 'no'";
+ }
+ if ( !contentHnd->processingInstruction( "xml", value ) ) {
+ d->error = contentHnd->errorString();
+ goto parseError;
+ }
+ } else {
+ if ( !contentHnd->processingInstruction( name(), string() ) ) {
+ d->error = contentHnd->errorString();
+ goto parseError;
+ }
+ }
+ }
+ // XML declaration only on first position possible
+ xmldecl_possible = FALSE;
+ break;
+ case Done:
+ return TRUE;
+ case -1:
+ d->error = XMLERR_ERRORPARSINGELEMENT;
+ goto parseError;
+ }
+
+ }
+
+ return TRUE;
+
+parseError:
+ reportParseError();
+ return FALSE;
+}
+
+/*!
+ Parse an element [39].
+
+ Precondition: the opening '<' is already read.
+*/
+bool QXmlSimpleReader::parseElement()
+{
+ static QString uri, lname, prefix;
+ static bool t;
+
+ const signed char Init = 0;
+ const signed char ReadName = 1;
+ const signed char Ws1 = 2;
+ const signed char STagEnd = 3;
+ const signed char STagEnd2 = 4;
+ const signed char ETagBegin = 5;
+ const signed char ETagBegin2 = 6;
+ const signed char Ws2 = 7;
+ const signed char EmptyTag = 8;
+ const signed char Attribute = 9;
+ const signed char Ws3 = 10;
+ const signed char Done = 11;
+
+ const signed char InpWs = 0; // whitespace
+ const signed char InpNameBe = 1; // is_NameBeginning()
+ const signed char InpGt = 2; // >
+ const signed char InpSlash = 3; // /
+ const signed char InpUnknown = 4;
+
+ // use some kind of state machine for parsing
+ static signed char table[11][5] = {
+ /* InpWs InpNameBe InpGt InpSlash InpUnknown */
+ { -1, ReadName, -1, -1, -1 }, // Init
+ { Ws1, Attribute, STagEnd, EmptyTag, -1 }, // ReadName
+ { -1, Attribute, STagEnd, EmptyTag, -1 }, // Ws1
+ { STagEnd2, STagEnd2, STagEnd2, STagEnd2, STagEnd2 }, // STagEnd
+ { -1, -1, -1, ETagBegin, -1 }, // STagEnd2
+ { -1, ETagBegin2, -1, -1, -1 }, // ETagBegin
+ { Ws2, -1, Done, -1, -1 }, // ETagBegin2
+ { -1, -1, Done, -1, -1 }, // Ws2
+ { -1, -1, Done, -1, -1 }, // EmptyTag
+ { Ws3, Attribute, STagEnd, EmptyTag, -1 }, // Attribute
+ { -1, Attribute, STagEnd, EmptyTag, -1 } // Ws3
+ };
+ signed char state = Init;
+ signed char input;
+ bool parseOk = TRUE;
+
+ while ( TRUE ) {
+
+ // read input
+ if ( atEnd() ) {
+ d->error = XMLERR_UNEXPECTEDEOF;
+ goto parseError;
+ }
+ if ( is_S(c) ) {
+ input = InpWs;
+ } else if ( is_NameBeginning(c) ) {
+ input = InpNameBe;
+ } else if ( c == '>' ) {
+ input = InpGt;
+ } else if ( c == '/' ) {
+ input = InpSlash;
+ } else {
+ input = InpUnknown;
+ }
+ // get new state
+//qDebug( "%d -%d(%c)-> %d", state, input, c.latin1(), table[state][input] );
+ state = table[state][input];
+
+ // in some cases do special actions depending on state
+ switch ( state ) {
+ case ReadName:
+ parseOk = parseName();
+ break;
+ case Ws1:
+ case Ws2:
+ case Ws3:
+ eat_ws();
+ break;
+ case STagEnd:
+ // call the handler
+ if ( contentHnd ) {
+ if ( d->useNamespaces ) {
+ d->namespaceSupport.processName( tags.top(), FALSE, uri, lname );
+ t = contentHnd->startElement( uri, lname, tags.top(), d->attList );
+ } else {
+ t = contentHnd->startElement( "", "", tags.top(), d->attList );
+ }
+ if ( !t ) {
+ d->error = contentHnd->errorString();
+ goto parseError;
+ }
+ }
+ next();
+ break;
+ case STagEnd2:
+ parseOk = parseContent();
+ break;
+ case ETagBegin:
+ next();
+ break;
+ case ETagBegin2:
+ // get the name of the tag
+ parseOk = parseName();
+ break;
+ case EmptyTag:
+ if ( tags.isEmpty() ) {
+ d->error = XMLERR_TAGMISMATCH;
+ goto parseError;
+ }
+ if ( !parseElementEmptyTag( t, uri, lname ) )
+ goto parseError;
+ // next character
+ next();
+ break;
+ case Attribute:
+ // get name and value of attribute
+ parseOk = parseAttribute();
+ break;
+ case Done:
+ next();
+ break;
+ }
+ // no input is read after this
+ switch ( state ) {
+ case ReadName:
+ if ( !parseOk ) {
+ d->error = XMLERR_ERRORPARSINGNAME;
+ goto parseError;
+ }
+ // store it on the stack
+ tags.push( name() );
+ // empty the attributes
+ d->attList.qnameList.clear();
+ d->attList.uriList.clear();
+ d->attList.localnameList.clear();
+ d->attList.valueList.clear();
+ // namespace support?
+ if ( d->useNamespaces ) {
+ d->namespaceSupport.pushContext();
+ }
+ break;
+ case STagEnd2:
+ if ( !parseOk ) {
+ d->error = XMLERR_ERRORPARSINGCONTENT;
+ goto parseError;
+ }
+ break;
+ case ETagBegin2:
+ if ( !parseOk ) {
+ d->error = XMLERR_ERRORPARSINGNAME;
+ goto parseError;
+ }
+ if ( !parseElementETagBegin2() )
+ goto parseError;
+ break;
+ case Attribute:
+ if ( !parseOk ) {
+ d->error = XMLERR_ERRORPARSINGATTRIBUTE;
+ goto parseError;
+ }
+ if ( !parseElementAttribute( prefix, uri, lname ) )
+ goto parseError;
+ break;
+ case Done:
+ return TRUE;
+ case -1:
+ d->error = XMLERR_ERRORPARSINGELEMENT;
+ goto parseError;
+ }
+
+ }
+
+ return TRUE;
+
+parseError:
+ reportParseError();
+ return FALSE;
+}
+/*!
+ Helper to break down the size of the code in the case statement.
+ Return FALSE on error, otherwise TRUE.
+*/
+bool QXmlSimpleReader::parseElementEmptyTag( bool &t, QString &uri, QString &lname )
+{
+ // pop the stack and call the handler
+ if ( contentHnd ) {
+ // report startElement first...
+ if ( d->useNamespaces ) {
+ d->namespaceSupport.processName( tags.top(), FALSE, uri, lname );
+ t = contentHnd->startElement( uri, lname, tags.top(), d->attList );
+ } else {
+ t = contentHnd->startElement( "", "", tags.top(), d->attList );
+ }
+ if ( !t ) {
+ d->error = contentHnd->errorString();
+ return FALSE;
+ }
+ // ... followed by endElement
+ // ### missing namespace support!
+ if ( !contentHnd->endElement( "","",tags.pop() ) ) {
+ d->error = contentHnd->errorString();
+ return FALSE;
+ }
+ // namespace support?
+ if ( d->useNamespaces ) {
+ QStringList prefixesBefore, prefixesAfter;
+ if ( contentHnd ) {
+ prefixesBefore = d->namespaceSupport.prefixes();
+ }
+ d->namespaceSupport.popContext();
+ // call the handler for prefix mapping
+ if ( contentHnd ) {
+ prefixesAfter = d->namespaceSupport.prefixes();
+ for ( QStringList::Iterator it = prefixesBefore.begin(); it != prefixesBefore.end(); ++it ) {
+ if ( prefixesAfter.contains(*it) == 0 ) {
+ if ( !contentHnd->endPrefixMapping( *it ) ) {
+ d->error = contentHnd->errorString();
+ return FALSE;
+ }
+ }
+ }
+ }
+ }
+ } else {
+ tags.pop();
+ }
+ return TRUE;
+}
+/*!
+ Helper to break down the size of the code in the case statement.
+ Return FALSE on error, otherwise TRUE.
+*/
+bool QXmlSimpleReader::parseElementETagBegin2()
+{
+
+ // pop the stack and compare it with the name
+ if ( tags.pop() != name() ) {
+ d->error = XMLERR_TAGMISMATCH;
+ return FALSE;
+ }
+ // call the handler
+ // ### missing namespace support!
+ if ( contentHnd ) {
+ if ( !contentHnd->endElement("","",name()) ) {
+ d->error = contentHnd->errorString();
+ return FALSE;
+ }
+ }
+ // namespace support?
+ if ( d->useNamespaces ) {
+ QStringList prefixesBefore, prefixesAfter;
+ if ( contentHnd ) {
+ prefixesBefore = d->namespaceSupport.prefixes();
+ }
+ d->namespaceSupport.popContext();
+ // call the handler for prefix mapping
+ if ( contentHnd ) {
+ prefixesAfter = d->namespaceSupport.prefixes();
+ for ( QStringList::Iterator it = prefixesBefore.begin(); it != prefixesBefore.end(); ++it ) {
+ if ( prefixesAfter.contains(*it) == 0 ) {
+ if ( !contentHnd->endPrefixMapping( *it ) ) {
+ d->error = contentHnd->errorString();
+ return FALSE;
+ }
+ }
+ }
+ }
+ }
+ return TRUE;
+}
+/*!
+ Helper to break down the size of the code in the case statement.
+ Return FALSE on error, otherwise TRUE.
+*/
+bool QXmlSimpleReader::parseElementAttribute( QString &prefix, QString &uri, QString &lname )
+{
+ // add the attribute to the list
+ if ( d->useNamespaces ) {
+ // is it a namespace declaration?
+ d->namespaceSupport.splitName( name(), prefix, lname );
+ if ( prefix == "xmlns" ) {
+ // namespace declaration
+ d->namespaceSupport.setPrefix( lname, string() );
+ if ( d->useNamespacePrefixes ) {
+ d->attList.qnameList.append( name() );
+ d->attList.uriList.append( "" );
+ d->attList.localnameList.append( "" );
+ d->attList.valueList.append( string() );
+ }
+ // call the handler for prefix mapping
+ if ( contentHnd ) {
+ if ( !contentHnd->startPrefixMapping( lname, string() ) ) {
+ d->error = contentHnd->errorString();
+ return FALSE;
+ }
+ }
+ } else {
+ // no namespace delcaration
+ d->namespaceSupport.processName( name(), TRUE, uri, lname );
+ d->attList.qnameList.append( name() );
+ d->attList.uriList.append( uri );
+ d->attList.localnameList.append( lname );
+ d->attList.valueList.append( string() );
+ }
+ } else {
+ // no namespace support
+ d->attList.qnameList.append( name() );
+ d->attList.uriList.append( "" );
+ d->attList.localnameList.append( "" );
+ d->attList.valueList.append( string() );
+ }
+ return TRUE;
+}
+
+/*!
+ Parse a content [43].
+
+ A content is only used between tags. If a end tag is found the < is already
+ read and the head stand on the '/' of the end tag '</name>'.
+*/
+bool QXmlSimpleReader::parseContent()
+{
+ bool charDataRead = FALSE;
+
+ const signed char Init = 0;
+ const signed char ChD = 1; // CharData
+ const signed char ChD1 = 2; // CharData help state
+ const signed char ChD2 = 3; // CharData help state
+ const signed char Ref = 4; // Reference
+ const signed char Lt = 5; // '<' read
+ const signed char PI = 6; // PI
+ const signed char Elem = 7; // Element
+ const signed char Em = 8; // '!' read
+ const signed char Com = 9; // Comment
+ const signed char CDS = 10; // CDSect
+ const signed char CDS1 = 11; // read a CDSect
+ const signed char CDS2 = 12; // read a CDSect (help state)
+ const signed char CDS3 = 13; // read a CDSect (help state)
+ const signed char Done = 14; // finished reading content
+
+ const signed char InpLt = 0; // <
+ const signed char InpGt = 1; // >
+ const signed char InpSlash = 2; // /
+ const signed char InpQMark = 3; // ?
+ const signed char InpEMark = 4; // !
+ const signed char InpAmp = 5; // &
+ const signed char InpDash = 6; // -
+ const signed char InpOpenB = 7; // [
+ const signed char InpCloseB = 8; // ]
+ const signed char InpUnknown = 9;
+
+ static signed char mapCLT2FSMChar[] = {
+ InpUnknown, // white space
+ InpUnknown, // %
+ InpAmp, // &
+ InpGt, // >
+ InpLt, // <
+ InpSlash, // /
+ InpQMark, // ?
+ InpEMark, // !
+ InpDash, // -
+ InpCloseB, // ]
+ InpOpenB, // [
+ InpUnknown, // =
+ InpUnknown, // "
+ InpUnknown, // '
+ InpUnknown // unknown
+ };
+
+ // use some kind of state machine for parsing
+ static signed char const table[14][10] = {
+ /* InpLt InpGt InpSlash InpQMark InpEMark InpAmp InpDash InpOpenB InpCloseB InpUnknown */
+ { Lt, ChD, ChD, ChD, ChD, Ref, ChD, ChD, ChD1, ChD }, // Init
+ { Lt, ChD, ChD, ChD, ChD, Ref, ChD, ChD, ChD1, ChD }, // ChD
+ { Lt, ChD, ChD, ChD, ChD, Ref, ChD, ChD, ChD2, ChD }, // ChD1
+ { Lt, -1, ChD, ChD, ChD, Ref, ChD, ChD, ChD2, ChD }, // ChD2
+ { Lt, ChD, ChD, ChD, ChD, Ref, ChD, ChD, ChD, ChD }, // Ref (same as Init)
+ { -1, -1, Done, PI, Em, -1, -1, -1, -1, Elem }, // Lt
+ { Lt, ChD, ChD, ChD, ChD, Ref, ChD, ChD, ChD, ChD }, // PI (same as Init)
+ { Lt, ChD, ChD, ChD, ChD, Ref, ChD, ChD, ChD, ChD }, // Elem (same as Init)
+ { -1, -1, -1, -1, -1, -1, Com, CDS, -1, -1 }, // Em
+ { Lt, ChD, ChD, ChD, ChD, Ref, ChD, ChD, ChD, ChD }, // Com (same as Init)
+ { CDS1, CDS1, CDS1, CDS1, CDS1, CDS1, CDS1, CDS1, CDS2, CDS1 }, // CDS
+ { CDS1, CDS1, CDS1, CDS1, CDS1, CDS1, CDS1, CDS1, CDS2, CDS1 }, // CDS1
+ { CDS1, CDS1, CDS1, CDS1, CDS1, CDS1, CDS1, CDS1, CDS3, CDS1 }, // CDS2
+ { CDS1, Init, CDS1, CDS1, CDS1, CDS1, CDS1, CDS1, CDS3, CDS1 } // CDS3
+ };
+ signed char state = Init;
+ signed char input;
+ bool parseOk = TRUE;
+
+ while ( TRUE ) {
+
+ // get input (use lookup-table instead of nested ifs for performance
+ // reasons)
+ if ( atEnd() ) {
+ d->error = XMLERR_UNEXPECTEDEOF;
+ goto parseError;
+ }
+ if ( c.row() ) {
+ input = InpUnknown;
+ } else {
+ input = mapCLT2FSMChar[ charLookupTable[ c.cell() ] ];
+ }
+
+ // set state according to input
+ state = table[state][input];
+
+ // do some actions according to state
+ switch ( state ) {
+ case Init:
+ // next character
+ next();
+ break;
+ case ChD:
+ // on first call: clear string
+ if ( !charDataRead ) {
+ charDataRead = TRUE;
+ stringClear();
+ }
+ stringAddC();
+ next();
+ break;
+ case ChD1:
+ // on first call: clear string
+ if ( !charDataRead ) {
+ charDataRead = TRUE;
+ stringClear();
+ }
+ stringAddC();
+ next();
+ break;
+ case ChD2:
+ stringAddC();
+ next();
+ break;
+ case Ref:
+ // reference may be CharData; so clear string to be safe
+ if ( !charDataRead) {
+ charDataRead = TRUE;
+ stringClear();
+ }
+ // parse reference
+ parseOk = parseReference( charDataRead, InContent );
+ break;
+ case Lt:
+ // call the handler for CharData
+ if ( contentHnd ) {
+ if ( charDataRead ) {
+ if ( d->reportWhitespaceCharData || !string().simplifyWhiteSpace().isEmpty() ) {
+ if ( !contentHnd->characters( string() ) ) {
+ d->error = contentHnd->errorString();
+ goto parseError;
+ }
+ }
+ }
+ }
+ charDataRead = FALSE;
+ // next character
+ next();
+ break;
+ case PI:
+ parseOk = parsePI();
+ break;
+ case Elem:
+ parseOk = parseElement();
+ break;
+ case Em:
+ // next character
+ next();
+ break;
+ case Com:
+ parseOk = parseComment();
+ break;
+ case CDS:
+ parseOk = parseString( "[CDATA[" );
+ break;
+ case CDS1:
+ // read one character and add it
+ stringAddC();
+ next();
+ break;
+ case CDS2:
+ // skip ']'
+ next();
+ break;
+ case CDS3:
+ // skip ']'...
+ next();
+ break;
+ }
+ // no input is read after this
+ switch ( state ) {
+ case Ref:
+ if ( !parseOk ) {
+ d->error = XMLERR_ERRORPARSINGREFERENCE;
+ goto parseError;
+ }
+ break;
+ case PI:
+ if ( !parseOk ) {
+ d->error = XMLERR_ERRORPARSINGPI;
+ goto parseError;
+ }
+ // call the handler
+ if ( contentHnd ) {
+ if ( !contentHnd->processingInstruction(name(),string()) ) {
+ d->error = contentHnd->errorString();
+ goto parseError;
+ }
+ }
+ break;
+ case Elem:
+ if ( !parseOk ) {
+ d->error = XMLERR_ERRORPARSINGELEMENT;
+ goto parseError;
+ }
+ break;
+ case Com:
+ if ( !parseOk ) {
+ d->error = XMLERR_ERRORPARSINGCOMMENT;
+ goto parseError;
+ }
+ if ( lexicalHnd ) {
+ if ( !lexicalHnd->comment( string() ) ) {
+ d->error = lexicalHnd->errorString();
+ goto parseError;
+ }
+ }
+ break;
+ case CDS:
+ if( !parseOk ) {
+ d->error = XMLERR_CDSECTHEADEREXPECTED;
+ goto parseError;
+ }
+ // empty string
+ stringClear();
+ break;
+ case CDS2:
+ if (c != ']') {
+ stringAddC( ']' );
+ }
+ break;
+ case CDS3:
+ // test if this skipping was legal
+ if ( c == '>' ) {
+ // the end of the CDSect
+ if ( lexicalHnd ) {
+ if ( !lexicalHnd->startCDATA() ) {
+ d->error = lexicalHnd->errorString();
+ goto parseError;
+ }
+ }
+ if ( contentHnd ) {
+ if ( !contentHnd->characters( string() ) ) {
+ d->error = contentHnd->errorString();
+ goto parseError;
+ }
+ }
+ if ( lexicalHnd ) {
+ if ( !lexicalHnd->endCDATA() ) {
+ d->error = lexicalHnd->errorString();
+ goto parseError;
+ }
+ }
+ } else if (c == ']') {
+ // three or more ']'
+ stringAddC( ']' );
+ } else {
+ // after ']]' comes another character
+ stringAddC( ']' );
+ stringAddC( ']' );
+ }
+ break;
+ case Done:
+ // call the handler for CharData
+ if ( contentHnd ) {
+ if ( charDataRead ) {
+ if ( d->reportWhitespaceCharData || !string().simplifyWhiteSpace().isEmpty() ) {
+ if ( !contentHnd->characters( string() ) ) {
+ d->error = contentHnd->errorString();
+ goto parseError;
+ }
+ }
+ }
+ }
+ // Done
+ return TRUE;
+ case -1:
+ // Error
+ d->error = XMLERR_ERRORPARSINGCONTENT;
+ goto parseError;
+ }
+
+ }
+
+ return TRUE;
+
+parseError:
+ reportParseError();
+ return FALSE;
+}
+
+/*!
+ Parse Misc [27].
+*/
+bool QXmlSimpleReader::parseMisc()
+{
+ const signed char Init = 0;
+ const signed char Lt = 1; // '<' was read
+ const signed char Comment = 2; // read comment
+ const signed char eatWS = 3; // eat whitespaces
+ const signed char PI = 4; // read PI
+ const signed char Comment2 = 5; // read comment
+
+ const signed char InpWs = 0; // S
+ const signed char InpLt = 1; // <
+ const signed char InpQm = 2; // ?
+ const signed char InpEm = 3; // !
+ const signed char InpUnknown = 4;
+
+ // use some kind of state machine for parsing
+ static signed char table[3][5] = {
+ /* InpWs InpLt InpQm InpEm InpUnknown */
+ { eatWS, Lt, -1, -1, -1 }, // Init
+ { -1, -1, PI, Comment, -1 }, // Lt
+ { -1, -1, -1, -1, Comment2 } // Comment
+ };
+ signed char state = Init;
+ signed char input;
+ bool parseOk = TRUE;
+
+ while ( TRUE ) {
+
+ // get input
+ if ( atEnd() ) {
+ d->error = XMLERR_UNEXPECTEDEOF;
+ goto parseError;
+ }
+ if ( is_S(c) ) {
+ input = InpWs;
+ } else if ( c == '<' ) {
+ input = InpLt;
+ } else if ( c == '?' ) {
+ input = InpQm;
+ } else if ( c == '!' ) {
+ input = InpEm;
+ } else {
+ input = InpUnknown;
+ }
+
+ // set state according to input
+ state = table[state][input];
+
+ // do some actions according to state
+ switch ( state ) {
+ case eatWS:
+ eat_ws();
+ break;
+ case Lt:
+ next();
+ break;
+ case PI:
+ parseOk = parsePI();
+ break;
+ case Comment:
+ next();
+ break;
+ case Comment2:
+ parseOk = parseComment();
+ break;
+ }
+ // no input is read after this
+ switch ( state ) {
+ case eatWS:
+ return TRUE;
+ case PI:
+ if ( !parseOk ) {
+ d->error = XMLERR_ERRORPARSINGPI;
+ goto parseError;
+ }
+ return TRUE;
+ case Comment2:
+ if ( !parseOk ) {
+ d->error = XMLERR_ERRORPARSINGCOMMENT;
+ goto parseError;
+ }
+ if ( lexicalHnd ) {
+ if ( !lexicalHnd->comment( string() ) ) {
+ d->error = lexicalHnd->errorString();
+ goto parseError;
+ }
+ }
+ return TRUE;
+ case -1:
+ // Error
+ d->error = XMLERR_UNEXPECTEDCHARACTER;
+ goto parseError;
+ }
+
+ }
+
+ return TRUE;
+
+parseError:
+ reportParseError();
+ return FALSE;
+}
+
+/*!
+ Parse a processing instruction [16].
+
+ If xmldec is TRUE, it tries to parse a PI or a XML declaration [23].
+
+ Precondition: the beginning '<' of the PI is already read and the head stand
+ on the '?' of '<?'.
+
+ If this funktion was successful, the head-position is on the first
+ character after the PI.
+*/
+bool QXmlSimpleReader::parsePI( bool xmldecl )
+{
+ const signed char Init = 0;
+ const signed char QmI = 1; // ? was read
+ const signed char Name = 2; // read Name
+ const signed char XMLDecl = 3; // read XMLDecl
+ const signed char Ws1 = 4; // eat ws after "xml" of XMLDecl
+ const signed char PI = 5; // read PI
+ const signed char Ws2 = 6; // eat ws after Name of PI
+ const signed char Version = 7; // read versionInfo
+ const signed char Ws3 = 8; // eat ws after versionInfo
+ const signed char EorSD = 9; // read EDecl or SDDecl
+ const signed char Ws4 = 10; // eat ws after EDecl or SDDecl
+ const signed char SD = 11; // read SDDecl
+ const signed char Ws5 = 12; // eat ws after SDDecl
+ const signed char ADone = 13; // almost done
+ const signed char Char = 14; // Char was read
+ const signed char Qm = 15; // Qm was read
+ const signed char Done = 16; // finished reading content
+
+ const signed char InpWs = 0; // whitespace
+ const signed char InpNameBe = 1; // is_nameBeginning()
+ const signed char InpGt = 2; // >
+ const signed char InpQm = 3; // ?
+ const signed char InpUnknown = 4;
+
+ // use some kind of state machine for parsing
+ static signed char table[16][5] = {
+ /* InpWs, InpNameBe InpGt InpQm InpUnknown */
+ { -1, -1, -1, QmI, -1 }, // Init
+ { -1, Name, -1, -1, -1 }, // QmI
+ { -1, -1, -1, -1, -1 }, // Name (this state is left not through input)
+ { Ws1, -1, -1, -1, -1 }, // XMLDecl
+ { -1, Version, -1, -1, -1 }, // Ws1
+ { Ws2, -1, -1, Qm, -1 }, // PI
+ { Char, Char, Char, Qm, Char }, // Ws2
+ { Ws3, -1, -1, ADone, -1 }, // Version
+ { -1, EorSD, -1, ADone, -1 }, // Ws3
+ { Ws4, -1, -1, ADone, -1 }, // EorSD
+ { -1, SD, -1, ADone, -1 }, // Ws4
+ { Ws5, -1, -1, ADone, -1 }, // SD
+ { -1, -1, -1, ADone, -1 }, // Ws5
+ { -1, -1, Done, -1, -1 }, // ADone
+ { Char, Char, Char, Qm, Char }, // Char
+ { Char, Char, Done, Qm, Char }, // Qm
+ };
+ signed char state = Init;
+ signed char input;
+ bool parseOk = TRUE;
+
+ while ( TRUE ) {
+
+ // get input
+ if ( atEnd() ) {
+ d->error = XMLERR_UNEXPECTEDEOF;
+ goto parseError;
+ }
+ if ( is_S(c) ) {
+ input = InpWs;
+ } else if ( is_NameBeginning(c) ) {
+ input = InpNameBe;
+ } else if ( c == '>' ) {
+ input = InpGt;
+ } else if ( c == '?' ) {
+ input = InpQm;
+ } else {
+ input = InpUnknown;
+ }
+
+ // set state according to input
+ state = table[state][input];
+
+ // do some actions according to state
+ switch ( state ) {
+ case QmI:
+ next();
+ break;
+ case Name:
+ parseOk = parseName();
+ break;
+ case Ws1:
+ case Ws2:
+ case Ws3:
+ case Ws4:
+ case Ws5:
+ eat_ws();
+ break;
+ case Version:
+ parseOk = parseAttribute();
+ break;
+ case EorSD:
+ parseOk = parseAttribute();
+ break;
+ case SD:
+ // get the SDDecl (syntax like an attribute)
+ if ( d->standalone != QXmlSimpleReaderPrivate::Unknown ) {
+ // already parsed the standalone declaration
+ d->error = XMLERR_UNEXPECTEDCHARACTER;
+ goto parseError;
+ }
+ parseOk = parseAttribute();
+ break;
+ case ADone:
+ next();
+ break;
+ case Char:
+ stringAddC();
+ next();
+ break;
+ case Qm:
+ // skip the '?'
+ next();
+ break;
+ case Done:
+ next();
+ break;
+ }
+ // no input is read after this
+ switch ( state ) {
+ case Name:
+ if ( !parseOk ) {
+ d->error = XMLERR_ERRORPARSINGNAME;
+ goto parseError;
+ }
+ // test what name was read and determine the next state
+ // (not very beautiful, I admit)
+ if ( name().lower() == "xml" ) {
+ if ( xmldecl && name()=="xml" ) {
+ state = XMLDecl;
+ } else {
+ d->error = XMLERR_INVALIDNAMEFORPI;
+ goto parseError;
+ }
+ } else {
+ state = PI;
+ stringClear();
+ }
+ break;
+ case Version:
+ // get version (syntax like an attribute)
+ if ( !parseOk ) {
+ d->error = XMLERR_VERSIONEXPECTED;
+ goto parseError;
+ }
+ if ( name() != "version" ) {
+ d->error = XMLERR_VERSIONEXPECTED;
+ goto parseError;
+ }
+ d->xmlVersion = string();
+ break;
+ case EorSD:
+ // get the EDecl or SDDecl (syntax like an attribute)
+ if ( !parseOk ) {
+ d->error = XMLERR_EDECLORSDDECLEXPECTED;
+ goto parseError;
+ }
+ if ( name() == "standalone" ) {
+ if ( string()=="yes" ) {
+ d->standalone = QXmlSimpleReaderPrivate::Yes;
+ } else if ( string()=="no" ) {
+ d->standalone = QXmlSimpleReaderPrivate::No;
+ } else {
+ d->error = XMLERR_WRONGVALUEFORSDECL;
+ goto parseError;
+ }
+ } else if ( name() == "encoding" ) {
+ d->encoding = string();
+ } else {
+ d->error = XMLERR_EDECLORSDDECLEXPECTED;
+ goto parseError;
+ }
+ break;
+ case SD:
+ if ( !parseOk ) {
+ d->error = XMLERR_SDDECLEXPECTED;
+ goto parseError;
+ }
+ if ( name() != "standalone" ) {
+ d->error = XMLERR_SDDECLEXPECTED;
+ goto parseError;
+ }
+ if ( string()=="yes" ) {
+ d->standalone = QXmlSimpleReaderPrivate::Yes;
+ } else if ( string()=="no" ) {
+ d->standalone = QXmlSimpleReaderPrivate::No;
+ } else {
+ d->error = XMLERR_WRONGVALUEFORSDECL;
+ goto parseError;
+ }
+ break;
+ case Qm:
+ // test if the skipping was legal
+ if ( c != '>' ) {
+ stringAddC( '?' );
+ }
+ break;
+ case Done:
+ return TRUE;
+ case -1:
+ // Error
+ d->error = XMLERR_UNEXPECTEDCHARACTER;
+ goto parseError;
+ }
+
+ }
+
+ return TRUE;
+
+parseError:
+ reportParseError();
+ return FALSE;
+}
+
+/*!
+ Parse a document type definition (doctypedecl [28]).
+
+ Precondition: the beginning '<!' of the doctype is already read the head
+ stands on the 'D' of '<!DOCTYPE'.
+
+ If this funktion was successful, the head-position is on the first
+ character after the document type definition.
+*/
+bool QXmlSimpleReader::parseDoctype()
+{
+ // some init-stuff
+ d->systemId = "";
+ d->publicId = "";
+
+ const signed char Init = 0;
+ const signed char Doctype = 1; // read the doctype
+ const signed char Ws1 = 2; // eat_ws
+ const signed char Doctype2 = 3; // read the doctype, part 2
+ const signed char Ws2 = 4; // eat_ws
+ const signed char Sys = 5; // read SYSTEM
+ const signed char Ws3 = 6; // eat_ws
+ const signed char MP = 7; // markupdecl or PEReference
+ const signed char PER = 8; // PERReference
+ const signed char Mup = 9; // markupdecl
+ const signed char Ws4 = 10; // eat_ws
+ const signed char MPE = 11; // end of markupdecl or PEReference
+ const signed char Done = 12;
+
+ const signed char InpWs = 0;
+ const signed char InpD = 1; // 'D'
+ const signed char InpS = 2; // 'S' or 'P'
+ const signed char InpOB = 3; // [
+ const signed char InpCB = 4; // ]
+ const signed char InpPer = 5; // %
+ const signed char InpGt = 6; // >
+ const signed char InpUnknown = 7;
+
+ // use some kind of state machine for parsing
+ static signed char table[12][8] = {
+ /* InpWs, InpD InpS InpOB InpCB InpPer InpGt InpUnknown */
+ { -1, Doctype, -1, -1, -1, -1, -1, -1 }, // Init
+ { Ws1, Doctype2, Doctype2, -1, -1, -1, -1, Doctype2 }, // Doctype
+ { -1, Doctype2, Doctype2, -1, -1, -1, -1, Doctype2 }, // Ws1
+ { Ws2, -1, Sys, MP, -1, -1, Done, -1 }, // Doctype2
+ { -1, -1, Sys, MP, -1, -1, Done, -1 }, // Ws2
+ { Ws3, -1, -1, MP, -1, -1, Done, -1 }, // Sys
+ { -1, -1, -1, MP, -1, -1, Done, -1 }, // Ws3
+ { -1, -1, -1, -1, MPE, PER, -1, Mup }, // MP
+ { Ws4, -1, -1, -1, MPE, PER, -1, Mup }, // PER
+ { Ws4, -1, -1, -1, MPE, PER, -1, Mup }, // Mup
+ { -1, -1, -1, -1, MPE, PER, -1, Mup }, // Ws4
+ { -1, -1, -1, -1, -1, -1, Done, -1 } // MPE
+ };
+ signed char state = Init;
+ signed char input;
+ bool parseOk = TRUE;
+
+ while ( TRUE ) {
+
+ // get input
+ if ( atEnd() ) {
+ d->error = XMLERR_UNEXPECTEDEOF;
+ goto parseError;
+ }
+ if ( is_S(c) ) {
+ input = InpWs;
+ } else if ( c == 'D' ) {
+ input = InpD;
+ } else if ( c == 'S' ) {
+ input = InpS;
+ } else if ( c == 'P' ) {
+ input = InpS;
+ } else if ( c == '[' ) {
+ input = InpOB;
+ } else if ( c == ']' ) {
+ input = InpCB;
+ } else if ( c == '%' ) {
+ input = InpPer;
+ } else if ( c == '>' ) {
+ input = InpGt;
+ } else {
+ input = InpUnknown;
+ }
+
+ // set state according to input
+ state = table[state][input];
+
+ // do some actions according to state
+ switch ( state ) {
+ case Doctype:
+ parseOk = parseString( "DOCTYPE" );
+ break;
+ case Ws1:
+ case Ws2:
+ case Ws3:
+ case Ws4:
+ eat_ws();
+ break;
+ case Doctype2:
+ parseName();
+ break;
+ case Sys:
+ parseOk = parseExternalID();
+ break;
+ case MP:
+ next_eat_ws();
+ break;
+ case PER:
+ parseOk = parsePEReference( InDTD );
+ break;
+ case Mup:
+ parseOk = parseMarkupdecl();
+ break;
+ case MPE:
+ next_eat_ws();
+ break;
+ case Done:
+ if ( lexicalHnd ) {
+ if ( !lexicalHnd->endDTD() ) {
+ d->error = lexicalHnd->errorString();
+ goto parseError;
+ }
+ }
+ next();
+ break;
+ }
+ // no input is read after this
+ switch ( state ) {
+ case Doctype:
+ if ( !parseOk ) {
+ d->error = XMLERR_ERRORPARSINGDOCTYPE;
+ goto parseError;
+ }
+ if ( !is_S(c) ) {
+ d->error = XMLERR_ERRORPARSINGDOCTYPE;
+ goto parseError;
+ }
+ break;
+ case Doctype2:
+ d->doctype = name();
+ if ( lexicalHnd ) {
+ if ( !lexicalHnd->startDTD( d->doctype, d->publicId, d->systemId ) ) {
+ d->error = lexicalHnd->errorString();
+ goto parseError;
+ }
+ }
+ break;
+ case Sys:
+ if ( !parseOk ) {
+ d->error = XMLERR_ERRORPARSINGDOCTYPE;
+ goto parseError;
+ }
+ break;
+ case PER:
+ if ( !parseOk ) {
+ d->error = XMLERR_ERRORPARSINGDOCTYPE;
+ goto parseError;
+ }
+ break;
+ case Mup:
+ if ( !parseOk ) {
+ d->error = XMLERR_ERRORPARSINGDOCTYPE;
+ goto parseError;
+ }
+ break;
+ case Done:
+ return TRUE;
+ case -1:
+ // Error
+ d->error = XMLERR_ERRORPARSINGDOCTYPE;
+ goto parseError;
+ }
+
+ }
+
+ return TRUE;
+
+parseError:
+ reportParseError();
+ return FALSE;
+}
+
+/*!
+ Parse a ExternalID [75].
+
+ If allowPublicID is TRUE parse ExternalID [75] or PublicID [83].
+*/
+bool QXmlSimpleReader::parseExternalID( bool allowPublicID )
+{
+ // some init-stuff
+ d->systemId = "";
+ d->publicId = "";
+
+ const signed char Init = 0;
+ const signed char Sys = 1; // parse 'SYSTEM'
+ const signed char SysWS = 2; // parse the whitespace after 'SYSTEM'
+ const signed char SysSQ = 3; // parse SystemLiteral with '
+ const signed char SysSQ2 = 4; // parse SystemLiteral with '
+ const signed char SysDQ = 5; // parse SystemLiteral with "
+ const signed char SysDQ2 = 6; // parse SystemLiteral with "
+ const signed char Pub = 7; // parse 'PUBLIC'
+ const signed char PubWS = 8; // parse the whitespace after 'PUBLIC'
+ const signed char PubSQ = 9; // parse PubidLiteral with '
+ const signed char PubSQ2 = 10; // parse PubidLiteral with '
+ const signed char PubDQ = 11; // parse PubidLiteral with "
+ const signed char PubDQ2 = 12; // parse PubidLiteral with "
+ const signed char PubE = 13; // finished parsing the PubidLiteral
+ const signed char PubWS2 = 14; // parse the whitespace after the PubidLiteral
+ const signed char PDone = 15; // done if allowPublicID is TRUE
+ const signed char Done = 16;
+
+ const signed char InpSQ = 0; // '
+ const signed char InpDQ = 1; // "
+ const signed char InpS = 2; // S
+ const signed char InpP = 3; // P
+ const signed char InpWs = 4; // white space
+ const signed char InpUnknown = 5;
+
+ // use some kind of state machine for parsing
+ static signed char table[15][6] = {
+ /* InpSQ InpDQ InpS InpP InpWs InpUnknown */
+ { -1, -1, Sys, Pub, -1, -1 }, // Init
+ { -1, -1, -1, -1, SysWS, -1 }, // Sys
+ { SysSQ, SysDQ, -1, -1, -1, -1 }, // SysWS
+ { Done, SysSQ2, SysSQ2, SysSQ2, SysSQ2, SysSQ2 }, // SysSQ
+ { Done, SysSQ2, SysSQ2, SysSQ2, SysSQ2, SysSQ2 }, // SysSQ2
+ { SysDQ2, Done, SysDQ2, SysDQ2, SysDQ2, SysDQ2 }, // SysDQ
+ { SysDQ2, Done, SysDQ2, SysDQ2, SysDQ2, SysDQ2 }, // SysDQ2
+ { -1, -1, -1, -1, PubWS, -1 }, // Pub
+ { PubSQ, PubDQ, -1, -1, -1, -1 }, // PubWS
+ { PubE, -1, PubSQ2, PubSQ2, PubSQ2, PubSQ2 }, // PubSQ
+ { PubE, -1, PubSQ2, PubSQ2, PubSQ2, PubSQ2 }, // PubSQ2
+ { -1, PubE, PubDQ2, PubDQ2, PubDQ2, PubDQ2 }, // PubDQ
+ { -1, PubE, PubDQ2, PubDQ2, PubDQ2, PubDQ2 }, // PubDQ2
+ { PDone, PDone, PDone, PDone, PubWS2, PDone }, // PubE
+ { SysSQ, SysDQ, PDone, PDone, PDone, PDone } // PubWS2
+ };
+ signed char state = Init;
+ signed char input;
+ bool parseOk = TRUE;
+
+ while ( TRUE ) {
+
+ // get input
+ if ( atEnd() ) {
+ d->error = XMLERR_UNEXPECTEDEOF;
+ goto parseError;
+ }
+ if ( is_S(c) ) {
+ input = InpWs;
+ } else if ( c == '\'' ) {
+ input = InpSQ;
+ } else if ( c == '"' ) {
+ input = InpDQ;
+ } else if ( c == 'S' ) {
+ input = InpS;
+ } else if ( c == 'P' ) {
+ input = InpP;
+ } else {
+ input = InpUnknown;
+ }
+
+ // set state according to input
+ state = table[state][input];
+
+ // do some actions according to state
+ switch ( state ) {
+ case Sys:
+ parseOk = parseString( "SYSTEM" );
+ break;
+ case SysWS:
+ eat_ws();
+ break;
+ case SysSQ:
+ case SysDQ:
+ stringClear();
+ next();
+ break;
+ case SysSQ2:
+ case SysDQ2:
+ stringAddC();
+ next();
+ break;
+ case Pub:
+ parseOk = parseString( "PUBLIC" );
+ break;
+ case PubWS:
+ eat_ws();
+ break;
+ case PubSQ:
+ case PubDQ:
+ stringClear();
+ next();
+ break;
+ case PubSQ2:
+ case PubDQ2:
+ stringAddC();
+ next();
+ break;
+ case PubE:
+ next();
+ break;
+ case PubWS2:
+ d->publicId = string();
+ eat_ws();
+ break;
+ case Done:
+ d->systemId = string();
+ next();
+ break;
+ }
+ // no input is read after this
+ switch ( state ) {
+ case Sys:
+ if( !parseOk ) {
+ d->error = XMLERR_UNEXPECTEDCHARACTER;
+ goto parseError;
+ }
+ break;
+ case Pub:
+ if( !parseOk ) {
+ d->error = XMLERR_UNEXPECTEDCHARACTER;
+ goto parseError;
+ }
+ break;
+ case PDone:
+ if ( allowPublicID ) {
+ d->publicId = string();
+ return TRUE;
+ } else {
+ d->error = XMLERR_UNEXPECTEDCHARACTER;
+ goto parseError;
+ }
+ break;
+ case Done:
+ return TRUE;
+ case -1:
+ // Error
+ d->error = XMLERR_UNEXPECTEDCHARACTER;
+ goto parseError;
+ }
+
+ }
+
+ return TRUE;
+
+parseError:
+ reportParseError();
+ return FALSE;
+}
+
+/*!
+ Parse a markupdecl [29].
+*/
+bool QXmlSimpleReader::parseMarkupdecl()
+{
+ const signed char Init = 0;
+ const signed char Lt = 1; // < was read
+ const signed char Em = 2; // ! was read
+ const signed char CE = 3; // E was read
+ const signed char Qm = 4; // ? was read
+ const signed char Dash = 5; // - was read
+ const signed char CA = 6; // A was read
+ const signed char CEL = 7; // EL was read
+ const signed char CEN = 8; // EN was read
+ const signed char CN = 9; // N was read
+ const signed char Done = 10;
+
+ const signed char InpLt = 0; // <
+ const signed char InpQm = 1; // ?
+ const signed char InpEm = 2; // !
+ const signed char InpDash = 3; // -
+ const signed char InpA = 4; // A
+ const signed char InpE = 5; // E
+ const signed char InpL = 6; // L
+ const signed char InpN = 7; // N
+ const signed char InpUnknown = 8;
+
+ // use some kind of state machine for parsing
+ static signed char table[4][9] = {
+ /* InpLt InpQm InpEm InpDash InpA InpE InpL InpN InpUnknown */
+ { Lt, -1, -1, -1, -1, -1, -1, -1, -1 }, // Init
+ { -1, Qm, Em, -1, -1, -1, -1, -1, -1 }, // Lt
+ { -1, -1, -1, Dash, CA, CE, -1, CN, -1 }, // Em
+ { -1, -1, -1, -1, -1, -1, CEL, CEN, -1 } // CE
+ };
+ signed char state = Init;
+ signed char input;
+ bool parseOk = TRUE;
+
+ while ( TRUE ) {
+
+ // get input
+ if ( atEnd() ) {
+ d->error = XMLERR_UNEXPECTEDEOF;
+ goto parseError;
+ }
+ if ( c == '<' ) {
+ input = InpLt;
+ } else if ( c == '?' ) {
+ input = InpQm;
+ } else if ( c == '!' ) {
+ input = InpEm;
+ } else if ( c == '-' ) {
+ input = InpDash;
+ } else if ( c == 'A' ) {
+ input = InpA;
+ } else if ( c == 'E' ) {
+ input = InpE;
+ } else if ( c == 'L' ) {
+ input = InpL;
+ } else if ( c == 'N' ) {
+ input = InpN;
+ } else {
+ input = InpUnknown;
+ }
+
+ // set state according to input
+ state = table[state][input];
+
+ // do some actions according to state
+ switch ( state ) {
+ case Lt:
+ next();
+ break;
+ case Em:
+ next();
+ break;
+ case CE:
+ next();
+ break;
+ case Qm:
+ parseOk = parsePI();
+ break;
+ case Dash:
+ parseOk = parseComment();
+ break;
+ case CA:
+ parseOk = parseAttlistDecl();
+ break;
+ case CEL:
+ parseOk = parseElementDecl();
+ break;
+ case CEN:
+ parseOk = parseEntityDecl();
+ break;
+ case CN:
+ parseOk = parseNotationDecl();
+ break;
+ }
+ // no input is read after this
+ switch ( state ) {
+ case Qm:
+ if ( !parseOk ) {
+ d->error = XMLERR_ERRORPARSINGPI;
+ goto parseError;
+ }
+ return TRUE;
+ case Dash:
+ if ( !parseOk ) {
+ d->error = XMLERR_ERRORPARSINGCOMMENT;
+ goto parseError;
+ }
+ if ( lexicalHnd ) {
+ if ( !lexicalHnd->comment( string() ) ) {
+ d->error = lexicalHnd->errorString();
+ goto parseError;
+ }
+ }
+ return TRUE;
+ case CA:
+ if ( !parseOk ) {
+ d->error = XMLERR_ERRORPARSINGATTLISTDECL;
+ goto parseError;
+ }
+ return TRUE;
+ case CEL:
+ if ( !parseOk ) {
+ d->error = XMLERR_ERRORPARSINGELEMENTDECL;
+ goto parseError;
+ }
+ return TRUE;
+ case CEN:
+ if ( !parseOk ) {
+ d->error = XMLERR_ERRORPARSINGENTITYDECL;
+ goto parseError;
+ }
+ return TRUE;
+ case CN:
+ if ( !parseOk ) {
+ d->error = XMLERR_ERRORPARSINGNOTATIONDECL;
+ goto parseError;
+ }
+ return TRUE;
+ case Done:
+ return TRUE;
+ case -1:
+ // Error
+ d->error = XMLERR_LETTEREXPECTED;
+ goto parseError;
+ }
+
+ }
+
+ return TRUE;
+
+parseError:
+ reportParseError();
+ return FALSE;
+}
+
+/*!
+ Parse a PEReference [69]
+*/
+bool QXmlSimpleReader::parsePEReference( EntityRecognitionContext context )
+{
+ const signed char Init = 0;
+ const signed char Next = 1;
+ const signed char Name = 2;
+ const signed char Done = 3;
+
+ const signed char InpSemi = 0; // ;
+ const signed char InpPer = 1; // %
+ const signed char InpUnknown = 2;
+
+ // use some kind of state machine for parsing
+ static signed char table[3][3] = {
+ /* InpSemi InpPer InpUnknown */
+ { -1, Next, -1 }, // Init
+ { -1, -1, Name }, // Next
+ { Done, -1, -1 } // Name
+ };
+ signed char state = Init;
+ signed char input;
+ bool parseOk = TRUE;
+
+ while ( TRUE ) {
+
+ // get input
+ if ( atEnd() ) {
+ d->error = XMLERR_UNEXPECTEDEOF;
+ goto parseError;
+ }
+ if ( c == ';' ) {
+ input = InpSemi;
+ } else if ( c == '%' ) {
+ input = InpPer;
+ } else {
+ input = InpUnknown;
+ }
+
+ // set state according to input
+ state = table[state][input];
+
+ // do some actions according to state
+ switch ( state ) {
+ case Next:
+ next();
+ break;
+ case Name:
+ parseOk = parseName( TRUE );
+ break;
+ case Done:
+ next();
+ break;
+ }
+ // no input is read after this
+ switch ( state ) {
+ case Name:
+ if ( !parseOk ) {
+ d->error = XMLERR_ERRORPARSINGNAME;
+ goto parseError;
+ }
+ if ( d->parameterEntities.find( ref() ) == d->parameterEntities.end() ) {
+ // ### skip it???
+ if ( contentHnd ) {
+ if ( !contentHnd->skippedEntity( QString("%") + ref() ) ) {
+ d->error = contentHnd->errorString();
+ goto parseError;
+ }
+ }
+ } else {
+ if ( context == InEntityValue ) {
+ // Included in literal
+ xmlRef = d->parameterEntities.find( ref() )
+ .data().replace( QRegExp("\""), "&quot;" ).replace( QRegExp("'"), "&apos;" )
+ + xmlRef;
+ } else if ( context == InDTD ) {
+ // Included as PE ### correct???
+ xmlRef = QString(" ") +
+ d->parameterEntities.find( ref() ).data() +
+ QString(" ") + xmlRef;
+ }
+ }
+ break;
+ case Done:
+ return TRUE;
+ case -1:
+ // Error
+ d->error = XMLERR_LETTEREXPECTED;
+ goto parseError;
+ }
+
+ }
+
+ return TRUE;
+
+parseError:
+ reportParseError();
+ return FALSE;
+}
+
+/*!
+ Parse a AttlistDecl [52].
+
+ Precondition: the beginning '<!' is already read and the head
+ stands on the 'A' of '<!ATTLIST'
+*/
+bool QXmlSimpleReader::parseAttlistDecl()
+{
+ const signed char Init = 0;
+ const signed char Attlist = 1; // parse the string "ATTLIST"
+ const signed char Ws = 2; // whitespace read
+ const signed char Name = 3; // parse name
+ const signed char Ws1 = 4; // whitespace read
+ const signed char Attdef = 5; // parse the AttDef
+ const signed char Ws2 = 6; // whitespace read
+ const signed char Atttype = 7; // parse the AttType
+ const signed char Ws3 = 8; // whitespace read
+ const signed char DDecH = 9; // DefaultDecl with #
+ const signed char DefReq = 10; // parse the string "REQUIRED"
+ const signed char DefImp = 11; // parse the string "IMPLIED"
+ const signed char DefFix = 12; // parse the string "FIXED"
+ const signed char Attval = 13; // parse the AttValue
+ const signed char Ws4 = 14; // whitespace read
+ const signed char Done = 15;
+
+ const signed char InpWs = 0; // white space
+ const signed char InpGt = 1; // >
+ const signed char InpHash = 2; // #
+ const signed char InpA = 3; // A
+ const signed char InpI = 4; // I
+ const signed char InpF = 5; // F
+ const signed char InpR = 6; // R
+ const signed char InpUnknown = 7;
+
+ // use some kind of state machine for parsing
+ static signed char table[15][8] = {
+ /* InpWs InpGt InpHash InpA InpI InpF InpR InpUnknown */
+ { -1, -1, -1, Attlist, -1, -1, -1, -1 }, // Init
+ { Ws, -1, -1, -1, -1, -1, -1, -1 }, // Attlist
+ { -1, -1, -1, Name, Name, Name, Name, Name }, // Ws
+ { Ws1, Done, Attdef, Attdef, Attdef, Attdef, Attdef, Attdef }, // Name
+ { -1, Done, Attdef, Attdef, Attdef, Attdef, Attdef, Attdef }, // Ws1
+ { Ws2, -1, -1, -1, -1, -1, -1, -1 }, // Attdef
+ { -1, Atttype, Atttype, Atttype, Atttype, Atttype, Atttype, Atttype }, // Ws2
+ { Ws3, -1, -1, -1, -1, -1, -1, -1 }, // Attype
+ { -1, Attval, DDecH, Attval, Attval, Attval, Attval, Attval }, // Ws3
+ { -1, -1, -1, -1, DefImp, DefFix, DefReq, -1 }, // DDecH
+ { Ws4, Ws4, -1, -1, -1, -1, -1, -1 }, // DefReq
+ { Ws4, Ws4, -1, -1, -1, -1, -1, -1 }, // DefImp
+ { Ws3, -1, -1, -1, -1, -1, -1, -1 }, // DefFix
+ { Ws4, Ws4, -1, -1, -1, -1, -1, -1 }, // Attval
+ { -1, Done, Attdef, Attdef, Attdef, Attdef, Attdef, Attdef } // Ws4
+ };
+ signed char state = Init;
+ signed char input;
+ bool parseOk = TRUE;
+
+ while ( TRUE ) {
+
+ // get input
+ if ( atEnd() ) {
+ d->error = XMLERR_UNEXPECTEDEOF;
+ goto parseError;
+ }
+ if ( is_S(c) ) {
+ input = InpWs;
+ } else if ( c == '>' ) {
+ input = InpGt;
+ } else if ( c == '#' ) {
+ input = InpHash;
+ } else if ( c == 'A' ) {
+ input = InpA;
+ } else if ( c == 'I' ) {
+ input = InpI;
+ } else if ( c == 'F' ) {
+ input = InpF;
+ } else if ( c == 'R' ) {
+ input = InpR;
+ } else {
+ input = InpUnknown;
+ }
+
+ // set state according to input
+ state = table[state][input];
+
+ // do some actions according to state
+ switch ( state ) {
+ case Attlist:
+ parseOk = parseString( "ATTLIST" );
+ break;
+ case Ws:
+ case Ws1:
+ case Ws2:
+ case Ws3:
+ eat_ws();
+ break;
+ case Name:
+ parseOk = parseName();
+ break;
+ case Attdef:
+ parseOk = parseName();
+ break;
+ case Atttype:
+ parseOk = parseAttType();
+ break;
+ case DDecH:
+ next();
+ break;
+ case DefReq:
+ parseOk = parseString( "REQUIRED" );
+ break;
+ case DefImp:
+ parseOk = parseString( "IMPLIED" );
+ break;
+ case DefFix:
+ parseOk = parseString( "FIXED" );
+ break;
+ case Attval:
+ parseOk = parseAttValue();
+ break;
+ case Ws4:
+ if ( declHnd ) {
+ // TODO: not all values are computed yet...
+ if ( !declHnd->attributeDecl( d->attDeclEName, d->attDeclAName, "", "", "" ) ) {
+ d->error = declHnd->errorString();
+ goto parseError;
+ }
+ }
+ eat_ws();
+ break;
+ case Done:
+ next();
+ break;
+ }
+ // no input is read after this
+ switch ( state ) {
+ case Attlist:
+ if( !parseOk ) {
+ d->error = XMLERR_UNEXPECTEDCHARACTER;
+ goto parseError;
+ }
+ break;
+ case Name:
+ if ( !parseOk ) {
+ d->error = XMLERR_ERRORPARSINGNAME;
+ goto parseError;
+ }
+ d->attDeclEName = name();
+ break;
+ case Attdef:
+ if ( !parseOk ) {
+ d->error = XMLERR_ERRORPARSINGNAME;
+ goto parseError;
+ }
+ d->attDeclAName = name();
+ break;
+ case Atttype:
+ if ( !parseOk ) {
+ d->error = XMLERR_ERRORPARSINGATTTYPE;
+ goto parseError;
+ }
+ break;
+ case DefReq:
+ if( !parseOk ) {
+ d->error = XMLERR_UNEXPECTEDCHARACTER;
+ goto parseError;
+ }
+ break;
+ case DefImp:
+ if( !parseOk ) {
+ d->error = XMLERR_UNEXPECTEDCHARACTER;
+ goto parseError;
+ }
+ break;
+ case DefFix:
+ if( !parseOk ) {
+ d->error = XMLERR_UNEXPECTEDCHARACTER;
+ goto parseError;
+ }
+ break;
+ case Attval:
+ if ( !parseOk ) {
+ d->error = XMLERR_ERRORPARSINGATTVALUE;
+ goto parseError;
+ }
+ break;
+ case Done:
+ return TRUE;
+ case -1:
+ // Error
+ d->error = XMLERR_LETTEREXPECTED;
+ goto parseError;
+ }
+
+ }
+
+ return TRUE;
+
+parseError:
+ reportParseError();
+ return FALSE;
+}
+
+/*!
+ Parse a AttType [54]
+*/
+bool QXmlSimpleReader::parseAttType()
+{
+ const signed char Init = 0;
+ const signed char ST = 1; // StringType
+ const signed char TTI = 2; // TokenizedType starting with 'I'
+ const signed char TTI2 = 3; // TokenizedType helpstate
+ const signed char TTI3 = 4; // TokenizedType helpstate
+ const signed char TTE = 5; // TokenizedType starting with 'E'
+ const signed char TTEY = 6; // TokenizedType starting with 'ENTITY'
+ const signed char TTEI = 7; // TokenizedType starting with 'ENTITI'
+ const signed char N = 8; // N read (TokenizedType or Notation)
+ const signed char TTNM = 9; // TokenizedType starting with 'NM'
+ const signed char TTNM2 = 10; // TokenizedType helpstate
+ const signed char NO = 11; // Notation
+ const signed char NO2 = 12; // Notation helpstate
+ const signed char NO3 = 13; // Notation helpstate
+ const signed char NOName = 14; // Notation, read name
+ const signed char NO4 = 15; // Notation helpstate
+ const signed char EN = 16; // Enumeration
+ const signed char ENNmt = 17; // Enumeration, read Nmtoken
+ const signed char EN2 = 18; // Enumeration helpstate
+ const signed char ADone = 19; // almost done (make next and accept)
+ const signed char Done = 20;
+
+ const signed char InpWs = 0; // whitespace
+ const signed char InpOp = 1; // (
+ const signed char InpCp = 2; // )
+ const signed char InpPipe = 3; // |
+ const signed char InpC = 4; // C
+ const signed char InpE = 5; // E
+ const signed char InpI = 6; // I
+ const signed char InpM = 7; // M
+ const signed char InpN = 8; // N
+ const signed char InpO = 9; // O
+ const signed char InpR = 10; // R
+ const signed char InpS = 11; // S
+ const signed char InpY = 12; // Y
+ const signed char InpUnknown = 13;
+
+ // use some kind of state machine for parsing
+ static signed char table[19][14] = {
+ /* InpWs InpOp InpCp InpPipe InpC InpE InpI InpM InpN InpO InpR InpS InpY InpUnknown */
+ { -1, EN, -1, -1, ST, TTE, TTI, -1, N, -1, -1, -1, -1, -1 }, // Init
+ { Done, Done, Done, Done, Done, Done, Done, Done, Done, Done, Done, Done, Done, Done }, // ST
+ { Done, Done, Done, Done, Done, Done, Done, Done, Done, Done, TTI2, Done, Done, Done }, // TTI
+ { Done, Done, Done, Done, Done, Done, Done, Done, Done, Done, Done, TTI3, Done, Done }, // TTI2
+ { Done, Done, Done, Done, Done, Done, Done, Done, Done, Done, Done, Done, Done, Done }, // TTI3
+ { -1, -1, -1, -1, -1, -1, TTEI, -1, -1, -1, -1, -1, TTEY, -1 }, // TTE
+ { Done, Done, Done, Done, Done, Done, Done, Done, Done, Done, Done, Done, Done, Done }, // TTEY
+ { Done, Done, Done, Done, Done, Done, Done, Done, Done, Done, Done, Done, Done, Done }, // TTEI
+ { -1, -1, -1, -1, -1, -1, -1, TTNM, -1, NO, -1, -1, -1, -1 }, // N
+ { Done, Done, Done, Done, Done, Done, Done, Done, Done, Done, Done, TTNM2, Done, Done }, // TTNM
+ { Done, Done, Done, Done, Done, Done, Done, Done, Done, Done, Done, Done, Done, Done }, // TTNM2
+ { NO2, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1 }, // NO
+ { -1, NO3, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1 }, // NO2
+ { NOName, NOName, NOName, NOName, NOName, NOName, NOName, NOName, NOName, NOName, NOName, NOName, NOName, NOName }, // NO3
+ { NO4, -1, ADone, NO3, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1 }, // NOName
+ { -1, -1, ADone, NO3, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1 }, // NO4
+ { -1, -1, ENNmt, -1, ENNmt, ENNmt, ENNmt, ENNmt, ENNmt, ENNmt, ENNmt, ENNmt, ENNmt, ENNmt }, // EN
+ { EN2, -1, ADone, EN, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1 }, // ENNmt
+ { -1, -1, ADone, EN, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1 } // EN2
+ };
+ signed char state = Init;
+ signed char input;
+ bool parseOk = TRUE;
+
+ while ( TRUE ) {
+
+ // get input
+ if ( atEnd() ) {
+ d->error = XMLERR_UNEXPECTEDEOF;
+ goto parseError;
+ }
+ if ( is_S(c) ) {
+ input = InpWs;
+ } else if ( c == '(' ) {
+ input = InpOp;
+ } else if ( c == ')' ) {
+ input = InpCp;
+ } else if ( c == '|' ) {
+ input = InpPipe;
+ } else if ( c == 'C' ) {
+ input = InpC;
+ } else if ( c == 'E' ) {
+ input = InpE;
+ } else if ( c == 'I' ) {
+ input = InpI;
+ } else if ( c == 'M' ) {
+ input = InpM;
+ } else if ( c == 'N' ) {
+ input = InpN;
+ } else if ( c == 'O' ) {
+ input = InpO;
+ } else if ( c == 'R' ) {
+ input = InpR;
+ } else if ( c == 'S' ) {
+ input = InpS;
+ } else if ( c == 'Y' ) {
+ input = InpY;
+ } else {
+ input = InpUnknown;
+ }
+
+ // set state according to input
+ state = table[state][input];
+
+ // do some actions according to state
+ switch ( state ) {
+ case ST:
+ parseOk = parseString( "CDATA" );
+ break;
+ case TTI:
+ parseOk = parseString( "ID" );
+ break;
+ case TTI2:
+ parseOk = parseString( "REF" );
+ break;
+ case TTI3:
+ next(); // S
+ break;
+ case TTE:
+ parseOk = parseString( "ENTIT" );
+ break;
+ case TTEY:
+ next(); // Y
+ break;
+ case TTEI:
+ parseOk = parseString( "IES" );
+ break;
+ case N:
+ next(); // N
+ break;
+ case TTNM:
+ parseOk = parseString( "MTOKEN" );
+ break;
+ case TTNM2:
+ next(); // S
+ break;
+ case NO:
+ parseOk = parseString( "OTATION" );
+ break;
+ case NO2:
+ eat_ws();
+ break;
+ case NO3:
+ next_eat_ws();
+ break;
+ case NOName:
+ parseOk = parseName();
+ break;
+ case NO4:
+ eat_ws();
+ break;
+ case EN:
+ next_eat_ws();
+ break;
+ case ENNmt:
+ parseOk = parseNmtoken();
+ break;
+ case EN2:
+ eat_ws();
+ break;
+ case ADone:
+ next();
+ break;
+ }
+ // no input is read after this
+ switch ( state ) {
+ case ST:
+ if( !parseOk ) {
+ d->error = XMLERR_UNEXPECTEDCHARACTER;
+ goto parseError;
+ }
+ break;
+ case TTI:
+ if( !parseOk ) {
+ d->error = XMLERR_UNEXPECTEDCHARACTER;
+ goto parseError;
+ }
+ break;
+ case TTI2:
+ if( !parseOk ) {
+ d->error = XMLERR_UNEXPECTEDCHARACTER;
+ goto parseError;
+ }
+ break;
+ case TTE:
+ if( !parseOk ) {
+ d->error = XMLERR_UNEXPECTEDCHARACTER;
+ goto parseError;
+ }
+ break;
+ case TTEI:
+ if( !parseOk ) {
+ d->error = XMLERR_UNEXPECTEDCHARACTER;
+ goto parseError;
+ }
+ break;
+ case TTNM:
+ if( !parseOk ) {
+ d->error = XMLERR_UNEXPECTEDCHARACTER;
+ goto parseError;
+ }
+ break;
+ case NO:
+ if( !parseOk ) {
+ d->error = XMLERR_UNEXPECTEDCHARACTER;
+ goto parseError;
+ }
+ break;
+ case NOName:
+ if ( !parseOk ) {
+ d->error = XMLERR_ERRORPARSINGNAME;
+ goto parseError;
+ }
+ break;
+ case ENNmt:
+ if ( !parseOk ) {
+ d->error = XMLERR_ERRORPARSINGNMTOKEN;
+ goto parseError;
+ }
+ break;
+ case ADone:
+ return TRUE;
+ case Done:
+ return TRUE;
+ case -1:
+ // Error
+ d->error = XMLERR_LETTEREXPECTED;
+ goto parseError;
+ }
+
+ }
+
+ return TRUE;
+
+parseError:
+ reportParseError();
+ return FALSE;
+}
+
+/*!
+ Parse a AttValue [10]
+
+ Precondition: the head stands on the beginning " or '
+
+ If this function was successful, the head stands on the first
+ character after the closing " or ' and the value of the attribute
+ is in string().
+*/
+bool QXmlSimpleReader::parseAttValue()
+{
+ bool tmp;
+
+ const signed char Init = 0;
+ const signed char Dq = 1; // double quotes were read
+ const signed char DqRef = 2; // read references in double quotes
+ const signed char DqC = 3; // signed character read in double quotes
+ const signed char Sq = 4; // single quotes were read
+ const signed char SqRef = 5; // read references in single quotes
+ const signed char SqC = 6; // signed character read in single quotes
+ const signed char Done = 7;
+
+ const signed char InpDq = 0; // "
+ const signed char InpSq = 1; // '
+ const signed char InpAmp = 2; // &
+ const signed char InpLt = 3; // <
+ const signed char InpUnknown = 4;
+
+ // use some kind of state machine for parsing
+ static signed char table[7][5] = {
+ /* InpDq InpSq InpAmp InpLt InpUnknown */
+ { Dq, Sq, -1, -1, -1 }, // Init
+ { Done, DqC, DqRef, -1, DqC }, // Dq
+ { Done, DqC, DqRef, -1, DqC }, // DqRef
+ { Done, DqC, DqRef, -1, DqC }, // DqC
+ { SqC, Done, SqRef, -1, SqC }, // Sq
+ { SqC, Done, SqRef, -1, SqC }, // SqRef
+ { SqC, Done, SqRef, -1, SqC } // SqRef
+ };
+ signed char state = Init;
+ signed char input;
+ bool parseOk = TRUE;
+
+ while ( TRUE ) {
+
+ // get input
+ if ( atEnd() ) {
+ d->error = XMLERR_UNEXPECTEDEOF;
+ goto parseError;
+ }
+ if ( c == '"' ) {
+ input = InpDq;
+ } else if ( c == '\'' ) {
+ input = InpSq;
+ } else if ( c == '&' ) {
+ input = InpAmp;
+ } else if ( c == '<' ) {
+ input = InpLt;
+ } else {
+ input = InpUnknown;
+ }
+
+ // set state according to input
+ state = table[state][input];
+
+ // do some actions according to state
+ switch ( state ) {
+ case Dq:
+ case Sq:
+ stringClear();
+ next();
+ break;
+ case DqRef:
+ case SqRef:
+ parseOk = parseReference( tmp, InAttributeValue );
+ break;
+ case DqC:
+ case SqC:
+ stringAddC();
+ next();
+ break;
+ case Done:
+ next();
+ break;
+ }
+ // no input is read after this
+ switch ( state ) {
+ case DqRef:
+ case SqRef:
+ if ( !parseOk ) {
+ d->error = XMLERR_ERRORPARSINGREFERENCE;
+ goto parseError;
+ }
+ break;
+ case Done:
+ return TRUE;
+ case -1:
+ // Error
+ d->error = XMLERR_UNEXPECTEDCHARACTER;
+ goto parseError;
+ }
+
+ }
+
+ return TRUE;
+
+parseError:
+ reportParseError();
+ return FALSE;
+}
+
+/*!
+ Parse a elementdecl [45].
+
+ Precondition: the beginning '<!E' is already read and the head
+ stands on the 'L' of '<!ELEMENT'
+*/
+bool QXmlSimpleReader::parseElementDecl()
+{
+ const signed char Init = 0;
+ const signed char Elem = 1; // parse the beginning string
+ const signed char Ws1 = 2; // whitespace required
+ const signed char Nam = 3; // parse Name
+ const signed char Ws2 = 4; // whitespace required
+ const signed char Empty = 5; // read EMPTY
+ const signed char Any = 6; // read ANY
+ const signed char Cont = 7; // read contentspec (except ANY or EMPTY)
+ const signed char Mix = 8; // read Mixed
+ const signed char Mix2 = 9; //
+ const signed char Mix3 = 10; //
+ const signed char MixN1 = 11; //
+ const signed char MixN2 = 12; //
+ const signed char MixN3 = 13; //
+ const signed char MixN4 = 14; //
+ const signed char Cp = 15; // parse cp
+ const signed char Cp2 = 16; //
+ const signed char WsD = 17; // eat whitespace before Done
+ const signed char Done = 18;
+
+ const signed char InpWs = 0;
+ const signed char InpGt = 1; // >
+ const signed char InpPipe = 2; // |
+ const signed char InpOp = 3; // (
+ const signed char InpCp = 4; // )
+ const signed char InpHash = 5; // #
+ const signed char InpQm = 6; // ?
+ const signed char InpAst = 7; // *
+ const signed char InpPlus = 8; // +
+ const signed char InpA = 9; // A
+ const signed char InpE = 10; // E
+ const signed char InpL = 11; // L
+ const signed char InpUnknown = 12;
+
+ // use some kind of state machine for parsing
+ static signed char table[18][13] = {
+ /* InpWs InpGt InpPipe InpOp InpCp InpHash InpQm InpAst InpPlus InpA InpE InpL InpUnknown */
+ { -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, Elem, -1 }, // Init
+ { Ws1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1 }, // Elem
+ { -1, -1, -1, -1, -1, -1, -1, -1, -1, Nam, Nam, Nam, Nam }, // Ws1
+ { Ws2, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1 }, // Nam
+ { -1, -1, -1, Cont, -1, -1, -1, -1, -1, Any, Empty, -1, -1 }, // Ws2
+ { WsD, Done, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1 }, // Empty
+ { WsD, Done, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1 }, // Any
+ { -1, -1, -1, Cp, Cp, Mix, -1, -1, -1, Cp, Cp, Cp, Cp }, // Cont
+ { Mix2, -1, MixN1, -1, Mix3, -1, -1, -1, -1, -1, -1, -1, -1 }, // Mix
+ { -1, -1, MixN1, -1, Mix3, -1, -1, -1, -1, -1, -1, -1, -1 }, // Mix2
+ { WsD, Done, -1, -1, -1, -1, -1, WsD, -1, -1, -1, -1, -1 }, // Mix3
+ { -1, -1, -1, -1, -1, -1, -1, -1, -1, MixN2, MixN2, MixN2, MixN2 }, // MixN1
+ { MixN3, -1, MixN1, -1, MixN4, -1, -1, -1, -1, -1, -1, -1, -1 }, // MixN2
+ { -1, -1, MixN1, -1, MixN4, -1, -1, -1, -1, -1, -1, -1, -1 }, // MixN3
+ { -1, -1, -1, -1, -1, -1, -1, WsD, -1, -1, -1, -1, -1 }, // MixN4
+ { WsD, Done, -1, -1, -1, -1, Cp2, Cp2, Cp2, -1, -1, -1, -1 }, // Cp
+ { WsD, Done, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1 }, // Cp2
+ { -1, Done, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1 } // WsD
+ };
+ signed char state = Init;
+ signed char input;
+ bool parseOk = TRUE;
+
+ while ( TRUE ) {
+
+ // read input
+ if ( atEnd() ) {
+ d->error = XMLERR_UNEXPECTEDEOF;
+ goto parseError;
+ }
+ if ( is_S(c) ) {
+ input = InpWs;
+ } else if ( c == '>' ) {
+ input = InpGt;
+ } else if ( c == '|' ) {
+ input = InpPipe;
+ } else if ( c == '(' ) {
+ input = InpOp;
+ } else if ( c == ')' ) {
+ input = InpCp;
+ } else if ( c == '#' ) {
+ input = InpHash;
+ } else if ( c == '?' ) {
+ input = InpQm;
+ } else if ( c == '*' ) {
+ input = InpAst;
+ } else if ( c == '+' ) {
+ input = InpPlus;
+ } else if ( c == 'A' ) {
+ input = InpA;
+ } else if ( c == 'E' ) {
+ input = InpE;
+ } else if ( c == 'L' ) {
+ input = InpL;
+ } else {
+ input = InpUnknown;
+ }
+ // get new state
+//qDebug( "%d -%d(%c)-> %d", state, input, c.latin1(), table[state][input] );
+ state = table[state][input];
+
+ // in some cases do special actions depending on state
+ switch ( state ) {
+ case Elem:
+ parseOk = parseString( "LEMENT" );
+ break;
+ case Ws1:
+ eat_ws();
+ break;
+ case Nam:
+ parseOk = parseName();
+ break;
+ case Ws2:
+ eat_ws();
+ break;
+ case Empty:
+ parseOk = parseString( "EMPTY" );
+ break;
+ case Any:
+ parseOk = parseString( "ANY" );
+ break;
+ case Cont:
+ next_eat_ws();
+ break;
+ case Mix:
+ parseOk = parseString( "#PCDATA" );
+ break;
+ case Mix2:
+ eat_ws();
+ break;
+ case Mix3:
+ next();
+ break;
+ case MixN1:
+ next_eat_ws();
+ break;
+ case MixN2:
+ parseOk = parseName();
+ break;
+ case MixN3:
+ eat_ws();
+ break;
+ case MixN4:
+ next();
+ break;
+ case Cp:
+ parseOk = parseChoiceSeq();
+ break;
+ case Cp2:
+ next();
+ break;
+ case WsD:
+ next_eat_ws();
+ break;
+ case Done:
+ next();
+ break;
+ }
+ // no input is read after this
+ switch ( state ) {
+ case Elem:
+ if( !parseOk ) {
+ d->error = XMLERR_UNEXPECTEDCHARACTER;
+ goto parseError;
+ }
+ break;
+ case Nam:
+ if ( !parseOk ) {
+ d->error = XMLERR_ERRORPARSINGNAME;
+ goto parseError;
+ }
+ break;
+ case Empty:
+ if( !parseOk ) {
+ d->error = XMLERR_UNEXPECTEDCHARACTER;
+ goto parseError;
+ }
+ break;
+ case Any:
+ if( !parseOk ) {
+ d->error = XMLERR_UNEXPECTEDCHARACTER;
+ goto parseError;
+ }
+ break;
+ case Mix:
+ if( !parseOk ) {
+ d->error = XMLERR_UNEXPECTEDCHARACTER;
+ goto parseError;
+ }
+ break;
+ case MixN2:
+ if ( !parseOk ) {
+ d->error = XMLERR_ERRORPARSINGNAME;
+ goto parseError;
+ }
+ break;
+ case Cp:
+ if ( !parseOk ) {
+ d->error = XMLERR_ERRORPARSINGCHOICE;
+ goto parseError;
+ }
+ break;
+ case Done:
+ return TRUE;
+ case -1:
+ d->error = XMLERR_UNEXPECTEDCHARACTER;
+ goto parseError;
+ }
+
+ }
+
+ return TRUE;
+
+parseError:
+ reportParseError();
+ return FALSE;
+}
+
+/*!
+ Parse a NotationDecl [82].
+
+ Precondition: the beginning '<!' is already read and the head
+ stands on the 'N' of '<!NOTATION'
+*/
+bool QXmlSimpleReader::parseNotationDecl()
+{
+ const signed char Init = 0;
+ const signed char Not = 1; // read NOTATION
+ const signed char Ws1 = 2; // eat whitespaces
+ const signed char Nam = 3; // read Name
+ const signed char Ws2 = 4; // eat whitespaces
+ const signed char ExtID = 5; // parse ExternalID
+ const signed char Ws3 = 6; // eat whitespaces
+ const signed char Done = 7;
+
+ const signed char InpWs = 0;
+ const signed char InpGt = 1; // >
+ const signed char InpN = 2; // N
+ const signed char InpUnknown = 3;
+
+ // use some kind of state machine for parsing
+ static signed char table[7][4] = {
+ /* InpWs InpGt InpN InpUnknown */
+ { -1, -1, Not, -1 }, // Init
+ { Ws1, -1, -1, -1 }, // Not
+ { -1, -1, Nam, Nam }, // Ws1
+ { Ws2, Done, -1, -1 }, // Nam
+ { -1, Done, ExtID, ExtID }, // Ws2
+ { Ws3, Done, -1, -1 }, // ExtID
+ { -1, Done, -1, -1 } // Ws3
+ };
+ signed char state = Init;
+ signed char input;
+ bool parseOk = TRUE;
+
+ while ( TRUE ) {
+
+ // get input
+ if ( atEnd() ) {
+ d->error = XMLERR_UNEXPECTEDEOF;
+ goto parseError;
+ }
+ if ( is_S(c) ) {
+ input = InpWs;
+ } else if ( c == '>' ) {
+ input = InpGt;
+ } else if ( c == 'N' ) {
+ input = InpN;
+ } else {
+ input = InpUnknown;
+ }
+
+ // set state according to input
+ state = table[state][input];
+
+ // do some actions according to state
+ switch ( state ) {
+ case Not:
+ parseOk = parseString( "NOTATION" );
+ break;
+ case Ws1:
+ eat_ws();
+ break;
+ case Nam:
+ parseOk = parseName();
+ break;
+ case Ws2:
+ eat_ws();
+ break;
+ case ExtID:
+ parseOk = parseExternalID( TRUE );
+ break;
+ case Ws3:
+ eat_ws();
+ break;
+ case Done:
+ next();
+ break;
+ }
+ // no input is read after this
+ switch ( state ) {
+ case Not:
+ if ( !parseOk ) {
+ d->error = XMLERR_UNEXPECTEDCHARACTER;
+ goto parseError;
+ }
+ break;
+ case Nam:
+ if ( !parseOk ) {
+ d->error = XMLERR_ERRORPARSINGNAME;
+ goto parseError;
+ }
+ break;
+ case ExtID:
+ if ( !parseOk ) {
+ d->error = XMLERR_ERRORPARSINGEXTERNALID;
+ goto parseError;
+ }
+ // call the handler
+ if ( dtdHnd ) {
+ if ( !dtdHnd->notationDecl( name(), d->publicId, d->systemId ) ) {
+ d->error = dtdHnd->errorString();
+ goto parseError;
+ }
+ }
+ break;
+ case Done:
+ return TRUE;
+ case -1:
+ // Error
+ d->error = XMLERR_UNEXPECTEDCHARACTER;
+ goto parseError;
+ }
+
+ }
+
+ return TRUE;
+
+parseError:
+ reportParseError();
+ return FALSE;
+}
+
+/*!
+ Parse choice [49] or seq [50].
+
+ Precondition: the beginning '('S? is already read and the head
+ stands on the first non-whitespace character after it.
+*/
+bool QXmlSimpleReader::parseChoiceSeq()
+{
+ const signed char Init = 0;
+ const signed char Ws1 = 1; // eat whitespace
+ const signed char CS = 2; // choice or set
+ const signed char Ws2 = 3; // eat whitespace
+ const signed char More = 4; // more cp to read
+ const signed char Name = 5; // read name
+ const signed char Done = 6; //
+
+ const signed char InpWs = 0; // S
+ const signed char InpOp = 1; // (
+ const signed char InpCp = 2; // )
+ const signed char InpQm = 3; // ?
+ const signed char InpAst = 4; // *
+ const signed char InpPlus = 5; // +
+ const signed char InpPipe = 6; // |
+ const signed char InpComm = 7; // ,
+ const signed char InpUnknown = 8;
+
+ // use some kind of state machine for parsing
+ static signed char table[6][9] = {
+ /* InpWs InpOp InpCp InpQm InpAst InpPlus InpPipe InpComm InpUnknown */
+ { -1, Ws1, -1, -1, -1, -1, -1, -1, Name }, // Init
+ { -1, CS, -1, -1, -1, -1, -1, -1, CS }, // Ws1
+ { Ws2, -1, Done, Ws2, Ws2, Ws2, More, More, -1 }, // CS
+ { -1, -1, Done, -1, -1, -1, More, More, -1 }, // Ws2
+ { -1, Ws1, -1, -1, -1, -1, -1, -1, Name }, // More (same as Init)
+ { Ws2, -1, Done, Ws2, Ws2, Ws2, More, More, -1 } // Name (same as CS)
+ };
+ signed char state = Init;
+ signed char input;
+ bool parseOk = TRUE;
+
+ while ( TRUE ) {
+
+ // get input
+ if ( atEnd() ) {
+ d->error = XMLERR_UNEXPECTEDEOF;
+ goto parseError;
+ }
+ if ( is_S(c) ) {
+ input = InpWs;
+ } else if ( c == '(' ) {
+ input = InpOp;
+ } else if ( c == ')' ) {
+ input = InpCp;
+ } else if ( c == '?' ) {
+ input = InpQm;
+ } else if ( c == '*' ) {
+ input = InpAst;
+ } else if ( c == '+' ) {
+ input = InpPlus;
+ } else if ( c == '|' ) {
+ input = InpPipe;
+ } else if ( c == ',' ) {
+ input = InpComm;
+ } else {
+ input = InpUnknown;
+ }
+
+ // set state according to input
+ state = table[state][input];
+
+ // do some actions according to state
+ switch ( state ) {
+ case Ws1:
+ next_eat_ws();
+ break;
+ case CS:
+ parseOk = parseChoiceSeq();
+ break;
+ case Ws2:
+ next_eat_ws();
+ break;
+ case More:
+ next_eat_ws();
+ break;
+ case Name:
+ parseOk = parseName();
+ break;
+ case Done:
+ next();
+ break;
+ }
+ // no input is read after this
+ switch ( state ) {
+ case CS:
+ if ( !parseOk ) {
+ d->error = XMLERR_ERRORPARSINGCHOICE;
+ goto parseError;
+ }
+ break;
+ case Name:
+ if ( !parseOk ) {
+ d->error = XMLERR_ERRORPARSINGNAME;
+ goto parseError;
+ }
+ break;
+ case Done:
+ return TRUE;
+ case -1:
+ // Error
+ d->error = XMLERR_UNEXPECTEDCHARACTER;
+ goto parseError;
+ }
+
+ }
+
+ return TRUE;
+
+parseError:
+ reportParseError();
+ return FALSE;
+}
+
+/*!
+ Parse a EntityDecl [70].
+
+ Precondition: the beginning '<!E' is already read and the head
+ stand on the 'N' of '<!ENTITY'
+*/
+bool QXmlSimpleReader::parseEntityDecl()
+{
+ const signed char Init = 0;
+ const signed char Ent = 1; // parse "ENTITY"
+ const signed char Ws1 = 2; // white space read
+ const signed char Name = 3; // parse name
+ const signed char Ws2 = 4; // white space read
+ const signed char EValue = 5; // parse entity value
+ const signed char ExtID = 6; // parse ExternalID
+ const signed char Ws3 = 7; // white space read
+ const signed char Ndata = 8; // parse "NDATA"
+ const signed char Ws4 = 9; // white space read
+ const signed char NNam = 10; // parse name
+ const signed char PEDec = 11; // parse PEDecl
+ const signed char Ws6 = 12; // white space read
+ const signed char PENam = 13; // parse name
+ const signed char Ws7 = 14; // white space read
+ const signed char PEVal = 15; // parse entity value
+ const signed char PEEID = 16; // parse ExternalID
+ const signed char WsE = 17; // white space read
+ const signed char Done = 18;
+
+ const signed char InpWs = 0; // white space
+ const signed char InpPer = 1; // %
+ const signed char InpQuot = 2; // " or '
+ const signed char InpGt = 3; // >
+ const signed char InpN = 4; // N
+ const signed char InpUnknown = 5;
+
+ // use some kind of state machine for parsing
+ static signed char table[18][6] = {
+ /* InpWs InpPer InpQuot InpGt InpN InpUnknown */
+ { -1, -1, -1, -1, Ent, -1 }, // Init
+ { Ws1, -1, -1, -1, -1, -1 }, // Ent
+ { -1, PEDec, -1, -1, Name, Name }, // Ws1
+ { Ws2, -1, -1, -1, -1, -1 }, // Name
+ { -1, -1, EValue, -1, -1, ExtID }, // Ws2
+ { WsE, -1, -1, Done, -1, -1 }, // EValue
+ { Ws3, -1, -1, Done, -1, -1 }, // ExtID
+ { -1, -1, -1, Done, Ndata, -1 }, // Ws3
+ { Ws4, -1, -1, -1, -1, -1 }, // Ndata
+ { -1, -1, -1, -1, NNam, NNam }, // Ws4
+ { WsE, -1, -1, Done, -1, -1 }, // NNam
+ { Ws6, -1, -1, -1, -1, -1 }, // PEDec
+ { -1, -1, -1, -1, PENam, PENam }, // Ws6
+ { Ws7, -1, -1, -1, -1, -1 }, // PENam
+ { -1, -1, PEVal, -1, -1, PEEID }, // Ws7
+ { WsE, -1, -1, Done, -1, -1 }, // PEVal
+ { WsE, -1, -1, Done, -1, -1 }, // PEEID
+ { -1, -1, -1, Done, -1, -1 } // WsE
+ };
+ signed char state = Init;
+ signed char input;
+ bool parseOk = TRUE;
+
+ while ( TRUE ) {
+
+ // get input
+ if ( atEnd() ) {
+ d->error = XMLERR_UNEXPECTEDEOF;
+ goto parseError;
+ }
+ if ( is_S(c) ) {
+ input = InpWs;
+ } else if ( c == '%' ) {
+ input = InpPer;
+ } else if ( c == '"' || c == '\'' ) {
+ input = InpQuot;
+ } else if ( c == '>' ) {
+ input = InpGt;
+ } else if ( c == 'N' ) {
+ input = InpN;
+ } else {
+ input = InpUnknown;
+ }
+
+ // set state according to input
+ state = table[state][input];
+
+ // do some actions according to state
+ switch ( state ) {
+ case Ent:
+ parseOk = parseString( "NTITY" );
+ break;
+ case Ws1:
+ eat_ws();
+ break;
+ case Name:
+ parseOk = parseName();
+ break;
+ case Ws2:
+ eat_ws();
+ break;
+ case EValue:
+ parseOk = parseEntityValue();
+ break;
+ case ExtID:
+ parseOk = parseExternalID();
+ break;
+ case Ws3:
+ eat_ws();
+ break;
+ case Ndata:
+ parseOk = parseString( "NDATA" );
+ break;
+ case Ws4:
+ eat_ws();
+ break;
+ case NNam:
+ parseOk = parseName( TRUE );
+ break;
+ case PEDec:
+ next();
+ break;
+ case Ws6:
+ eat_ws();
+ break;
+ case PENam:
+ parseOk = parseName();
+ break;
+ case Ws7:
+ eat_ws();
+ break;
+ case PEVal:
+ parseOk = parseEntityValue();
+ break;
+ case PEEID:
+ parseOk = parseExternalID();
+ break;
+ case WsE:
+ eat_ws();
+ break;
+ case Done:
+ next();
+ break;
+ }
+ // no input is read after this
+ switch ( state ) {
+ case Ent:
+ if ( !parseOk ) {
+ d->error = XMLERR_UNEXPECTEDCHARACTER;
+ goto parseError;
+ }
+ break;
+ case Name:
+ if ( !parseOk ) {
+ d->error = XMLERR_ERRORPARSINGNAME;
+ goto parseError;
+ }
+ break;
+ case EValue:
+ if ( !parseOk ) {
+ d->error = XMLERR_ERRORPARSINGENTITYVALUE;
+ goto parseError;
+ }
+ if ( d->entities.find( name() ) == d->entities.end() &&
+ d->externEntities.find( name() ) == d->externEntities.end() ) {
+ d->entities.insert( name(), string() );
+ if ( declHnd ) {
+ if ( !declHnd->internalEntityDecl( name(), string() ) ) {
+ d->error = declHnd->errorString();
+ goto parseError;
+ }
+ }
+ }
+ break;
+ case ExtID:
+ if ( !parseOk ) {
+ d->error = XMLERR_ERRORPARSINGEXTERNALID;
+ goto parseError;
+ }
+ if ( d->entities.find( name() ) == d->entities.end() &&
+ d->externEntities.find( name() ) == d->externEntities.end() ) {
+ d->externEntities.insert( name(), QXmlSimpleReaderPrivate::ExternEntity( d->publicId, d->systemId, "" ) );
+ if ( declHnd ) {
+ if ( !declHnd->externalEntityDecl( name(), d->publicId, d->systemId ) ) {
+ d->error = declHnd->errorString();
+ goto parseError;
+ }
+ }
+ }
+ break;
+ case Ndata:
+ if ( !parseOk ) {
+ d->error = XMLERR_UNEXPECTEDCHARACTER;
+ goto parseError;
+ }
+ break;
+ case NNam:
+ if ( !parseOk ) {
+ d->error = XMLERR_ERRORPARSINGNAME;
+ goto parseError;
+ }
+ if ( d->entities.find( name() ) == d->entities.end() &&
+ d->externEntities.find( name() ) == d->externEntities.end() ) {
+ d->externEntities.insert( name(), QXmlSimpleReaderPrivate::ExternEntity( d->publicId, d->systemId, ref() ) );
+ if ( declHnd ) {
+ if ( !declHnd->externalEntityDecl( name(), d->publicId, d->systemId ) ) {
+ d->error = declHnd->errorString();
+ goto parseError;
+ }
+ }
+ }
+ break;
+ case PENam:
+ if ( !parseOk ) {
+ d->error = XMLERR_ERRORPARSINGNAME;
+ goto parseError;
+ }
+ break;
+ case PEVal:
+ if ( !parseOk ) {
+ d->error = XMLERR_ERRORPARSINGENTITYVALUE;
+ goto parseError;
+ }
+ if ( d->parameterEntities.find( name() ) == d->parameterEntities.end() &&
+ d->externParameterEntities.find( name() ) == d->externParameterEntities.end() ) {
+ d->parameterEntities.insert( name(), string() );
+ if ( declHnd ) {
+ if ( !declHnd->internalEntityDecl( QString("%")+name(), string() ) ) {
+ d->error = declHnd->errorString();
+ goto parseError;
+ }
+ }
+ }
+ break;
+ case PEEID:
+ if ( !parseOk ) {
+ d->error = XMLERR_ERRORPARSINGEXTERNALID;
+ goto parseError;
+ }
+ if ( d->parameterEntities.find( name() ) == d->parameterEntities.end() &&
+ d->externParameterEntities.find( name() ) == d->externParameterEntities.end() ) {
+ d->externParameterEntities.insert( name(), QXmlSimpleReaderPrivate::ExternParameterEntity( d->publicId, d->systemId ) );
+ if ( declHnd ) {
+ if ( !declHnd->externalEntityDecl( QString("%")+name(), d->publicId, d->systemId ) ) {
+ d->error = declHnd->errorString();
+ goto parseError;
+ }
+ }
+ }
+ break;
+ case Done:
+ return TRUE;
+ case -1:
+ // Error
+ d->error = XMLERR_LETTEREXPECTED;
+ goto parseError;
+ }
+
+ }
+
+ return TRUE;
+
+parseError:
+ reportParseError();
+ return FALSE;
+}
+
+/*!
+ Parse a EntityValue [9]
+*/
+bool QXmlSimpleReader::parseEntityValue()
+{
+ bool tmp;
+
+ const signed char Init = 0;
+ const signed char Dq = 1; // EntityValue is double quoted
+ const signed char DqC = 2; // signed character
+ const signed char DqPER = 3; // PERefence
+ const signed char DqRef = 4; // Reference
+ const signed char Sq = 5; // EntityValue is double quoted
+ const signed char SqC = 6; // signed character
+ const signed char SqPER = 7; // PERefence
+ const signed char SqRef = 8; // Reference
+ const signed char Done = 9;
+
+ const signed char InpDq = 0; // "
+ const signed char InpSq = 1; // '
+ const signed char InpAmp = 2; // &
+ const signed char InpPer = 3; // %
+ const signed char InpUnknown = 4;
+
+ // use some kind of state machine for parsing
+ static signed char table[9][5] = {
+ /* InpDq InpSq InpAmp InpPer InpUnknown */
+ { Dq, Sq, -1, -1, -1 }, // Init
+ { Done, DqC, DqRef, DqPER, DqC }, // Dq
+ { Done, DqC, DqRef, DqPER, DqC }, // DqC
+ { Done, DqC, DqRef, DqPER, DqC }, // DqPER
+ { Done, DqC, DqRef, DqPER, DqC }, // DqRef
+ { SqC, Done, SqRef, SqPER, SqC }, // Sq
+ { SqC, Done, SqRef, SqPER, SqC }, // SqC
+ { SqC, Done, SqRef, SqPER, SqC }, // SqPER
+ { SqC, Done, SqRef, SqPER, SqC } // SqRef
+ };
+ signed char state = Init;
+ signed char input;
+ bool parseOk = TRUE;
+
+ while ( TRUE ) {
+
+ // get input
+ if ( atEnd() ) {
+ d->error = XMLERR_UNEXPECTEDEOF;
+ goto parseError;
+ }
+ if ( c == '"' ) {
+ input = InpDq;
+ } else if ( c == '\'' ) {
+ input = InpSq;
+ } else if ( c == '&' ) {
+ input = InpAmp;
+ } else if ( c == '%' ) {
+ input = InpPer;
+ } else {
+ input = InpUnknown;
+ }
+
+ // set state according to input
+ state = table[state][input];
+
+ // do some actions according to state
+ switch ( state ) {
+ case Dq:
+ case Sq:
+ stringClear();
+ next();
+ break;
+ case DqC:
+ case SqC:
+ stringAddC();
+ next();
+ break;
+ case DqPER:
+ case SqPER:
+ parseOk = parsePEReference( InEntityValue );
+ break;
+ case DqRef:
+ case SqRef:
+ parseOk = parseReference( tmp, InEntityValue );
+ break;
+ case Done:
+ next();
+ break;
+ }
+ // no input is read after this
+ switch ( state ) {
+ case DqPER:
+ case SqPER:
+ if ( !parseOk ) {
+ d->error = XMLERR_ERRORPARSINGDOCTYPE;
+ goto parseError;
+ }
+ break;
+ case DqRef:
+ case SqRef:
+ if ( !parseOk ) {
+ d->error = XMLERR_ERRORPARSINGREFERENCE;
+ goto parseError;
+ }
+ break;
+ case Done:
+ return TRUE;
+ case -1:
+ // Error
+ d->error = XMLERR_LETTEREXPECTED;
+ goto parseError;
+ }
+
+ }
+
+ return TRUE;
+
+parseError:
+ reportParseError();
+ return FALSE;
+}
+
+/*!
+ Parse a comment [15].
+
+ Precondition: the beginning '<!' of the comment is already read and the head
+ stands on the first '-' of '<!--'.
+
+ If this funktion was successful, the head-position is on the first
+ character after the comment.
+*/
+bool QXmlSimpleReader::parseComment()
+{
+ const signed char Init = 0;
+ const signed char Dash1 = 1; // the first dash was read
+ const signed char Dash2 = 2; // the second dash was read
+ const signed char Com = 3; // read comment
+ const signed char Com2 = 4; // read comment (help state)
+ const signed char ComE = 5; // finished reading comment
+ const signed char Done = 6;
+
+ const signed char InpDash = 0; // -
+ const signed char InpGt = 1; // >
+ const signed char InpUnknown = 2;
+
+ // use some kind of state machine for parsing
+ static signed char table[6][3] = {
+ /* InpDash InpGt InpUnknown */
+ { Dash1, -1, -1 }, // Init
+ { Dash2, -1, -1 }, // Dash1
+ { Com2, Com, Com }, // Dash2
+ { Com2, Com, Com }, // Com
+ { ComE, Com, Com }, // Com2
+ { -1, Done, -1 } // ComE
+ };
+ signed char state = Init;
+ signed char input;
+
+ while ( TRUE ) {
+
+ // get input
+ if ( atEnd() ) {
+ d->error = XMLERR_UNEXPECTEDEOF;
+ goto parseError;
+ }
+ if ( c == '-' ) {
+ input = InpDash;
+ } else if ( c == '>' ) {
+ input = InpGt;
+ } else {
+ input = InpUnknown;
+ }
+
+ // set state according to input
+ state = table[state][input];
+
+ // do some actions according to state
+ switch ( state ) {
+ case Dash1:
+ next();
+ break;
+ case Dash2:
+ next();
+ break;
+ case Com:
+ stringAddC();
+ next();
+ break;
+ case Com2:
+ next();
+ break;
+ case ComE:
+ next();
+ break;
+ case Done:
+ next();
+ break;
+ }
+ // no input is read after this
+ switch ( state ) {
+ case Dash2:
+ stringClear();
+ break;
+ case Com2:
+ // if next character is not a dash than don't skip it
+ if ( c != '-' ) {
+ stringAddC( '-' );
+ }
+ break;
+ case Done:
+ return TRUE;
+ case -1:
+ // Error
+ d->error = XMLERR_ERRORPARSINGCOMMENT;
+ goto parseError;
+ }
+
+ }
+
+ return TRUE;
+
+parseError:
+ reportParseError();
+ return FALSE;
+}
+
+/*!
+ Parse a Attribute [41].
+
+ Precondition: the head stands on the first character of the name of the
+ attribute (i.e. all whitespaces are already parsed).
+
+ The head stand on the next character after the end quotes. The variable name
+ contains the name of the attribute and the variable string contains the value
+ of the attribute.
+*/
+bool QXmlSimpleReader::parseAttribute()
+{
+ const signed char Init = 0;
+ const signed char PName = 1; // parse name
+ const signed char Ws = 2; // eat ws
+ const signed char Eq = 3; // the '=' was read
+ const signed char Quotes = 4; // " or ' were read
+
+ const signed char InpNameBe = 0;
+ const signed char InpEq = 1; // =
+ const signed char InpDq = 2; // "
+ const signed char InpSq = 3; // '
+ const signed char InpUnknown = 4;
+
+ // use some kind of state machine for parsing
+ static signed char table[4][5] = {
+ /* InpNameBe InpEq InpDq InpSq InpUnknown */
+ { PName, -1, -1, -1, -1 }, // Init
+ { -1, Eq, -1, -1, Ws }, // PName
+ { -1, Eq, -1, -1, -1 }, // Ws
+ { -1, -1, Quotes, Quotes, -1 } // Eq
+ };
+ signed char state = Init;
+ signed char input;
+ bool parseOk = TRUE;
+
+ while ( TRUE ) {
+
+ // get input
+ if ( atEnd() ) {
+ d->error = XMLERR_UNEXPECTEDEOF;
+ goto parseError;
+ }
+ if ( is_NameBeginning(c) ) {
+ input = InpNameBe;
+ } else if ( c == '=' ) {
+ input = InpEq;
+ } else if ( c == '"' ) {
+ input = InpDq;
+ } else if ( c == '\'' ) {
+ input = InpSq;
+ } else {
+ input = InpUnknown;
+ }
+
+ // set state according to input
+ state = table[state][input];
+
+ // do some actions according to state
+ switch ( state ) {
+ case PName:
+ parseOk = parseName();
+ break;
+ case Ws:
+ eat_ws();
+ break;
+ case Eq:
+ next_eat_ws();
+ break;
+ case Quotes:
+ parseOk = parseAttValue();
+ break;
+ }
+ // no input is read after this
+ switch ( state ) {
+ case PName:
+ if ( !parseOk ) {
+ d->error = XMLERR_ERRORPARSINGNAME;
+ goto parseError;
+ }
+ break;
+ case Quotes:
+ if ( !parseOk ) {
+ d->error = XMLERR_ERRORPARSINGATTVALUE;
+ goto parseError;
+ }
+ // Done
+ return TRUE;
+ case -1:
+ // Error
+ d->error = XMLERR_UNEXPECTEDCHARACTER;
+ goto parseError;
+ }
+
+ }
+
+ return TRUE;
+
+parseError:
+ reportParseError();
+ return FALSE;
+}
+
+/*!
+ Parse a Name [5] and store the name in name or ref (if useRef is TRUE).
+*/
+bool QXmlSimpleReader::parseName( bool useRef )
+{
+ const signed char Init = 0;
+ const signed char Name1 = 1; // parse first signed character of the name
+ const signed char Name = 2; // parse name
+ const signed char Done = 3;
+
+ const signed char InpNameBe = 0; // name beginning signed characters
+ const signed char InpNameCh = 1; // NameChar without InpNameBe
+ const signed char InpUnknown = 2;
+
+ // use some kind of state machine for parsing
+ static signed char table[3][3] = {
+ /* InpNameBe InpNameCh InpUnknown */
+ { Name1, -1, -1 }, // Init
+ { Name, Name, Done }, // Name1
+ { Name, Name, Done } // Name
+ };
+ signed char state = Init;
+ signed char input;
+
+ while ( TRUE ) {
+
+ // get input
+ if ( atEnd() ) {
+ d->error = XMLERR_UNEXPECTEDEOF;
+ goto parseError;
+ }
+ if ( is_NameBeginning(c) ) {
+ input = InpNameBe;
+ } else if ( is_NameChar(c) ) {
+ input = InpNameCh;
+ } else {
+ input = InpUnknown;
+ }
+
+ // set state according to input
+ state = table[state][input];
+
+ // do some actions according to state
+ switch ( state ) {
+ case Name1:
+ if ( useRef ) {
+ refClear();
+ refAddC();
+ } else {
+ nameClear();
+ nameAddC();
+ }
+ next();
+ break;
+ case Name:
+ if ( useRef ) {
+ refAddC();
+ } else {
+ nameAddC();
+ }
+ next();
+ break;
+ }
+ // no input is read after this
+ switch ( state ) {
+ case Done:
+ return TRUE;
+ case -1:
+ // Error
+ d->error = XMLERR_LETTEREXPECTED;
+ goto parseError;
+ }
+
+ }
+
+ return TRUE;
+
+parseError:
+ reportParseError();
+ return FALSE;
+}
+
+/*!
+ Parse a Nmtoken [7] and store the name in name.
+*/
+bool QXmlSimpleReader::parseNmtoken()
+{
+ const signed char Init = 0;
+ const signed char NameF = 1;
+ const signed char Name = 2;
+ const signed char Done = 3;
+
+ const signed char InpNameCh = 0; // NameChar without InpNameBe
+ const signed char InpUnknown = 1;
+
+ // use some kind of state machine for parsing
+ static signed char table[3][2] = {
+ /* InpNameCh InpUnknown */
+ { NameF, -1 }, // Init
+ { Name, Done }, // NameF
+ { Name, Done } // Name
+ };
+ signed char state = Init;
+ signed char input;
+
+ while ( TRUE ) {
+
+ // get input
+ if ( atEnd() ) {
+ d->error = XMLERR_UNEXPECTEDEOF;
+ goto parseError;
+ }
+ if ( is_NameChar(c) ) {
+ input = InpNameCh;
+ } else {
+ input = InpUnknown;
+ }
+
+ // set state according to input
+ state = table[state][input];
+
+ // do some actions according to state
+ switch ( state ) {
+ case NameF:
+ nameClear();
+ nameAddC();
+ next();
+ break;
+ case Name:
+ nameAddC();
+ next();
+ break;
+ }
+ // no input is read after this
+ switch ( state ) {
+ case Done:
+ return TRUE;
+ case -1:
+ // Error
+ d->error = XMLERR_LETTEREXPECTED;
+ goto parseError;
+ }
+
+ }
+
+ return TRUE;
+
+parseError:
+ reportParseError();
+ return FALSE;
+}
+
+/*!
+ Parse a Reference [67].
+
+ charDataRead is set to TRUE if the reference must not be parsed. The
+ character(s) which the reference mapped to are appended to string. The
+ head stands on the first character after the reference.
+
+ charDataRead is set to FALSE if the reference must be parsed. The
+ charachter(s) which the reference mapped to are inserted at the reference
+ position. The head stands on the first character of the replacement).
+*/
+bool QXmlSimpleReader::parseReference( bool &charDataRead, EntityRecognitionContext context )
+{
+ // temporary variables
+ uint tmp;
+ bool ok;
+
+ const signed char Init = 0;
+ const signed char SRef = 1; // start of a reference
+ const signed char ChRef = 2; // parse CharRef
+ const signed char ChDec = 3; // parse CharRef decimal
+ const signed char ChHexS = 4; // start CharRef hexadecimal
+ const signed char ChHex = 5; // parse CharRef hexadecimal
+ const signed char Name = 6; // parse name
+ const signed char DoneD = 7; // done CharRef decimal
+ const signed char DoneH = 8; // done CharRef hexadecimal
+ const signed char DoneN = 9; // done EntityRef
+
+ const signed char InpAmp = 0; // &
+ const signed char InpSemi = 1; // ;
+ const signed char InpHash = 2; // #
+ const signed char InpX = 3; // x
+ const signed char InpNum = 4; // 0-9
+ const signed char InpHex = 5; // a-f A-F
+ const signed char InpUnknown = 6;
+
+ // use some kind of state machine for parsing
+ static signed char table[8][7] = {
+ /* InpAmp InpSemi InpHash InpX InpNum InpHex InpUnknown */
+ { SRef, -1, -1, -1, -1, -1, -1 }, // Init
+ { -1, -1, ChRef, Name, Name, Name, Name }, // SRef
+ { -1, -1, -1, ChHexS, ChDec, -1, -1 }, // ChRef
+ { -1, DoneD, -1, -1, ChDec, -1, -1 }, // ChDec
+ { -1, -1, -1, -1, ChHex, ChHex, -1 }, // ChHexS
+ { -1, DoneH, -1, -1, ChHex, ChHex, -1 }, // ChHex
+ { -1, DoneN, -1, -1, -1, -1, -1 } // Name
+ };
+ signed char state = Init;
+ signed char input;
+
+ while ( TRUE ) {
+
+ // get input
+ if ( atEnd() ) {
+ d->error = XMLERR_UNEXPECTEDEOF;
+ goto parseError;
+ }
+ if ( c.row() ) {
+ input = InpUnknown;
+ } else if ( c.cell() == '&' ) {
+ input = InpAmp;
+ } else if ( c.cell() == ';' ) {
+ input = InpSemi;
+ } else if ( c.cell() == '#' ) {
+ input = InpHash;
+ } else if ( c.cell() == 'x' ) {
+ input = InpX;
+ } else if ( '0' <= c.cell() && c.cell() <= '9' ) {
+ input = InpNum;
+ } else if ( 'a' <= c.cell() && c.cell() <= 'f' ) {
+ input = InpHex;
+ } else if ( 'A' <= c.cell() && c.cell() <= 'F' ) {
+ input = InpHex;
+ } else {
+ input = InpUnknown;
+ }
+
+ // set state according to input
+ state = table[state][input];
+
+ // do some actions according to state
+ switch ( state ) {
+ case SRef:
+ refClear();
+ next();
+ break;
+ case ChRef:
+ next();
+ break;
+ case ChDec:
+ refAddC();
+ next();
+ break;
+ case ChHexS:
+ next();
+ break;
+ case ChHex:
+ refAddC();
+ next();
+ break;
+ case Name:
+ // read the name into the ref
+ parseName( TRUE );
+ break;
+ case DoneD:
+ tmp = ref().toUInt( &ok, 10 );
+ if ( ok ) {
+ stringAddC( QChar(tmp) );
+ } else {
+ d->error = XMLERR_ERRORPARSINGREFERENCE;
+ goto parseError;
+ }
+ charDataRead = TRUE;
+ next();
+ break;
+ case DoneH:
+ tmp = ref().toUInt( &ok, 16 );
+ if ( ok ) {
+ stringAddC( QChar(tmp) );
+ } else {
+ d->error = XMLERR_ERRORPARSINGREFERENCE;
+ goto parseError;
+ }
+ charDataRead = TRUE;
+ next();
+ break;
+ case DoneN:
+ if ( ref() == "amp" ) {
+ if ( context == InEntityValue ) {
+ // Bypassed
+ stringAddC( '&' ); stringAddC( 'a' ); stringAddC( 'm' ); stringAddC( 'p' ); stringAddC( ';' );
+ } else {
+ // Included or Included in literal
+ stringAddC( '&' );
+ }
+ charDataRead = TRUE;
+ } else if ( ref() == "lt" ) {
+ if ( context == InEntityValue ) {
+ // Bypassed
+ stringAddC( '&' ); stringAddC( 'l' ); stringAddC( 't' ); stringAddC( ';' );
+ } else {
+ // Included or Included in literal
+ stringAddC( '<' );
+ }
+ charDataRead = TRUE;
+ } else if ( ref() == "gt" ) {
+ if ( context == InEntityValue ) {
+ // Bypassed
+ stringAddC( '&' ); stringAddC( 'g' ); stringAddC( 't' ); stringAddC( ';' );
+ } else {
+ // Included or Included in literal
+ stringAddC( '>' );
+ }
+ charDataRead = TRUE;
+ } else if ( ref() == "apos" ) {
+ if ( context == InEntityValue ) {
+ // Bypassed
+ stringAddC( '&' ); stringAddC( 'a' ); stringAddC( 'p' ); stringAddC( 'o' ); stringAddC( 's' ); stringAddC( ';' );
+ } else {
+ // Included or Included in literal
+ stringAddC( '\'' );
+ }
+ charDataRead = TRUE;
+ } else if ( ref() == "quot" ) {
+ if ( context == InEntityValue ) {
+ // Bypassed
+ stringAddC( '&' ); stringAddC( 'q' ); stringAddC( 'u' ); stringAddC( 'o' ); stringAddC( 't' ); stringAddC( ';' );
+ } else {
+ // Included or Included in literal
+ stringAddC( '"' );
+ }
+ charDataRead = TRUE;
+ } else {
+ QMap<QString,QString>::Iterator it;
+ it = d->entities.find( ref() );
+ if ( it != d->entities.end() ) {
+ // "Internal General"
+ switch ( context ) {
+ case InContent:
+ // Included
+ xmlRef = it.data() + xmlRef;
+ charDataRead = FALSE;
+ break;
+ case InAttributeValue:
+ // Included in literal
+ xmlRef = it.data().replace( QRegExp("\""), "&quot;" ).replace( QRegExp("'"), "&apos;" )
+ + xmlRef;
+ charDataRead = FALSE;
+ break;
+ case InEntityValue:
+ {
+ // Bypassed
+ stringAddC( '&' );
+ for ( int i=0; i<(int)ref().length(); i++ ) {
+ stringAddC( ref()[i] );
+ }
+ stringAddC( ';');
+ charDataRead = TRUE;
+ }
+ break;
+ case InDTD:
+ // Forbidden
+ d->error = XMLERR_INTERNALGENERALENTITYINDTD;
+ charDataRead = FALSE;
+ break;
+ }
+ } else {
+ QMap<QString,QXmlSimpleReaderPrivate::ExternEntity>::Iterator itExtern;
+ itExtern = d->externEntities.find( ref() );
+ if ( itExtern != d->externEntities.end() ) {
+ // "External Parsed General"
+ switch ( context ) {
+ case InContent:
+ // Included if validating
+ if ( contentHnd ) {
+ if ( !contentHnd->skippedEntity( ref() ) ) {
+ d->error = contentHnd->errorString();
+ goto parseError;
+ }
+ }
+ charDataRead = FALSE;
+ break;
+ case InAttributeValue:
+ // Forbidden
+ d->error = XMLERR_EXTERNALGENERALENTITYINAV;
+ charDataRead = FALSE;
+ break;
+ case InEntityValue:
+ {
+ // Bypassed
+ stringAddC( '&' );
+ for ( int i=0; i<(int)ref().length(); i++ ) {
+ stringAddC( ref()[i] );
+ }
+ stringAddC( ';');
+ charDataRead = TRUE;
+ }
+ break;
+ case InDTD:
+ // Forbidden
+ d->error = XMLERR_EXTERNALGENERALENTITYINDTD;
+ charDataRead = FALSE;
+ break;
+ }
+ } else {
+ // "Unparsed" ### or is the definition of unparsed entities different?
+ if ( context == InEntityValue ) {
+ // Bypassed
+ // (this does not conform with the table 4.4 of the XML specification;
+ // on the other hand: in this case it is not really an unparsed entity)
+ stringAddC( '&' );
+ for ( int i=0; i<(int)ref().length(); i++ ) {
+ stringAddC( ref()[i] );
+ }
+ stringAddC( ';');
+ charDataRead = TRUE;
+ } else {
+#if 0
+ // Forbidden
+ d->error = XMLERR_UNPARSEDENTITYREFERENCE;
+ goto parseError;
+ charDataRead = FALSE;
+#else
+ // ### skip it???
+ if ( contentHnd ) {
+ if ( !contentHnd->skippedEntity( ref() ) ) {
+ d->error = contentHnd->errorString();
+ goto parseError;
+ }
+ }
+#endif
+ }
+ }
+ }
+ }
+ next();
+ break;
+ }
+ // no input is read after this
+ switch ( state ) {
+ case DoneD:
+ return TRUE;
+ case DoneH:
+ return TRUE;
+ case DoneN:
+ return TRUE;
+ case -1:
+ // Error
+ d->error = XMLERR_ERRORPARSINGREFERENCE;
+ goto parseError;
+ }
+
+ }
+
+ return TRUE;
+
+parseError:
+ reportParseError();
+ return FALSE;
+}
+
+/*!
+ Parse over a simple string.
+
+ After the string was successfully parsed, the head is on the first
+ character after the string.
+*/
+bool QXmlSimpleReader::parseString( const QString& s )
+{
+ signed char Done = s.length();
+
+ const signed char InpCharExpected = 0; // the character that was expected
+ const signed char InpUnknown = 1;
+
+ signed char state = 0; // state in this function is the position in the string s
+ signed char input;
+
+ while ( TRUE ) {
+
+ // get input
+ if ( atEnd() ) {
+ d->error = XMLERR_UNEXPECTEDEOF;
+ goto parseError;
+ }
+ if ( c == s[(int)state] ) {
+ input = InpCharExpected;
+ } else {
+ input = InpUnknown;
+ }
+
+ // set state according to input
+ if ( input == InpCharExpected ) {
+ state++;
+ } else {
+ // Error
+ d->error = XMLERR_UNEXPECTEDCHARACTER;
+ goto parseError;
+ }
+
+ // do some actions according to state
+ next();
+ // no input is read after this
+ if ( state == Done ) {
+ return TRUE;
+ }
+
+ }
+
+ return TRUE;
+
+parseError:
+ reportParseError();
+ return FALSE;
+}
+
+
+/*
+ Init the data values.
+*/
+void QXmlSimpleReader::init( const QXmlInputSource& i )
+{
+ xml = i.data();
+ xmlLength = xml.length();
+ xmlRef = "";
+
+ d->externParameterEntities.clear();
+ d->parameterEntities.clear();
+ d->externEntities.clear();
+ d->entities.clear();
+
+ tags.clear();
+
+ d->doctype = "";
+ d->xmlVersion = "";
+ d->encoding = "";
+ d->standalone = QXmlSimpleReaderPrivate::Unknown;
+
+ lineNr = 0;
+ columnNr = -1;
+ pos = 0;
+ next();
+ d->error = XMLERR_OK;
+}
+
+void QXmlSimpleReader::reportParseError()
+{
+ if ( errorHnd )
+ errorHnd->fatalError( QXmlParseException( d->error, columnNr+1, lineNr+1 ) );
+}
+
+#endif //QT_NO_XML
l com"> * * Results: * None. * * Side effects: * The font's objRefCount is incremented and the internal rep of the copy * is set to point to it. * *--------------------------------------------------------------------------- */ static void DupFontObjProc( Tcl_Obj *srcObjPtr, /* The object we are copying from. */ Tcl_Obj *dupObjPtr) /* The object we are copying to. */ { TkFont *fontPtr = srcObjPtr->internalRep.twoPtrValue.ptr1; dupObjPtr->typePtr = srcObjPtr->typePtr; dupObjPtr->internalRep.twoPtrValue.ptr1 = fontPtr; dupObjPtr->internalRep.twoPtrValue.ptr2 = srcObjPtr->internalRep.twoPtrValue.ptr2; if (fontPtr != NULL) { fontPtr->objRefCount++; } } /* *--------------------------------------------------------------------------- * * Tk_FontId -- * * Given a font, return an opaque handle that should be selected into the * XGCValues structure in order to get the constructed gc to use this * font. This function would go away if the XGCValues structure were * replaced with a TkGCValues structure. * * Results: * As above. * * Side effects: * None. * *--------------------------------------------------------------------------- */ Font Tk_FontId( Tk_Font tkfont) /* Font that is going to be selected into * GC. */ { TkFont *fontPtr = (TkFont *) tkfont; return fontPtr->fid; } /* *--------------------------------------------------------------------------- * * Tk_GetFontMetrics -- * * Returns overall ascent and descent metrics for the given font. These * values can be used to space multiple lines of text and to align the * baselines of text in different fonts. * * Results: * If *heightPtr is non-NULL, it is filled with the overall height of the * font, which is the sum of the ascent and descent. If *ascentPtr or * *descentPtr is non-NULL, they are filled with the ascent and/or * descent information for the font. * * Side effects: * None. * *--------------------------------------------------------------------------- */ void Tk_GetFontMetrics( Tk_Font tkfont, /* Font in which metrics are calculated. */ Tk_FontMetrics *fmPtr) /* Pointer to structure in which font metrics * for tkfont will be stored. */ { TkFont *fontPtr = (TkFont *) tkfont; fmPtr->ascent = fontPtr->fm.ascent; fmPtr->descent = fontPtr->fm.descent; fmPtr->linespace = fontPtr->fm.ascent + fontPtr->fm.descent; } /* *--------------------------------------------------------------------------- * * Tk_PostscriptFontName -- * * Given a Tk_Font, return the name of the corresponding Postscript font. * * Results: * The return value is the pointsize of the given Tk_Font. The name of * the Postscript font is appended to dsPtr. * * Side effects: * If the font does not exist on the printer, the print job will fail at * print time. Given a "reasonable" Postscript printer, the following * Tk_Font font families should print correctly: * * Avant Garde, Arial, Bookman, Courier, Courier New, Geneva, * Helvetica, Monaco, New Century Schoolbook, New York, * Palatino, Symbol, Times, Times New Roman, Zapf Chancery, * and Zapf Dingbats. * * Any other Tk_Font font families may not print correctly because the * computed Postscript font name may be incorrect. * *--------------------------------------------------------------------------- */ int Tk_PostscriptFontName( Tk_Font tkfont, /* Font in which text will be printed. */ Tcl_DString *dsPtr) /* Pointer to an initialized Tcl_DString to * which the name of the Postscript font that * corresponds to tkfont will be appended. */ { TkFont *fontPtr = (TkFont *) tkfont; Tk_Uid family, weightString, slantString; char *src, *dest; int upper, len; len = Tcl_DStringLength(dsPtr); /* * Convert the case-insensitive Tk_Font family name to the case-sensitive * Postscript family name. Take out any spaces and capitalize the first * letter of each word. */ family = fontPtr->fa.family; if (strncasecmp(family, "itc ", 4) == 0) { family = family + 4; } if ((strcasecmp(family, "Arial") == 0) || (strcasecmp(family, "Geneva") == 0)) { family = "Helvetica"; } else if ((strcasecmp(family, "Times New Roman") == 0) || (strcasecmp(family, "New York") == 0)) { family = "Times"; } else if ((strcasecmp(family, "Courier New") == 0) || (strcasecmp(family, "Monaco") == 0)) { family = "Courier"; } else if (strcasecmp(family, "AvantGarde") == 0) { family = "AvantGarde"; } else if (strcasecmp(family, "ZapfChancery") == 0) { family = "ZapfChancery"; } else if (strcasecmp(family, "ZapfDingbats") == 0) { family = "ZapfDingbats"; } else { int ch; /* * Inline, capitalize the first letter of each word, lowercase the * rest of the letters in each word, and then take out the spaces * between the words. This may make the DString shorter, which is safe * to do. */ Tcl_DStringAppend(dsPtr, family, -1); src = dest = Tcl_DStringValue(dsPtr) + len; upper = 1; for (; *src != '\0'; ) { while (isspace(UCHAR(*src))) { /* INTL: ISO space */ src++; upper = 1; } src += TkUtfToUniChar(src, &ch); if (upper) { ch = Tcl_UniCharToUpper(ch); upper = 0; } else { ch = Tcl_UniCharToLower(ch); } dest += TkUniCharToUtf(ch, dest); } *dest = '\0'; Tcl_DStringSetLength(dsPtr, dest - Tcl_DStringValue(dsPtr)); family = Tcl_DStringValue(dsPtr) + len; } if (family != Tcl_DStringValue(dsPtr) + len) { Tcl_DStringAppend(dsPtr, family, -1); family = Tcl_DStringValue(dsPtr) + len; } if (strcasecmp(family, "NewCenturySchoolbook") == 0) { Tcl_DStringSetLength(dsPtr, len); Tcl_DStringAppend(dsPtr, "NewCenturySchlbk", -1); family = Tcl_DStringValue(dsPtr) + len; } /* * Get the string to use for the weight. */ weightString = NULL; if (fontPtr->fa.weight == TK_FW_NORMAL) { if (strcmp(family, "Bookman") == 0) { weightString = "Light"; } else if (strcmp(family, "AvantGarde") == 0) { weightString = "Book"; } else if (strcmp(family, "ZapfChancery") == 0) { weightString = "Medium"; } } else { if ((strcmp(family, "Bookman") == 0) || (strcmp(family, "AvantGarde") == 0)) { weightString = "Demi"; } else { weightString = "Bold"; } } /* * Get the string to use for the slant. */ slantString = NULL; if (fontPtr->fa.slant == TK_FS_ROMAN) { /* Do nothing */ } else if ((strcmp(family, "Helvetica") == 0) || (strcmp(family, "Courier") == 0) || (strcmp(family, "AvantGarde") == 0)) { slantString = "Oblique"; } else { slantString = "Italic"; } /* * The string "Roman" needs to be added to some fonts that are not bold * and not italic. */ if ((slantString == NULL) && (weightString == NULL)) { if ((strcmp(family, "Times") == 0) || (strcmp(family, "NewCenturySchlbk") == 0) || (strcmp(family, "Palatino") == 0)) { Tcl_DStringAppend(dsPtr, "-Roman", -1); } } else { Tcl_DStringAppend(dsPtr, "-", -1); if (weightString != NULL) { Tcl_DStringAppend(dsPtr, weightString, -1); } if (slantString != NULL) { Tcl_DStringAppend(dsPtr, slantString, -1); } } return fontPtr->fa.size; } /* *--------------------------------------------------------------------------- * * Tk_TextWidth -- * * A wrapper function for the more complicated interface of * Tk_MeasureChars. Computes how much space the given simple string * needs. * * Results: * The return value is the width (in pixels) of the given string. * * Side effects: * None. * *--------------------------------------------------------------------------- */ int Tk_TextWidth( Tk_Font tkfont, /* Font in which text will be measured. */ const char *string, /* String whose width will be computed. */ int numBytes) /* Number of bytes to consider from string, or * < 0 for strlen(). */ { int width; if (numBytes < 0) { numBytes = strlen(string); } Tk_MeasureChars(tkfont, string, numBytes, -1, 0, &width); return width; } /* *--------------------------------------------------------------------------- * * Tk_UnderlineChars, TkUnderlineCharsInContext -- * * These procedures draw an underline for a given range of characters in * a given string. They don't draw the characters (which are assumed to * have been displayed previously); they just draw the underline. These * procedures would mainly be used to quickly underline a few characters * without having to construct an underlined font. To produce properly * underlined text, the appropriate underlined font should be constructed * and used. * * Results: * None. * * Side effects: * Information gets displayed in "drawable". * *---------------------------------------------------------------------- */ void Tk_UnderlineChars( Display *display, /* Display on which to draw. */ Drawable drawable, /* Window or pixmap in which to draw. */ GC gc, /* Graphics context for actually drawing * line. */ Tk_Font tkfont, /* Font used in GC; must have been allocated * by Tk_GetFont(). Used for character * dimensions, etc. */ const char *string, /* String containing characters to be * underlined or overstruck. */ int x, int y, /* Coordinates at which first character of * string is drawn. */ int firstByte, /* Index of first byte of first character. */ int lastByte) /* Index of first byte after the last * character. */ { TkUnderlineCharsInContext(display, drawable, gc, tkfont, string, lastByte, x, y, firstByte, lastByte); } void TkUnderlineCharsInContext( Display *display, /* Display on which to draw. */ Drawable drawable, /* Window or pixmap in which to draw. */ GC gc, /* Graphics context for actually drawing * line. */ Tk_Font tkfont, /* Font used in GC; must have been allocated * by Tk_GetFont(). Used for character * dimensions, etc. */ const char *string, /* String containing characters to be * underlined or overstruck. */ int numBytes, /* Number of bytes in string. */ int x, int y, /* Coordinates at which the first character of * the whole string would be drawn. */ int firstByte, /* Index of first byte of first character. */ int lastByte) /* Index of first byte after the last * character. */ { TkFont *fontPtr = (TkFont *) tkfont; int startX, endX; TkpMeasureCharsInContext(tkfont, string, numBytes, 0, firstByte, -1, 0, &startX); TkpMeasureCharsInContext(tkfont, string, numBytes, 0, lastByte, -1, 0, &endX); XFillRectangle(display, drawable, gc, x + startX, y + fontPtr->underlinePos, (unsigned) (endX - startX), (unsigned) fontPtr->underlineHeight); } /* *--------------------------------------------------------------------------- * * Tk_ComputeTextLayout -- * * Computes the amount of screen space needed to display a multi-line, * justified string of text. Records all the measurements that were done * to determine to size and positioning of the individual lines of text; * this information can be used by the Tk_DrawTextLayout() function to * display the text quickly (without remeasuring it). * * This function is useful for simple widgets that want to display * single-font, multi-line text and want Tk to handle the details. * * Results: * The return value is a Tk_TextLayout token that holds the measurement * information for the given string. The token is only valid for the * given string. If the string is freed, the token is no longer valid and * must also be freed. To free the token, call Tk_FreeTextLayout(). * * The dimensions of the screen area needed to display the text are * stored in *widthPtr and *heightPtr. * * Side effects: * Memory is allocated to hold the measurement information. * *--------------------------------------------------------------------------- */ Tk_TextLayout Tk_ComputeTextLayout( Tk_Font tkfont, /* Font that will be used to display text. */ const char *string, /* String whose dimensions are to be * computed. */ int numChars, /* Number of characters to consider from * string, or < 0 for strlen(). */ int wrapLength, /* Longest permissible line length, in pixels. * <= 0 means no automatic wrapping: just let * lines get as long as needed. */ Tk_Justify justify, /* How to justify lines. */ int flags, /* Flag bits OR-ed together. TK_IGNORE_TABS * means that tab characters should not be * expanded. TK_IGNORE_NEWLINES means that * newline characters should not cause a line * break. */ int *widthPtr, /* Filled with width of string. */ int *heightPtr) /* Filled with height of string. */ { TkFont *fontPtr = (TkFont *) tkfont; const char *start, *end, *special; int n, y, bytesThisChunk, maxChunks, curLine, layoutHeight; int baseline, height, curX, newX, maxWidth, *lineLengths; TextLayout *layoutPtr; LayoutChunk *chunkPtr; const TkFontMetrics *fmPtr; Tcl_DString lineBuffer; Tcl_DStringInit(&lineBuffer); if ((fontPtr == NULL) || (string == NULL)) { if (widthPtr != NULL) { *widthPtr = 0; } if (heightPtr != NULL) { *heightPtr = 0; } return NULL; } fmPtr = &fontPtr->fm; height = fmPtr->ascent + fmPtr->descent; if (numChars < 0) { numChars = Tcl_NumUtfChars(string, -1); } if (wrapLength == 0) { wrapLength = -1; } maxChunks = 1; layoutPtr = ckalloc(sizeof(TextLayout) + (maxChunks-1) * sizeof(LayoutChunk)); layoutPtr->tkfont = tkfont; layoutPtr->string = string; layoutPtr->numChunks = 0; baseline = fmPtr->ascent; maxWidth = 0; /* * Divide the string up into simple strings and measure each string. */ curX = 0; end = Tcl_UtfAtIndex(string, numChars); special = string; flags &= TK_IGNORE_TABS | TK_IGNORE_NEWLINES; flags |= TK_WHOLE_WORDS | TK_AT_LEAST_ONE; for (start = string; start < end; ) { if (start >= special) { /* * Find the next special character in the string. * * INTL: Note that it is safe to increment by byte, because we are * looking for 7-bit characters that will appear unchanged in * UTF-8. At some point we may need to support the full Unicode * whitespace set. */ for (special = start; special < end; special++) { if (!(flags & TK_IGNORE_NEWLINES)) { if ((*special == '\n') || (*special == '\r')) { break; } } if (!(flags & TK_IGNORE_TABS)) { if (*special == '\t') { break; } } } } /* * Special points at the next special character (or the end of the * string). Process characters between start and special. */ chunkPtr = NULL; if (start < special) { bytesThisChunk = Tk_MeasureChars(tkfont, start, special - start, wrapLength - curX, flags, &newX); newX += curX; flags &= ~TK_AT_LEAST_ONE; if (bytesThisChunk > 0) { chunkPtr = NewChunk(&layoutPtr, &maxChunks, start, bytesThisChunk, curX, newX, baseline); start += bytesThisChunk; curX = newX; } } if ((start == special) && (special < end)) { /* * Handle the special character. * * INTL: Special will be pointing at a 7-bit character so we can * safely treat it as a single byte. */ chunkPtr = NULL; if (*special == '\t') { newX = curX + fontPtr->tabWidth; newX -= newX % fontPtr->tabWidth; NewChunk(&layoutPtr, &maxChunks, start, 1, curX, newX, baseline)->numDisplayChars = -1; start++; curX = newX; flags &= ~TK_AT_LEAST_ONE; if ((start < end) && ((wrapLength <= 0) || (newX <= wrapLength))) { /* * More chars can still fit on this line. */ continue; } } else { NewChunk(&layoutPtr, &maxChunks, start, 1, curX, curX, baseline)->numDisplayChars = -1; start++; goto wrapLine; } } /* * No more characters are going to go on this line, either because no * more characters can fit or there are no more characters left. * Consume all extra spaces at end of line. */ while ((start < end) && isspace(UCHAR(*start))) { /* INTL: ISO space */ if (!(flags & TK_IGNORE_NEWLINES)) { if ((*start == '\n') || (*start == '\r')) { break; } } if (!(flags & TK_IGNORE_TABS)) { if (*start == '\t') { break; } } start++; } if (chunkPtr != NULL) { const char *end; /* * Append all the extra spaces on this line to the end of the last * text chunk. This is a little tricky because we are switching * back and forth between characters and bytes. */ end = chunkPtr->start + chunkPtr->numBytes; bytesThisChunk = start - end; if (bytesThisChunk > 0) { bytesThisChunk = Tk_MeasureChars(tkfont, end, bytesThisChunk, -1, 0, &chunkPtr->totalWidth); chunkPtr->numBytes += bytesThisChunk; chunkPtr->numChars += Tcl_NumUtfChars(end, bytesThisChunk); chunkPtr->totalWidth += curX; } } wrapLine: flags |= TK_AT_LEAST_ONE; /* * Save current line length, then move current position to start of * next line. */ if (curX > maxWidth) { maxWidth = curX; } /* * Remember width of this line, so that all chunks on this line can be * centered or right justified, if necessary. */ Tcl_DStringAppend(&lineBuffer, (char *) &curX, sizeof(curX)); curX = 0; baseline += height; } /* * If last line ends with a newline, then we need to make a 0 width chunk * on the next line. Otherwise "Hello" and "Hello\n" are the same height. */ if ((layoutPtr->numChunks > 0) && !(flags & TK_IGNORE_NEWLINES)) { if (layoutPtr->chunks[layoutPtr->numChunks - 1].start[0] == '\n') { chunkPtr = NewChunk(&layoutPtr, &maxChunks, start, 0, curX, curX, baseline); chunkPtr->numDisplayChars = -1; Tcl_DStringAppend(&lineBuffer, (char *) &curX, sizeof(curX)); baseline += height; } } layoutPtr->width = maxWidth; layoutHeight = baseline - fmPtr->ascent; if (layoutPtr->numChunks == 0) { layoutHeight = height; /* * This fake chunk is used by the other functions so that they can * pretend that there is a chunk with no chars in it, which makes the * coding simpler. */ layoutPtr->numChunks = 1; layoutPtr->chunks[0].start = string; layoutPtr->chunks[0].numBytes = 0; layoutPtr->chunks[0].numChars = 0; layoutPtr->chunks[0].numDisplayChars = -1; layoutPtr->chunks[0].x = 0; layoutPtr->chunks[0].y = fmPtr->ascent; layoutPtr->chunks[0].totalWidth = 0; layoutPtr->chunks[0].displayWidth = 0; } else { /* * Using maximum line length, shift all the chunks so that the lines * are all justified correctly. */ curLine = 0; chunkPtr = layoutPtr->chunks; y = chunkPtr->y; lineLengths = (int *) Tcl_DStringValue(&lineBuffer); for (n = 0; n < layoutPtr->numChunks; n++) { int extra; if (chunkPtr->y != y) { curLine++; y = chunkPtr->y; } extra = maxWidth - lineLengths[curLine]; if (justify == TK_JUSTIFY_CENTER) { chunkPtr->x += extra / 2; } else if (justify == TK_JUSTIFY_RIGHT) { chunkPtr->x += extra; } chunkPtr++; } } if (widthPtr != NULL) { *widthPtr = layoutPtr->width; } if (heightPtr != NULL) { *heightPtr = layoutHeight; } Tcl_DStringFree(&lineBuffer); return (Tk_TextLayout) layoutPtr; } /* *--------------------------------------------------------------------------- * * Tk_FreeTextLayout -- * * This function is called to release the storage associated with a * Tk_TextLayout when it is no longer needed. * * Results: * None. * * Side effects: * Memory is freed. * *--------------------------------------------------------------------------- */ void Tk_FreeTextLayout( Tk_TextLayout textLayout) /* The text layout to be released. */ { TextLayout *layoutPtr = (TextLayout *) textLayout; if (layoutPtr != NULL) { ckfree(layoutPtr); } } /* *--------------------------------------------------------------------------- * * Tk_DrawTextLayout -- * * Use the information in the Tk_TextLayout token to display a * multi-line, justified string of text. * * This function is useful for simple widgets that need to display * single-font, multi-line text and want Tk to handle the details. * * Results: * None. * * Side effects: * Text drawn on the screen. * *--------------------------------------------------------------------------- */ void Tk_DrawTextLayout( Display *display, /* Display on which to draw. */ Drawable drawable, /* Window or pixmap in which to draw. */ GC gc, /* Graphics context to use for drawing * text. */ Tk_TextLayout layout, /* Layout information, from a previous call to * Tk_ComputeTextLayout(). */ int x, int y, /* Upper-left hand corner of rectangle in * which to draw (pixels). */ int firstChar, /* The index of the first character to draw * from the given text item. 0 specfies the * beginning. */ int lastChar) /* The index just after the last character to * draw from the given text item. A number < 0 * means to draw all characters. */ { TextLayout *layoutPtr = (TextLayout *) layout; int i, numDisplayChars, drawX; const char *firstByte, *lastByte; LayoutChunk *chunkPtr; if (layoutPtr == NULL) { return; } if (lastChar < 0) { lastChar = 100000000; } chunkPtr = layoutPtr->chunks; for (i = 0; i < layoutPtr->numChunks; i++) { numDisplayChars = chunkPtr->numDisplayChars; if ((numDisplayChars > 0) && (firstChar < numDisplayChars)) { if (firstChar <= 0) { drawX = 0; firstChar = 0; firstByte = chunkPtr->start; } else { firstByte = Tcl_UtfAtIndex(chunkPtr->start, firstChar); Tk_MeasureChars(layoutPtr->tkfont, chunkPtr->start, firstByte - chunkPtr->start, -1, 0, &drawX); } if (lastChar < numDisplayChars) { numDisplayChars = lastChar; } lastByte = Tcl_UtfAtIndex(chunkPtr->start, numDisplayChars); Tk_DrawChars(display, drawable, gc, layoutPtr->tkfont, firstByte, lastByte - firstByte, x+chunkPtr->x+drawX, y+chunkPtr->y); } firstChar -= chunkPtr->numChars; lastChar -= chunkPtr->numChars; if (lastChar <= 0) { break; } chunkPtr++; } } void TkDrawAngledTextLayout( Display *display, /* Display on which to draw. */ Drawable drawable, /* Window or pixmap in which to draw. */ GC gc, /* Graphics context to use for drawing * text. */ Tk_TextLayout layout, /* Layout information, from a previous call to * Tk_ComputeTextLayout(). */ int x, int y, /* Upper-left hand corner of rectangle in * which to draw (pixels). */ double angle, int firstChar, /* The index of the first character to draw * from the given text item. 0 specfies the * beginning. */ int lastChar) /* The index just after the last character to * draw from the given text item. A number < 0 * means to draw all characters. */ { TextLayout *layoutPtr = (TextLayout *) layout; int i, numDisplayChars, drawX; const char *firstByte, *lastByte; LayoutChunk *chunkPtr; double sinA = sin(angle * PI/180.0), cosA = cos(angle * PI/180.0); if (layoutPtr == NULL) { return; } if (lastChar < 0) { lastChar = 100000000; } chunkPtr = layoutPtr->chunks; for (i = 0; i < layoutPtr->numChunks; i++) { numDisplayChars = chunkPtr->numDisplayChars; if ((numDisplayChars > 0) && (firstChar < numDisplayChars)) { double dx, dy; if (firstChar <= 0) { drawX = 0; firstChar = 0; firstByte = chunkPtr->start; } else { firstByte = Tcl_UtfAtIndex(chunkPtr->start, firstChar); Tk_MeasureChars(layoutPtr->tkfont, chunkPtr->start, firstByte - chunkPtr->start, -1, 0, &drawX); } if (lastChar < numDisplayChars) { numDisplayChars = lastChar; } lastByte = Tcl_UtfAtIndex(chunkPtr->start, numDisplayChars); dx = cosA * (chunkPtr->x + drawX) + sinA * (chunkPtr->y); dy = -sinA * (chunkPtr->x + drawX) + cosA * (chunkPtr->y); if (angle == 0.0) { Tk_DrawChars(display, drawable, gc, layoutPtr->tkfont, firstByte, lastByte - firstByte, (int)(x + dx), (int)(y + dy)); } else { TkDrawAngledChars(display, drawable, gc, layoutPtr->tkfont, firstByte, lastByte - firstByte, x+dx, y+dy, angle); } } firstChar -= chunkPtr->numChars; lastChar -= chunkPtr->numChars; if (lastChar <= 0) { break; } chunkPtr++; } } /* *--------------------------------------------------------------------------- * * Tk_UnderlineTextLayout -- * * Use the information in the Tk_TextLayout token to display an underline * below an individual character. This function does not draw the text, * just the underline. * * This function is useful for simple widgets that need to display * single-font, multi-line text with an individual character underlined * and want Tk to handle the details. To display larger amounts of * underlined text, construct and use an underlined font. * * Results: * None. * * Side effects: * Underline drawn on the screen. * *--------------------------------------------------------------------------- */ void Tk_UnderlineTextLayout( Display *display, /* Display on which to draw. */ Drawable drawable, /* Window or pixmap in which to draw. */ GC gc, /* Graphics context to use for drawing text. */ Tk_TextLayout layout, /* Layout information, from a previous call to * Tk_ComputeTextLayout(). */ int x, int y, /* Upper-left hand corner of rectangle in * which to draw (pixels). */ int underline) /* Index of the single character to underline, * or -1 for no underline. */ { int xx, yy, width, height; if ((Tk_CharBbox(layout, underline, &xx, &yy, &width, &height) != 0) && (width != 0)) { TextLayout *layoutPtr = (TextLayout *) layout; TkFont *fontPtr = (TkFont *) layoutPtr->tkfont; XFillRectangle(display, drawable, gc, x + xx, y + yy + fontPtr->fm.ascent + fontPtr->underlinePos, (unsigned) width, (unsigned) fontPtr->underlineHeight); } } void TkUnderlineAngledTextLayout( Display *display, /* Display on which to draw. */ Drawable drawable, /* Window or pixmap in which to draw. */ GC gc, /* Graphics context to use for drawing * text. */ Tk_TextLayout layout, /* Layout information, from a previous call to * Tk_ComputeTextLayout(). */ int x, int y, /* Upper-left hand corner of rectangle in * which to draw (pixels). */ double angle, int underline) /* Index of the single character to underline, * or -1 for no underline. */ { int xx, yy, width, height; if (angle == 0.0) { Tk_UnderlineTextLayout(display, drawable, gc, layout, x,y, underline); return; } if ((Tk_CharBbox(layout, underline, &xx, &yy, &width, &height) != 0) && (width != 0)) { TextLayout *layoutPtr = (TextLayout *) layout; TkFont *fontPtr = (TkFont *) layoutPtr->tkfont; double sinA = sin(angle*PI/180), cosA = cos(angle*PI/180); double dy = yy + fontPtr->fm.ascent + fontPtr->underlinePos; XPoint points[5]; /* * Note that we're careful to only round a double value once, which * minimizes roundoff errors. */ points[0].x = x + ROUND16(xx*cosA + dy*sinA); points[0].y = y + ROUND16(dy*cosA - xx*sinA); points[1].x = x + ROUND16(xx*cosA + dy*sinA + width*cosA); points[1].y = y + ROUND16(dy*cosA - xx*sinA - width*sinA); if (fontPtr->underlineHeight == 1) { /* * Thin underlines look better when rotated when drawn as a line * rather than a rectangle; the rasterizer copes better. */ XDrawLines(display, drawable, gc, points, 2, CoordModeOrigin); } else { points[2].x = x + ROUND16(xx*cosA + dy*sinA + width*cosA + fontPtr->underlineHeight*sinA); points[2].y = y + ROUND16(dy*cosA - xx*sinA - width*sinA + fontPtr->underlineHeight*cosA); points[3].x = x + ROUND16(xx*cosA + dy*sinA + fontPtr->underlineHeight*sinA); points[3].y = y + ROUND16(dy*cosA - xx*sinA + fontPtr->underlineHeight*cosA); points[4].x = points[0].x; points[4].y = points[0].y; XFillPolygon(display, drawable, gc, points, 5, Complex, CoordModeOrigin); XDrawLines(display, drawable, gc, points, 5, CoordModeOrigin); } } } /* *--------------------------------------------------------------------------- * * Tk_PointToChar -- * * Use the information in the Tk_TextLayout token to determine the * character closest to the given point. The point must be specified with * respect to the upper-left hand corner of the text layout, which is * considered to be located at (0, 0). * * Any point whose y-value is less that 0 will be considered closest to * the first character in the text layout; any point whose y-value is * greater than the height of the text layout will be considered closest * to the last character in the text layout. * * Any point whose x-value is less than 0 will be considered closest to * the first character on that line; any point whose x-value is greater * than the width of the text layout will be considered closest to the * last character on that line. * * Results: * The return value is the index of the character that was closest to the * point. Given a text layout with no characters, the value 0 will always * be returned, referring to a hypothetical zero-width placeholder * character. * * Side effects: * None. * *--------------------------------------------------------------------------- */ int Tk_PointToChar( Tk_TextLayout layout, /* Layout information, from a previous call to * Tk_ComputeTextLayout(). */ int x, int y) /* Coordinates of point to check, with respect * to the upper-left corner of the text * layout. */ { TextLayout *layoutPtr = (TextLayout *) layout; LayoutChunk *chunkPtr, *lastPtr; TkFont *fontPtr; int i, n, dummy, baseline, pos, numChars; if (y < 0) { /* * Point lies above any line in this layout. Return the index of the * first char. */ return 0; } /* * Find which line contains the point. */ fontPtr = (TkFont *) layoutPtr->tkfont; lastPtr = chunkPtr = layoutPtr->chunks; numChars = 0; for (i = 0; i < layoutPtr->numChunks; i++) { baseline = chunkPtr->y; if (y < baseline + fontPtr->fm.descent) { if (x < chunkPtr->x) { /* * Point is to the left of all chunks on this line. Return the * index of the first character on this line. */ return numChars; } if (x >= layoutPtr->width) { /* * If point lies off right side of the text layout, return the * last char in the last chunk on this line. Without this, it * might return the index of the first char that was located * outside of the text layout. */ x = INT_MAX; } /* * Examine all chunks on this line to see which one contains the * specified point. */ lastPtr = chunkPtr; while ((i < layoutPtr->numChunks) && (chunkPtr->y == baseline)) { if (x < chunkPtr->x + chunkPtr->totalWidth) { /* * Point falls on one of the characters in this chunk. */ if (chunkPtr->numDisplayChars < 0) { /* * This is a special chunk that encapsulates a single * tab or newline char. */ return numChars; } n = Tk_MeasureChars((Tk_Font) fontPtr, chunkPtr->start, chunkPtr->numBytes, x - chunkPtr->x, 0, &dummy); return numChars + Tcl_NumUtfChars(chunkPtr->start, n); } numChars += chunkPtr->numChars; lastPtr = chunkPtr; chunkPtr++; i++; } /* * Point is to the right of all chars in all the chunks on this * line. Return the index just past the last char in the last * chunk on this line. */ pos = numChars; if (i < layoutPtr->numChunks) { pos--; } return pos; } numChars += chunkPtr->numChars; lastPtr = chunkPtr; chunkPtr++; } /* * Point lies below any line in this text layout. Return the index just * past the last char. */ return (lastPtr->start + lastPtr->numChars) - layoutPtr->string; } /* *--------------------------------------------------------------------------- * * Tk_CharBbox -- * * Use the information in the Tk_TextLayout token to return the bounding * box for the character specified by index. * * The width of the bounding box is the advance width of the character, * and does not include and left- or right-bearing. Any character that * extends partially outside of the text layout is considered to be * truncated at the edge. Any character which is located completely * outside of the text layout is considered to be zero-width and pegged * against the edge. * * The height of the bounding box is the line height for this font, * extending from the top of the ascent to the bottom of the descent. * Information about the actual height of the individual letter is not * available. * * A text layout that contains no characters is considered to contain a * single zero-width placeholder character. * * Results: * The return value is 0 if the index did not specify a character in the * text layout, or non-zero otherwise. In that case, *bbox is filled with * the bounding box of the character. * * Side effects: * None. * *--------------------------------------------------------------------------- */ int Tk_CharBbox( Tk_TextLayout layout, /* Layout information, from a previous call to * Tk_ComputeTextLayout(). */ int index, /* The index of the character whose bbox is * desired. */ int *xPtr, int *yPtr, /* Filled with the upper-left hand corner, in * pixels, of the bounding box for the * character specified by index, if * non-NULL. */ int *widthPtr, int *heightPtr) /* Filled with the width and height of the * bounding box for the character specified by * index, if non-NULL. */ { TextLayout *layoutPtr = (TextLayout *) layout; LayoutChunk *chunkPtr; int i, x = 0, w; Tk_Font tkfont; TkFont *fontPtr; const char *end; if (index < 0) { return 0; } chunkPtr = layoutPtr->chunks; tkfont = layoutPtr->tkfont; fontPtr = (TkFont *) tkfont; for (i = 0; i < layoutPtr->numChunks; i++) { if (chunkPtr->numDisplayChars < 0) { if (index == 0) { x = chunkPtr->x; w = chunkPtr->totalWidth; goto check; } } else if (index < chunkPtr->numChars) { end = Tcl_UtfAtIndex(chunkPtr->start, index); if (xPtr != NULL) { Tk_MeasureChars(tkfont, chunkPtr->start, end - chunkPtr->start, -1, 0, &x); x += chunkPtr->x; } if (widthPtr != NULL) { Tk_MeasureChars(tkfont, end, Tcl_UtfNext(end) - end, -1, 0, &w); } goto check; } index -= chunkPtr->numChars; chunkPtr++; } if (index != 0) { return 0; } /* * Special case to get location just past last char in layout. */ chunkPtr--; x = chunkPtr->x + chunkPtr->totalWidth; w = 0; /* * Ensure that the bbox lies within the text layout. This forces all chars * that extend off the right edge of the text layout to have truncated * widths, and all chars that are completely off the right edge of the * text layout to peg to the edge and have 0 width. */ check: if (yPtr != NULL) { *yPtr = chunkPtr->y - fontPtr->fm.ascent; } if (heightPtr != NULL) { *heightPtr = fontPtr->fm.ascent + fontPtr->fm.descent; } if (x > layoutPtr->width) { x = layoutPtr->width; } if (xPtr != NULL) { *xPtr = x; } if (widthPtr != NULL) { if (x + w > layoutPtr->width) { w = layoutPtr->width - x; } *widthPtr = w; } return 1; } /* *--------------------------------------------------------------------------- * * Tk_DistanceToTextLayout -- * * Computes the distance in pixels from the given point to the given text * layout. Non-displaying space characters that occur at the end of * individual lines in the text layout are ignored for hit detection * purposes. * * Results: * The return value is 0 if the point (x, y) is inside the text layout. * If the point isn't inside the text layout then the return value is the * distance in pixels from the point to the text item. * * Side effects: * None. * *--------------------------------------------------------------------------- */ int Tk_DistanceToTextLayout( Tk_TextLayout layout, /* Layout information, from a previous call * to Tk_ComputeTextLayout(). */ int x, int y) /* Coordinates of point to check, with respect * to the upper-left corner of the text layout * (in pixels). */ { int i, x1, x2, y1, y2, xDiff, yDiff, dist, minDist, ascent, descent; TextLayout *layoutPtr = (TextLayout *) layout; LayoutChunk *chunkPtr; TkFont *fontPtr; fontPtr = (TkFont *) layoutPtr->tkfont; ascent = fontPtr->fm.ascent; descent = fontPtr->fm.descent; minDist = 0; chunkPtr = layoutPtr->chunks; for (i = 0; i < layoutPtr->numChunks; i++) { if (chunkPtr->start[0] == '\n') { /* * Newline characters are not counted when computing distance (but * tab characters would still be considered). */ chunkPtr++; continue; } x1 = chunkPtr->x; y1 = chunkPtr->y - ascent; x2 = chunkPtr->x + chunkPtr->displayWidth; y2 = chunkPtr->y + descent; if (x < x1) { xDiff = x1 - x; } else if (x >= x2) { xDiff = x - x2 + 1; } else { xDiff = 0; } if (y < y1) { yDiff = y1 - y; } else if (y >= y2) { yDiff = y - y2 + 1; } else { yDiff = 0; } if ((xDiff == 0) && (yDiff == 0)) { return 0; } dist = (int) hypot((double) xDiff, (double) yDiff); if ((dist < minDist) || (minDist == 0)) { minDist = dist; } chunkPtr++; } return minDist; } /* *--------------------------------------------------------------------------- * * Tk_IntersectTextLayout -- * * Determines whether a text layout lies entirely inside, entirely * outside, or overlaps a given rectangle. Non-displaying space * characters that occur at the end of individual lines in the text * layout are ignored for intersection calculations. * * Results: * The return value is -1 if the text layout is entirely outside of the * rectangle, 0 if it overlaps, and 1 if it is entirely inside of the * rectangle. * * Side effects: * None. * *--------------------------------------------------------------------------- */ int Tk_IntersectTextLayout( Tk_TextLayout layout, /* Layout information, from a previous call to * Tk_ComputeTextLayout(). */ int x, int y, /* Upper-left hand corner, in pixels, of * rectangular area to compare with text * layout. Coordinates are with respect to the * upper-left hand corner of the text layout * itself. */ int width, int height) /* The width and height of the above * rectangular area, in pixels. */ { int result, i, x1, y1, x2, y2; TextLayout *layoutPtr = (TextLayout *) layout; LayoutChunk *chunkPtr; TkFont *fontPtr; int left, top, right, bottom; /* * Scan the chunks one at a time, seeing whether each is entirely in, * entirely out, or overlapping the rectangle. If an overlap is detected, * return immediately; otherwise wait until all chunks have been processed * and see if they were all inside or all outside. */ chunkPtr = layoutPtr->chunks; fontPtr = (TkFont *) layoutPtr->tkfont; left = x; top = y; right = x + width; bottom = y + height; result = 0; for (i = 0; i < layoutPtr->numChunks; i++) { if ((chunkPtr->start[0] == '\n') || (chunkPtr->numBytes == 0)) { /* * Newline characters and empty chunks are not counted when * computing area intersection (but tab characters would still be * considered). */ chunkPtr++; continue; } x1 = chunkPtr->x; y1 = chunkPtr->y - fontPtr->fm.ascent; x2 = chunkPtr->x + chunkPtr->displayWidth; y2 = chunkPtr->y + fontPtr->fm.descent; if ((right < x1) || (left >= x2) || (bottom < y1) || (top >= y2)) { if (result == 1) { return 0; } result = -1; } else if ((x1 < left) || (x2 >= right) || (y1 < top) || (y2 >= bottom)) { return 0; } else if (result == -1) { return 0; } else { result = 1; } chunkPtr++; } return result; } /* *--------------------------------------------------------------------------- * * TkIntersectAngledTextLayout -- * * Determines whether a text layout that has been turned by an angle * about its top-left coordinae lies entirely inside, entirely outside, * or overlaps a given rectangle. Non-displaying space characters that * occur at the end of individual lines in the text layout are ignored * for intersection calculations. * * Results: * The return value is -1 if the text layout is entirely outside of the * rectangle, 0 if it overlaps, and 1 if it is entirely inside of the * rectangle. * * Side effects: * None. * *--------------------------------------------------------------------------- */ static inline int PointInQuadrilateral( double qx[], double qy[], double x, double y) { int i; for (i=0 ; i<4 ; i++) { double sideDX = qx[(i+1)%4] - qx[i]; double sideDY = qy[(i+1)%4] - qy[i]; double dx = x - qx[i]; double dy = y - qy[i]; if (sideDX*dy < sideDY*dx) { return 0; } } return 1; } static inline int SidesIntersect( double ax1, double ay1, double ax2, double ay2, double bx1, double by1, double bx2, double by2) { #if 0 /* http://www.freelunchdesign.com/cgi-bin/codwiki.pl?DiscussionTopics/CollideMeUpBaby */ double a1, b1, c1, a2, b2, c2, r1, r2, r3, r4, denom; a1 = ay2 - ay1; b1 = ax1 - ax2; c1 = (ax2 * ay1) - (ax1 * ay2); r3 = (a1 * bx1) + (b1 * by1) + c1; r4 = (a1 * bx2) + (b1 * by2) + c1; if ((r3 != 0.0) && (r4 != 0.0) && (r3*r4 > 0.0)) { return 0; } a2 = by2 - by1; b2 = bx1 - bx2; c2 = (bx2 * by1) - (bx1 * by2); r1 = (a2 * ax1) + (b2 * ay1) + c2; r2 = (a2 * ax2) + (b2 * ay2) + c2; if ((r1 != 0.0) && (r2 != 0.0) && (r1*r2 > 0.0)) { return 0; } denom = (a1 * b2) - (a2 * b1); return (denom != 0.0); #else /* * A more efficient version. Two line segments intersect if, when seen * from the perspective of one line, the two endpoints of the other * segment lie on opposite sides of the line, and vice versa. "Lie on * opposite sides" is computed by taking the cross products and seeing if * they are of opposite signs. */ double dx, dy, dx1, dy1; dx = ax2 - ax1; dy = ay2 - ay1; dx1 = bx1 - ax1; dy1 = by1 - ay1; if ((dx*dy1-dy*dx1 > 0.0) == (dx*(by2-ay1)-dy*(bx2-ax1) > 0.0)) { return 0; } dx = bx2 - bx1; dy = by2 - by1; if ((dy*dx1-dx*dy1 > 0.0) == (dx*(ay2-by1)-dy*(ax2-bx1) > 0.0)) { return 0; } return 1; #endif } int TkIntersectAngledTextLayout( Tk_TextLayout layout, /* Layout information, from a previous call to * Tk_ComputeTextLayout(). */ int x, int y, /* Upper-left hand corner, in pixels, of * rectangular area to compare with text * layout. Coordinates are with respect to the * upper-left hand corner of the text layout * itself. */ int width, int height, /* The width and height of the above * rectangular area, in pixels. */ double angle) { int i, x1, y1, x2, y2; TextLayout *layoutPtr; LayoutChunk *chunkPtr; TkFont *fontPtr; double c = cos(angle * PI/180.0), s = sin(angle * PI/180.0); double rx[4], ry[4]; if (angle == 0.0) { return Tk_IntersectTextLayout(layout, x, y, width, height); } /* * Compute the coordinates of the rectangle, rotated into text layout * space. */ rx[0] = x*c - y*s; ry[0] = y*c + x*s; rx[1] = (x+width)*c - y*s; ry[1] = y*c + (x+width)*s; rx[2] = (x+width)*c - (y+height)*s; ry[2] = (y+height)*c + (x+width)*s; rx[3] = x*c - (y+height)*s; ry[3] = (y+height)*c + x*s; /* * Want to know if all chunks are inside the rectangle, or if there is any * overlap. First, we check to see if all chunks are inside; if and only * if they are, we're in the "inside" case. */ layoutPtr = (TextLayout *) layout; chunkPtr = layoutPtr->chunks; fontPtr = (TkFont *) layoutPtr->tkfont; for (i=0 ; i<layoutPtr->numChunks ; i++,chunkPtr++) { if (chunkPtr->start[0] == '\n') { /* * Newline characters are not counted when computing area * intersection (but tab characters would still be considered). */ continue; } x1 = chunkPtr->x; y1 = chunkPtr->y - fontPtr->fm.ascent; x2 = chunkPtr->x + chunkPtr->displayWidth; y2 = chunkPtr->y + fontPtr->fm.descent; if ( !PointInQuadrilateral(rx, ry, x1, y1) || !PointInQuadrilateral(rx, ry, x2, y1) || !PointInQuadrilateral(rx, ry, x2, y2) || !PointInQuadrilateral(rx, ry, x1, y2)) { goto notInside; } } return 1; /* * Next, check to see if all the points of the rectangle are inside a * single chunk; if they are, we're in an "overlap" case. */ notInside: chunkPtr = layoutPtr->chunks; for (i=0 ; i<layoutPtr->numChunks ; i++,chunkPtr++) { double cx[4], cy[4]; if (chunkPtr->start[0] == '\n') { /* * Newline characters are not counted when computing area * intersection (but tab characters would still be considered). */ continue; } cx[0] = cx[3] = chunkPtr->x; cy[0] = cy[1] = chunkPtr->y - fontPtr->fm.ascent; cx[1] = cx[2] = chunkPtr->x + chunkPtr->displayWidth; cy[2] = cy[3] = chunkPtr->y + fontPtr->fm.descent; if ( !PointInQuadrilateral(cx, cy, rx[0], ry[0]) || !PointInQuadrilateral(cx, cy, rx[1], ry[1]) || !PointInQuadrilateral(cx, cy, rx[2], ry[2]) || !PointInQuadrilateral(cx, cy, rx[3], ry[3])) { goto notReverseInside; } } return 0; /* * If we're overlapping now, we must be partially in and out of at least * one chunk. If that is the case, there must be one line segment of the * rectangle that is touching or crossing a line segment of a chunk. */ notReverseInside: chunkPtr = layoutPtr->chunks; for (i=0 ; i<layoutPtr->numChunks ; i++,chunkPtr++) { int j; if (chunkPtr->start[0] == '\n') { /* * Newline characters are not counted when computing area * intersection (but tab characters would still be considered). */ continue; } x1 = chunkPtr->x; y1 = chunkPtr->y - fontPtr->fm.ascent; x2 = chunkPtr->x + chunkPtr->displayWidth; y2 = chunkPtr->y + fontPtr->fm.descent; for (j=0 ; j<4 ; j++) { int k = (j+1) % 4; if ( SidesIntersect(rx[j],ry[j], rx[k],ry[k], x1,y1, x2,y1) || SidesIntersect(rx[j],ry[j], rx[k],ry[k], x2,y1, x2,y2) || SidesIntersect(rx[j],ry[j], rx[k],ry[k], x2,y2, x1,y2) || SidesIntersect(rx[j],ry[j], rx[k],ry[k], x1,y2, x1,y1)) { return 0; } } } /* * They must be wholly non-overlapping. */ return -1; } /* *--------------------------------------------------------------------------- * * Tk_TextLayoutToPostscript -- * * Outputs the contents of a text layout in Postscript format. The set of * lines in the text layout will be rendered by the user supplied * Postscript function. The function should be of the form: * * justify x y string function -- * * Justify is -1, 0, or 1, depending on whether the following string * should be left, center, or right justified, x and y is the location * for the origin of the string, string is the sequence of characters to * be printed, and function is the name of the caller-provided function; * the function should leave nothing on the stack. * * The meaning of the origin of the string (x and y) depends on the * justification. For left justification, x is where the left edge of the * string should appear. For center justification, x is where the center * of the string should appear. And for right justification, x is where * the right edge of the string should appear. This behavior is necessary * because, for example, right justified text on the screen is justified * with screen metrics. The same string needs to be justified with * printer metrics on the printer to appear in the correct place with * respect to other similarly justified strings. In all circumstances, y * is the location of the baseline for the string. * * Results: * The interp's result is modified to hold the Postscript code that will * render the text layout. * * Side effects: * None. * *--------------------------------------------------------------------------- */ void Tk_TextLayoutToPostscript( Tcl_Interp *interp, /* Filled with Postscript code. */ Tk_TextLayout layout) /* The layout to be rendered. */ { TextLayout *layoutPtr = (TextLayout *) layout; LayoutChunk *chunkPtr = layoutPtr->chunks; int baseline = chunkPtr->y; Tcl_Obj *psObj = Tcl_NewObj(); int i, j, len; const char *p, *glyphname; char uindex[5], c, *ps; int ch; Tcl_AppendToObj(psObj, "[(", -1); for (i = 0; i < layoutPtr->numChunks; i++, chunkPtr++) { if (baseline != chunkPtr->y) { Tcl_AppendToObj(psObj, ")]\n[(", -1); baseline = chunkPtr->y; } if (chunkPtr->numDisplayChars <= 0) { if (chunkPtr->start[0] == '\t') { Tcl_AppendToObj(psObj, "\\t", -1); } continue; } for (p=chunkPtr->start, j=0; j<chunkPtr->numDisplayChars; j++) { /* * INTL: We only handle symbols that have an encoding as a glyph * from the standard set defined by Adobe. The rest get punted. * Eventually this should be revised to handle more sophsticiated * international postscript fonts. */ p += TkUtfToUniChar(p, &ch); if ((ch == '(') || (ch == ')') || (ch == '\\') || (ch < 0x20)) { /* * Tricky point: the "03" is necessary in the sprintf below, * so that a full three digits of octal are always generated. * Without the "03", a number following this sequence could be * interpreted by Postscript as part of this sequence. */ Tcl_AppendPrintfToObj(psObj, "\\%03o", ch); continue; } else if (ch <= 0x7f) { /* * Normal ASCII character. */ c = (char) ch; Tcl_AppendToObj(psObj, &c, 1); continue; } /* * This character doesn't belong to the ASCII character set, so we * use the full glyph name. */ if (ch > 0xffff) { goto noMapping; } sprintf(uindex, "%04X", ch); /* endianness? */ glyphname = Tcl_GetVar2(interp, "::tk::psglyphs", uindex, 0); if (glyphname) { ps = Tcl_GetStringFromObj(psObj, &len); if (ps[len-1] == '(') { /* * In-place edit. Ewww! */ ps[len-1] = '/'; } else { Tcl_AppendToObj(psObj, ")/", -1); } Tcl_AppendToObj(psObj, glyphname, -1); Tcl_AppendToObj(psObj, "(", -1); } else { /* * No known mapping for the character into the space of * PostScript glyphs. Ignore it. :-( */ noMapping: ; #ifdef TK_DEBUG_POSTSCRIPT_OUTPUT fprintf(stderr, "Warning: no mapping to PostScript " "glyphs for \\u%04x\n", ch); #endif } } } Tcl_AppendToObj(psObj, ")]\n", -1); Tcl_AppendObjToObj(Tcl_GetObjResult(interp), psObj); Tcl_DecrRefCount(psObj); } /* *--------------------------------------------------------------------------- * * ConfigAttributesObj -- * * Process command line options to fill in fields of a properly * initialized font attributes structure. * * Results: * A standard Tcl return value. If TCL_ERROR is returned, an error * message will be left in interp's result object. * * Side effects: * The fields of the font attributes structure get filled in with * information from argc/argv. If an error occurs while parsing, the font * attributes structure will contain all modifications specified in the * command line options up to the point of the error. * *--------------------------------------------------------------------------- */ static int ConfigAttributesObj( Tcl_Interp *interp, /* Interp for error return. */ Tk_Window tkwin, /* For display on which font will be used. */ int objc, /* Number of elements in argv. */ Tcl_Obj *const objv[], /* Command line options. */ TkFontAttributes *faPtr) /* Font attributes structure whose fields are * to be modified. Structure must already be * properly initialized. */ { int i, n, index; Tcl_Obj *optionPtr, *valuePtr; const char *value; for (i = 0; i < objc; i += 2) { optionPtr = objv[i]; if (Tcl_GetIndexFromObj(interp, optionPtr, fontOpt, "option", 1, &index) != TCL_OK) { return TCL_ERROR; } if ((i+2 >= objc) && (objc & 1)) { /* * This test occurs after Tcl_GetIndexFromObj() so that "font * create xyz -xyz" will return the error message that "-xyz" is a * bad option, rather than that the value for "-xyz" is missing. */ if (interp != NULL) { Tcl_SetObjResult(interp, Tcl_ObjPrintf( "value for \"%s\" option missing", Tcl_GetString(optionPtr))); Tcl_SetErrorCode(interp, "TK", "FONT", "NO_ATTRIBUTE", NULL); } return TCL_ERROR; } valuePtr = objv[i + 1]; switch (index) { case FONT_FAMILY: value = Tcl_GetString(valuePtr); faPtr->family = Tk_GetUid(value); break; case FONT_SIZE: if (Tcl_GetIntFromObj(interp, valuePtr, &n) != TCL_OK) { return TCL_ERROR; } faPtr->size = n; break; case FONT_WEIGHT: n = TkFindStateNumObj(interp, optionPtr, weightMap, valuePtr); if (n == TK_FW_UNKNOWN) { return TCL_ERROR; } faPtr->weight = n; break; case FONT_SLANT: n = TkFindStateNumObj(interp, optionPtr, slantMap, valuePtr); if (n == TK_FS_UNKNOWN) { return TCL_ERROR; } faPtr->slant = n; break; case FONT_UNDERLINE: if (Tcl_GetBooleanFromObj(interp, valuePtr, &n) != TCL_OK) { return TCL_ERROR; } faPtr->underline = n; break; case FONT_OVERSTRIKE: if (Tcl_GetBooleanFromObj(interp, valuePtr, &n) != TCL_OK) { return TCL_ERROR; } faPtr->overstrike = n; break; } } return TCL_OK; } /* *--------------------------------------------------------------------------- * * GetAttributeInfoObj -- * * Return information about the font attributes as a Tcl list. * * Results: * The return value is TCL_OK if the objPtr was non-NULL and specified a * valid font attribute, TCL_ERROR otherwise. If TCL_OK is returned, the * interp's result object is modified to hold a description of either the * current value of a single option, or a list of all options and their * current values for the given font attributes. If TCL_ERROR is * returned, the interp's result is set to an error message describing * that the objPtr did not refer to a valid option. * * Side effects: * None. * *--------------------------------------------------------------------------- */ static int GetAttributeInfoObj( Tcl_Interp *interp, /* Interp to hold result. */ const TkFontAttributes *faPtr, /* The font attributes to inspect. */ Tcl_Obj *objPtr) /* If non-NULL, indicates the single option * whose value is to be returned. Otherwise * information is returned for all options. */ { int i, index, start, end; const char *str; Tcl_Obj *valuePtr, *resultPtr = NULL; start = 0; end = FONT_NUMFIELDS; if (objPtr != NULL) { if (Tcl_GetIndexFromObj(interp, objPtr, fontOpt, "option", TCL_EXACT, &index) != TCL_OK) { return TCL_ERROR; } start = index; end = index + 1; } valuePtr = NULL; if (objPtr == NULL) { resultPtr = Tcl_NewObj(); } for (i = start; i < end; i++) { switch (i) { case FONT_FAMILY: str = faPtr->family; valuePtr = Tcl_NewStringObj(str, ((str == NULL) ? 0 : -1)); break; case FONT_SIZE: valuePtr = Tcl_NewIntObj(faPtr->size); break; case FONT_WEIGHT: str = TkFindStateString(weightMap, faPtr->weight); valuePtr = Tcl_NewStringObj(str, -1); break; case FONT_SLANT: str = TkFindStateString(slantMap, faPtr->slant); valuePtr = Tcl_NewStringObj(str, -1); break; case FONT_UNDERLINE: valuePtr = Tcl_NewBooleanObj(faPtr->underline); break; case FONT_OVERSTRIKE: valuePtr = Tcl_NewBooleanObj(faPtr->overstrike); break; } if (objPtr != NULL) { Tcl_SetObjResult(interp, valuePtr); return TCL_OK; } Tcl_ListObjAppendElement(NULL, resultPtr, Tcl_NewStringObj(fontOpt[i], -1)); Tcl_ListObjAppendElement(NULL, resultPtr, valuePtr); } Tcl_SetObjResult(interp, resultPtr); return TCL_OK; } /* *--------------------------------------------------------------------------- * * ParseFontNameObj -- * * Converts a object into a set of font attributes that can be used to * construct a font. * * The string rep of the object can be one of the following forms: * XLFD (see X documentation) * "family [size] [style1 [style2 ...]" * "-option value [-option value ...]" * * Results: * The return value is TCL_ERROR if the object was syntactically invalid. * In that case an error message is left in interp's result object. * Otherwise, fills the font attribute buffer with the values parsed from * the string and returns TCL_OK; * * Side effects: * None. * *--------------------------------------------------------------------------- */ static int ParseFontNameObj( Tcl_Interp *interp, /* Interp for error return. */ Tk_Window tkwin, /* For display on which font is used. */ Tcl_Obj *objPtr, /* Parseable font description object. */ TkFontAttributes *faPtr) /* Filled with attributes parsed from font * name. Any attributes that were not * specified in font name are filled with * default values. */ { char *dash; int objc, result, i, n; Tcl_Obj **objv; const char *string; TkInitFontAttributes(faPtr); string = Tcl_GetString(objPtr); if (*string == '-') { /* * This may be an XLFD or an "-option value" string. * * If the string begins with "-*" or a "-foundry-family-*" pattern, * then consider it an XLFD. */ if (string[1] == '*') { goto xlfd; } dash = strchr(string + 1, '-'); if ((dash != NULL) && !isspace(UCHAR(dash[-1]))) { /* INTL: ISO space */ goto xlfd; } if (Tcl_ListObjGetElements(interp, objPtr, &objc, &objv) != TCL_OK) { return TCL_ERROR; } return ConfigAttributesObj(interp, tkwin, objc, objv, faPtr); } if (*string == '*') { /* * This is appears to be an XLFD. Under Unix, all valid XLFDs were * already handled by TkpGetNativeFont. If we are here, either we have * something that initially looks like an XLFD but isn't or we have * encountered an XLFD on Windows or Mac. */ xlfd: result = TkFontParseXLFD(string, faPtr, NULL); if (result == TCL_OK) { return TCL_OK; } /* * If the string failed to parse but was considered to be a XLFD * then it may be a "-option value" string with a hyphenated family * name as per bug 2791352 */ if (Tcl_ListObjGetElements(interp, objPtr, &objc, &objv) != TCL_OK) { return TCL_ERROR; } if (ConfigAttributesObj(interp, tkwin, objc, objv, faPtr) == TCL_OK) { return TCL_OK; } } /* * Wasn't an XLFD or "-option value" string. Try it as a "font size style" * list. */ if ((Tcl_ListObjGetElements(NULL, objPtr, &objc, &objv) != TCL_OK) || (objc < 1)) { if (interp != NULL) { Tcl_SetObjResult(interp, Tcl_ObjPrintf( "font \"%s\" doesn't exist", string)); Tcl_SetErrorCode(interp, "TK", "LOOKUP", "FONT", string, NULL); } return TCL_ERROR; } faPtr->family = Tk_GetUid(Tcl_GetString(objv[0])); if (objc > 1) { if (Tcl_GetIntFromObj(interp, objv[1], &n) != TCL_OK) { return TCL_ERROR; } faPtr->size = n; } i = 2; if (objc == 3) { if (Tcl_ListObjGetElements(interp, objv[2], &objc, &objv) != TCL_OK) { return TCL_ERROR; } i = 0; } for ( ; i < objc; i++) { n = TkFindStateNumObj(NULL, NULL, weightMap, objv[i]); if (n != TK_FW_UNKNOWN) { faPtr->weight = n; continue; } n = TkFindStateNumObj(NULL, NULL, slantMap, objv[i]); if (n != TK_FS_UNKNOWN) { faPtr->slant = n; continue; } n = TkFindStateNumObj(NULL, NULL, underlineMap, objv[i]); if (n != 0) { faPtr->underline = n; continue; } n = TkFindStateNumObj(NULL, NULL, overstrikeMap, objv[i]); if (n != 0) { faPtr->overstrike = n; continue; } /* * Unknown style. */ if (interp != NULL) { Tcl_SetObjResult(interp, Tcl_ObjPrintf( "unknown font style \"%s\"", Tcl_GetString(objv[i]))); Tcl_SetErrorCode(interp, "TK", "LOOKUP", "FONT_STYLE", Tcl_GetString(objv[i]), NULL); } return TCL_ERROR; } return TCL_OK; } /* *--------------------------------------------------------------------------- * * NewChunk -- * * Helper function for Tk_ComputeTextLayout(). Encapsulates a measured * set of characters in a chunk that can be quickly drawn. * * Results: * A pointer to the new chunk in the text layout. * * Side effects: * The text layout is reallocated to hold more chunks as necessary. * * Currently, Tk_ComputeTextLayout() stores contiguous ranges of "normal" * characters in a chunk, along with individual tab and newline chars in * their own chunks. All characters in the text layout are accounted for. * *--------------------------------------------------------------------------- */ static LayoutChunk * NewChunk( TextLayout **layoutPtrPtr, int *maxPtr, const char *start, int numBytes, int curX, int newX, int y) { TextLayout *layoutPtr; LayoutChunk *chunkPtr; int maxChunks, numChars; size_t s; layoutPtr = *layoutPtrPtr; maxChunks = *maxPtr; if (layoutPtr->numChunks == maxChunks) { maxChunks *= 2; s = sizeof(TextLayout) + ((maxChunks - 1) * sizeof(LayoutChunk)); layoutPtr = ckrealloc(layoutPtr, s); *layoutPtrPtr = layoutPtr; *maxPtr = maxChunks; } numChars = Tcl_NumUtfChars(start, numBytes); chunkPtr = &layoutPtr->chunks[layoutPtr->numChunks]; chunkPtr->start = start; chunkPtr->numBytes = numBytes; chunkPtr->numChars = numChars; chunkPtr->numDisplayChars = numChars; chunkPtr->x = curX; chunkPtr->y = y; chunkPtr->totalWidth = newX - curX; chunkPtr->displayWidth = newX - curX; layoutPtr->numChunks++; return chunkPtr; } /* *--------------------------------------------------------------------------- * * TkFontParseXLFD -- * * Break up a fully specified XLFD into a set of font attributes. * * Results: * Return value is TCL_ERROR if string was not a fully specified XLFD. * Otherwise, fills font attribute buffer with the values parsed from the * XLFD and returns TCL_OK. * * Side effects: * None. * *--------------------------------------------------------------------------- */ int TkFontParseXLFD( const char *string, /* Parseable font description string. */ TkFontAttributes *faPtr, /* Filled with attributes parsed from font * name. Any attributes that were not * specified in font name are filled with * default values. */ TkXLFDAttributes *xaPtr) /* Filled with X-specific attributes parsed * from font name. Any attributes that were * not specified in font name are filled with * default values. May be NULL if such * information is not desired. */ { char *src; const char *str; int i, j; char *field[XLFD_NUMFIELDS + 2]; Tcl_DString ds; TkXLFDAttributes xa; if (xaPtr == NULL) { xaPtr = &xa; } TkInitFontAttributes(faPtr); TkInitXLFDAttributes(xaPtr); memset(field, '\0', sizeof(field)); str = string; if (*str == '-') { str++; } Tcl_DStringInit(&ds); Tcl_DStringAppend(&ds, str, -1); src = Tcl_DStringValue(&ds); field[0] = src; for (i = 0; *src != '\0'; src++) { if (!(*src & 0x80) && Tcl_UniCharIsUpper(UCHAR(*src))) { *src = (char) Tcl_UniCharToLower(UCHAR(*src)); } if (*src == '-') { i++; if (i == XLFD_NUMFIELDS) { continue; } *src = '\0'; field[i] = src + 1; if (i > XLFD_NUMFIELDS) { break; } } } /* * An XLFD of the form -adobe-times-medium-r-*-12-*-* is pretty common, * but it is (strictly) malformed, because the first * is eliding both the * Setwidth and the Addstyle fields. If the Addstyle field is a number, * then assume the above incorrect form was used and shift all the rest of * the fields right by one, so the number gets interpreted as a pixelsize. * This fix is so that we don't get a million reports that "it works under * X (as a native font name), but gives a syntax error under Windows (as a * parsed set of attributes)". */ if ((i > XLFD_ADD_STYLE) && FieldSpecified(field[XLFD_ADD_STYLE])) { if (atoi(field[XLFD_ADD_STYLE]) != 0) { for (j = XLFD_NUMFIELDS - 1; j >= XLFD_ADD_STYLE; j--) { field[j + 1] = field[j]; } field[XLFD_ADD_STYLE] = NULL; i++; } } /* * Bail if we don't have enough of the fields (up to pointsize). */ if (i < XLFD_FAMILY) { Tcl_DStringFree(&ds); return TCL_ERROR; } if (FieldSpecified(field[XLFD_FOUNDRY])) { xaPtr->foundry = Tk_GetUid(field[XLFD_FOUNDRY]); } if (FieldSpecified(field[XLFD_FAMILY])) { faPtr->family = Tk_GetUid(field[XLFD_FAMILY]); } if (FieldSpecified(field[XLFD_WEIGHT])) { faPtr->weight = TkFindStateNum(NULL, NULL, xlfdWeightMap, field[XLFD_WEIGHT]); } if (FieldSpecified(field[XLFD_SLANT])) { xaPtr->slant = TkFindStateNum(NULL, NULL, xlfdSlantMap, field[XLFD_SLANT]); if (xaPtr->slant == TK_FS_ROMAN) { faPtr->slant = TK_FS_ROMAN; } else { faPtr->slant = TK_FS_ITALIC; } } if (FieldSpecified(field[XLFD_SETWIDTH])) { xaPtr->setwidth = TkFindStateNum(NULL, NULL, xlfdSetwidthMap, field[XLFD_SETWIDTH]); } /* XLFD_ADD_STYLE ignored. */ /* * Pointsize in tenths of a point, but treat it as tenths of a pixel for * historical compatibility. */ faPtr->size = 12; if (FieldSpecified(field[XLFD_POINT_SIZE])) { if (field[XLFD_POINT_SIZE][0] == '[') { /* * Some X fonts have the point size specified as follows: * * [ N1 N2 N3 N4 ] * * where N1 is the point size (in points, not decipoints!), and * N2, N3, and N4 are some additional numbers that I don't know * the purpose of, so I ignore them. */ faPtr->size = atoi(field[XLFD_POINT_SIZE] + 1); } else if (Tcl_GetInt(NULL, field[XLFD_POINT_SIZE], &faPtr->size) == TCL_OK) { faPtr->size /= 10; } else { return TCL_ERROR; } } /* * Pixel height of font. If specified, overrides pointsize. */ if (FieldSpecified(field[XLFD_PIXEL_SIZE])) { if (field[XLFD_PIXEL_SIZE][0] == '[') { /* * Some X fonts have the pixel size specified as follows: * * [ N1 N2 N3 N4 ] * * where N1 is the pixel size, and where N2, N3, and N4 are some * additional numbers that I don't know the purpose of, so I * ignore them. */ faPtr->size = atoi(field[XLFD_PIXEL_SIZE] + 1); } else if (Tcl_GetInt(NULL, field[XLFD_PIXEL_SIZE], &faPtr->size) != TCL_OK) { return TCL_ERROR; } } faPtr->size = -faPtr->size; /* XLFD_RESOLUTION_X ignored. */ /* XLFD_RESOLUTION_Y ignored. */ /* XLFD_SPACING ignored. */ /* XLFD_AVERAGE_WIDTH ignored. */ if (FieldSpecified(field[XLFD_CHARSET])) { xaPtr->charset = Tk_GetUid(field[XLFD_CHARSET]); } else { xaPtr->charset = Tk_GetUid("iso8859-1"); } Tcl_DStringFree(&ds); return TCL_OK; } /* *--------------------------------------------------------------------------- * * FieldSpecified -- * * Helper function for TkParseXLFD(). Determines if a field in the XLFD * was set to a non-null, non-don't-care value. * * Results: * The return value is 0 if the field in the XLFD was not set and should * be ignored, non-zero otherwise. * * Side effects: * None. * *--------------------------------------------------------------------------- */ static int FieldSpecified( const char *field) /* The field of the XLFD to check. Strictly * speaking, only when the string is "*" does * it mean don't-care. However, an unspecified * or question mark is also interpreted as * don't-care. */ { char ch; if (field == NULL) { return 0; } ch = field[0]; return (ch != '*' && ch != '?'); } /* *--------------------------------------------------------------------------- * * TkFontGetPixels -- * * Given a font size specification (as described in the TkFontAttributes * structure) return the number of pixels it represents. * * Results: * As above. * * Side effects: * None. * *--------------------------------------------------------------------------- */ int TkFontGetPixels( Tk_Window tkwin, /* For point->pixel conversion factor. */ int size) /* Font size. */ { double d; if (size < 0) { return -size; } d = size * 25.4 / 72.0; d *= WidthOfScreen(Tk_Screen(tkwin)); d /= WidthMMOfScreen(Tk_Screen(tkwin)); return (int) (d + 0.5); } /* *--------------------------------------------------------------------------- * * TkFontGetPoints -- * * Given a font size specification (as described in the TkFontAttributes * structure) return the number of points it represents. * * Results: * As above. * * Side effects: * None. * *--------------------------------------------------------------------------- */ int TkFontGetPoints( Tk_Window tkwin, /* For pixel->point conversion factor. */ int size) /* Font size. */ { double d; if (size >= 0) { return size; } d = -size * 72.0 / 25.4; d *= WidthMMOfScreen(Tk_Screen(tkwin)); d /= WidthOfScreen(Tk_Screen(tkwin)); return (int) (d + 0.5); } /* *------------------------------------------------------------------------- * * TkFontGetAliasList -- * * Given a font name, find the list of all aliases for that font name. * One of the names in this list will probably be the name that this * platform expects when asking for the font. * * Results: * As above. The return value is NULL if the font name has no aliases. * * Side effects: * None. * *------------------------------------------------------------------------- */ const char *const * TkFontGetAliasList( const char *faceName) /* Font name to test for aliases. */ { int i, j; for (i = 0; fontAliases[i] != NULL; i++) { for (j = 0; fontAliases[i][j] != NULL; j++) { if (strcasecmp(faceName, fontAliases[i][j]) == 0) { return fontAliases[i]; } } } return NULL; } /* *------------------------------------------------------------------------- * * TkFontGetFallbacks -- * * Get the list of font fallbacks that the platform-specific code can use * to try to find the closest matching font the name requested. * * Results: * As above. * * Side effects: * None. * *------------------------------------------------------------------------- */ const char *const *const * TkFontGetFallbacks(void) { return fontFallbacks; } /* *------------------------------------------------------------------------- * * TkFontGetGlobalClass -- * * Get the list of fonts to try if the requested font name does not * exist and no fallbacks for that font name could be used either. * The names in this list are considered preferred over all the other * font names in the system when looking for a last-ditch fallback. * * Results: * As above. * * Side effects: * None. * *------------------------------------------------------------------------- */ const char *const * TkFontGetGlobalClass(void) { return globalFontClass; } /* *------------------------------------------------------------------------- * * TkFontGetSymbolClass -- * * Get the list of fonts that are symbolic; used if the operating system * cannot apriori identify symbolic fonts on its own. * * Results: * As above. * * Side effects: * None. * *------------------------------------------------------------------------- */ const char *const * TkFontGetSymbolClass(void) { return symbolClass; } /* *---------------------------------------------------------------------- * * TkDebugFont -- * * This function returns debugging information about a font. * * Results: * The return value is a list with one sublist for each TkFont * corresponding to "name". Each sublist has two elements that contain * the resourceRefCount and objRefCount fields from the TkFont structure. * * Side effects: * None. * *---------------------------------------------------------------------- */ Tcl_Obj * TkDebugFont( Tk_Window tkwin, /* The window in which the font will be used * (not currently used). */ const char *name) /* Name of the desired color. */ { TkFont *fontPtr; Tcl_HashEntry *hashPtr; Tcl_Obj *resultPtr, *objPtr; resultPtr = Tcl_NewObj(); hashPtr = Tcl_FindHashEntry( &((TkWindow *) tkwin)->mainPtr->fontInfoPtr->fontCache, name); if (hashPtr != NULL) { fontPtr = Tcl_GetHashValue(hashPtr); if (fontPtr == NULL) { Tcl_Panic("TkDebugFont found empty hash table entry"); } for ( ; (fontPtr != NULL); fontPtr = fontPtr->nextPtr) { objPtr = Tcl_NewObj(); Tcl_ListObjAppendElement(NULL, objPtr, Tcl_NewIntObj(fontPtr->resourceRefCount)); Tcl_ListObjAppendElement(NULL, objPtr, Tcl_NewIntObj(fontPtr->objRefCount)); Tcl_ListObjAppendElement(NULL, resultPtr, objPtr); } } return resultPtr; } /* *---------------------------------------------------------------------- * * TkFontGetFirstTextLayout -- * * This function returns the first chunk of a Tk_TextLayout, i.e. until * the first font change on the first line (or the whole first line if * there is no such font change). * * Results: * The return value is the byte length of the chunk, the chunk itself is * copied into dst and its Tk_Font into font. * * Side effects: * None. * *---------------------------------------------------------------------- */ int TkFontGetFirstTextLayout( Tk_TextLayout layout, /* Layout information, from a previous call to * Tk_ComputeTextLayout(). */ Tk_Font *font, char *dst) { TextLayout *layoutPtr = (TextLayout *) layout; LayoutChunk *chunkPtr; int numBytesInChunk; if ((layoutPtr == NULL) || (layoutPtr->numChunks == 0) || (layoutPtr->chunks->numDisplayChars <= 0)) { dst[0] = '\0'; return 0; } chunkPtr = layoutPtr->chunks; numBytesInChunk = chunkPtr->numBytes; strncpy(dst, chunkPtr->start, (size_t) numBytesInChunk); *font = layoutPtr->tkfont; return numBytesInChunk; } /* * Local Variables: * mode: c * c-basic-offset: 4 * fill-column: 78 * End: */