diff --git a/src/corelib/serialization/qxmlstream.cpp b/src/corelib/serialization/qxmlstream.cpp index 598b9f0..885354f 100644 --- src/corelib/serialization/qxmlstream.cpp +++ src/corelib/serialization/qxmlstream.cpp @@ -126,7 +126,7 @@ addData() or by waiting for it to arrive on the device(). \value UnexpectedElementError The parser encountered an element - that was different to those it expected. + or token that was different to those it expected. */ @@ -263,13 +263,34 @@ QXmlStreamReader is a well-formed XML 1.0 parser that does \e not include external parsed entities. As long as no error occurs, the - application code can thus be assured that the data provided by the - stream reader satisfies the W3C's criteria for well-formed XML. For - example, you can be certain that all tags are indeed nested and - closed properly, that references to internal entities have been - replaced with the correct replacement text, and that attributes have - been normalized or added according to the internal subset of the - DTD. + application code can thus be assured, that + \list + \li the data provided by the stream reader satisfies the W3C's + criteria for well-formed XML, + \li tokens are provided in a valid order. + \endlist + + Unless QXmlStreamReader raises an error, it guarantees the following: + \list + \li All tags are nested and closed properly. + \li References to internal entities have been replaced with the + correct replacement text. + \li Attributes have been normalized or added according to the + internal subset of the \l DTD. + \li Tokens of type \l StartDocument happen before all others, + aside from comments and processing instructions. + \li At most one DOCTYPE element (a token of type \l DTD) is present. + \li If present, the DOCTYPE appears before all other elements, + aside from StartDocument, comments and processing instructions. + \endlist + + In particular, once any token of type \l StartElement, \l EndElement, + \l Characters, \l EntityReference or \l EndDocument is seen, no + tokens of type StartDocument or DTD will be seen. If one is present in + the input stream, out of order, an error is raised. + + \note The token types \l Comment and \l ProcessingInstruction may appear + anywhere in the stream. If an error occurs while parsing, atEnd() and hasError() return true, and error() returns the error that occurred. The functions @@ -572,6 +593,7 @@ d->token = -1; return readNext(); } + d->checkToken(); return d->type; } @@ -691,6 +713,13 @@ 0, 8, 16, 30, 42, 55, 66, 77, 85, 89, 105, 0 }; +static const char QXmlStreamReader_XmlContextString[] = + "Prolog\0" + "Body\0"; + +static const short QXmlStreamReader_XmlContextString_indices[] = { + 0, 7 +}; /*! \property QXmlStreamReader::namespaceProcessing @@ -727,6 +756,16 @@ QXmlStreamReader_tokenTypeString_indices[d->type]); } +/*! + \internal + \return \param ctxt (Prolog/Body) as a string. + */ +QString contextString(QXmlStreamReaderPrivate::XmlContext ctxt) +{ + return QLatin1String(QXmlStreamReader_XmlContextString + + QXmlStreamReader_XmlContextString_indices[static_cast(ctxt)]); +} + #endif // QT_NO_XMLSTREAMREADER QXmlStreamPrivateTagStack::QXmlStreamPrivateTagStack() @@ -813,6 +852,8 @@ type = QXmlStreamReader::NoToken; error = QXmlStreamReader::NoError; + currentContext = XmlContext::Prolog; + foundDTD = false; } /* @@ -3730,6 +3771,91 @@ } } +static constexpr bool isTokenAllowedInContext(QXmlStreamReader::TokenType type, + QXmlStreamReaderPrivate::XmlContext loc) +{ + switch (type) { + case QXmlStreamReader::StartDocument: + case QXmlStreamReader::DTD: + return loc == QXmlStreamReaderPrivate::XmlContext::Prolog; + + case QXmlStreamReader::StartElement: + case QXmlStreamReader::EndElement: + case QXmlStreamReader::Characters: + case QXmlStreamReader::EntityReference: + case QXmlStreamReader::EndDocument: + return loc == QXmlStreamReaderPrivate::XmlContext::Body; + + case QXmlStreamReader::Comment: + case QXmlStreamReader::ProcessingInstruction: + return true; + + case QXmlStreamReader::NoToken: + case QXmlStreamReader::Invalid: + return false; + } + return false; +} + +/*! + \internal + \brief QXmlStreamReader::isValidToken + \return \c true if \param type is a valid token type. + \return \c false if \param type is an unexpected token, + which indicates a non-well-formed or invalid XML stream. + */ +bool QXmlStreamReaderPrivate::isValidToken(QXmlStreamReader::TokenType type) +{ + // Don't change currentContext, if Invalid or NoToken occur in the prolog + if (type == QXmlStreamReader::Invalid || type == QXmlStreamReader::NoToken) + return false; + + // If a token type gets rejected in the body, there is no recovery + const bool result = isTokenAllowedInContext(type, currentContext); + if (result || currentContext == XmlContext::Body) + return result; + + // First non-Prolog token observed => switch context to body and check again. + currentContext = XmlContext::Body; + return isTokenAllowedInContext(type, currentContext); +} + +/*! + \internal + Checks token type and raises an error, if it is invalid + in the current context (prolog/body). + */ +void QXmlStreamReaderPrivate::checkToken() +{ + Q_Q(QXmlStreamReader); + + // The token type must be consumed, to keep track if the body has been reached. + const XmlContext context = currentContext; + const bool ok = isValidToken(type); + + // Do nothing if an error has been raised already (going along with an unexpected token) + if (error != QXmlStreamReader::Error::NoError) + return; + + if (!ok) { + raiseError(QXmlStreamReader::UnexpectedElementError, + QLatin1String("Unexpected token type %1 in %2.") + .arg(q->tokenString(), contextString(context))); + return; + } + + if (type != QXmlStreamReader::DTD) + return; + + // Raise error on multiple DTD tokens + if (foundDTD) { + raiseError(QXmlStreamReader::UnexpectedElementError, + QLatin1String("Found second DTD token in %1.").arg(contextString(context))); + } else { + foundDTD = true; + } +} + /*! \fn bool QXmlStreamAttributes::hasAttribute(const QString &qualifiedName) const \since 4.5 diff --git a/src/corelib/serialization/qxmlstream_p.h b/src/corelib/serialization/qxmlstream_p.h index a277753..c33f1cd 100644 --- src/corelib/serialization/qxmlstream_p.h +++ src/corelib/serialization/qxmlstream_p.h @@ -270,6 +270,17 @@ QStringDecoder decoder; bool atEnd; + enum class XmlContext + { + Prolog, + Body, + }; + + XmlContext currentContext = XmlContext::Prolog; + bool foundDTD = false; + bool isValidToken(QXmlStreamReader::TokenType type); + void checkToken(); + /*! \sa setType() */