QDomDocument: Add a way to enable spacing-only text nodes

Added a parse option that can be passed to setContent(), to specify that spacing-only text nodes must be preserved. [ChangeLog][QtXml][QDomDocument] Spacing-only text nodes can now be preserved by passing the ParseOption::PreserveSpacingOnlyNodes option to setContent(). Fixes: QTBUG-104130 Fixes: QTBUG-89690 Task-number: QTBUG-90003 Change-Id: Id43730ce5b79a856c4b434d1f1d4dd7c49c25f31 Reviewed-by: Edward Welbourne <[email protected]> Reviewed-by: Thiago Macieira <[email protected]>
petrmanek · Aug 4, 2022 · 26a73e1 · 26a73e1
1 parent 7ceba9c
commit 26a73e1
Show file tree

Hide file tree

Showing 6 changed files with 87 additions and 8 deletions.
diff --git a/src/xml/doc/src/qt6-changes.qdoc b/src/xml/doc/src/qt6-changes.qdoc
@@ -73,6 +73,15 @@
     If you use QDomDocument and rely on any of these, you must update
     your code and XML documents accordingly.
 
+    \section3 Spacing-only text nodes
+
+    By default, text nodes containing only spacing characters are stripped
+    and won't appear in the QDomDocument. The Qt 5 way of changing this behavior
+    was using the QDomDocument::setContent() overload that allowed a \c QXmlReader
+    to be supplied. That overload was removed in Qt 6.0, but since Qt 6.5,
+    you can pass QDomDocument::ParseOption::PreserveSpacingOnlyNodes as a parse
+    option, to specify that spacing-only text nodes must be preserved.
+
     \section2 Qt Core5 compatibility library
 
     If your application or library cannot be ported right now, the \l

diff --git a/src/xml/dom/qdom.cpp b/src/xml/dom/qdom.cpp
@@ -6221,6 +6221,8 @@ bool QDomDocument::setContent(QXmlStreamReader *reader, bool namespaceProcessing
 
     \value Default No parse options are set.
     \value UseNamespaceProcessing Namespace processing is enabled.
+    \value PreserveSpacingOnlyNodes Text nodes containing only spacing
+           characters are preserved.
 
     \sa setContent()
 */
@@ -6307,7 +6309,9 @@ bool QDomDocument::setContent(QXmlStreamReader *reader, bool namespaceProcessing
     string if the element or attribute has no prefix.
 
     Text nodes consisting only of whitespace are stripped and won't
-    appear in the QDomDocument.
+    appear in the QDomDocument. Since Qt 6.5, one can pass
+    QDomDocument::ParseOption::PreserveSpacingOnlyNodes as a parse
+    option, to specify that spacing-only text nodes must be preserved.
 
     \include qdom.cpp entity-refs
 

diff --git a/src/xml/dom/qdom.h b/src/xml/dom/qdom.h
@@ -268,6 +268,7 @@ class Q_XML_EXPORT QDomDocument : public QDomNode
     enum class ParseOption {
         Default = 0x00,
         UseNamespaceProcessing = 0x01,
+        PreserveSpacingOnlyNodes = 0x02,
     };
     Q_DECLARE_FLAGS(ParseOptions, ParseOption)
 

diff --git a/src/xml/dom/qdomhelpers.cpp b/src/xml/dom/qdomhelpers.cpp
@@ -349,13 +349,14 @@ bool QDomParser::parseBody()
             }
             break;
         case QXmlStreamReader::Characters:
-            if (!reader->isWhitespace()) { // Skip the content consisting of only whitespaces
-                if (reader->isCDATA() || !reader->text().trimmed().isEmpty()) {
-                    if (!domBuilder.characters(reader->text().toString(), reader->isCDATA())) {
-                        domBuilder.fatalError(QDomParser::tr(
-                                "Error occurred while processing the element content"));
-                        return false;
-                    }
+            // Skip the content if it contains only spacing characters,
+            // unless it's CDATA or PreserveSpacingOnlyNodes was specified.
+            if (reader->isCDATA() || domBuilder.preserveSpacingOnlyNodes()
+                || !(reader->isWhitespace() || reader->text().trimmed().isEmpty())) {
+                if (!domBuilder.characters(reader->text().toString(), reader->isCDATA())) {
+                    domBuilder.fatalError(
+                            QDomParser::tr("Error occurred while processing the element content"));
+                    return false;
                 }
             }
             break;

diff --git a/src/xml/dom/qdomhelpers_p.h b/src/xml/dom/qdomhelpers_p.h
@@ -56,6 +56,9 @@ class QDomBuilder
     void fatalError(const QString &message);
     QDomDocument::ParseResult result() const { return parseResult; }
 
+    bool preserveSpacingOnlyNodes() const
+    { return parseOptions & QDomDocument::ParseOption::PreserveSpacingOnlyNodes; }
+
 private:
     QString dtdInternalSubset(const QString &dtd);
 

diff --git a/tests/auto/xml/dom/qdom/tst_qdom.cpp b/tests/auto/xml/dom/qdom/tst_qdom.cpp
@@ -32,6 +32,8 @@ private slots:
     void setContent();
     void setContentOverloads();
     void parseOptions();
+    void spacingOnlyNodes_data() const;
+    void spacingOnlyNodes() const;
     void parseResult();
     void toString_01_data();
     void toString_01();
@@ -292,6 +294,65 @@ void tst_QDom::parseOptions()
     }
 }
 
+void tst_QDom::spacingOnlyNodes_data() const
+{
+    QTest::addColumn<QString>("input");
+    QTest::addColumn<QString>("expected");
+    QTest::addColumn<QDomDocument::ParseOption>("options");
+
+    QTest::newRow("spacing-only-remove")
+            << u"<a> \t \n \r</a>"_s
+            << u"<a/>"_s
+            << QDomDocument::ParseOption::Default;
+    // \r is translated to \n, see https://www.w3.org/TR/xml11/#sec-line-ends
+    QTest::newRow("spacing-only-preserve")
+            << u"<a> \t \n \r</a>"_s
+            << u"<a> \t \n \n</a>"_s
+            << QDomDocument::ParseOption::PreserveSpacingOnlyNodes;
+    QTest::newRow("mixed-text-remove")
+            << u"<a> abc \t \n \r</a>"_s
+            << u"<a> abc \t \n \n</a>"_s
+            << QDomDocument::ParseOption::Default;
+    QTest::newRow("mixed-text-preserve")
+            << u"<a> abc \t \n \r</a>"_s
+            << u"<a> abc \t \n \n</a>"_s
+            << QDomDocument::ParseOption::PreserveSpacingOnlyNodes;
+
+    // QDomDocument treats all chacarcters below as spaces (see QTBUG-105348)
+    static constexpr char16_t spaces[] = {
+        QChar::Space, QChar::Tabulation, QChar::LineFeed,
+        QChar::CarriageReturn, QChar::Nbsp,
+        0x2002, // EN SPACE
+        0x2003, // EM SPACE
+        0x2009  // THIN SPACE
+    };
+
+    for (char16_t space : spaces) {
+        QTest::addRow("spacing-remove-u%04x", space)
+                << u"<a>"_s + space + u"</a>"_s
+                << u"<a/>"_s
+                << QDomDocument::ParseOption::Default;
+
+        // \r is translated to \n, see https://www.w3.org/TR/xml11/#sec-line-ends
+        char16_t expected = (space == QChar::CarriageReturn) ? char16_t(QChar::LineFeed) : space;
+        QTest::addRow("spacing-preserve-u%04x", space)
+                << u"<a>"_s + space + u"</a>"_s
+                << u"<a>"_s + expected + u"</a>"_s
+                << QDomDocument::ParseOption::PreserveSpacingOnlyNodes;
+    }
+}
+
+void tst_QDom::spacingOnlyNodes() const
+{
+    QFETCH(QString, input);
+    QFETCH(QString, expected);
+    QFETCH(QDomDocument::ParseOption, options);
+
+    QDomDocument doc;
+    QVERIFY(doc.setContent(input, options));
+    QCOMPARE(doc.toString(-1), expected);
+}
+
 void tst_QDom::parseResult()
 {
     QString input = u"<doc xmlns:b='http://example.com/'>"