From da0d7f61c851431d14430684c62345bc23dbf001 Mon Sep 17 00:00:00 2001
From: Sona Kurazyan <sona.kurazyan@qt.io>
Date: Fri, 17 Jun 2022 14:43:17 +0200
Subject: QDom: Stop treating non-BMP characters as invalid

According to https://www.w3.org/TR/REC-xml/#NT-Char unicode characters
within the range of [#x10000-#x10FFFF] are considered to be valid, so
fix the check for valid characters accordingly. This requires changing
the loop over the input QString to iterate over code points (instead of
code units).

Fixes: QTBUG-104362
Pick-to: 6.4 6.3 6.2 5.15
Change-Id: I7dcf5cad05265a54882807a50522d28b647e06ee
Reviewed-by: Marc Mutz <marc.mutz@qt.io>
---
 tests/auto/xml/dom/qdom/tst_qdom.cpp | 22 ++++++++++++++++++++++
 1 file changed, 22 insertions(+)

(limited to 'tests')

diff --git a/tests/auto/xml/dom/qdom/tst_qdom.cpp b/tests/auto/xml/dom/qdom/tst_qdom.cpp
index e1c2b12ab5..f05020f61c 100644
--- a/tests/auto/xml/dom/qdom/tst_qdom.cpp
+++ b/tests/auto/xml/dom/qdom/tst_qdom.cpp
@@ -9,6 +9,7 @@
 #include <QFile>
 #include <QList>
 #include <QRegularExpression>
+#include <QScopeGuard>
 #include <QTextStream>
 #include <QTest>
 #include <QtXml>
@@ -62,6 +63,7 @@ private slots:
     void invalidQualifiedName();
     void invalidCharData_data();
     void invalidCharData();
+    void nonBMPCharacters();
 
     void roundTripAttributes() const;
     void roundTripCDATA() const;
@@ -1342,6 +1344,10 @@ void tst_QDom::invalidCharData_data()
     QTest::newRow( "f<o&o" )   << QString("f<o&o")     << true  << true  << true  << QString("f<o&o");
     QTest::newRow( "empty" )   << QString()            << true  << true  << true  << QString();
     QTest::newRow("f\\x07o\\x02")<< QString("f\x07o\x02")<< true  << true  << false << QString("fo");
+
+    const QChar pair[2] = { QChar(0xdc00), QChar(0xe000) };
+    QString invalid(pair, 2);
+    QTest::newRow("\\xdc00\\xe000") << invalid << true << true << false << invalid.last(1);
 }
 
 void tst_QDom::invalidCharData()
@@ -1385,6 +1391,22 @@ void tst_QDom::invalidCharData()
     }
 }
 
+void tst_QDom::nonBMPCharacters()
+{
+    const auto invalidDataPolicy = QDomImplementation::invalidDataPolicy();
+    auto resetInvalidDataPolicy = qScopeGuard(
+            [invalidDataPolicy] { QDomImplementation::setInvalidDataPolicy(invalidDataPolicy); });
+    QDomImplementation::setInvalidDataPolicy(QDomImplementation::DropInvalidChars);
+
+    const QString input = u"<text>Supplementary Plane: 𝄞 😂 🀄 🀶 🃪 🃋</text>"_qs;
+
+    QString errorMsg;
+    QDomDocument doc;
+    doc.setContent(input, &errorMsg);
+    QVERIFY(errorMsg.isEmpty());
+    QCOMPARE(doc.toString(-1), input);
+}
+
 void tst_QDom::roundTripAttributes() const
 {
     /* Create an attribute via the QDom API with weird whitespace content. */
-- 
cgit v1.2.3