Add support of high unicodes in QUrl.

Fix some typos in the test data (all 3 typos are already
fixed in the original testsuite).

Add incorrect utf-8 data (sub)test from nameprep testsuite.

Reviewed-by: Andreas
Merge-Request: 605

Merge-request: 605
Reviewed-by: Andreas Aardal Hanssen <andreas.aardal.hanssen@nokia.com>

2 files changed, 55 insertions, 13 deletions

diff --git a/tests/auto/qurl/tst_qurl.cpp b/tests/auto/qurl/tst_qurl.cpp
index d055b6b..b2b28bd 100644
--- a/tests/auto/qurl/tst_qurl.cpp
+++ b/tests/auto/qurl/tst_qurl.cpp
@@ -161,6 +161,8 @@ private slots:
     void idna_testsuite();
     void nameprep_testsuite_data();
     void nameprep_testsuite();
+    void nameprep_highcodes_data();
+    void nameprep_highcodes();
     void ace_testsuite_data();
     void ace_testsuite();
     void std3violations_data();
@@ -2912,7 +2914,6 @@ void tst_QUrl::nameprep_testsuite_data()
         << QString() << 0 << 0;
 
     QTest::newRow("Case folding 8bit U+00DF (german sharp s)")
-//        << QString::fromUtf8("\xC3\xDF") ### typo in the original testsuite
         << QString::fromUtf8("\xC3\x9F")
         << QString("ss")
         << QString() << 0 << 0;
@@ -2943,7 +2944,8 @@ void tst_QUrl::nameprep_testsuite_data()
         << QString() << 0 << 0;
 
     QTest::newRow("Self-reverting case folding U+01F0 and normalization")
-        << QString::fromUtf8("\xC7\xF0")
+//        << QString::fromUtf8("\xC7\xF0") ### typo in the original testsuite
+        << QString::fromUtf8("\xC7\xB0")
         << QString::fromUtf8("\xC7\xB0")
         << QString() << 0 << 0;
 
@@ -3118,13 +3120,13 @@ void tst_QUrl::nameprep_testsuite_data()
         << QString("Nameprep") << STRINGPREP_NO_UNASSIGNED << STRINGPREP_CONTAINS_UNASSIGNED;
 
     QTest::newRow("Larger test (shrinking)")
-        << QString::fromUtf8("X\xC2\xAD\xC3\xDF\xC4\xB0\xE2\x84\xA1\x6a\xcc\x8c\xc2\xa0\xc2"
+        << QString::fromUtf8("X\xC2\xAD\xC3\x9F\xC4\xB0\xE2\x84\xA1\x6a\xcc\x8c\xc2\xa0\xc2"
                              "\xaa\xce\xb0\xe2\x80\x80")
         << QString::fromUtf8("xssi\xcc\x87""tel\xc7\xb0 a\xce\xb0 ")
         << QString("Nameprep") << 0 << 0;
 
     QTest::newRow("Larger test (expanding)")
-        << QString::fromUtf8("X\xC3\xDF\xe3\x8c\x96\xC4\xB0\xE2\x84\xA1\xE2\x92\x9F\xE3\x8c\x80")
+        << QString::fromUtf8("X\xC3\x9F\xe3\x8c\x96\xC4\xB0\xE2\x84\xA1\xE2\x92\x9F\xE3\x8c\x80")
         << QString::fromUtf8("xss\xe3\x82\xad\xe3\x83\xad\xe3\x83\xa1\xe3\x83\xbc\xe3\x83\x88"
                              "\xe3\x83\xab""i\xcc\x87""tel\x28""d\x29\xe3\x82\xa2\xe3\x83\x91"
                              "\xe3\x83\xbc\xe3\x83\x88")
@@ -3145,20 +3147,58 @@ void tst_QUrl::nameprep_testsuite()
     QFETCH(QString, out);
     QFETCH(QString, profile);
 
-    QEXPECT_FAIL("Case folding U+2121 U+33C6 U+1D7BB",
-                 ">0xffff unicode points are not supported", Continue);
-    QEXPECT_FAIL("Self-reverting case folding U+01F0 and normalization",
-                 "Investigate further", Continue);
     QEXPECT_FAIL("Left-to-right mark U+200E",
                  "Investigate further", Continue);
     QEXPECT_FAIL("Deprecated U+202A",
                  "Investigate further", Continue);
     QEXPECT_FAIL("Language tagging character U+E0001",
                  "Investigate further", Continue);
-    QEXPECT_FAIL("Larger test (shrinking)",
-                 "Investigate further", Continue);
-    QEXPECT_FAIL("Larger test (expanding)",
-                 "Investigate further", Continue);
+    qt_nameprep(&in, 0);
+    QCOMPARE(in, out);
+#endif
+}
+
+void tst_QUrl::nameprep_highcodes_data()
+{
+    QTest::addColumn<QString>("in");
+    QTest::addColumn<QString>("out");
+    QTest::addColumn<QString>("profile");
+    QTest::addColumn<int>("flags");
+    QTest::addColumn<int>("rc");
+
+    {
+        QChar st[] = { '-', 0xd801, 0xdc1d, 'a' };
+        QChar se[] = { '-', 0xd801, 0xdc45, 'a' };
+        QTest::newRow("highcodes (U+1041D)")
+            << QString(st, sizeof(st)/sizeof(st[0]))
+            << QString(se, sizeof(se)/sizeof(se[0]))
+            << QString() << 0 << 0;
+    }
+    {
+        QChar st[] = { 0x011C, 0xd835, 0xdf6e, 0x0110 };
+        QChar se[] = { 0x011D, 0x03C9, 0x0111 };
+        QTest::newRow("highcodes (U+1D76E)")
+            << QString(st, sizeof(st)/sizeof(st[0]))
+            << QString(se, sizeof(se)/sizeof(se[0]))
+            << QString() << 0 << 0;
+    }
+    {
+        QChar st[] = { 'D', 0xdb40, 0xdc20, 'o', 0xd834, 0xdd7a, '\'', 0x2060, 'h' };
+        QChar se[] = { 'd', 'o', '\'', 'h' };
+        QTest::newRow("highcodes (D, U+E0020, o, U+1D17A, ', U+2060, h)")
+            << QString(st, sizeof(st)/sizeof(st[0]))
+            << QString(se, sizeof(se)/sizeof(se[0]))
+            << QString() << 0 << 0;
+    }
+}
+
+void tst_QUrl::nameprep_highcodes()
+{
+#ifdef QT_BUILD_INTERNAL
+    QFETCH(QString, in);
+    QFETCH(QString, out);
+    QFETCH(QString, profile);
+
     qt_nameprep(&in, 0);
     QCOMPARE(in, out);
 #endif
diff --git a/tests/auto/utf8/tst_utf8.cpp b/tests/auto/utf8/tst_utf8.cpp
index 7bbbfab..9b6b8c1 100644
--- a/tests/auto/utf8/tst_utf8.cpp
+++ b/tests/auto/utf8/tst_utf8.cpp
@@ -210,7 +210,9 @@ void tst_Utf8::invalidUtf8_data()
     QTest::addColumn<QByteArray>("utf8");
 
     QTest::newRow("1char") << QByteArray("\x80");
-    QTest::newRow("2chars") << QByteArray("\xC2\xC0");
+    QTest::newRow("2chars-1") << QByteArray("\xC2\xC0");
+    QTest::newRow("2chars-2") << QByteArray("\xC3\xDF");
+    QTest::newRow("2chars-3") << QByteArray("\xC7\xF0");
     QTest::newRow("3chars-1") << QByteArray("\xE0\xA0\xC0");
     QTest::newRow("3chars-2") << QByteArray("\xE0\xC0\xA0");
     QTest::newRow("4chars-1") << QByteArray("\xF0\x90\x80\xC0");