Adds QUrl::fromUserInput, gathered from QWebView::guessUrlFromString.

Reviewed-by: Thiago Macieira

3 files changed, 142 insertions, 0 deletions

diff --git a/src/corelib/io/qurl.cpp b/src/corelib/io/qurl.cpp
index face923..c9a4cf1 100644
--- a/src/corelib/io/qurl.cpp
+++ b/src/corelib/io/qurl.cpp
@@ -172,6 +172,8 @@
 #include "private/qunicodetables_p.h"
 #include "qatomic.h"
 #include "qbytearray.h"
+#include "qdir.h"
+#include "qfile.h"
 #include "qlist.h"
 #ifndef QT_NO_REGEXP
 #include "qregexp.h"
@@ -5547,6 +5549,79 @@ QUrl QUrl::fromEncoded(const QByteArray &input, ParsingMode parsingMode)
 }
 
 /*!
+    Returns a valid URL from a user supplied \a userInput string if one can be
+    deducted. In the case that is not possible, an invalid QUrl() is returned.
+
+    \since 4.6
+
+    Most applications that can browse the web, allow the user to input a URL
+    in the form of a plain string. This string can be manually typed into
+    a location bar, obtained from the clipboard, or passed in via command
+    line arguments.
+
+    When the string is not already a valid URL, a best guess is performed,
+    making various web related assumptions.
+
+    In the case the string corresponds to a valid file path on the system,
+    a file:// URL is constructed, using QUrl::fromLocalFile().
+
+    If that is not the case, an attempt is made to turn the string into a
+    http:// or ftp:// URL. The latter in the case the string starts with
+    'ftp'. The result is then passed through QUrl's tolerant parser, and
+    in the case or success, a valid QUrl is returned, or else a QUrl().
+
+    \section1 Examples:
+
+    \list
+    \o qt.nokia.com becomes http://qt.nokia.com
+    \o ftp.qt.nokia.com becomes ftp://ftp.qt.nokia.com
+    \o localhost becomes http://localhost
+    \o /home/user/test.html becomes file:///home/user/test.html (if exists)
+    \endlist
+
+    \section2 Tips to avoid erroneous character conversion when dealing with
+    URLs and strings:
+
+    \list
+    \o When creating an URL QString from a QByteArray or a char*, always use
+       QString::fromUtf8().
+    \o Favor the use of QUrl::fromEncoded() and QUrl::toEncoded() instead of
+       QUrl(string) and QUrl::toString() when converting QUrl to/from string.
+    \endlist
+*/
+QUrl QUrl::fromUserInput(const QString &userInput)
+{
+    QString trimmedString = userInput.trimmed();
+
+    // Absolute files
+    if (QDir::isAbsolutePath(trimmedString))
+        return QUrl::fromLocalFile(trimmedString);
+
+    // Check the most common case of a valid url with scheme and host first
+    QUrl url = QUrl::fromEncoded(trimmedString.toUtf8(), QUrl::TolerantMode);
+    if (url.isValid() && !url.scheme().isEmpty() && !url.host().isEmpty())
+        return url;
+
+    // If the string is missing the scheme or the scheme is not valid, prepend a scheme
+    QString scheme = url.scheme();
+    if (scheme.isEmpty() || scheme.contains(QLatin1Char('.')) || scheme == QLatin1String("localhost")) {
+        // Do not do anything for strings such as "foo", only "foo.com"
+        int dotIndex = trimmedString.indexOf(QLatin1Char('.'));
+        if (dotIndex != -1 || trimmedString.startsWith(QLatin1String("localhost"))) {
+            const QString hostscheme = trimmedString.left(dotIndex).toLower();
+            QByteArray scheme = (hostscheme == QLatin1String("ftp")) ? "ftp" : "http";
+            trimmedString = QLatin1String(scheme) + QLatin1String("://") + trimmedString;
+        }
+        url = QUrl::fromEncoded(trimmedString.toUtf8(), QUrl::TolerantMode);
+    }
+
+    if (url.isValid())
+        return url;
+
+    return QUrl();
+}
+
+/*!
     Returns a decoded copy of \a input. \a input is first decoded from
     percent encoding, then converted from UTF-8 to unicode.
 */
diff --git a/src/corelib/io/qurl.h b/src/corelib/io/qurl.h
index b00074a..f76d345 100644
--- a/src/corelib/io/qurl.h
+++ b/src/corelib/io/qurl.h
@@ -189,6 +189,8 @@ public:
     static QUrl fromEncoded(const QByteArray &url, ParsingMode mode);
     // ### Qt 5: merge the two fromEncoded() functions, with mode = TolerantMode
 
+    static QUrl fromUserInput(const QString &userInput);
+
     void detach();
     bool isDetached() const;
 
diff --git a/tests/auto/qurl/tst_qurl.cpp b/tests/auto/qurl/tst_qurl.cpp
index 8899176..fb3cf0e 100644
--- a/tests/auto/qurl/tst_qurl.cpp
+++ b/tests/auto/qurl/tst_qurl.cpp
@@ -186,6 +186,8 @@ private slots:
     void resolvedWithAbsoluteSchemes_data() const;
     void binaryData_data();
     void binaryData();
+    void fromUserInput_data();
+    void fromUserInput();
     void task_199967();
     void task_240612();
 
@@ -3637,6 +3639,69 @@ void tst_QUrl::binaryData()
     QCOMPARE(url2, url);
 }
 
+void tst_QUrl::fromUserInput_data()
+{
+    QTest::addColumn<QString>("string");
+    QTest::addColumn<QUrl>("url");
+
+    // Null
+    QTest::newRow("null") << QString() << QUrl();
+
+    // File
+    QDirIterator it(QDir::homePath());
+    QString fileString;
+    int c = 0;
+    while (it.hasNext()) {
+        it.next();
+        QTest::newRow(QString("file-%1").arg(c++).toLatin1()) << it.filePath() << QUrl::fromLocalFile(it.filePath());
+    }
+
+    // basic latin1
+    QTest::newRow("unicode-0") << QString::fromUtf8("\xC3\xA5.com/") << QUrl::fromEncoded(QString::fromUtf8("http://\xC3\xA5.com/").toUtf8(), QUrl::TolerantMode);
+    // unicode
+    QTest::newRow("unicode-1") << QString::fromUtf8("\xCE\xBB.com/") << QUrl::fromEncoded(QString::fromUtf8("http://\xCE\xBB.com/").toUtf8(), QUrl::TolerantMode);
+
+    // no scheme
+    QTest::newRow("add scheme-0") << "webkit.org" << QUrl("http://webkit.org");
+    QTest::newRow("add scheme-1") << "www.webkit.org" << QUrl("http://www.webkit.org");
+    QTest::newRow("add scheme-2") << "ftp.webkit.org" << QUrl("ftp://ftp.webkit.org");
+    QTest::newRow("add scheme-3") << "webkit" << QUrl("webkit");
+
+    // QUrl's tolerant parser should already handle this
+    QTest::newRow("not-encoded-0") << "http://webkit.org/test page.html" << QUrl("http://webkit.org/test%20page.html");
+
+    // Make sure the :80, i.e. port doesn't screw anything up
+    QUrl portUrl("http://webkit.org");
+    portUrl.setPort(80);
+    QTest::newRow("port-0") << "webkit.org:80" << portUrl;
+    QTest::newRow("port-1") << "http://webkit.org:80" << portUrl;
+
+    // mailto doesn't have a ://, but is valid
+    QUrl mailto("somebody@somewhere.net");
+    mailto.setScheme("mailto");
+    QTest::newRow("mailto") << "mailto:somebody@somewhere.net" << mailto;
+
+    // misc
+    QTest::newRow("localhost-0") << "localhost" << QUrl("http://localhost");
+    QTest::newRow("localhost-1") << "localhost:80" << QUrl("http://localhost:80");
+    QTest::newRow("spaces-0") << "  http://webkit.org/test page.html " << QUrl("http://webkit.org/test%20page.html");
+    QTest::newRow("trash-0") << "webkit.org/test?someData=42%&someOtherData=abcde#anchor" << QUrl::fromEncoded("http://webkit.org/test?someData=42%25&someOtherData=abcde#anchor");
+
+    // FYI: The scheme in the resulting url user
+    QUrl authUrl("user:pass@domain.com");
+    QTest::newRow("misc-1") << "user:pass@domain.com" << authUrl;
+}
+
+// public static QUrl guessUrlFromString(QString const& string)
+void tst_QUrl::fromUserInput()
+{
+    QFETCH(QString, string);
+    QFETCH(QUrl, url);
+
+    QUrl guessedUrl = QUrl::fromUserInput(string);
+    QCOMPARE(guessedUrl, url);
+}
+
 void tst_QUrl::task_199967()
 {
     {