Initial rework of matchLink

resources/resources.qrc

@@ -66,6 +66,7 @@
         <file>images/buttons/unban.png</file>
         <file>images/buttons/unmod.png</file>
         <file>images/emote_dark.svg</file>
+        <file>tlds.txt</file>
     </qresource>
     <qresource prefix="/qt/etc">
         <file>qt.conf</file>

src/messages/MessageBuilder.cpp

@@ -43,9 +43,44 @@ void MessageBuilder::appendTimestamp(const QTime &time)
 
 QString MessageBuilder::matchLink(const QString &string)
 {
-    static QRegularExpression linkRegex("[[:ascii:]]*\\.[a-zA-Z]+\\/?[[:ascii:]]*");
+    QFile tldFile(":/tlds.txt");
+    tldFile.open(QFile::ReadOnly);
+    QTextStream t1(&tldFile);
+    t1.setCodec("UTF-8");
+    QString tldData = t1.readAll();
+    tldData.replace("\n", "|");
+    const QString urlRegExp = "^"
+    // protocol identifier
+    "(?:(?:https?|ftps?)://)?"
+    // user:pass authentication
+    "(?:\\S+(?::\\S*)?@)?"
+    "(?:"
+    // IP address dotted notation octets
+    // excludes loopback network 0.0.0.0
+    // excludes reserved space >= 224.0.0.0
+    // excludes network & broacast addresses
+    // (first & last IP address of each class)
+    "(?:[1-9]\\d?|1\\d\\d|2[01]\\d|22[0-3])"
+    "(?:\\.(?:1?\\d{1,2}|2[0-4]\\d|25[0-5])){2}"
+    "(?:\\.(?:[1-9]\\d?|1\\d\\d|2[0-4]\\d|25[0-4]))"
+    "|"
+    // host name
+    "(?:(?:[_a-z\\x{00a1}-\\x{ffff}0-9]-*)*[a-z\\x{00a1}-\\x{ffff}0-9]+)"
+    // domain name
+    "(?:\\.(?:[a-z\\x{00a1}-\\x{ffff}0-9]-*)*[a-z\\x{00a1}-\\x{ffff}0-9]+)*"
+    // TLD identifier
+    //"(?:\\.(?:[a-z\\x{00a1}-\\x{ffff}]{2,}))"
+    "(?:[\\.](?:" + tldData + "))"
+    "\\.?"
+    ")"
+    // port number
+    "(?::\\d{2,5})?"
+    // resource path
+    "(?:[/?#]\\S*)?"
+    "$";
+    static QRegularExpression linkRegex(urlRegExp, QRegularExpression::CaseInsensitiveOption);
     static QRegularExpression httpRegex("\\bhttps?://");
-
+    static QRegularExpression ftpRegex("\\bftps?://");
     auto match = linkRegex.match(string);
 
     if (!match.hasMatch()) {
@@ -55,7 +90,9 @@ QString MessageBuilder::matchLink(const QString &string)
     QString captured = match.captured();
 
     if (!captured.contains(httpRegex)) {
-        captured.insert(0, "http://");
+        if (!captured.contains(ftpRegex)) {
+            captured.insert(0, "http://");
+        }
     }
 
     return captured;

tools/get-tlds-update.sh

@@ -0,0 +1,13 @@
+#!/bin/sh
+
+#Download the official list of active TLDs from IANA
+#Remove the first line that contains data not needed.
+#Put everything that can be into lowercase.
+#Output the result to a file.
+curl -s 'https://data.iana.org/TLD/tlds-alpha-by-domain.txt' | sed -e '1d' -e 's/\(.*\)/\L\1/' > tlds.txt
+
+#Get the TLDs in punycode format.
+#Convert the punycode to Unicode.
+#Append the results to the current file.
+sed -n -e '/^xn--/p' tlds.txt | idn2 -d >> tlds.txt
+mv tlds.txt ../resources/tlds.txt