[prev in list] [next in list] [prev in thread] [next in thread] 

List:       kde-commits
Subject:    [messagelib] webengineviewer/src/checkphishingurl: Continue to implement safe browsing support
From:       Montel Laurent <montel () kde ! org>
Date:       2016-11-30 20:20:02
Message-ID: E1cCBMI-0002OC-72 () code ! kde ! org
[Download RAW message or body]

Git commit 31eda8911e2f58cbec5a905bedd1f6d5bf757310 by Montel Laurent.
Committed on 30/11/2016 at 20:19.
Pushed by mlaurent into branch 'master'.

Continue to implement safe browsing support

M  +3    -2    webengineviewer/src/checkphishingurl/autotests/localdatabasefiletest.cpp
 M  +24   -16   webengineviewer/src/checkphishingurl/localdatabasefile.cpp
M  +61   -0    webengineviewer/src/checkphishingurl/urlhashing.cpp
M  +2    -0    webengineviewer/src/checkphishingurl/urlhashing.h

https://commits.kde.org/messagelib/31eda8911e2f58cbec5a905bedd1f6d5bf757310

diff --git a/webengineviewer/src/checkphishingurl/autotests/localdatabasefiletest.cpp \
b/webengineviewer/src/checkphishingurl/autotests/localdatabasefiletest.cpp index \
                a02ee59..aee882a 100644
--- a/webengineviewer/src/checkphishingurl/autotests/localdatabasefiletest.cpp
+++ b/webengineviewer/src/checkphishingurl/autotests/localdatabasefiletest.cpp
@@ -58,10 +58,11 @@ void LocalDataBaseFileTest::shouldCheckHashBinaryFile_data()
     QTest::addColumn<QByteArray>("hash");
     QTest::addColumn<QByteArray>("resultHash");
     QTest::addColumn<bool>("found");
-    QTest::newRow("nohash") << QByteArrayLiteral("foo") << QByteArray() << false;
+    QTest::newRow("nohash") << QByteArrayLiteral("foo") << QByteArrayLiteral("efgh") \
                << true;
     QTest::newRow("foundhash") << QByteArrayLiteral("1111") << \
                QByteArrayLiteral("1111") << true;
     QTest::newRow("foundhash1") << QByteArrayLiteral("11111") << \
                QByteArrayLiteral("1111") << true;
-    QTest::newRow("foundhash2") << QByteArrayLiteral("HGsse") << \
QByteArrayLiteral("1111") << true; +    QTest::newRow("foundhash2") << \
QByteArrayLiteral("HGsse") << QByteArrayLiteral("54321") << true; +    \
QTest::newRow("foundhash3") << QByteArrayLiteral("1112") << QByteArrayLiteral("1111") \
<< true;  }
 
 void LocalDataBaseFileTest::shouldCheckHashBinaryFile()
diff --git a/webengineviewer/src/checkphishingurl/localdatabasefile.cpp \
b/webengineviewer/src/checkphishingurl/localdatabasefile.cpp index 064f328..154ec2f \
                100644
--- a/webengineviewer/src/checkphishingurl/localdatabasefile.cpp
+++ b/webengineviewer/src/checkphishingurl/localdatabasefile.cpp
@@ -142,23 +142,31 @@ QByteArray LocalDataBaseFile::searchHash(const QByteArray \
&hashToSearch)  const int numHash = getUint64(4);
     int begin = 0;
     int end = numHash - 1;
-    //QByteArray previousValue;
-    while (begin <= end) {
-        const int medium = (begin + end) / 2;
-        const int off = posListOffset + 8 * medium;
-        const int hashOffset = getUint64(off);
-        const char *hashCharStar = getCharStar(hashOffset);
-        const int cmp = qstrcmp(hashCharStar, hashToSearch.constData());
-        qCWarning(WEBENGINEVIEWER_LOG) << "search " << hashToSearch << " begin " << \
                begin << " end " << end << " hashCharStar" <<hashCharStar;
-        if (cmp < 0) {
-            begin = medium + 1;
-        } else if (cmp > 0) {
-            end = medium - 1;
-        } else {
-            return QByteArray(hashCharStar);
-        }
+    QByteArray previousValue;
+    if (end > 0) {
+        QByteArray currentValue;
+        do {
+            previousValue = currentValue;
+            const int medium = (begin + end) / 2;
+            const int off = posListOffset + 8 * medium;
+            const int hashOffset = getUint64(off);
+            const char *hashCharStar = getCharStar(hashOffset);
+            const int cmp = qstrcmp(hashCharStar, hashToSearch.constData());
+            currentValue = QByteArray(hashCharStar);
+            qCWarning(WEBENGINEVIEWER_LOG) << "search " << hashToSearch << " begin " \
<< begin << " end " << end << " hashCharStar" <<hashCharStar; +            if (end == \
begin) { +                return currentValue;
+            }
+            if (cmp < 0) {
+                begin = medium + 1;
+            } else if (cmp > 0) {
+                end = medium - 1;
+            } else {
+                return currentValue;
+            }
+        } while (begin <= end);
     }
-    return QByteArray();
+    return previousValue;
 }
 
 bool LocalDataBaseFile::shouldCheck() const
diff --git a/webengineviewer/src/checkphishingurl/urlhashing.cpp \
b/webengineviewer/src/checkphishingurl/urlhashing.cpp index e9bf7a4..8887b25 100644
--- a/webengineviewer/src/checkphishingurl/urlhashing.cpp
+++ b/webengineviewer/src/checkphishingurl/urlhashing.cpp
@@ -18,6 +18,7 @@
 */
 
 #include "urlhashing.h"
+#include <QDebug>
 
 using namespace WebEngineViewer;
 
@@ -37,9 +38,34 @@ QString UrlHashing::canonicalizeUrl()
     if (mUrl.isEmpty()) {
         return {};
     }
+    QString path = mUrl.path();
     if (mUrl.path().isEmpty()) {
         mUrl.setPath(QStringLiteral("/"));
+    } else {
+        // First, remove tab (0x09), CR (0x0d), and LF (0x0a) characters from the \
URL. Do not remove escape sequences for these characters (e.g. '%0a'). +        \
path.remove(QLatin1Char('\t')); +        path.remove(QLatin1Char('\r'));
+        path.remove(QLatin1Char('\n'));
+        mUrl.setPath(path);
     }
+    // Remove all leading and trailing dots.
+#if 0
+    QString hostname = mUrl.host();
+    qDebug() << " hostname" << hostname;
+    while(!hostname.isEmpty() && hostname.at(0) == QLatin1Char('.')) {
+        hostname.remove(0, 1);
+    }
+    qDebug() << "111111 hostname" << hostname;
+    for (int i = hostname.length(); i >= 0; --i) {
+        if (hostname.at(i) == QLatin1Char('.')) {
+            hostname.remove(i);
+        } else {
+            break;
+        }
+    }
+    qDebug() << "2222222 hostname" << hostname;
+    mUrl.setHost(hostname);
+#endif
     mUrl.setPort(-1);
 
     return QString::fromLatin1(mUrl.toEncoded(QUrl::RemoveFragment|QUrl::NormalizePathSegments|QUrl::EncodeUnicode));
 @@ -54,3 +80,38 @@ QStringList UrlHashing::generateHostsToCheck()
 {
     return {};
 }
+
+QByteArray UrlHashing::hashComputation()
+{
+#if 0
+    Unit Test (in pseudo-C)
+
+    // Example B1 from FIPS-180-2
+    string input1 = "abc";
+    string output1 = TruncatedSha256Prefix(input1, 32);
+    int expected1[] = { 0xba, 0x78, 0x16, 0xbf };
+    assert(output1.size() == 4);  // 4 bytes == 32 bits
+    for (int i = 0; i < output1.size(); i++) assert(output1[i] == expected1[i]);
+
+    // Example B2 from FIPS-180-2
+    string input2 = "abcdbcdecdefdefgefghfghighijhijkijkljklmklmnlmnomnopnopq";
+    string output2 = TruncatedSha256Prefix(input2, 48);
+    int expected2[] = { 0x24, 0x8d, 0x6a, 0x61, 0xd2, 0x06 };
+    assert(output2.size() == 6);
+    for (int i = 0; i < output2.size(); i++) assert(output2[i] == expected2[i]);
+
+    // Example B3 from FIPS-180-2
+    string input3(1000000, 'a');  // 'a' repeated a million times
+    string output3 = TruncatedSha256Prefix(input3, 96);
+    int expected3[] = { 0xcd, 0xc7, 0x6e, 0x5c, 0x99, 0x14, 0xfb, 0x92,
+                        0x81, 0xa1, 0xc7, 0xe2 };
+    assert(output3.size() == 12);
+    for (int i = 0; i < output3.size(); i++) assert(output3[i] == expected3[i]);
+#endif
+    return {};
+}
+
+QByteArray UrlHashing::hashPrefixComputation()
+{
+    return {};
+}
diff --git a/webengineviewer/src/checkphishingurl/urlhashing.h \
b/webengineviewer/src/checkphishingurl/urlhashing.h index 607fc35..3ad7e4c 100644
--- a/webengineviewer/src/checkphishingurl/urlhashing.h
+++ b/webengineviewer/src/checkphishingurl/urlhashing.h
@@ -36,6 +36,8 @@ public:
     QStringList generatePathsToCheck();
     QStringList generateHostsToCheck();
 
+    QByteArray hashComputation();
+    QByteArray hashPrefixComputation();
 private:
     QUrl mUrl;
 };


[prev in list] [next in list] [prev in thread] [next in thread]