messageviewer/spamheaderanalyzer.cpp

   1 /*  -*- mode: C++; c-file-style: "gnu" -*-
   2     spamheaderanalyzer.cpp
   3
   4     This file is part of KMail, the KDE mail client.
   5     Copyright (c) 2004 Patrick Audley <paudley@blackcat.ca>
   6     Copyright (c) 2004 Ingo Kloecker <kloecker@kde.org>
   7
   8     KMail is free software; you can redistribute it and/or modify
   9     it under the terms of the GNU General Public License as published by
  10     the Free Software Foundation; either version 2 of the License, or
  11     (at your option) any later version.
  12
  13     KMail is distributed in the hope that it will be useful, but
  14     WITHOUT ANY WARRANTY; without even the implied warranty of
  15     MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
  16     GNU General Public License for more details.
  17
  18     You should have received a copy of the GNU General Public License
  19     along with this program; if not, write to the Free Software
  20     Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301  USA
  21
  22     In addition, as a special exception, the copyright holders give
  23     permission to link the code of this program with any edition of
  24     the Qt library by Trolltech AS, Norway (or with modified versions
  25     of Qt that use the same license as Qt), and distribute linked
  26     combinations including the two.  You must obey the GNU General
  27     Public License in all respects for all of the code used other than
  28     Qt.  If you modify this file, you may extend this exception to
  29     your version of the file, but you are not obligated to do so.  If
  30     you do not wish to do so, delete this exception statement from
  31     your version.
  32 */
  33
  34
  35 #include <config-messageviewer.h>
  36
  37 #include "spamheaderanalyzer.h"
  38
  39 #include "antispamconfig.h"
  40
  41 #include <kmime/kmime_message.h>
  42 #include <kmime/kmime_headers.h>
  43
  44 #include <kdebug.h>
  45
  46 #include <boost/shared_ptr.hpp>
  47
  48 using namespace MessageViewer;
  49
  50 // static
  51 SpamScores SpamHeaderAnalyzer::getSpamScores( KMime::Message *message ) {
  52   SpamScores scores;
  53   const SpamAgents agents = AntiSpamConfig::instance()->uniqueAgents();
  54   SpamAgents::const_iterator end( agents.end() );
  55   for ( SpamAgents::const_iterator it = agents.begin(); it != end; ++it ) {
  56     float score = -2.0;
  57
  58     SpamError spamError = noError;
  59
  60     // Skip bogus agents
  61     if ( (*it).scoreType() == SpamAgentNone )
  62       continue;
  63
  64     // Do we have the needed score field for this agent?
  65     KMime::Headers::Base *header= message->headerByType( (*it).header() );
  66     if ( !header )
  67       continue;
  68
  69     const QString mField = header->asUnicodeString();
  70
  71     if ( mField.isEmpty() )
  72       continue;
  73
  74     QString scoreString;
  75     bool scoreValid = false;
  76
  77     if ( (*it).scoreType() != SpamAgentBool ) {
  78       // Can we extract the score?
  79       QRegExp scorePattern = (*it).scorePattern();
  80       if ( scorePattern.indexIn( mField ) != -1 ) {
  81         scoreString = scorePattern.cap( 1 );
  82         scoreValid = true;
  83       }
  84     } else
  85       scoreValid = true;
  86
  87     if ( !scoreValid ) {
  88       spamError = couldNotFindTheScoreField;
  89       kDebug() << "Score could not be extracted from header '"
  90                << mField << "'";
  91     } else {
  92       bool floatValid = false;
  93       switch ( (*it).scoreType() ) {
  94         case SpamAgentNone:
  95           spamError = errorExtractingAgentString;
  96           break;
  97
  98         case SpamAgentBool:
  99           if( (*it).scorePattern().indexIn( mField ) == -1 )
 100             score = 0.0;
 101           else
 102             score = 100.0;
 103           break;
 104
 105         case SpamAgentFloat:
 106           score = scoreString.toFloat( &floatValid );
 107           if ( !floatValid ) {
 108             spamError = couldNotConverScoreToFloat;
 109             kDebug() << "Score (" << scoreString << ") is no number";
 110           }
 111           else
 112             score *= 100.0;
 113           break;
 114
 115         case SpamAgentFloatLarge:
 116           score = scoreString.toFloat( &floatValid );
 117           if ( !floatValid ) {
 118             spamError = couldNotConverScoreToFloat;
 119             kDebug() << "Score (" << scoreString << ") is no number";
 120           }
 121           break;
 122
 123         case SpamAgentAdjustedFloat:
 124           score = scoreString.toFloat( &floatValid );
 125           if ( !floatValid ) {
 126             spamError = couldNotConverScoreToFloat;
 127             kDebug() << "Score (" << scoreString << ") is no number";
 128             break;
 129           }
 130
 131           // Find the threshold value.
 132           QString thresholdString;
 133           QRegExp thresholdPattern = (*it).thresholdPattern();
 134           if ( thresholdPattern.indexIn( mField ) != -1 ) {
 135             thresholdString = thresholdPattern.cap( 1 );
 136           }
 137           else {
 138             spamError = couldNotFindTheThresholdField;
 139             kDebug() << "Threshold could not be extracted from header '"
 140                      << mField << "'";
 141             break;
 142           }
 143           float threshold = thresholdString.toFloat( &floatValid );
 144           if ( !floatValid || ( threshold <= 0.0 ) ) {
 145             spamError = couldNotConvertThresholdToFloatOrThresholdIsNegative;
 146             kDebug() << "Threshold (" << thresholdString << ") is no"
 147                      << "number or is negative";
 148             break;
 149           }
 150
 151           // Normalize the score. Anything below 0 means 0%, anything above
 152           // threshold mean 100%. Values between 0 and threshold are mapped
 153           // linearily to 0% - 100%.
 154           if ( score < 0.0 )
 155             score = 0.0;
 156           else if ( score > threshold )
 157             score = 100.0;
 158           else
 159             score = score / threshold * 100.0;
 160           break;
 161         }
 162     }
 163     //Find the confidence
 164     float confidence = -2.0;
 165     QString confidenceString = "-2.0";
 166     bool confidenceValid = false;
 167     // Do we have the needed confidence field for this agent?
 168     QByteArray confidenceHeaderName = (*it).confidenceHeader();
 169     QString mCField;
 170     if( !confidenceHeaderName.isEmpty() )
 171     {
 172       KMime::Headers::Base *cHeader = message->headerByType( confidenceHeaderName );
 173       if ( cHeader )
 174       {
 175         mCField = cHeader->asUnicodeString();
 176         if ( ! mCField.isEmpty() ) {
 177           // Can we extract the confidence?
 178           QRegExp cScorePattern = (*it).confidencePattern();
 179           if ( cScorePattern.indexIn( mCField ) != -1 ) {
 180             confidenceString = cScorePattern.cap( 1 );
 181           }
 182           confidence = confidenceString.toFloat( &confidenceValid );
 183           if( !confidenceValid) {
 184             spamError = couldNotConvertConfidenceToFloat;
 185             kDebug() << "Unable to convert confidence to float:" << confidenceString;
 186           }
 187         }
 188       }
 189     }
 190     scores.append( SpamScore( (*it).name(), spamError, score, confidence*100, mField, mCField ) );
 191   }
 192
 193   return scores;
 194 }