| 1 |
/* ============================================================================ |
| 2 |
* Douglas Thrift's Search Engine License |
| 3 |
* |
| 4 |
* Copyright (C) 2002-2004, Douglas Thrift. All Rights Reserved. |
| 5 |
* Redistribution and use in source and binary forms, with or without |
| 6 |
* modification, are permitted provided that the following conditions are met: |
| 7 |
* |
| 8 |
* 1. Redistributions of source code must retain the above copyright notice, |
| 9 |
* this list of conditions and the following disclaimer. |
| 10 |
* |
| 11 |
* 2. Redistributions in binary form must reproduce the above copyright notice, |
| 12 |
* this list of conditions and the following disclaimer in the documentation |
| 13 |
* and/or other materials provided with the distribution. |
| 14 |
* |
| 15 |
* 3. The end-user documentation included with the redistribution, if any, must |
| 16 |
* include the following acknowledgment: |
| 17 |
* |
| 18 |
* "This product includes software developed by Douglas Thrift |
| 19 |
* (http://computers.douglasthrift.net/searchengine/)." |
| 20 |
* |
| 21 |
* Alternately, this acknowledgment may appear in the software itself, if |
| 22 |
* and wherever such third-party acknowledgments normally appear. |
| 23 |
* |
| 24 |
* 4. The names "Douglas Thrift" and "Douglas Thrift's Search Engine" must not |
| 25 |
* be used to endorse or promote products derived from this software without |
| 26 |
* specific prior written permission. For written permission, please visit |
| 27 |
* http://www.douglasthrift.net/contact.cgi for contact information. |
| 28 |
* |
| 29 |
* 5. Products derived from this software may not be called "Douglas Thrift's |
| 30 |
* Search Engine", nor may "Douglas Thrift's Search Engine" appear in their |
| 31 |
* name, without prior written permission. |
| 32 |
* |
| 33 |
* THIS SOFTWARE IS PROVIDED "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, |
| 34 |
* INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND |
| 35 |
* FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE |
| 36 |
* COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, |
| 37 |
* INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT |
| 38 |
* LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, |
| 39 |
* OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF |
| 40 |
* LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING |
| 41 |
* NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, |
| 42 |
* EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. |
| 43 |
* ============================================================================ |
| 44 |
*/ |
| 45 |
// Douglas Thrift's Search Engine Ranker |
| 46 |
// |
| 47 |
// Douglas Thrift |
| 48 |
// |
| 49 |
// $Id$ |
| 50 |
|
| 51 |
#ifndef _Ranker_hpp_ |
| 52 |
#define _Ranker_hpp_ |
| 53 |
|
| 54 |
#include "Search.hpp" |
| 55 |
#include "Page.hpp" |
| 56 |
|
| 57 |
class Ranker : public Page |
| 58 |
{ |
| 59 |
private: |
| 60 |
enum state { all, title, url, text }; |
| 61 |
enum { sampleMax = 160 }; |
| 62 |
string sample; |
| 63 |
map<unsigned, unsigned> occurrencesTitle; |
| 64 |
map<unsigned, unsigned> occurrencesDescription; |
| 65 |
map<unsigned, unsigned> occurrencesText; |
| 66 |
unsigned value; |
| 67 |
vector<string> required; |
| 68 |
vector<string> excluded; |
| 69 |
vector<string> eitherOr; |
| 70 |
unsigned requiredValue; |
| 71 |
unsigned excludedValue; |
| 72 |
unsigned eitherOrValue; |
| 73 |
string site; |
| 74 |
state allIn; |
| 75 |
string lowerURL; |
| 76 |
string lowerAddress; |
| 77 |
string lowerTitle; |
| 78 |
string lowerText; |
| 79 |
void rank(); |
| 80 |
void checkRequired(); |
| 81 |
void checkExcluded(); |
| 82 |
void checkEitherOr(); |
| 83 |
unsigned find(string word, const string& where); |
| 84 |
unsigned find(string word, const string& where, map<unsigned, unsigned>& |
| 85 |
occurrences); |
| 86 |
unsigned phrase(const string& phrase, const string& where); |
| 87 |
unsigned phrase(const string& phrase, const string& where, map<unsigned, |
| 88 |
unsigned>& occurrences); |
| 89 |
unsigned phrase(const vector<string>& words, unsigned word, unsigned& |
| 90 |
begin, bool start, const string& where); |
| 91 |
unsigned phrase(const vector<string>& words, unsigned word, unsigned& |
| 92 |
begin, bool start, const string& where, map<unsigned, unsigned>& |
| 93 |
occurrences); |
| 94 |
unsigned evaluate(vector<unsigned>& ins); |
| 95 |
void decrap(string& crap); |
| 96 |
public: |
| 97 |
Ranker(Page& page); |
| 98 |
~Ranker() {} |
| 99 |
void rank(vector<string> query); |
| 100 |
void setSample(); |
| 101 |
string getTitle(); |
| 102 |
string getDescription(); |
| 103 |
string& getSample() { return sample; } |
| 104 |
bool operator==(const unsigned number) const; |
| 105 |
bool operator==(const Ranker& ranker) const; |
| 106 |
bool operator!=(const unsigned number) const; |
| 107 |
bool operator!=(const Ranker& ranker) const; |
| 108 |
bool operator<(const unsigned number) const; |
| 109 |
bool operator<(const Ranker& ranker) const; |
| 110 |
bool operator>(const unsigned number) const; |
| 111 |
bool operator>(const Ranker& ranker) const; |
| 112 |
bool operator<=(const unsigned number) const |
| 113 |
{ |
| 114 |
return *this == number || *this < number; |
| 115 |
} |
| 116 |
bool operator<=(const Ranker& ranker) const |
| 117 |
{ |
| 118 |
return *this == ranker || *this < ranker; |
| 119 |
} |
| 120 |
bool operator>=(const unsigned number) const |
| 121 |
{ |
| 122 |
return *this == number || *this > number; |
| 123 |
} |
| 124 |
bool operator>=(const Ranker& ranker) const |
| 125 |
{ |
| 126 |
return *this == ranker || *this > ranker; |
| 127 |
} |
| 128 |
// friends: |
| 129 |
friend bool operator==(const unsigned number, const Ranker& ranker) |
| 130 |
{ |
| 131 |
return ranker == number; |
| 132 |
} |
| 133 |
friend bool operator!=(const unsigned number, const Ranker& ranker) |
| 134 |
{ |
| 135 |
return ranker != number; |
| 136 |
} |
| 137 |
friend bool operator>(const unsigned number, const Ranker& ranker) |
| 138 |
{ |
| 139 |
return ranker < number; |
| 140 |
} |
| 141 |
friend bool operator<(const unsigned number, const Ranker& ranker) |
| 142 |
{ |
| 143 |
return ranker > number; |
| 144 |
} |
| 145 |
friend bool operator>=(const unsigned number, const Ranker& ranker) |
| 146 |
{ |
| 147 |
return ranker <= number; |
| 148 |
} |
| 149 |
friend bool operator<=(const unsigned number, const Ranker& ranker) |
| 150 |
{ |
| 151 |
return ranker >= number; |
| 152 |
} |
| 153 |
}; |
| 154 |
|
| 155 |
#endif // _Ranker_h_ |
