[15123] | 1 | /*
|
---|
| 2 | * GS2MGPPSearch.java
|
---|
| 3 | * Copyright (C) 2002 New Zealand Digital Library, http://www.nzdl.org
|
---|
| 4 | *
|
---|
| 5 | * This program is free software; you can redistribute it and/or modify
|
---|
| 6 | * the Free Software Foundation; either version 2 of the License, or
|
---|
| 7 | * (at your option) any later version.
|
---|
| 8 | *
|
---|
| 9 | * This program is distributed in the hope that it will be useful,
|
---|
| 10 | * but WITHOUT ANY WARRANTY; without even the implied warranty of
|
---|
| 11 | * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
|
---|
| 12 | * GNU General Public License for more details.
|
---|
| 13 | *
|
---|
| 14 | * You should have received a copy of the GNU General Public License
|
---|
| 15 | * along with this program; if not, write to the Free Software
|
---|
| 16 | * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
|
---|
| 17 | */
|
---|
| 18 | package org.greenstone.gsdl3.service;
|
---|
| 19 |
|
---|
| 20 |
|
---|
| 21 | // Greenstone classes
|
---|
| 22 | import org.greenstone.mgpp.*;
|
---|
| 23 | import org.greenstone.gsdl3.util.*;
|
---|
| 24 | import org.w3c.dom.Element;
|
---|
| 25 |
|
---|
| 26 | import java.util.Vector;
|
---|
| 27 | import java.util.ArrayList;
|
---|
| 28 | import java.util.Collections;
|
---|
| 29 | import org.apache.log4j.*;
|
---|
| 30 |
|
---|
| 31 | /**
|
---|
| 32 | *
|
---|
| 33 | * @author <a href="mailto:[email protected]">Shaoqun Wu</a>
|
---|
| 34 | */
|
---|
| 35 |
|
---|
| 36 | public class GoogleNgramMGPPSearch
|
---|
| 37 | extends GS2MGPPSearch {
|
---|
| 38 | static Logger logger = Logger.getLogger(org.greenstone.gsdl3.service.GoogleNgramMGPPSearch.class.getName());
|
---|
[28181] | 39 |
|
---|
| 40 | protected String default_max_docs = "-1";
|
---|
| 41 | protected String default_hits_per_page = "30";
|
---|
[15123] | 42 | /** constructor */
|
---|
| 43 | public GoogleNgramMGPPSearch(){
|
---|
[15326] | 44 | super();
|
---|
[15123] | 45 | }
|
---|
| 46 |
|
---|
| 47 | /** configure this service */
|
---|
| 48 | public boolean configure(Element info, Element extra_info) {
|
---|
| 49 | if (!super.configure(info, extra_info)){
|
---|
| 50 | return false;
|
---|
| 51 | }
|
---|
| 52 |
|
---|
| 53 | this.default_max_docs = "-1";
|
---|
| 54 | this.default_hits_per_page = "30";
|
---|
| 55 | this.does_stem = false;
|
---|
| 56 | this.does_paging = true;
|
---|
| 57 | return true;
|
---|
| 58 | }
|
---|
| 59 |
|
---|
| 60 | // sort the doc_nums by their frequency
|
---|
| 61 | protected String [] getDocIDs(Object query_result) {
|
---|
| 62 | try{
|
---|
| 63 | Vector docs = ((MGPPQueryResult)query_result).getDocs();
|
---|
| 64 | //ArrayList docList_past = new ArrayList();
|
---|
| 65 | //ArrayList docList_future = new ArrayList();
|
---|
| 66 | //ArrayList docList_present = new ArrayList();
|
---|
| 67 |
|
---|
[25635] | 68 | ArrayList<DocWrapper> docList = new ArrayList<DocWrapper>();
|
---|
[15123] | 69 |
|
---|
| 70 | for (int d = 0; d < docs.size(); d++) {
|
---|
| 71 | String num = Long.toString((((MGPPDocInfo) docs.elementAt(d)).num_));
|
---|
| 72 | String doc_id = internalNum2OID(num);
|
---|
[24394] | 73 | DBInfo dbInfo = this.gs_doc_db.getInfo(doc_id);
|
---|
[15326] | 74 | String fre = (String)dbInfo.getInfo("Frequency");
|
---|
| 75 | String tense = (String)dbInfo.getInfo("Tense");
|
---|
[15123] | 76 |
|
---|
| 77 | if(!fre.equals("")){
|
---|
| 78 | // if (tense.equals("past")){
|
---|
| 79 | // docList_past.add(new DocWrapper(num,Integer.parseInt(fre),tense));
|
---|
| 80 | // }
|
---|
| 81 | // else{
|
---|
| 82 | // if (tense.equals("future")){
|
---|
| 83 | // docList_future.add(new DocWrapper(num,Integer.parseInt(fre),tense));
|
---|
| 84 | // }
|
---|
| 85 | // else{
|
---|
| 86 | // if(tense.equals("present")){
|
---|
| 87 | // docList_present.add(new DocWrapper(num,Integer.parseInt(fre),tense));
|
---|
| 88 | // }
|
---|
| 89 | // }
|
---|
| 90 | //}
|
---|
| 91 | docList.add(new DocWrapper(num,Integer.parseInt(fre),tense));
|
---|
| 92 | }
|
---|
| 93 |
|
---|
| 94 | }
|
---|
| 95 |
|
---|
| 96 |
|
---|
| 97 | //Collections.sort(docList_past);
|
---|
| 98 | //Collections.sort(docList_future);
|
---|
| 99 | //Collections.sort(docList_present);
|
---|
| 100 |
|
---|
| 101 | Collections.sort(docList);
|
---|
| 102 | int i_pa = 0;
|
---|
| 103 | int i_f = 0;
|
---|
| 104 | int i_pre = 0;
|
---|
| 105 |
|
---|
| 106 | //String [] doc_nums = new String [docList_past.size()+docList_future.size()+docList_present.size()];
|
---|
| 107 | String [] doc_nums = new String [docList.size()];
|
---|
| 108 | int interval = 10;
|
---|
| 109 |
|
---|
| 110 | for(int d = 0; d < doc_nums.length; d++){
|
---|
| 111 |
|
---|
| 112 | // for(;i_pre < docList_present.size() && interval > 0;i_pre++){
|
---|
| 113 | // doc_nums[d] = ((DocWrapper)docList_present.get(i_pre)).num;
|
---|
| 114 | // d++;
|
---|
| 115 | // interval--;
|
---|
| 116 | // }
|
---|
| 117 |
|
---|
| 118 | // interval = 10+interval;
|
---|
| 119 |
|
---|
| 120 | // for(;i_pa < docList_past.size() && interval > 0;i_pa++){
|
---|
| 121 | // doc_nums[d] = ((DocWrapper)docList_past.get(i_pa)).num;
|
---|
| 122 | // d++;
|
---|
| 123 | // interval--;
|
---|
| 124 | // }
|
---|
| 125 |
|
---|
| 126 |
|
---|
| 127 | // interval = 10+interval;
|
---|
| 128 |
|
---|
| 129 | // for(;i_f < docList_future.size() && interval > 0;i_f++){
|
---|
| 130 | // doc_nums[d] = ((DocWrapper)docList_future.get(i_f)).num;
|
---|
| 131 | // d++;
|
---|
| 132 | // interval--;
|
---|
| 133 | // }
|
---|
| 134 |
|
---|
| 135 | // interval = 10;
|
---|
| 136 |
|
---|
[25635] | 137 | doc_nums[d] = docList.get(d).num;
|
---|
[15123] | 138 |
|
---|
| 139 | }
|
---|
| 140 |
|
---|
| 141 | return doc_nums;
|
---|
| 142 | }
|
---|
| 143 | catch(Exception e){
|
---|
| 144 | e.printStackTrace();
|
---|
| 145 | }
|
---|
| 146 |
|
---|
| 147 | return null;
|
---|
| 148 | }
|
---|
| 149 |
|
---|
| 150 | static class DocWrapper implements Comparable{
|
---|
| 151 | public int fre = 0;
|
---|
| 152 | public String num = "";
|
---|
| 153 | public String tense = "";
|
---|
| 154 |
|
---|
| 155 |
|
---|
| 156 | public DocWrapper(String num, int fre, String tense){
|
---|
| 157 | this.fre = fre;
|
---|
| 158 | this.num = num;
|
---|
| 159 | this.tense = tense;
|
---|
| 160 | }
|
---|
| 161 |
|
---|
| 162 | public int compareTo(Object o){
|
---|
| 163 |
|
---|
| 164 | if (!(o instanceof DocWrapper)) return -1;
|
---|
| 165 | DocWrapper docIn = (DocWrapper)o;
|
---|
| 166 | if (num.equals(docIn.num)){
|
---|
| 167 | return 0;
|
---|
| 168 | }
|
---|
| 169 |
|
---|
| 170 | if (fre > docIn.fre) return -1;
|
---|
| 171 | return 1;
|
---|
| 172 | }
|
---|
| 173 |
|
---|
| 174 | public boolean equals(Object o){
|
---|
| 175 | if (!(o instanceof DocWrapper)) return false;
|
---|
| 176 | DocWrapper docIn = (DocWrapper)o;
|
---|
| 177 | if (num.equals(docIn.num)){
|
---|
| 178 | return true;
|
---|
| 179 | }
|
---|
| 180 | return false;
|
---|
| 181 | }
|
---|
| 182 |
|
---|
| 183 |
|
---|
| 184 | }
|
---|
| 185 |
|
---|
| 186 |
|
---|
| 187 | }
|
---|
| 188 |
|
---|
| 189 |
|
---|