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());
|
---|
39 | protected GDBMWrapper gdbm_src = null;
|
---|
40 |
|
---|
41 | /** constructor */
|
---|
42 | public GoogleNgramMGPPSearch(){
|
---|
43 | super();
|
---|
44 | gdbm_src = new GDBMWrapper();
|
---|
45 |
|
---|
46 | }
|
---|
47 |
|
---|
48 | /** configure this service */
|
---|
49 | public boolean configure(Element info, Element extra_info) {
|
---|
50 | if (!super.configure(info, extra_info)){
|
---|
51 | return false;
|
---|
52 | }
|
---|
53 |
|
---|
54 | // Open GDBM database for querying
|
---|
55 | String gdbm_db_file = GSFile.GDBMDatabaseFile(this.site_home, this.cluster_name, this.index_stem);
|
---|
56 | if (!this.gdbm_src.openDatabase(gdbm_db_file, GDBMWrapper.READER)) {
|
---|
57 | logger.error("Could not open GDBM database!");
|
---|
58 | return false;
|
---|
59 | }
|
---|
60 | this.default_max_docs = "-1";
|
---|
61 | this.default_hits_per_page = "30";
|
---|
62 | this.does_stem = false;
|
---|
63 | this.does_paging = true;
|
---|
64 | return true;
|
---|
65 | }
|
---|
66 |
|
---|
67 | // sort the doc_nums by their frequency
|
---|
68 | protected String [] getDocIDs(Object query_result) {
|
---|
69 | try{
|
---|
70 | Vector docs = ((MGPPQueryResult)query_result).getDocs();
|
---|
71 | //ArrayList docList_past = new ArrayList();
|
---|
72 | //ArrayList docList_future = new ArrayList();
|
---|
73 | //ArrayList docList_present = new ArrayList();
|
---|
74 |
|
---|
75 | ArrayList docList = new ArrayList();
|
---|
76 |
|
---|
77 | for (int d = 0; d < docs.size(); d++) {
|
---|
78 | String num = Long.toString((((MGPPDocInfo) docs.elementAt(d)).num_));
|
---|
79 | String doc_id = internalNum2OID(num);
|
---|
80 | DBInfo gdbmInfo = this.gdbm_src.getInfo(doc_id);
|
---|
81 | String fre = (String)gdbmInfo.getInfo("Frequency");
|
---|
82 | String tense = (String)gdbmInfo.getInfo("Tense");
|
---|
83 |
|
---|
84 | if(!fre.equals("")){
|
---|
85 | // if (tense.equals("past")){
|
---|
86 | // docList_past.add(new DocWrapper(num,Integer.parseInt(fre),tense));
|
---|
87 | // }
|
---|
88 | // else{
|
---|
89 | // if (tense.equals("future")){
|
---|
90 | // docList_future.add(new DocWrapper(num,Integer.parseInt(fre),tense));
|
---|
91 | // }
|
---|
92 | // else{
|
---|
93 | // if(tense.equals("present")){
|
---|
94 | // docList_present.add(new DocWrapper(num,Integer.parseInt(fre),tense));
|
---|
95 | // }
|
---|
96 | // }
|
---|
97 | //}
|
---|
98 | docList.add(new DocWrapper(num,Integer.parseInt(fre),tense));
|
---|
99 | }
|
---|
100 |
|
---|
101 | }
|
---|
102 |
|
---|
103 |
|
---|
104 | //Collections.sort(docList_past);
|
---|
105 | //Collections.sort(docList_future);
|
---|
106 | //Collections.sort(docList_present);
|
---|
107 |
|
---|
108 | Collections.sort(docList);
|
---|
109 | int i_pa = 0;
|
---|
110 | int i_f = 0;
|
---|
111 | int i_pre = 0;
|
---|
112 |
|
---|
113 | //String [] doc_nums = new String [docList_past.size()+docList_future.size()+docList_present.size()];
|
---|
114 | String [] doc_nums = new String [docList.size()];
|
---|
115 | int interval = 10;
|
---|
116 |
|
---|
117 | for(int d = 0; d < doc_nums.length; d++){
|
---|
118 |
|
---|
119 | // for(;i_pre < docList_present.size() && interval > 0;i_pre++){
|
---|
120 | // doc_nums[d] = ((DocWrapper)docList_present.get(i_pre)).num;
|
---|
121 | // d++;
|
---|
122 | // interval--;
|
---|
123 | // }
|
---|
124 |
|
---|
125 | // interval = 10+interval;
|
---|
126 |
|
---|
127 | // for(;i_pa < docList_past.size() && interval > 0;i_pa++){
|
---|
128 | // doc_nums[d] = ((DocWrapper)docList_past.get(i_pa)).num;
|
---|
129 | // d++;
|
---|
130 | // interval--;
|
---|
131 | // }
|
---|
132 |
|
---|
133 |
|
---|
134 | // interval = 10+interval;
|
---|
135 |
|
---|
136 | // for(;i_f < docList_future.size() && interval > 0;i_f++){
|
---|
137 | // doc_nums[d] = ((DocWrapper)docList_future.get(i_f)).num;
|
---|
138 | // d++;
|
---|
139 | // interval--;
|
---|
140 | // }
|
---|
141 |
|
---|
142 | // interval = 10;
|
---|
143 |
|
---|
144 | doc_nums[d] = ((DocWrapper)docList.get(d)).num;
|
---|
145 |
|
---|
146 | }
|
---|
147 |
|
---|
148 | return doc_nums;
|
---|
149 | }
|
---|
150 | catch(Exception e){
|
---|
151 | e.printStackTrace();
|
---|
152 | }
|
---|
153 |
|
---|
154 | return null;
|
---|
155 | }
|
---|
156 |
|
---|
157 | static class DocWrapper implements Comparable{
|
---|
158 | public int fre = 0;
|
---|
159 | public String num = "";
|
---|
160 | public String tense = "";
|
---|
161 |
|
---|
162 |
|
---|
163 | public DocWrapper(String num, int fre, String tense){
|
---|
164 | this.fre = fre;
|
---|
165 | this.num = num;
|
---|
166 | this.tense = tense;
|
---|
167 | }
|
---|
168 |
|
---|
169 | public int compareTo(Object o){
|
---|
170 |
|
---|
171 | if (!(o instanceof DocWrapper)) return -1;
|
---|
172 | DocWrapper docIn = (DocWrapper)o;
|
---|
173 | if (num.equals(docIn.num)){
|
---|
174 | return 0;
|
---|
175 | }
|
---|
176 |
|
---|
177 | if (fre > docIn.fre) return -1;
|
---|
178 | return 1;
|
---|
179 | }
|
---|
180 |
|
---|
181 | public boolean equals(Object o){
|
---|
182 | if (!(o instanceof DocWrapper)) return false;
|
---|
183 | DocWrapper docIn = (DocWrapper)o;
|
---|
184 | if (num.equals(docIn.num)){
|
---|
185 | return true;
|
---|
186 | }
|
---|
187 | return false;
|
---|
188 | }
|
---|
189 |
|
---|
190 |
|
---|
191 | }
|
---|
192 |
|
---|
193 |
|
---|
194 | }
|
---|
195 |
|
---|
196 |
|
---|