1 |
1831
|
tao
|
/**
|
2 |
|
|
* '$RCSfile$'
|
3 |
2366
|
sgarg
|
* Purpose: A Class that represents a structured query, and can be
|
4 |
|
|
* constructed from an XML serialization conforming to
|
5 |
|
|
* pathquery.dtd. The printSQL() method can be used to print
|
6 |
1831
|
tao
|
* a SQL serialization of the query.
|
7 |
|
|
* Copyright: 2000 Regents of the University of California and the
|
8 |
|
|
* National Center for Ecological Analysis and Synthesis
|
9 |
|
|
* Authors: Matt Jones
|
10 |
|
|
*
|
11 |
|
|
* '$Author$'
|
12 |
|
|
* '$Date$'
|
13 |
|
|
* '$Revision$'
|
14 |
|
|
*
|
15 |
|
|
* This program is free software; you can redistribute it and/or modify
|
16 |
|
|
* it under the terms of the GNU General Public License as published by
|
17 |
|
|
* the Free Software Foundation; either version 2 of the License, or
|
18 |
|
|
* (at your option) any later version.
|
19 |
|
|
*
|
20 |
|
|
* This program is distributed in the hope that it will be useful,
|
21 |
|
|
* but WITHOUT ANY WARRANTY; without even the implied warranty of
|
22 |
|
|
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
|
23 |
|
|
* GNU General Public License for more details.
|
24 |
|
|
*
|
25 |
|
|
* You should have received a copy of the GNU General Public License
|
26 |
|
|
* along with this program; if not, write to the Free Software
|
27 |
|
|
* Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
|
28 |
|
|
*/
|
29 |
|
|
|
30 |
|
|
package edu.ucsb.nceas.metacat;
|
31 |
|
|
|
32 |
|
|
import edu.ucsb.nceas.dbadapter.*;
|
33 |
|
|
|
34 |
|
|
import java.io.*;
|
35 |
|
|
import java.util.Hashtable;
|
36 |
|
|
import java.util.Stack;
|
37 |
|
|
import java.util.Vector;
|
38 |
|
|
import java.util.Enumeration;
|
39 |
|
|
|
40 |
2677
|
sgarg
|
import org.apache.log4j.Logger;
|
41 |
|
|
|
42 |
1831
|
tao
|
/** a utility class that represents a group of terms in a query */
|
43 |
|
|
public class QueryGroup {
|
44 |
|
|
private String operator = null; // indicates how query terms are combined
|
45 |
|
|
private Vector children = null; // the list of query terms and groups
|
46 |
|
|
private int countPercentageSearchItem = 0;
|
47 |
3225
|
tao
|
private Vector queryTermsWithSameValue = null;//this two dimension vectors.
|
48 |
|
|
//will hold query terms which has same search value.
|
49 |
|
|
private Vector queryTerms = null;//this vector only holds query terms without same search value.
|
50 |
3226
|
tao
|
private Vector queryGroupsChildren = null;
|
51 |
2677
|
sgarg
|
private static Logger logMetacat = Logger.getLogger(QueryGroup.class);
|
52 |
|
|
|
53 |
2366
|
sgarg
|
/**
|
54 |
|
|
* construct a new QueryGroup
|
55 |
1831
|
tao
|
*
|
56 |
2366
|
sgarg
|
* @param operator the boolean conector used to connect query terms
|
57 |
1831
|
tao
|
* in this query group
|
58 |
|
|
*/
|
59 |
|
|
public QueryGroup(String operator) {
|
60 |
|
|
this.operator = operator;
|
61 |
|
|
children = new Vector();
|
62 |
3225
|
tao
|
queryTermsWithSameValue = new Vector();
|
63 |
|
|
queryTerms = new Vector();
|
64 |
3226
|
tao
|
queryGroupsChildren = new Vector();
|
65 |
1831
|
tao
|
}
|
66 |
|
|
|
67 |
2366
|
sgarg
|
/**
|
68 |
1831
|
tao
|
* Add a child QueryGroup to this QueryGroup
|
69 |
|
|
*
|
70 |
|
|
* @param qgroup the query group to be added to the list of terms
|
71 |
|
|
*/
|
72 |
|
|
public void addChild(QueryGroup qgroup) {
|
73 |
3226
|
tao
|
//children.add((Object)qgroup);
|
74 |
|
|
queryGroupsChildren.add(qgroup);
|
75 |
1831
|
tao
|
}
|
76 |
|
|
|
77 |
|
|
/**
|
78 |
|
|
* Add a child QueryTerm to this QueryGroup
|
79 |
|
|
*
|
80 |
|
|
* @param qterm the query term to be added to the list of terms
|
81 |
|
|
*/
|
82 |
|
|
public void addChild(QueryTerm qterm) {
|
83 |
3226
|
tao
|
// children.add((Object)qterm);
|
84 |
3225
|
tao
|
handleNewQueryTerms(qterm);
|
85 |
|
|
|
86 |
1831
|
tao
|
}
|
87 |
|
|
|
88 |
|
|
/**
|
89 |
|
|
* Retrieve an Enumeration of query terms for this QueryGroup
|
90 |
|
|
*/
|
91 |
|
|
public Enumeration getChildren() {
|
92 |
|
|
return children.elements();
|
93 |
|
|
}
|
94 |
2366
|
sgarg
|
|
95 |
1831
|
tao
|
public int getPercentageSymbolCount()
|
96 |
|
|
{
|
97 |
|
|
return countPercentageSearchItem;
|
98 |
|
|
}
|
99 |
2366
|
sgarg
|
|
100 |
1831
|
tao
|
/**
|
101 |
|
|
* create a SQL serialization of the query that this instance represents
|
102 |
|
|
*/
|
103 |
|
|
public String printSQL(boolean useXMLIndex) {
|
104 |
|
|
StringBuffer self = new StringBuffer();
|
105 |
2373
|
sgarg
|
StringBuffer queryString = new StringBuffer();
|
106 |
|
|
|
107 |
1831
|
tao
|
boolean first = true;
|
108 |
2677
|
sgarg
|
|
109 |
1831
|
tao
|
Enumeration en= getChildren();
|
110 |
|
|
while (en.hasMoreElements()) {
|
111 |
|
|
Object qobject = en.nextElement();
|
112 |
|
|
if (qobject instanceof QueryGroup) {
|
113 |
2635
|
sgarg
|
QueryGroup qg = (QueryGroup)qobject;
|
114 |
2634
|
sgarg
|
String queryGroupSQL = qg.printSQL(useXMLIndex);
|
115 |
2677
|
sgarg
|
logMetacat.info("In QueryGroup.printSQL.. found a QueryGroup: "
|
116 |
|
|
+ queryGroupSQL);
|
117 |
2635
|
sgarg
|
|
118 |
2634
|
sgarg
|
if (first) {
|
119 |
|
|
first = false;
|
120 |
|
|
} else {
|
121 |
|
|
if(!queryString.toString().equals("") && queryGroupSQL != null &&!queryGroupSQL.equals("")){
|
122 |
2503
|
sgarg
|
queryString.append(" " + operator + " ");
|
123 |
2677
|
sgarg
|
}
|
124 |
|
|
}
|
125 |
|
|
queryString.append(queryGroupSQL);
|
126 |
|
|
|
127 |
|
|
// count percerntage number
|
128 |
|
|
int count = qg.getPercentageSymbolCount();
|
129 |
|
|
countPercentageSearchItem = countPercentageSearchItem + count;
|
130 |
1831
|
tao
|
} else if (qobject instanceof QueryTerm) {
|
131 |
|
|
QueryTerm qt = (QueryTerm)qobject;
|
132 |
2373
|
sgarg
|
String termQueryString = qt.printSQL(useXMLIndex);
|
133 |
2677
|
sgarg
|
logMetacat.info("In QueryGroup.printSQL.. found a QueryGroup: "
|
134 |
|
|
+ termQueryString);
|
135 |
2366
|
sgarg
|
if(!(qt.getSearchMode().equals("contains") && qt.getValue().equals("%"))){
|
136 |
2634
|
sgarg
|
if (first) {
|
137 |
2503
|
sgarg
|
first = false;
|
138 |
|
|
} else {
|
139 |
|
|
if(!queryString.toString().equals("")){
|
140 |
|
|
queryString.append(" " + operator + " ");
|
141 |
|
|
}
|
142 |
|
|
}
|
143 |
2373
|
sgarg
|
queryString.append(termQueryString);
|
144 |
2366
|
sgarg
|
}
|
145 |
1831
|
tao
|
// count percerntage number
|
146 |
|
|
int count = qt.getPercentageSymbolCount();
|
147 |
|
|
countPercentageSearchItem = countPercentageSearchItem + count;
|
148 |
|
|
} else {
|
149 |
|
|
System.err.println("qobject wrong type: fatal error");
|
150 |
|
|
}
|
151 |
|
|
}
|
152 |
2373
|
sgarg
|
|
153 |
|
|
if(!queryString.toString().equals("")){
|
154 |
|
|
self.append("(");
|
155 |
|
|
self.append(queryString.toString());
|
156 |
|
|
self.append(")");
|
157 |
|
|
}
|
158 |
2677
|
sgarg
|
|
159 |
|
|
logMetacat.info("In QueryGroup.printSQL.. final query returned is: "
|
160 |
|
|
+ self.toString());
|
161 |
1831
|
tao
|
return self.toString();
|
162 |
|
|
}
|
163 |
3226
|
tao
|
|
164 |
|
|
|
165 |
|
|
/*
|
166 |
|
|
* If every query term in a queryGroup share a search value, we should
|
167 |
|
|
* use a new query to replace the original query term query in order to
|
168 |
|
|
* improve performance
|
169 |
|
|
*/
|
170 |
|
|
private String printSQLStringWithSameSearchValue()
|
171 |
|
|
{
|
172 |
|
|
String sql ="";
|
173 |
|
|
String value ="";
|
174 |
|
|
StringBuffer sqlBuff = new StringBuffer();
|
175 |
|
|
Vector pathVector = new Vector();
|
176 |
|
|
int index =0;
|
177 |
|
|
|
178 |
|
|
if (queryTermsWithSameValue != null)
|
179 |
|
|
{
|
180 |
|
|
QueryTerm term1 = (QueryTerm)queryTermsWithSameValue.elementAt(0);
|
181 |
|
|
value = term1.getValue();
|
182 |
|
|
boolean first = true;
|
183 |
|
|
sqlBuff.append("SELECT DISTINCT docid FROM xml_path_index WHERE UPPER(nodedata) LIKE '%");
|
184 |
|
|
sqlBuff.append(value);
|
185 |
|
|
sqlBuff.append("%' AND path IN (");
|
186 |
|
|
//gets every path in query term object
|
187 |
|
|
for (int i=0; i<queryTermsWithSameValue.size(); i++)
|
188 |
|
|
{
|
189 |
|
|
QueryTerm term = (QueryTerm)queryTermsWithSameValue.elementAt(i);
|
190 |
|
|
value = term.getValue();
|
191 |
|
|
String path = term.getPathExpression();
|
192 |
|
|
if (path != null && !path.equals(""))
|
193 |
|
|
{
|
194 |
|
|
if (first)
|
195 |
|
|
{
|
196 |
|
|
first = false;
|
197 |
|
|
sqlBuff.append("'");
|
198 |
|
|
sqlBuff.append(path);
|
199 |
|
|
sqlBuff.append("'");
|
200 |
|
|
|
201 |
|
|
}
|
202 |
|
|
else
|
203 |
|
|
{
|
204 |
|
|
sqlBuff.append(",'");
|
205 |
|
|
sqlBuff.append(path);
|
206 |
|
|
sqlBuff.append("'");
|
207 |
|
|
}
|
208 |
|
|
index++;
|
209 |
|
|
}
|
210 |
|
|
}
|
211 |
|
|
sqlBuff.append(")");
|
212 |
|
|
|
213 |
|
|
}
|
214 |
|
|
if (index >0)
|
215 |
|
|
{
|
216 |
|
|
sql = sqlBuff.toString();
|
217 |
|
|
}
|
218 |
|
|
return sql;
|
219 |
|
|
}
|
220 |
1831
|
tao
|
|
221 |
|
|
/**
|
222 |
|
|
* create a String description of the query that this instance represents.
|
223 |
|
|
* This should become a way to get the XML serialization of the query.
|
224 |
|
|
*/
|
225 |
|
|
public String toString() {
|
226 |
|
|
StringBuffer self = new StringBuffer();
|
227 |
|
|
|
228 |
|
|
self.append(" (Query group operator=" + operator + "\n");
|
229 |
|
|
Enumeration en= getChildren();
|
230 |
|
|
while (en.hasMoreElements()) {
|
231 |
|
|
Object qobject = en.nextElement();
|
232 |
|
|
self.append(qobject);
|
233 |
|
|
}
|
234 |
|
|
self.append(" )\n");
|
235 |
|
|
return self.toString();
|
236 |
|
|
}
|
237 |
3225
|
tao
|
|
238 |
|
|
/*
|
239 |
|
|
* When a new QueryTerm come, first we need to compare it to
|
240 |
|
|
* the queryTerm vector, which contains queryTerm that doesn't
|
241 |
|
|
* have same search value to any other queryTerm. Here is algorithm.
|
242 |
|
|
* 1) If new QueryTerm find a QueryTerm in queryTerms which has same search value,
|
243 |
|
|
* them create a new vector which contain both QueryTerms and add the new vector
|
244 |
|
|
* to two-dimention vector queryTermsWithSameValue, and remove the QueryTerm which
|
245 |
|
|
* was in queryTerm.
|
246 |
|
|
* 2) If new QueryTerm couldn't find a QueryTerm in queryTerms which has same search value,
|
247 |
|
|
* then search queryTermsWithSameValue, to see if this vector already has the search value.
|
248 |
|
|
* 2.1) if has the search value, add the new QueryTerm to the queryTermsWithSameValue.
|
249 |
|
|
* 2.2) if hasn't, add the new QueryTerm to queryTerms vector.
|
250 |
|
|
*/
|
251 |
|
|
private void handleNewQueryTerms(QueryTerm newTerm)
|
252 |
|
|
{
|
253 |
|
|
if (newTerm != null)
|
254 |
|
|
{
|
255 |
|
|
for (int i=0; i<queryTerms.size(); i++)
|
256 |
|
|
{
|
257 |
|
|
QueryTerm term = (QueryTerm)queryTerms.elementAt(i);
|
258 |
|
|
if (term != null && term.hasSameSearchValue(newTerm))
|
259 |
|
|
{
|
260 |
|
|
// find a target which has same search value
|
261 |
|
|
Vector newSameValueVector = new Vector();
|
262 |
|
|
newSameValueVector.add(term);
|
263 |
|
|
newSameValueVector.addElement(newTerm);
|
264 |
|
|
queryTermsWithSameValue.add(newSameValueVector);
|
265 |
|
|
queryTerms.remove(i);
|
266 |
|
|
return;
|
267 |
|
|
}
|
268 |
|
|
}
|
269 |
|
|
// no same search value was found in queryTerms.
|
270 |
|
|
// then we need search queryTermsWithSameValue
|
271 |
|
|
for (int i=0; i<queryTermsWithSameValue.size(); i++)
|
272 |
|
|
{
|
273 |
|
|
Vector sameValueVec = (Vector)queryTermsWithSameValue.elementAt(i);
|
274 |
|
|
// we only compare the first query term
|
275 |
|
|
QueryTerm term = (QueryTerm)sameValueVec.elementAt(0);
|
276 |
|
|
if (term != null && term.hasSameSearchValue(newTerm))
|
277 |
|
|
{
|
278 |
|
|
sameValueVec.add(newTerm);
|
279 |
|
|
return;
|
280 |
|
|
}
|
281 |
|
|
}
|
282 |
|
|
// both queryTerms and queryTermsWithSameValue don't have the search value,
|
283 |
|
|
// add this newTerm to queryTerms
|
284 |
|
|
queryTerms.add(newTerm);
|
285 |
|
|
}
|
286 |
|
|
|
287 |
|
|
}
|
288 |
1831
|
tao
|
}
|