1
|
/**
|
2
|
* '$RCSfile$'
|
3
|
* 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
|
* 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
|
* Release: @release@
|
11
|
*
|
12
|
* '$Author: berkley $'
|
13
|
* '$Date: 2002-01-18 10:24:13 -0800 (Fri, 18 Jan 2002) $'
|
14
|
* '$Revision: 899 $'
|
15
|
*
|
16
|
* This program is free software; you can redistribute it and/or modify
|
17
|
* it under the terms of the GNU General Public License as published by
|
18
|
* the Free Software Foundation; either version 2 of the License, or
|
19
|
* (at your option) any later version.
|
20
|
*
|
21
|
* This program is distributed in the hope that it will be useful,
|
22
|
* but WITHOUT ANY WARRANTY; without even the implied warranty of
|
23
|
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
|
24
|
* GNU General Public License for more details.
|
25
|
*
|
26
|
* You should have received a copy of the GNU General Public License
|
27
|
* along with this program; if not, write to the Free Software
|
28
|
* Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
|
29
|
*/
|
30
|
|
31
|
package edu.ucsb.nceas.metacat;
|
32
|
|
33
|
import edu.ucsb.nceas.dbadapter.*;
|
34
|
|
35
|
import java.io.*;
|
36
|
import java.util.Stack;
|
37
|
import java.util.Vector;
|
38
|
import java.util.Enumeration;
|
39
|
|
40
|
import org.xml.sax.Attributes;
|
41
|
import org.xml.sax.InputSource;
|
42
|
import org.xml.sax.SAXException;
|
43
|
import org.xml.sax.SAXParseException;
|
44
|
import org.xml.sax.XMLReader;
|
45
|
import org.xml.sax.helpers.XMLReaderFactory;
|
46
|
import org.xml.sax.helpers.DefaultHandler;
|
47
|
|
48
|
/**
|
49
|
* A Class that represents a structured query, and can be
|
50
|
* constructed from an XML serialization conforming to @see pathquery.dtd.
|
51
|
* The printSQL() method can be used to print a SQL serialization of the query.
|
52
|
*/
|
53
|
public class QuerySpecification extends DefaultHandler {
|
54
|
|
55
|
/** flag determining whether extended query terms are present */
|
56
|
private boolean containsExtendedSQL=false;
|
57
|
/** Identifier for this query document */
|
58
|
private String meta_file_id;
|
59
|
/** Title of this query */
|
60
|
private String queryTitle;
|
61
|
/** List of document types to be returned using package back tracing */
|
62
|
private Vector returnDocList;
|
63
|
/** List of document types to be searched */
|
64
|
private Vector filterDocList;
|
65
|
/** List of fields to be returned in result set */
|
66
|
private Vector returnFieldList;
|
67
|
/** List of users owning documents to be searched */
|
68
|
private Vector ownerList;
|
69
|
/** List of sites/scopes used to constrain search */
|
70
|
private Vector siteList;
|
71
|
/** The root query group that contains the recursive query constraints */
|
72
|
private QueryGroup query = null;
|
73
|
|
74
|
// Query data structures used temporarily during XML parsing
|
75
|
private Stack elementStack;
|
76
|
private Stack queryStack;
|
77
|
private String currentValue;
|
78
|
private String currentPathexpr;
|
79
|
private String parserName = null;
|
80
|
private String accNumberSeparator = null;
|
81
|
private static final AbstractDatabase dbAdapter = MetaCatUtil.dbAdapter;
|
82
|
|
83
|
|
84
|
/**
|
85
|
* construct an instance of the QuerySpecification class
|
86
|
*
|
87
|
* @param queryspec the XML representation of the query (should conform
|
88
|
* to pathquery.dtd) as a Reader
|
89
|
* @param parserName the fully qualified name of a Java Class implementing
|
90
|
* the org.xml.sax.XMLReader interface
|
91
|
*/
|
92
|
public QuerySpecification( Reader queryspec, String parserName,
|
93
|
String accNumberSeparator ) throws IOException {
|
94
|
super();
|
95
|
|
96
|
// Initialize the class variables
|
97
|
returnDocList = new Vector();
|
98
|
filterDocList = new Vector();
|
99
|
elementStack = new Stack();
|
100
|
queryStack = new Stack();
|
101
|
returnFieldList = new Vector();
|
102
|
ownerList = new Vector();
|
103
|
siteList = new Vector();
|
104
|
this.parserName = parserName;
|
105
|
this.accNumberSeparator = accNumberSeparator;
|
106
|
|
107
|
// Initialize the parser and read the queryspec
|
108
|
XMLReader parser = initializeParser();
|
109
|
if (parser == null) {
|
110
|
System.err.println("SAX parser not instantiated properly.");
|
111
|
}
|
112
|
try {
|
113
|
parser.parse(new InputSource(queryspec));
|
114
|
} catch (SAXException e) {
|
115
|
System.err.println("error parsing data in " +
|
116
|
"QuerySpecification.QuerySpecification");
|
117
|
System.err.println(e.getMessage());
|
118
|
}
|
119
|
}
|
120
|
|
121
|
/**
|
122
|
* construct an instance of the QuerySpecification class
|
123
|
*
|
124
|
* @param queryspec the XML representation of the query (should conform
|
125
|
* to pathquery.dtd) as a String
|
126
|
* @param parserName the fully qualified name of a Java Class implementing
|
127
|
* the org.xml.sax.Parser interface
|
128
|
*/
|
129
|
public QuerySpecification( String queryspec, String parserName,
|
130
|
String accNumberSeparator) throws IOException {
|
131
|
this(new StringReader(queryspec), parserName, accNumberSeparator);
|
132
|
}
|
133
|
|
134
|
/** Main routine for testing */
|
135
|
static public void main(String[] args) {
|
136
|
|
137
|
if (args.length < 1) {
|
138
|
System.err.println("Wrong number of arguments!!!");
|
139
|
System.err.println("USAGE: java QuerySpecification <xmlfile>");
|
140
|
return;
|
141
|
} else {
|
142
|
int i = 0;
|
143
|
boolean useXMLIndex = true;
|
144
|
if ( args[i].equals( "-noindex" ) ) {
|
145
|
useXMLIndex = false;
|
146
|
i++;
|
147
|
}
|
148
|
String xmlfile = args[i];
|
149
|
|
150
|
try {
|
151
|
MetaCatUtil util = new MetaCatUtil();
|
152
|
FileReader xml = new FileReader(new File(xmlfile));
|
153
|
QuerySpecification qspec =
|
154
|
new QuerySpecification(xml, util.getOption("saxparser"),
|
155
|
util.getOption("accNumberSeparator"));
|
156
|
System.out.println(qspec.printSQL(useXMLIndex));
|
157
|
|
158
|
} catch (IOException e) {
|
159
|
System.err.println(e.getMessage());
|
160
|
}
|
161
|
|
162
|
}
|
163
|
}
|
164
|
|
165
|
/**
|
166
|
* Returns true if the parsed query contains and extended xml query
|
167
|
* (i.e. there is at least one <returnfield> in the pathquery document)
|
168
|
*/
|
169
|
public boolean containsExtendedSQL()
|
170
|
{
|
171
|
if(containsExtendedSQL)
|
172
|
{
|
173
|
return true;
|
174
|
}
|
175
|
else
|
176
|
{
|
177
|
return false;
|
178
|
}
|
179
|
}
|
180
|
|
181
|
/**
|
182
|
* Accessor method to return the identifier of this Query
|
183
|
*/
|
184
|
public String getIdentifier()
|
185
|
{
|
186
|
return meta_file_id;
|
187
|
}
|
188
|
|
189
|
/**
|
190
|
* method to set the identifier of this query
|
191
|
*/
|
192
|
public void setIdentifier(String id) {
|
193
|
this.meta_file_id = id;
|
194
|
}
|
195
|
|
196
|
/**
|
197
|
* Accessor method to return the title of this Query
|
198
|
*/
|
199
|
public String getQueryTitle()
|
200
|
{
|
201
|
return queryTitle;
|
202
|
}
|
203
|
|
204
|
/**
|
205
|
* method to set the title of this query
|
206
|
*/
|
207
|
public void setQueryTitle(String title)
|
208
|
{
|
209
|
this.queryTitle = title;
|
210
|
}
|
211
|
|
212
|
/**
|
213
|
* Accessor method to return a vector of the return document types as
|
214
|
* defined in the <returndoctype> tag in the pathquery dtd.
|
215
|
*/
|
216
|
public Vector getReturnDocList()
|
217
|
{
|
218
|
return this.returnDocList;
|
219
|
}
|
220
|
|
221
|
/**
|
222
|
* method to set the list of return docs of this query
|
223
|
*/
|
224
|
public void setReturnDocList(Vector returnDocList)
|
225
|
{
|
226
|
this.returnDocList = returnDocList;
|
227
|
}
|
228
|
|
229
|
/**
|
230
|
* Accessor method to return a vector of the filter doc types as
|
231
|
* defined in the <filterdoctype> tag in the pathquery dtd.
|
232
|
*/
|
233
|
public Vector getFilterDocList()
|
234
|
{
|
235
|
return this.filterDocList;
|
236
|
}
|
237
|
|
238
|
/**
|
239
|
* method to set the list of filter docs of this query
|
240
|
*/
|
241
|
public void setFilterDocList(Vector filterDocList)
|
242
|
{
|
243
|
this.filterDocList = filterDocList;
|
244
|
}
|
245
|
|
246
|
/**
|
247
|
* Accessor method to return a vector of the extended return fields as
|
248
|
* defined in the <returnfield> tag in the pathquery dtd.
|
249
|
*/
|
250
|
public Vector getReturnFieldList()
|
251
|
{
|
252
|
return this.returnFieldList;
|
253
|
}
|
254
|
|
255
|
/**
|
256
|
* method to set the list of fields to be returned by this query
|
257
|
*/
|
258
|
public void setReturnFieldList(Vector returnFieldList)
|
259
|
{
|
260
|
this.returnFieldList = returnFieldList;
|
261
|
}
|
262
|
|
263
|
/**
|
264
|
* Accessor method to return a vector of the owner fields as
|
265
|
* defined in the <owner> tag in the pathquery dtd.
|
266
|
*/
|
267
|
public Vector getOwnerList()
|
268
|
{
|
269
|
return this.ownerList;
|
270
|
}
|
271
|
|
272
|
/**
|
273
|
* method to set the list of owners used to constrain this query
|
274
|
*/
|
275
|
public void setOwnerList(Vector ownerList)
|
276
|
{
|
277
|
this.ownerList = ownerList;
|
278
|
}
|
279
|
|
280
|
/**
|
281
|
* Accessor method to return a vector of the site fields as
|
282
|
* defined in the <site> tag in the pathquery dtd.
|
283
|
*/
|
284
|
public Vector getSiteList()
|
285
|
{
|
286
|
return this.siteList;
|
287
|
}
|
288
|
|
289
|
/**
|
290
|
* method to set the list of sites used to constrain this query
|
291
|
*/
|
292
|
public void setSiteList(Vector siteList)
|
293
|
{
|
294
|
this.siteList = siteList;
|
295
|
}
|
296
|
|
297
|
/**
|
298
|
* get the QueryGroup used to express query constraints
|
299
|
*/
|
300
|
public QueryGroup getQueryGroup()
|
301
|
{
|
302
|
return query;
|
303
|
}
|
304
|
|
305
|
/**
|
306
|
* Set up the SAX parser for reading the XML serialized query
|
307
|
*/
|
308
|
private XMLReader initializeParser() {
|
309
|
XMLReader parser = null;
|
310
|
|
311
|
// Set up the SAX document handlers for parsing
|
312
|
try {
|
313
|
|
314
|
// Get an instance of the parser
|
315
|
parser = XMLReaderFactory.createXMLReader(parserName);
|
316
|
|
317
|
// Set the ContentHandler to this instance
|
318
|
parser.setContentHandler(this);
|
319
|
|
320
|
// Set the error Handler to this instance
|
321
|
parser.setErrorHandler(this);
|
322
|
|
323
|
} catch (Exception e) {
|
324
|
System.err.println("Error in QuerySpcecification.initializeParser " +
|
325
|
e.toString());
|
326
|
}
|
327
|
|
328
|
return parser;
|
329
|
}
|
330
|
|
331
|
/**
|
332
|
* callback method used by the SAX Parser when the start tag of an
|
333
|
* element is detected. Used in this context to parse and store
|
334
|
* the query information in class variables.
|
335
|
*/
|
336
|
public void startElement (String uri, String localName,
|
337
|
String qName, Attributes atts)
|
338
|
throws SAXException {
|
339
|
BasicNode currentNode = new BasicNode(localName);
|
340
|
// add attributes to BasicNode here
|
341
|
if (atts != null) {
|
342
|
int len = atts.getLength();
|
343
|
for (int i = 0; i < len; i++) {
|
344
|
currentNode.setAttribute(atts.getLocalName(i), atts.getValue(i));
|
345
|
}
|
346
|
}
|
347
|
|
348
|
elementStack.push(currentNode);
|
349
|
if (currentNode.getTagName().equals("querygroup")) {
|
350
|
QueryGroup currentGroup = new QueryGroup(
|
351
|
currentNode.getAttribute("operator"));
|
352
|
if (query == null) {
|
353
|
query = currentGroup;
|
354
|
} else {
|
355
|
QueryGroup parentGroup = (QueryGroup)queryStack.peek();
|
356
|
parentGroup.addChild(currentGroup);
|
357
|
}
|
358
|
queryStack.push(currentGroup);
|
359
|
}
|
360
|
}
|
361
|
|
362
|
/**
|
363
|
* callback method used by the SAX Parser when the end tag of an
|
364
|
* element is detected. Used in this context to parse and store
|
365
|
* the query information in class variables.
|
366
|
*/
|
367
|
public void endElement (String uri, String localName,
|
368
|
String qName) throws SAXException {
|
369
|
BasicNode leaving = (BasicNode)elementStack.pop();
|
370
|
if (leaving.getTagName().equals("queryterm")) {
|
371
|
boolean isCaseSensitive = (new Boolean(
|
372
|
leaving.getAttribute("casesensitive"))).booleanValue();
|
373
|
QueryTerm currentTerm = null;
|
374
|
if (currentPathexpr == null) {
|
375
|
currentTerm = new QueryTerm(isCaseSensitive,
|
376
|
leaving.getAttribute("searchmode"),currentValue);
|
377
|
} else {
|
378
|
currentTerm = new QueryTerm(isCaseSensitive,
|
379
|
leaving.getAttribute("searchmode"),currentValue,
|
380
|
currentPathexpr);
|
381
|
}
|
382
|
QueryGroup currentGroup = (QueryGroup)queryStack.peek();
|
383
|
currentGroup.addChild(currentTerm);
|
384
|
currentValue = null;
|
385
|
currentPathexpr = null;
|
386
|
} else if (leaving.getTagName().equals("querygroup")) {
|
387
|
QueryGroup leavingGroup = (QueryGroup)queryStack.pop();
|
388
|
}
|
389
|
}
|
390
|
|
391
|
/**
|
392
|
* callback method used by the SAX Parser when the text sequences of an
|
393
|
* xml stream are detected. Used in this context to parse and store
|
394
|
* the query information in class variables.
|
395
|
*/
|
396
|
public void characters(char ch[], int start, int length) {
|
397
|
|
398
|
String inputString = new String(ch, start, length);
|
399
|
BasicNode currentNode = (BasicNode)elementStack.peek();
|
400
|
String currentTag = currentNode.getTagName();
|
401
|
if (currentTag.equals("meta_file_id")) {
|
402
|
meta_file_id = inputString;
|
403
|
} else if (currentTag.equals("querytitle")) {
|
404
|
queryTitle = inputString;
|
405
|
} else if (currentTag.equals("value")) {
|
406
|
currentValue = inputString;
|
407
|
} else if (currentTag.equals("pathexpr")) {
|
408
|
currentPathexpr = inputString;
|
409
|
} else if (currentTag.equals("returndoctype")) {
|
410
|
returnDocList.add(inputString);
|
411
|
} else if (currentTag.equals("filterdoctype")) {
|
412
|
filterDocList.add(inputString);
|
413
|
} else if (currentTag.equals("returnfield")) {
|
414
|
returnFieldList.add(inputString);
|
415
|
containsExtendedSQL = true;
|
416
|
} else if (currentTag.equals("filterdoctype")) {
|
417
|
filterDocList.add(inputString);
|
418
|
} else if (currentTag.equals("owner")) {
|
419
|
ownerList.add(inputString);
|
420
|
} else if (currentTag.equals("site")) {
|
421
|
siteList.add(inputString);
|
422
|
}
|
423
|
}
|
424
|
|
425
|
/**
|
426
|
* create a SQL serialization of the query that this instance represents
|
427
|
*/
|
428
|
public String printSQL(boolean useXMLIndex) {
|
429
|
StringBuffer self = new StringBuffer();
|
430
|
|
431
|
self.append("SELECT docid,docname,doctype,");
|
432
|
self.append("date_created, date_updated, rev ");
|
433
|
self.append("FROM xml_documents WHERE docid IN (");
|
434
|
|
435
|
// This determines the documents that meet the query conditions
|
436
|
self.append(query.printSQL(useXMLIndex));
|
437
|
|
438
|
self.append(") ");
|
439
|
|
440
|
// Add SQL to filter for doctypes requested in the query
|
441
|
// This is an implicit OR for the list of doctypes. Only doctypes in this
|
442
|
// list will be searched if the tag is present
|
443
|
if (!filterDocList.isEmpty()) {
|
444
|
boolean firstdoctype = true;
|
445
|
self.append(" AND (");
|
446
|
Enumeration en = filterDocList.elements();
|
447
|
while (en.hasMoreElements()) {
|
448
|
String currentDoctype = (String)en.nextElement();
|
449
|
if (firstdoctype) {
|
450
|
firstdoctype = false;
|
451
|
self.append(" doctype = '" + currentDoctype + "'");
|
452
|
} else {
|
453
|
self.append(" OR doctype = '" + currentDoctype + "'");
|
454
|
}
|
455
|
}
|
456
|
self.append(") ");
|
457
|
}
|
458
|
|
459
|
// Add SQL to filter for owners requested in the query
|
460
|
// This is an implicit OR for the list of owners
|
461
|
if (!ownerList.isEmpty()) {
|
462
|
boolean first = true;
|
463
|
self.append(" AND (");
|
464
|
Enumeration en = ownerList.elements();
|
465
|
while (en.hasMoreElements()) {
|
466
|
String current = (String)en.nextElement();
|
467
|
if (first) {
|
468
|
first = false;
|
469
|
self.append(" user_owner = '" + current + "'");
|
470
|
} else {
|
471
|
self.append(" OR user_owner = '" + current + "'");
|
472
|
}
|
473
|
}
|
474
|
self.append(") ");
|
475
|
}
|
476
|
|
477
|
// Add SQL to filter for sites requested in the query
|
478
|
// This is an implicit OR for the list of sites
|
479
|
if (!siteList.isEmpty()) {
|
480
|
boolean first = true;
|
481
|
self.append(" AND (");
|
482
|
Enumeration en = siteList.elements();
|
483
|
while (en.hasMoreElements()) {
|
484
|
String current = (String)en.nextElement();
|
485
|
if (first) {
|
486
|
first = false;
|
487
|
self.append(" SUBSTR(docid, 1, INSTR(docid, '" +
|
488
|
accNumberSeparator + "')-1) = '" + current + "'");
|
489
|
} else {
|
490
|
self.append(" OR SUBSTR(docid, 1, INSTR(docid, '" +
|
491
|
accNumberSeparator + "')-1) = '" + current + "'");
|
492
|
}
|
493
|
}
|
494
|
self.append(") ");
|
495
|
}
|
496
|
//System.out.println(self.toString());
|
497
|
return self.toString();
|
498
|
}
|
499
|
|
500
|
/**
|
501
|
* This method prints sql based upon the <returnfield> tag in the
|
502
|
* pathquery document. This allows for customization of the
|
503
|
* returned fields
|
504
|
* @param doclist the list of document ids to search by
|
505
|
*/
|
506
|
public String printExtendedSQL(String doclist)
|
507
|
{
|
508
|
StringBuffer self = new StringBuffer();
|
509
|
self.append("select xml_nodes.docid, xml_index.path, xml_nodes.nodedata ");
|
510
|
self.append("from xml_index, xml_nodes where xml_index.nodeid=");
|
511
|
self.append("xml_nodes.parentnodeid and (xml_index.path like '");
|
512
|
boolean firstfield = true;
|
513
|
//put the returnfields into the query
|
514
|
//the for loop allows for multiple fields
|
515
|
for(int i=0; i<returnFieldList.size(); i++)
|
516
|
{
|
517
|
if(firstfield)
|
518
|
{
|
519
|
firstfield = false;
|
520
|
self.append((String)returnFieldList.elementAt(i));
|
521
|
self.append("' ");
|
522
|
}
|
523
|
else
|
524
|
{
|
525
|
self.append("or xml_index.path like '");
|
526
|
self.append((String)returnFieldList.elementAt(i));
|
527
|
self.append("' ");
|
528
|
}
|
529
|
}
|
530
|
self.append(") AND xml_nodes.docid in (");
|
531
|
//self.append(query.printSQL());
|
532
|
self.append(doclist);
|
533
|
self.append(")");
|
534
|
self.append(" AND xml_nodes.nodetype = 'TEXT'");
|
535
|
|
536
|
//System.out.println(self.toString());
|
537
|
return self.toString();
|
538
|
}
|
539
|
|
540
|
public static String printRelationSQL(String docid)
|
541
|
{
|
542
|
StringBuffer self = new StringBuffer();
|
543
|
self.append("select subject, relationship, object, subdoctype, ");
|
544
|
self.append("objdoctype from xml_relation ");
|
545
|
self.append("where docid like '").append(docid).append("'");
|
546
|
return self.toString();
|
547
|
}
|
548
|
|
549
|
/**
|
550
|
* Prints sql that returns all relations in the database.
|
551
|
*/
|
552
|
public static String printPackageSQL()
|
553
|
{
|
554
|
StringBuffer self = new StringBuffer();
|
555
|
self.append("select z.nodedata, x.nodedata, y.nodedata from ");
|
556
|
self.append("(select nodeid, parentnodeid from xml_index where path like ");
|
557
|
self.append("'triple/subject') s, (select nodeid, parentnodeid ");
|
558
|
self.append("from xml_index where path like ");
|
559
|
self.append("'triple/relationship') rel, ");
|
560
|
self.append("(select nodeid, parentnodeid from xml_index where path like ");
|
561
|
self.append("'triple/object') o, ");
|
562
|
self.append("xml_nodes x, xml_nodes y, xml_nodes z ");
|
563
|
self.append("where s.parentnodeid = rel.parentnodeid ");
|
564
|
self.append("and rel.parentnodeid = o.parentnodeid ");
|
565
|
self.append("and x.parentnodeid in (rel.nodeid) ");
|
566
|
self.append("and y.parentnodeid in (o.nodeid) ");
|
567
|
self.append("and z.parentnodeid in (s.nodeid) ");
|
568
|
//self.append("and z.nodedata like '%");
|
569
|
//self.append(docid);
|
570
|
//self.append("%'");
|
571
|
return self.toString();
|
572
|
}
|
573
|
|
574
|
/**
|
575
|
* Prints sql that returns all relations in the database that were input
|
576
|
* under a specific docid
|
577
|
* @param docid the docid to search for.
|
578
|
*/
|
579
|
public static String printPackageSQL(String docid)
|
580
|
{
|
581
|
StringBuffer self = new StringBuffer();
|
582
|
self.append("select z.nodedata, x.nodedata, y.nodedata from ");
|
583
|
self.append("(select nodeid, parentnodeid from xml_index where path like ");
|
584
|
self.append("'triple/subject') s, (select nodeid, parentnodeid ");
|
585
|
self.append("from xml_index where path like ");
|
586
|
self.append("'triple/relationship') rel, ");
|
587
|
self.append("(select nodeid, parentnodeid from xml_index where path like ");
|
588
|
self.append("'triple/object') o, ");
|
589
|
self.append("xml_nodes x, xml_nodes y, xml_nodes z ");
|
590
|
self.append("where s.parentnodeid = rel.parentnodeid ");
|
591
|
self.append("and rel.parentnodeid = o.parentnodeid ");
|
592
|
self.append("and x.parentnodeid in (rel.nodeid) ");
|
593
|
self.append("and y.parentnodeid in (o.nodeid) ");
|
594
|
self.append("and z.parentnodeid in (s.nodeid) ");
|
595
|
self.append("and z.docid like '").append(docid).append("'");
|
596
|
|
597
|
return self.toString();
|
598
|
}
|
599
|
|
600
|
/**
|
601
|
* Returns all of the relations that has a certain docid in the subject
|
602
|
* or the object.
|
603
|
*
|
604
|
* @param docid the docid to search for
|
605
|
*/
|
606
|
public static String printPackageSQL(String subDocidURL, String objDocidURL)
|
607
|
{
|
608
|
StringBuffer self = new StringBuffer();
|
609
|
self.append("select z.nodedata, x.nodedata, y.nodedata from ");
|
610
|
self.append("(select nodeid, parentnodeid from xml_index where path like ");
|
611
|
self.append("'triple/subject') s, (select nodeid, parentnodeid ");
|
612
|
self.append("from xml_index where path like ");
|
613
|
self.append("'triple/relationship') rel, ");
|
614
|
self.append("(select nodeid, parentnodeid from xml_index where path like ");
|
615
|
self.append("'triple/object') o, ");
|
616
|
self.append("xml_nodes x, xml_nodes y, xml_nodes z ");
|
617
|
self.append("where s.parentnodeid = rel.parentnodeid ");
|
618
|
self.append("and rel.parentnodeid = o.parentnodeid ");
|
619
|
self.append("and x.parentnodeid in (rel.nodeid) ");
|
620
|
self.append("and y.parentnodeid in (o.nodeid) ");
|
621
|
self.append("and z.parentnodeid in (s.nodeid) ");
|
622
|
self.append("and (z.nodedata like '");
|
623
|
self.append(subDocidURL);
|
624
|
self.append("' or y.nodedata like '");
|
625
|
self.append(objDocidURL);
|
626
|
self.append("')");
|
627
|
return self.toString();
|
628
|
}
|
629
|
|
630
|
public static String printGetDocByDoctypeSQL(String docid)
|
631
|
{
|
632
|
StringBuffer self = new StringBuffer();
|
633
|
|
634
|
self.append("SELECT docid,docname,doctype,");
|
635
|
self.append("date_created, date_updated ");
|
636
|
self.append("FROM xml_documents WHERE docid IN (");
|
637
|
self.append(docid).append(")");
|
638
|
return self.toString();
|
639
|
}
|
640
|
|
641
|
/**
|
642
|
* create a String description of the query that this instance represents.
|
643
|
* This should become a way to get the XML serialization of the query.
|
644
|
*/
|
645
|
public String toString() {
|
646
|
return "meta_file_id=" + meta_file_id + "\n" + query;
|
647
|
//DOCTITLE attr cleared from the db
|
648
|
// return "meta_file_id=" + meta_file_id + "\n" +
|
649
|
// "querytitle=" + querytitle + "\n" + query;
|
650
|
}
|
651
|
|
652
|
/** a utility class that represents a group of terms in a query */
|
653
|
private class QueryGroup {
|
654
|
private String operator = null; // indicates how query terms are combined
|
655
|
private Vector children = null; // the list of query terms and groups
|
656
|
|
657
|
/**
|
658
|
* construct a new QueryGroup
|
659
|
*
|
660
|
* @param operator the boolean conector used to connect query terms
|
661
|
* in this query group
|
662
|
*/
|
663
|
public QueryGroup(String operator) {
|
664
|
this.operator = operator;
|
665
|
children = new Vector();
|
666
|
}
|
667
|
|
668
|
/**
|
669
|
* Add a child QueryGroup to this QueryGroup
|
670
|
*
|
671
|
* @param qgroup the query group to be added to the list of terms
|
672
|
*/
|
673
|
public void addChild(QueryGroup qgroup) {
|
674
|
children.add((Object)qgroup);
|
675
|
}
|
676
|
|
677
|
/**
|
678
|
* Add a child QueryTerm to this QueryGroup
|
679
|
*
|
680
|
* @param qterm the query term to be added to the list of terms
|
681
|
*/
|
682
|
public void addChild(QueryTerm qterm) {
|
683
|
children.add((Object)qterm);
|
684
|
}
|
685
|
|
686
|
/**
|
687
|
* Retrieve an Enumeration of query terms for this QueryGroup
|
688
|
*/
|
689
|
public Enumeration getChildren() {
|
690
|
return children.elements();
|
691
|
}
|
692
|
|
693
|
/**
|
694
|
* create a SQL serialization of the query that this instance represents
|
695
|
*/
|
696
|
public String printSQL(boolean useXMLIndex) {
|
697
|
StringBuffer self = new StringBuffer();
|
698
|
boolean first = true;
|
699
|
|
700
|
self.append("(");
|
701
|
|
702
|
Enumeration en= getChildren();
|
703
|
while (en.hasMoreElements()) {
|
704
|
Object qobject = en.nextElement();
|
705
|
if (first) {
|
706
|
first = false;
|
707
|
} else {
|
708
|
self.append(" " + operator + " ");
|
709
|
}
|
710
|
if (qobject instanceof QueryGroup) {
|
711
|
QueryGroup qg = (QueryGroup)qobject;
|
712
|
self.append(qg.printSQL(useXMLIndex));
|
713
|
} else if (qobject instanceof QueryTerm) {
|
714
|
QueryTerm qt = (QueryTerm)qobject;
|
715
|
self.append(qt.printSQL(useXMLIndex));
|
716
|
} else {
|
717
|
System.err.println("qobject wrong type: fatal error");
|
718
|
}
|
719
|
}
|
720
|
self.append(") \n");
|
721
|
return self.toString();
|
722
|
}
|
723
|
|
724
|
/**
|
725
|
* create a String description of the query that this instance represents.
|
726
|
* This should become a way to get the XML serialization of the query.
|
727
|
*/
|
728
|
public String toString() {
|
729
|
StringBuffer self = new StringBuffer();
|
730
|
|
731
|
self.append(" (Query group operator=" + operator + "\n");
|
732
|
Enumeration en= getChildren();
|
733
|
while (en.hasMoreElements()) {
|
734
|
Object qobject = en.nextElement();
|
735
|
self.append(qobject);
|
736
|
}
|
737
|
self.append(" )\n");
|
738
|
return self.toString();
|
739
|
}
|
740
|
}
|
741
|
|
742
|
/** a utility class that represents a single term in a query */
|
743
|
private class QueryTerm {
|
744
|
private boolean casesensitive = false;
|
745
|
private String searchmode = null;
|
746
|
private String value = null;
|
747
|
private String pathexpr = null;
|
748
|
|
749
|
/**
|
750
|
* Construct a new instance of a query term for a free text search
|
751
|
* (using the value only)
|
752
|
*
|
753
|
* @param casesensitive flag indicating whether case is used to match
|
754
|
* @param searchmode determines what kind of substring match is performed
|
755
|
* (one of starts-with|ends-with|contains|matches-exactly)
|
756
|
* @param value the text value to match
|
757
|
*/
|
758
|
public QueryTerm(boolean casesensitive, String searchmode,
|
759
|
String value) {
|
760
|
this.casesensitive = casesensitive;
|
761
|
this.searchmode = searchmode;
|
762
|
this.value = value;
|
763
|
}
|
764
|
|
765
|
/**
|
766
|
* Construct a new instance of a query term for a structured search
|
767
|
* (matching the value only for those nodes in the pathexpr)
|
768
|
*
|
769
|
* @param casesensitive flag indicating whether case is used to match
|
770
|
* @param searchmode determines what kind of substring match is performed
|
771
|
* (one of starts-with|ends-with|contains|matches-exactly)
|
772
|
* @param value the text value to match
|
773
|
* @param pathexpr the hierarchical path to the nodes to be searched
|
774
|
*/
|
775
|
public QueryTerm(boolean casesensitive, String searchmode,
|
776
|
String value, String pathexpr) {
|
777
|
this(casesensitive, searchmode, value);
|
778
|
this.pathexpr = pathexpr;
|
779
|
}
|
780
|
|
781
|
/** determine if the QueryTerm is case sensitive */
|
782
|
public boolean isCaseSensitive() {
|
783
|
return casesensitive;
|
784
|
}
|
785
|
|
786
|
/** get the searchmode parameter */
|
787
|
public String getSearchMode() {
|
788
|
return searchmode;
|
789
|
}
|
790
|
|
791
|
/** get the Value parameter */
|
792
|
public String getValue() {
|
793
|
return value;
|
794
|
}
|
795
|
|
796
|
/** get the path expression parameter */
|
797
|
public String getPathExpression() {
|
798
|
return pathexpr;
|
799
|
}
|
800
|
|
801
|
/**
|
802
|
* create a SQL serialization of the query that this instance represents
|
803
|
*/
|
804
|
public String printSQL(boolean useXMLIndex) {
|
805
|
StringBuffer self = new StringBuffer();
|
806
|
|
807
|
// Uppercase the search string if case match is not important
|
808
|
String casevalue = null;
|
809
|
String nodedataterm = null;
|
810
|
|
811
|
if (casesensitive) {
|
812
|
nodedataterm = "nodedata";
|
813
|
casevalue = value;
|
814
|
} else {
|
815
|
nodedataterm = "UPPER(nodedata)";
|
816
|
casevalue = value.toUpperCase();
|
817
|
}
|
818
|
|
819
|
// Add appropriate wildcards to search string
|
820
|
//String searchvalue = null;
|
821
|
String searchexpr = null;
|
822
|
if (searchmode.equals("starts-with")) {
|
823
|
//searchvalue = casevalue + "%";
|
824
|
searchexpr = nodedataterm + " LIKE '" + casevalue + "%' ";
|
825
|
} else if (searchmode.equals("ends-with")) {
|
826
|
//searchvalue = "%" + casevalue;
|
827
|
searchexpr = nodedataterm + " LIKE '%" + casevalue + "' ";
|
828
|
} else if (searchmode.equals("contains")) {
|
829
|
//searchvalue = "%" + casevalue + "%";
|
830
|
searchexpr = nodedataterm + " LIKE '%" + casevalue + "%' ";
|
831
|
} else if (searchmode.equals("equals")) {
|
832
|
//searchvalue = casevalue;
|
833
|
searchexpr = nodedataterm + " = '" + casevalue + "' ";
|
834
|
} else if (searchmode.equals("isnot-equal")) {
|
835
|
//searchvalue = casevalue;
|
836
|
searchexpr = nodedataterm + " != '" + casevalue + "' ";
|
837
|
} else {
|
838
|
//searchvalue = casevalue;
|
839
|
String oper = null;
|
840
|
if (searchmode.equals("greater-than")) {
|
841
|
oper = ">";
|
842
|
} else if (searchmode.equals("greater-than-equals")) {
|
843
|
oper = ">=";
|
844
|
} else if (searchmode.equals("less-than")) {
|
845
|
oper = "<";
|
846
|
} else if (searchmode.equals("less-than-equals")) {
|
847
|
oper = "<=";
|
848
|
} else {
|
849
|
System.out.println("NOT expected case. NOT recognized operator: " +
|
850
|
searchmode);
|
851
|
return null;
|
852
|
}
|
853
|
try {
|
854
|
// it is number; numeric comparison
|
855
|
searchexpr = nodedataterm + " " + oper + " " +
|
856
|
new Double(casevalue) + " ";
|
857
|
} catch (NumberFormatException nfe) {
|
858
|
// these are characters; character comparison
|
859
|
searchexpr = nodedataterm + " " + oper + " '" + casevalue + "' ";
|
860
|
}
|
861
|
}
|
862
|
|
863
|
self.append("SELECT DISTINCT docid FROM xml_nodes WHERE \n");
|
864
|
//self.append(nodedataterm + " LIKE " + "'" + searchvalue + "' ");
|
865
|
self.append(searchexpr);
|
866
|
|
867
|
if (pathexpr != null) {
|
868
|
self.append("AND parentnodeid IN ");
|
869
|
// use XML Index
|
870
|
if ( useXMLIndex ) {
|
871
|
self.append("(SELECT nodeid FROM xml_index WHERE path LIKE " +
|
872
|
"'" + pathexpr + "') " );
|
873
|
// without using XML Index; using nested statements instead
|
874
|
} else {
|
875
|
self.append(useNestedStatements(pathexpr));
|
876
|
}
|
877
|
}
|
878
|
|
879
|
return self.toString();
|
880
|
}
|
881
|
|
882
|
/*
|
883
|
* Constraint the query with @pathexp without using the XML Index,
|
884
|
* but nested SQL statements instead. The query migth be slower.
|
885
|
*/
|
886
|
private String useNestedStatements(String pathexpr)
|
887
|
{
|
888
|
StringBuffer nestedStmts = new StringBuffer();
|
889
|
Vector nodes = new Vector();
|
890
|
String path = pathexpr;
|
891
|
int inx = 0;
|
892
|
|
893
|
do {
|
894
|
inx = path.lastIndexOf("/");
|
895
|
//System.out.println(path.substring(inx+1));
|
896
|
nodes.addElement(path.substring(inx+1));
|
897
|
path = path.substring(0, Math.abs(inx));
|
898
|
} while ( inx > 0 );
|
899
|
|
900
|
// nested statements
|
901
|
int i = 0;
|
902
|
for (i = 0; i < nodes.size()-1; i++) {
|
903
|
nestedStmts.append("(SELECT nodeid FROM xml_nodes" +
|
904
|
" WHERE nodename LIKE '" +
|
905
|
(String)nodes.elementAt(i) + "'" +
|
906
|
" AND parentnodeid IN ");
|
907
|
}
|
908
|
// for the last statement: it is without " AND parentnodeid IN "
|
909
|
nestedStmts.append("(SELECT nodeid FROM xml_nodes" +
|
910
|
" WHERE nodename LIKE '" +
|
911
|
(String)nodes.elementAt(i) + "'" );
|
912
|
// node.size() number of closing brackets
|
913
|
for (i = 0; i < nodes.size(); i++) {
|
914
|
nestedStmts.append(")");
|
915
|
}
|
916
|
|
917
|
|
918
|
//System.out.println(nestedStmts.toString());
|
919
|
return nestedStmts.toString();
|
920
|
}
|
921
|
|
922
|
/**
|
923
|
* create a String description of the query that this instance represents.
|
924
|
* This should become a way to get the XML serialization of the query.
|
925
|
*/
|
926
|
public String toString() {
|
927
|
|
928
|
return this.printSQL(true);
|
929
|
}
|
930
|
}
|
931
|
}
|