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: tao $'
|
13
|
* '$Date: 2002-11-21 11:40:37 -0800 (Thu, 21 Nov 2002) $'
|
14
|
* '$Revision: 1332 $'
|
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
|
private String userName = null;
|
84
|
private static final String PUBLIC = "public";
|
85
|
private String [] group = null;
|
86
|
|
87
|
|
88
|
/**
|
89
|
* construct an instance of the QuerySpecification class
|
90
|
*
|
91
|
* @param queryspec the XML representation of the query (should conform
|
92
|
* to pathquery.dtd) as a Reader
|
93
|
* @param parserName the fully qualified name of a Java Class implementing
|
94
|
* the org.xml.sax.XMLReader interface
|
95
|
*/
|
96
|
public QuerySpecification( Reader queryspec, String parserName,
|
97
|
String accNumberSeparator ) throws IOException {
|
98
|
super();
|
99
|
|
100
|
// Initialize the class variables
|
101
|
returnDocList = new Vector();
|
102
|
filterDocList = new Vector();
|
103
|
elementStack = new Stack();
|
104
|
queryStack = new Stack();
|
105
|
returnFieldList = new Vector();
|
106
|
ownerList = new Vector();
|
107
|
siteList = new Vector();
|
108
|
this.parserName = parserName;
|
109
|
this.accNumberSeparator = accNumberSeparator;
|
110
|
|
111
|
// Initialize the parser and read the queryspec
|
112
|
XMLReader parser = initializeParser();
|
113
|
if (parser == null) {
|
114
|
System.err.println("SAX parser not instantiated properly.");
|
115
|
}
|
116
|
try {
|
117
|
parser.parse(new InputSource(queryspec));
|
118
|
} catch (SAXException e) {
|
119
|
System.err.println("error parsing data in " +
|
120
|
"QuerySpecification.QuerySpecification");
|
121
|
System.err.println(e.getMessage());
|
122
|
}
|
123
|
}
|
124
|
|
125
|
/**
|
126
|
* construct an instance of the QuerySpecification class
|
127
|
*
|
128
|
* @param queryspec the XML representation of the query (should conform
|
129
|
* to pathquery.dtd) as a String
|
130
|
* @param parserName the fully qualified name of a Java Class implementing
|
131
|
* the org.xml.sax.Parser interface
|
132
|
*/
|
133
|
public QuerySpecification( String queryspec, String parserName,
|
134
|
String accNumberSeparator) throws IOException {
|
135
|
this(new StringReader(queryspec), parserName, accNumberSeparator);
|
136
|
}
|
137
|
|
138
|
/**
|
139
|
* Method to set user name
|
140
|
*
|
141
|
* @param myName the user name
|
142
|
*/
|
143
|
public void setUserName(String myName)
|
144
|
{
|
145
|
this.userName = myName;
|
146
|
}
|
147
|
|
148
|
/**
|
149
|
* Method to set user group
|
150
|
*
|
151
|
* @param myGroup the user group
|
152
|
*/
|
153
|
public void setGroup(String [] myGroup)
|
154
|
{
|
155
|
this.group = myGroup;
|
156
|
}
|
157
|
|
158
|
/*
|
159
|
* Method to get owner query. If it is owner it has all permission
|
160
|
*/
|
161
|
private String createOwerQuery()
|
162
|
{
|
163
|
String ownerQuery = null;
|
164
|
ownerQuery = "SELECT docid FROM xml_documents WHERE user_owner ='" +
|
165
|
PUBLIC + "'";
|
166
|
if (userName != null && !userName.equals(""))
|
167
|
{
|
168
|
ownerQuery = ownerQuery + " OR user_owner ='"+ userName +"'";
|
169
|
}
|
170
|
|
171
|
if (group != null)
|
172
|
{
|
173
|
for (int i = 0; i< group.length; i++)
|
174
|
{
|
175
|
String groupUint = group[i];
|
176
|
if (groupUint != null && !groupUint.equals(""))
|
177
|
{
|
178
|
ownerQuery = ownerQuery +" OR user_owner = '" + groupUint + "'";
|
179
|
}//if
|
180
|
}//for
|
181
|
}
|
182
|
MetaCatUtil.debugMessage("OwnerQuery: "+ownerQuery, 30);
|
183
|
return ownerQuery;
|
184
|
}
|
185
|
|
186
|
/*
|
187
|
* Method to create query for xml_access, this part is to get docid list which
|
188
|
* have a allow rule for a given user
|
189
|
*/
|
190
|
private String createAllowRuleQuery()
|
191
|
{
|
192
|
String allowQuery = null;
|
193
|
allowQuery ="SELECT docid from xml_access WHERE ";
|
194
|
// add allow rule for user name
|
195
|
if (userName != null && !userName.equals(""))
|
196
|
{
|
197
|
allowQuery = allowQuery +"(principal_name = '" + userName
|
198
|
+"' AND perm_type = 'allow'"
|
199
|
+" AND (permission='4' OR permission='7'))";
|
200
|
}
|
201
|
// add allow rule for public
|
202
|
allowQuery = allowQuery +"OR (principal_name = '" + PUBLIC
|
203
|
+"' AND perm_type = 'allow'"
|
204
|
+" AND (permission='4' OR permission='7'))";
|
205
|
|
206
|
// add allow rule for group
|
207
|
if (group != null)
|
208
|
{
|
209
|
for (int i = 0; i< group.length; i++)
|
210
|
{
|
211
|
String groupUint = group[i];
|
212
|
if (groupUint != null && !groupUint.equals(""))
|
213
|
{
|
214
|
allowQuery = allowQuery +" OR (principal_name = '" + groupUint
|
215
|
+"' AND perm_type = 'allow'"
|
216
|
+" AND (permission='4' OR permission='7'))";
|
217
|
}//if
|
218
|
}//for
|
219
|
}//if
|
220
|
MetaCatUtil.debugMessage("allow query is: "+ allowQuery, 30);
|
221
|
return allowQuery;
|
222
|
|
223
|
}
|
224
|
|
225
|
/*
|
226
|
* Method to create query for xml_access, this part is to get docid list which
|
227
|
* have a deny rule and perm_order is allowFirst for a given user. This means
|
228
|
* the user will be denied to read
|
229
|
*/
|
230
|
private String createDenyRuleQuery()
|
231
|
{
|
232
|
String denyQuery = null;
|
233
|
denyQuery ="SELECT docid from xml_access WHERE ";
|
234
|
// add deny rule for user name
|
235
|
if (userName != null && !userName.equals(""))
|
236
|
{
|
237
|
denyQuery = denyQuery +"(principal_name = '" + userName
|
238
|
+"' AND perm_type = 'deny' "
|
239
|
+"AND perm_order ='allowFirst'"
|
240
|
+" AND (permission='4' OR permission='7'))";
|
241
|
}
|
242
|
// add deny rule for public
|
243
|
denyQuery = denyQuery +"OR (principal_name = '" + PUBLIC
|
244
|
+"' AND perm_type = 'deny' "
|
245
|
+"AND perm_order ='allowFirst'"
|
246
|
+" AND (permission='4' OR permission='7'))";
|
247
|
|
248
|
// add allow rule for group
|
249
|
if (group != null)
|
250
|
{
|
251
|
for (int i = 0; i< group.length; i++)
|
252
|
{
|
253
|
String groupUint = group[i];
|
254
|
if (groupUint != null && !groupUint.equals(""))
|
255
|
{
|
256
|
denyQuery = denyQuery +" OR (principal_name = '" + groupUint
|
257
|
+"' AND perm_type = 'deny' "
|
258
|
+"AND perm_order ='allowFirst'"
|
259
|
+" AND (permission='4' OR permission='7'))";
|
260
|
}//if
|
261
|
}//for
|
262
|
}//if
|
263
|
MetaCatUtil.debugMessage("denyquery is: "+ denyQuery, 30);
|
264
|
return denyQuery;
|
265
|
|
266
|
}
|
267
|
|
268
|
/**
|
269
|
* Method to append a access control query to SQL. So in DBQuery class, we can
|
270
|
* get docid from both user specified query and access control query. We don't
|
271
|
* need to checking permission after we get the doclist. It will be good to
|
272
|
* performance
|
273
|
*
|
274
|
*/
|
275
|
public String getAccessQuery()
|
276
|
{
|
277
|
String accessQuery = null;
|
278
|
String onwer = createOwerQuery();
|
279
|
String allow = createAllowRuleQuery();
|
280
|
String deny = createDenyRuleQuery();
|
281
|
accessQuery = " AND (docid IN("+ onwer + ")";
|
282
|
accessQuery = accessQuery + " OR (docid IN (" + allow + ")"
|
283
|
+ " AND docid NOT IN ("+ deny + ")))";
|
284
|
MetaCatUtil.debugMessage("accessquery is: "+ accessQuery, 30);
|
285
|
return accessQuery;
|
286
|
}
|
287
|
|
288
|
/** Main routine for testing */
|
289
|
static public void main(String[] args) {
|
290
|
|
291
|
if (args.length < 1) {
|
292
|
System.err.println("Wrong number of arguments!!!");
|
293
|
System.err.println("USAGE: java QuerySpecification <xmlfile>");
|
294
|
return;
|
295
|
} else {
|
296
|
int i = 0;
|
297
|
boolean useXMLIndex = true;
|
298
|
if ( args[i].equals( "-noindex" ) ) {
|
299
|
useXMLIndex = false;
|
300
|
i++;
|
301
|
}
|
302
|
String xmlfile = args[i];
|
303
|
|
304
|
try {
|
305
|
MetaCatUtil util = new MetaCatUtil();
|
306
|
FileReader xml = new FileReader(new File(xmlfile));
|
307
|
QuerySpecification qspec =
|
308
|
new QuerySpecification(xml, util.getOption("saxparser"),
|
309
|
util.getOption("accNumberSeparator"));
|
310
|
System.out.println(qspec.printSQL(useXMLIndex));
|
311
|
|
312
|
} catch (IOException e) {
|
313
|
System.err.println(e.getMessage());
|
314
|
}
|
315
|
|
316
|
}
|
317
|
}
|
318
|
|
319
|
/**
|
320
|
* Returns true if the parsed query contains and extended xml query
|
321
|
* (i.e. there is at least one <returnfield> in the pathquery document)
|
322
|
*/
|
323
|
public boolean containsExtendedSQL()
|
324
|
{
|
325
|
if(containsExtendedSQL)
|
326
|
{
|
327
|
return true;
|
328
|
}
|
329
|
else
|
330
|
{
|
331
|
return false;
|
332
|
}
|
333
|
}
|
334
|
|
335
|
/**
|
336
|
* Accessor method to return the identifier of this Query
|
337
|
*/
|
338
|
public String getIdentifier()
|
339
|
{
|
340
|
return meta_file_id;
|
341
|
}
|
342
|
|
343
|
/**
|
344
|
* method to set the identifier of this query
|
345
|
*/
|
346
|
public void setIdentifier(String id) {
|
347
|
this.meta_file_id = id;
|
348
|
}
|
349
|
|
350
|
/**
|
351
|
* Accessor method to return the title of this Query
|
352
|
*/
|
353
|
public String getQueryTitle()
|
354
|
{
|
355
|
return queryTitle;
|
356
|
}
|
357
|
|
358
|
/**
|
359
|
* method to set the title of this query
|
360
|
*/
|
361
|
public void setQueryTitle(String title)
|
362
|
{
|
363
|
this.queryTitle = title;
|
364
|
}
|
365
|
|
366
|
/**
|
367
|
* Accessor method to return a vector of the return document types as
|
368
|
* defined in the <returndoctype> tag in the pathquery dtd.
|
369
|
*/
|
370
|
public Vector getReturnDocList()
|
371
|
{
|
372
|
return this.returnDocList;
|
373
|
}
|
374
|
|
375
|
/**
|
376
|
* method to set the list of return docs of this query
|
377
|
*/
|
378
|
public void setReturnDocList(Vector returnDocList)
|
379
|
{
|
380
|
this.returnDocList = returnDocList;
|
381
|
}
|
382
|
|
383
|
/**
|
384
|
* Accessor method to return a vector of the filter doc types as
|
385
|
* defined in the <filterdoctype> tag in the pathquery dtd.
|
386
|
*/
|
387
|
public Vector getFilterDocList()
|
388
|
{
|
389
|
return this.filterDocList;
|
390
|
}
|
391
|
|
392
|
/**
|
393
|
* method to set the list of filter docs of this query
|
394
|
*/
|
395
|
public void setFilterDocList(Vector filterDocList)
|
396
|
{
|
397
|
this.filterDocList = filterDocList;
|
398
|
}
|
399
|
|
400
|
/**
|
401
|
* Accessor method to return a vector of the extended return fields as
|
402
|
* defined in the <returnfield> tag in the pathquery dtd.
|
403
|
*/
|
404
|
public Vector getReturnFieldList()
|
405
|
{
|
406
|
return this.returnFieldList;
|
407
|
}
|
408
|
|
409
|
/**
|
410
|
* method to set the list of fields to be returned by this query
|
411
|
*/
|
412
|
public void setReturnFieldList(Vector returnFieldList)
|
413
|
{
|
414
|
this.returnFieldList = returnFieldList;
|
415
|
}
|
416
|
|
417
|
/**
|
418
|
* Accessor method to return a vector of the owner fields as
|
419
|
* defined in the <owner> tag in the pathquery dtd.
|
420
|
*/
|
421
|
public Vector getOwnerList()
|
422
|
{
|
423
|
return this.ownerList;
|
424
|
}
|
425
|
|
426
|
/**
|
427
|
* method to set the list of owners used to constrain this query
|
428
|
*/
|
429
|
public void setOwnerList(Vector ownerList)
|
430
|
{
|
431
|
this.ownerList = ownerList;
|
432
|
}
|
433
|
|
434
|
/**
|
435
|
* Accessor method to return a vector of the site fields as
|
436
|
* defined in the <site> tag in the pathquery dtd.
|
437
|
*/
|
438
|
public Vector getSiteList()
|
439
|
{
|
440
|
return this.siteList;
|
441
|
}
|
442
|
|
443
|
/**
|
444
|
* method to set the list of sites used to constrain this query
|
445
|
*/
|
446
|
public void setSiteList(Vector siteList)
|
447
|
{
|
448
|
this.siteList = siteList;
|
449
|
}
|
450
|
|
451
|
/**
|
452
|
* get the QueryGroup used to express query constraints
|
453
|
*/
|
454
|
public QueryGroup getQueryGroup()
|
455
|
{
|
456
|
return query;
|
457
|
}
|
458
|
|
459
|
/**
|
460
|
* Set up the SAX parser for reading the XML serialized query
|
461
|
*/
|
462
|
private XMLReader initializeParser() {
|
463
|
XMLReader parser = null;
|
464
|
|
465
|
// Set up the SAX document handlers for parsing
|
466
|
try {
|
467
|
|
468
|
// Get an instance of the parser
|
469
|
parser = XMLReaderFactory.createXMLReader(parserName);
|
470
|
|
471
|
// Set the ContentHandler to this instance
|
472
|
parser.setContentHandler(this);
|
473
|
|
474
|
// Set the error Handler to this instance
|
475
|
parser.setErrorHandler(this);
|
476
|
|
477
|
} catch (Exception e) {
|
478
|
System.err.println("Error in QuerySpcecification.initializeParser " +
|
479
|
e.toString());
|
480
|
}
|
481
|
|
482
|
return parser;
|
483
|
}
|
484
|
|
485
|
/**
|
486
|
* callback method used by the SAX Parser when the start tag of an
|
487
|
* element is detected. Used in this context to parse and store
|
488
|
* the query information in class variables.
|
489
|
*/
|
490
|
public void startElement (String uri, String localName,
|
491
|
String qName, Attributes atts)
|
492
|
throws SAXException {
|
493
|
BasicNode currentNode = new BasicNode(localName);
|
494
|
// add attributes to BasicNode here
|
495
|
if (atts != null) {
|
496
|
int len = atts.getLength();
|
497
|
for (int i = 0; i < len; i++) {
|
498
|
currentNode.setAttribute(atts.getLocalName(i), atts.getValue(i));
|
499
|
}
|
500
|
}
|
501
|
|
502
|
elementStack.push(currentNode);
|
503
|
if (currentNode.getTagName().equals("querygroup")) {
|
504
|
QueryGroup currentGroup = new QueryGroup(
|
505
|
currentNode.getAttribute("operator"));
|
506
|
if (query == null) {
|
507
|
query = currentGroup;
|
508
|
} else {
|
509
|
QueryGroup parentGroup = (QueryGroup)queryStack.peek();
|
510
|
parentGroup.addChild(currentGroup);
|
511
|
}
|
512
|
queryStack.push(currentGroup);
|
513
|
}
|
514
|
}
|
515
|
|
516
|
/**
|
517
|
* callback method used by the SAX Parser when the end tag of an
|
518
|
* element is detected. Used in this context to parse and store
|
519
|
* the query information in class variables.
|
520
|
*/
|
521
|
public void endElement (String uri, String localName,
|
522
|
String qName) throws SAXException {
|
523
|
BasicNode leaving = (BasicNode)elementStack.pop();
|
524
|
if (leaving.getTagName().equals("queryterm")) {
|
525
|
boolean isCaseSensitive = (new Boolean(
|
526
|
leaving.getAttribute("casesensitive"))).booleanValue();
|
527
|
QueryTerm currentTerm = null;
|
528
|
if (currentPathexpr == null) {
|
529
|
currentTerm = new QueryTerm(isCaseSensitive,
|
530
|
leaving.getAttribute("searchmode"),currentValue);
|
531
|
} else {
|
532
|
currentTerm = new QueryTerm(isCaseSensitive,
|
533
|
leaving.getAttribute("searchmode"),currentValue,
|
534
|
currentPathexpr);
|
535
|
}
|
536
|
QueryGroup currentGroup = (QueryGroup)queryStack.peek();
|
537
|
currentGroup.addChild(currentTerm);
|
538
|
currentValue = null;
|
539
|
currentPathexpr = null;
|
540
|
} else if (leaving.getTagName().equals("querygroup")) {
|
541
|
QueryGroup leavingGroup = (QueryGroup)queryStack.pop();
|
542
|
}
|
543
|
}
|
544
|
|
545
|
/**
|
546
|
* callback method used by the SAX Parser when the text sequences of an
|
547
|
* xml stream are detected. Used in this context to parse and store
|
548
|
* the query information in class variables.
|
549
|
*/
|
550
|
public void characters(char ch[], int start, int length) {
|
551
|
|
552
|
String inputString = new String(ch, start, length);
|
553
|
BasicNode currentNode = (BasicNode)elementStack.peek();
|
554
|
String currentTag = currentNode.getTagName();
|
555
|
if (currentTag.equals("meta_file_id")) {
|
556
|
meta_file_id = inputString;
|
557
|
} else if (currentTag.equals("querytitle")) {
|
558
|
queryTitle = inputString;
|
559
|
} else if (currentTag.equals("value")) {
|
560
|
currentValue = inputString;
|
561
|
} else if (currentTag.equals("pathexpr")) {
|
562
|
currentPathexpr = inputString;
|
563
|
} else if (currentTag.equals("returndoctype")) {
|
564
|
returnDocList.add(inputString);
|
565
|
} else if (currentTag.equals("filterdoctype")) {
|
566
|
filterDocList.add(inputString);
|
567
|
} else if (currentTag.equals("returnfield")) {
|
568
|
returnFieldList.add(inputString);
|
569
|
containsExtendedSQL = true;
|
570
|
} else if (currentTag.equals("filterdoctype")) {
|
571
|
filterDocList.add(inputString);
|
572
|
} else if (currentTag.equals("owner")) {
|
573
|
ownerList.add(inputString);
|
574
|
} else if (currentTag.equals("site")) {
|
575
|
siteList.add(inputString);
|
576
|
}
|
577
|
}
|
578
|
|
579
|
/**
|
580
|
* create a SQL serialization of the query that this instance represents
|
581
|
*/
|
582
|
public String printSQL(boolean useXMLIndex) {
|
583
|
StringBuffer self = new StringBuffer();
|
584
|
|
585
|
self.append("SELECT docid,docname,doctype,");
|
586
|
self.append("date_created, date_updated, rev ");
|
587
|
self.append("FROM xml_documents WHERE docid IN (");
|
588
|
|
589
|
// This determines the documents that meet the query conditions
|
590
|
self.append(query.printSQL(useXMLIndex));
|
591
|
|
592
|
self.append(") ");
|
593
|
|
594
|
// Add SQL to filter for doctypes requested in the query
|
595
|
// This is an implicit OR for the list of doctypes. Only doctypes in this
|
596
|
// list will be searched if the tag is present
|
597
|
if (!filterDocList.isEmpty()) {
|
598
|
boolean firstdoctype = true;
|
599
|
self.append(" AND (");
|
600
|
Enumeration en = filterDocList.elements();
|
601
|
while (en.hasMoreElements()) {
|
602
|
String currentDoctype = (String)en.nextElement();
|
603
|
if (firstdoctype) {
|
604
|
firstdoctype = false;
|
605
|
self.append(" doctype = '" + currentDoctype + "'");
|
606
|
} else {
|
607
|
self.append(" OR doctype = '" + currentDoctype + "'");
|
608
|
}
|
609
|
}
|
610
|
self.append(") ");
|
611
|
}
|
612
|
|
613
|
// Add SQL to filter for owners requested in the query
|
614
|
// This is an implicit OR for the list of owners
|
615
|
if (!ownerList.isEmpty()) {
|
616
|
boolean first = true;
|
617
|
self.append(" AND (");
|
618
|
Enumeration en = ownerList.elements();
|
619
|
while (en.hasMoreElements()) {
|
620
|
String current = (String)en.nextElement();
|
621
|
if (first) {
|
622
|
first = false;
|
623
|
self.append(" user_owner = '" + current + "'");
|
624
|
} else {
|
625
|
self.append(" OR user_owner = '" + current + "'");
|
626
|
}
|
627
|
}
|
628
|
self.append(") ");
|
629
|
}
|
630
|
|
631
|
// Add SQL to filter for sites requested in the query
|
632
|
// This is an implicit OR for the list of sites
|
633
|
if (!siteList.isEmpty()) {
|
634
|
boolean first = true;
|
635
|
self.append(" AND (");
|
636
|
Enumeration en = siteList.elements();
|
637
|
while (en.hasMoreElements()) {
|
638
|
String current = (String)en.nextElement();
|
639
|
if (first) {
|
640
|
first = false;
|
641
|
self.append(" SUBSTR(docid, 1, INSTR(docid, '" +
|
642
|
accNumberSeparator + "')-1) = '" + current + "'");
|
643
|
} else {
|
644
|
self.append(" OR SUBSTR(docid, 1, INSTR(docid, '" +
|
645
|
accNumberSeparator + "')-1) = '" + current + "'");
|
646
|
}
|
647
|
}
|
648
|
self.append(") ");
|
649
|
}
|
650
|
//System.out.println(self.toString());
|
651
|
return self.toString();
|
652
|
}
|
653
|
|
654
|
/**
|
655
|
* This method prints sql based upon the <returnfield> tag in the
|
656
|
* pathquery document. This allows for customization of the
|
657
|
* returned fields
|
658
|
* @param doclist the list of document ids to search by
|
659
|
*/
|
660
|
public String printExtendedSQL(String doclist)
|
661
|
{
|
662
|
StringBuffer self = new StringBuffer();
|
663
|
self.append("select xml_nodes.docid, xml_index.path, xml_nodes.nodedata ");
|
664
|
self.append("from xml_index, xml_nodes where xml_index.nodeid=");
|
665
|
self.append("xml_nodes.parentnodeid and (xml_index.path like '");
|
666
|
boolean firstfield = true;
|
667
|
//put the returnfields into the query
|
668
|
//the for loop allows for multiple fields
|
669
|
for(int i=0; i<returnFieldList.size(); i++)
|
670
|
{
|
671
|
if(firstfield)
|
672
|
{
|
673
|
firstfield = false;
|
674
|
self.append((String)returnFieldList.elementAt(i));
|
675
|
self.append("' ");
|
676
|
}
|
677
|
else
|
678
|
{
|
679
|
self.append("or xml_index.path like '");
|
680
|
self.append((String)returnFieldList.elementAt(i));
|
681
|
self.append("' ");
|
682
|
}
|
683
|
}
|
684
|
self.append(") AND xml_nodes.docid in (");
|
685
|
//self.append(query.printSQL());
|
686
|
self.append(doclist);
|
687
|
self.append(")");
|
688
|
self.append(" AND xml_nodes.nodetype = 'TEXT'");
|
689
|
|
690
|
//System.out.println(self.toString());
|
691
|
return self.toString();
|
692
|
}
|
693
|
|
694
|
public static String printRelationSQL(String docid)
|
695
|
{
|
696
|
StringBuffer self = new StringBuffer();
|
697
|
self.append("select subject, relationship, object, subdoctype, ");
|
698
|
self.append("objdoctype from xml_relation ");
|
699
|
self.append("where docid like '").append(docid).append("'");
|
700
|
return self.toString();
|
701
|
}
|
702
|
|
703
|
/**
|
704
|
* Prints sql that returns all relations in the database.
|
705
|
*/
|
706
|
public static String printPackageSQL()
|
707
|
{
|
708
|
StringBuffer self = new StringBuffer();
|
709
|
self.append("select z.nodedata, x.nodedata, y.nodedata from ");
|
710
|
self.append("(select nodeid, parentnodeid from xml_index where path like ");
|
711
|
self.append("'triple/subject') s, (select nodeid, parentnodeid ");
|
712
|
self.append("from xml_index where path like ");
|
713
|
self.append("'triple/relationship') rel, ");
|
714
|
self.append("(select nodeid, parentnodeid from xml_index where path like ");
|
715
|
self.append("'triple/object') o, ");
|
716
|
self.append("xml_nodes x, xml_nodes y, xml_nodes z ");
|
717
|
self.append("where s.parentnodeid = rel.parentnodeid ");
|
718
|
self.append("and rel.parentnodeid = o.parentnodeid ");
|
719
|
self.append("and x.parentnodeid in (rel.nodeid) ");
|
720
|
self.append("and y.parentnodeid in (o.nodeid) ");
|
721
|
self.append("and z.parentnodeid in (s.nodeid) ");
|
722
|
//self.append("and z.nodedata like '%");
|
723
|
//self.append(docid);
|
724
|
//self.append("%'");
|
725
|
return self.toString();
|
726
|
}
|
727
|
|
728
|
/**
|
729
|
* Prints sql that returns all relations in the database that were input
|
730
|
* under a specific docid
|
731
|
* @param docid the docid to search for.
|
732
|
*/
|
733
|
public static String printPackageSQL(String docid)
|
734
|
{
|
735
|
StringBuffer self = new StringBuffer();
|
736
|
self.append("select z.nodedata, x.nodedata, y.nodedata from ");
|
737
|
self.append("(select nodeid, parentnodeid from xml_index where path like ");
|
738
|
self.append("'triple/subject') s, (select nodeid, parentnodeid ");
|
739
|
self.append("from xml_index where path like ");
|
740
|
self.append("'triple/relationship') rel, ");
|
741
|
self.append("(select nodeid, parentnodeid from xml_index where path like ");
|
742
|
self.append("'triple/object') o, ");
|
743
|
self.append("xml_nodes x, xml_nodes y, xml_nodes z ");
|
744
|
self.append("where s.parentnodeid = rel.parentnodeid ");
|
745
|
self.append("and rel.parentnodeid = o.parentnodeid ");
|
746
|
self.append("and x.parentnodeid in (rel.nodeid) ");
|
747
|
self.append("and y.parentnodeid in (o.nodeid) ");
|
748
|
self.append("and z.parentnodeid in (s.nodeid) ");
|
749
|
self.append("and z.docid like '").append(docid).append("'");
|
750
|
|
751
|
return self.toString();
|
752
|
}
|
753
|
|
754
|
/**
|
755
|
* Returns all of the relations that has a certain docid in the subject
|
756
|
* or the object.
|
757
|
*
|
758
|
* @param docid the docid to search for
|
759
|
*/
|
760
|
public static String printPackageSQL(String subDocidURL, String objDocidURL)
|
761
|
{
|
762
|
StringBuffer self = new StringBuffer();
|
763
|
self.append("select z.nodedata, x.nodedata, y.nodedata from ");
|
764
|
self.append("(select nodeid, parentnodeid from xml_index where path like ");
|
765
|
self.append("'triple/subject') s, (select nodeid, parentnodeid ");
|
766
|
self.append("from xml_index where path like ");
|
767
|
self.append("'triple/relationship') rel, ");
|
768
|
self.append("(select nodeid, parentnodeid from xml_index where path like ");
|
769
|
self.append("'triple/object') o, ");
|
770
|
self.append("xml_nodes x, xml_nodes y, xml_nodes z ");
|
771
|
self.append("where s.parentnodeid = rel.parentnodeid ");
|
772
|
self.append("and rel.parentnodeid = o.parentnodeid ");
|
773
|
self.append("and x.parentnodeid in (rel.nodeid) ");
|
774
|
self.append("and y.parentnodeid in (o.nodeid) ");
|
775
|
self.append("and z.parentnodeid in (s.nodeid) ");
|
776
|
self.append("and (z.nodedata like '");
|
777
|
self.append(subDocidURL);
|
778
|
self.append("' or y.nodedata like '");
|
779
|
self.append(objDocidURL);
|
780
|
self.append("')");
|
781
|
return self.toString();
|
782
|
}
|
783
|
|
784
|
public static String printGetDocByDoctypeSQL(String docid)
|
785
|
{
|
786
|
StringBuffer self = new StringBuffer();
|
787
|
|
788
|
self.append("SELECT docid,docname,doctype,");
|
789
|
self.append("date_created, date_updated ");
|
790
|
self.append("FROM xml_documents WHERE docid IN (");
|
791
|
self.append(docid).append(")");
|
792
|
return self.toString();
|
793
|
}
|
794
|
|
795
|
/**
|
796
|
* create a String description of the query that this instance represents.
|
797
|
* This should become a way to get the XML serialization of the query.
|
798
|
*/
|
799
|
public String toString() {
|
800
|
return "meta_file_id=" + meta_file_id + "\n" + query;
|
801
|
//DOCTITLE attr cleared from the db
|
802
|
// return "meta_file_id=" + meta_file_id + "\n" +
|
803
|
// "querytitle=" + querytitle + "\n" + query;
|
804
|
}
|
805
|
|
806
|
/** a utility class that represents a group of terms in a query */
|
807
|
private class QueryGroup {
|
808
|
private String operator = null; // indicates how query terms are combined
|
809
|
private Vector children = null; // the list of query terms and groups
|
810
|
|
811
|
/**
|
812
|
* construct a new QueryGroup
|
813
|
*
|
814
|
* @param operator the boolean conector used to connect query terms
|
815
|
* in this query group
|
816
|
*/
|
817
|
public QueryGroup(String operator) {
|
818
|
this.operator = operator;
|
819
|
children = new Vector();
|
820
|
}
|
821
|
|
822
|
/**
|
823
|
* Add a child QueryGroup to this QueryGroup
|
824
|
*
|
825
|
* @param qgroup the query group to be added to the list of terms
|
826
|
*/
|
827
|
public void addChild(QueryGroup qgroup) {
|
828
|
children.add((Object)qgroup);
|
829
|
}
|
830
|
|
831
|
/**
|
832
|
* Add a child QueryTerm to this QueryGroup
|
833
|
*
|
834
|
* @param qterm the query term to be added to the list of terms
|
835
|
*/
|
836
|
public void addChild(QueryTerm qterm) {
|
837
|
children.add((Object)qterm);
|
838
|
}
|
839
|
|
840
|
/**
|
841
|
* Retrieve an Enumeration of query terms for this QueryGroup
|
842
|
*/
|
843
|
public Enumeration getChildren() {
|
844
|
return children.elements();
|
845
|
}
|
846
|
|
847
|
/**
|
848
|
* create a SQL serialization of the query that this instance represents
|
849
|
*/
|
850
|
public String printSQL(boolean useXMLIndex) {
|
851
|
StringBuffer self = new StringBuffer();
|
852
|
boolean first = true;
|
853
|
|
854
|
self.append("(");
|
855
|
|
856
|
Enumeration en= getChildren();
|
857
|
while (en.hasMoreElements()) {
|
858
|
Object qobject = en.nextElement();
|
859
|
if (first) {
|
860
|
first = false;
|
861
|
} else {
|
862
|
self.append(" " + operator + " ");
|
863
|
}
|
864
|
if (qobject instanceof QueryGroup) {
|
865
|
QueryGroup qg = (QueryGroup)qobject;
|
866
|
self.append(qg.printSQL(useXMLIndex));
|
867
|
} else if (qobject instanceof QueryTerm) {
|
868
|
QueryTerm qt = (QueryTerm)qobject;
|
869
|
self.append(qt.printSQL(useXMLIndex));
|
870
|
} else {
|
871
|
System.err.println("qobject wrong type: fatal error");
|
872
|
}
|
873
|
}
|
874
|
self.append(") \n");
|
875
|
return self.toString();
|
876
|
}
|
877
|
|
878
|
/**
|
879
|
* create a String description of the query that this instance represents.
|
880
|
* This should become a way to get the XML serialization of the query.
|
881
|
*/
|
882
|
public String toString() {
|
883
|
StringBuffer self = new StringBuffer();
|
884
|
|
885
|
self.append(" (Query group operator=" + operator + "\n");
|
886
|
Enumeration en= getChildren();
|
887
|
while (en.hasMoreElements()) {
|
888
|
Object qobject = en.nextElement();
|
889
|
self.append(qobject);
|
890
|
}
|
891
|
self.append(" )\n");
|
892
|
return self.toString();
|
893
|
}
|
894
|
}
|
895
|
|
896
|
/** a utility class that represents a single term in a query */
|
897
|
private class QueryTerm {
|
898
|
private boolean casesensitive = false;
|
899
|
private String searchmode = null;
|
900
|
private String value = null;
|
901
|
private String pathexpr = null;
|
902
|
|
903
|
/**
|
904
|
* Construct a new instance of a query term for a free text search
|
905
|
* (using the value only)
|
906
|
*
|
907
|
* @param casesensitive flag indicating whether case is used to match
|
908
|
* @param searchmode determines what kind of substring match is performed
|
909
|
* (one of starts-with|ends-with|contains|matches-exactly)
|
910
|
* @param value the text value to match
|
911
|
*/
|
912
|
public QueryTerm(boolean casesensitive, String searchmode,
|
913
|
String value) {
|
914
|
this.casesensitive = casesensitive;
|
915
|
this.searchmode = searchmode;
|
916
|
this.value = value;
|
917
|
}
|
918
|
|
919
|
/**
|
920
|
* Construct a new instance of a query term for a structured search
|
921
|
* (matching the value only for those nodes in the pathexpr)
|
922
|
*
|
923
|
* @param casesensitive flag indicating whether case is used to match
|
924
|
* @param searchmode determines what kind of substring match is performed
|
925
|
* (one of starts-with|ends-with|contains|matches-exactly)
|
926
|
* @param value the text value to match
|
927
|
* @param pathexpr the hierarchical path to the nodes to be searched
|
928
|
*/
|
929
|
public QueryTerm(boolean casesensitive, String searchmode,
|
930
|
String value, String pathexpr) {
|
931
|
this(casesensitive, searchmode, value);
|
932
|
this.pathexpr = pathexpr;
|
933
|
}
|
934
|
|
935
|
/** determine if the QueryTerm is case sensitive */
|
936
|
public boolean isCaseSensitive() {
|
937
|
return casesensitive;
|
938
|
}
|
939
|
|
940
|
/** get the searchmode parameter */
|
941
|
public String getSearchMode() {
|
942
|
return searchmode;
|
943
|
}
|
944
|
|
945
|
/** get the Value parameter */
|
946
|
public String getValue() {
|
947
|
return value;
|
948
|
}
|
949
|
|
950
|
/** get the path expression parameter */
|
951
|
public String getPathExpression() {
|
952
|
return pathexpr;
|
953
|
}
|
954
|
|
955
|
/**
|
956
|
* create a SQL serialization of the query that this instance represents
|
957
|
*/
|
958
|
public String printSQL(boolean useXMLIndex) {
|
959
|
StringBuffer self = new StringBuffer();
|
960
|
|
961
|
// Uppercase the search string if case match is not important
|
962
|
String casevalue = null;
|
963
|
String nodedataterm = null;
|
964
|
|
965
|
if (casesensitive) {
|
966
|
nodedataterm = "nodedata";
|
967
|
casevalue = value;
|
968
|
} else {
|
969
|
nodedataterm = "UPPER(nodedata)";
|
970
|
casevalue = value.toUpperCase();
|
971
|
}
|
972
|
|
973
|
// Add appropriate wildcards to search string
|
974
|
//String searchvalue = null;
|
975
|
String searchexpr = null;
|
976
|
if (searchmode.equals("starts-with")) {
|
977
|
//searchvalue = casevalue + "%";
|
978
|
searchexpr = nodedataterm + " LIKE '" + casevalue + "%' ";
|
979
|
} else if (searchmode.equals("ends-with")) {
|
980
|
//searchvalue = "%" + casevalue;
|
981
|
searchexpr = nodedataterm + " LIKE '%" + casevalue + "' ";
|
982
|
} else if (searchmode.equals("contains")) {
|
983
|
//searchvalue = "%" + casevalue + "%";
|
984
|
if (!casevalue.equals("%"))
|
985
|
{
|
986
|
searchexpr = nodedataterm + " LIKE '%" + casevalue + "%' ";
|
987
|
}
|
988
|
else
|
989
|
{
|
990
|
searchexpr = nodedataterm + " LIKE '" + casevalue + "' ";
|
991
|
}
|
992
|
} else if (searchmode.equals("equals")) {
|
993
|
//searchvalue = casevalue;
|
994
|
searchexpr = nodedataterm + " = '" + casevalue + "' ";
|
995
|
} else if (searchmode.equals("isnot-equal")) {
|
996
|
//searchvalue = casevalue;
|
997
|
searchexpr = nodedataterm + " != '" + casevalue + "' ";
|
998
|
} else {
|
999
|
//searchvalue = casevalue;
|
1000
|
String oper = null;
|
1001
|
if (searchmode.equals("greater-than")) {
|
1002
|
oper = ">";
|
1003
|
} else if (searchmode.equals("greater-than-equals")) {
|
1004
|
oper = ">=";
|
1005
|
} else if (searchmode.equals("less-than")) {
|
1006
|
oper = "<";
|
1007
|
} else if (searchmode.equals("less-than-equals")) {
|
1008
|
oper = "<=";
|
1009
|
} else {
|
1010
|
System.out.println("NOT expected case. NOT recognized operator: " +
|
1011
|
searchmode);
|
1012
|
return null;
|
1013
|
}
|
1014
|
try {
|
1015
|
// it is number; numeric comparison
|
1016
|
// but we need to make sure there is no string in node data
|
1017
|
String getRidOfString = " AND UPPER(nodedata) = LOWER(nodedata)" +
|
1018
|
" AND LTRIM(nodedata) != ' ' " +
|
1019
|
" AND nodedata IS NOT NULL ";
|
1020
|
searchexpr = nodedataterm + " " + oper + " " +
|
1021
|
new Double(casevalue) + " "+getRidOfString;
|
1022
|
} catch (NumberFormatException nfe) {
|
1023
|
// these are characters; character comparison
|
1024
|
searchexpr = nodedataterm + " " + oper + " '" + casevalue + "' ";
|
1025
|
}
|
1026
|
}
|
1027
|
|
1028
|
self.append("SELECT DISTINCT docid FROM xml_nodes WHERE \n");
|
1029
|
//self.append(nodedataterm + " LIKE " + "'" + searchvalue + "' ");
|
1030
|
self.append(searchexpr);
|
1031
|
|
1032
|
if (pathexpr != null) {
|
1033
|
self.append("AND parentnodeid IN ");
|
1034
|
// use XML Index
|
1035
|
if ( useXMLIndex ) {
|
1036
|
self.append("(SELECT nodeid FROM xml_index WHERE path LIKE " +
|
1037
|
"'" + pathexpr + "') " );
|
1038
|
// without using XML Index; using nested statements instead
|
1039
|
} else {
|
1040
|
self.append(useNestedStatements(pathexpr));
|
1041
|
}
|
1042
|
}
|
1043
|
|
1044
|
return self.toString();
|
1045
|
}
|
1046
|
|
1047
|
/*
|
1048
|
* Constraint the query with @pathexp without using the XML Index,
|
1049
|
* but nested SQL statements instead. The query migth be slower.
|
1050
|
*/
|
1051
|
private String useNestedStatements(String pathexpr)
|
1052
|
{
|
1053
|
StringBuffer nestedStmts = new StringBuffer();
|
1054
|
Vector nodes = new Vector();
|
1055
|
String path = pathexpr;
|
1056
|
int inx = 0;
|
1057
|
|
1058
|
do {
|
1059
|
inx = path.lastIndexOf("/");
|
1060
|
//System.out.println(path.substring(inx+1));
|
1061
|
nodes.addElement(path.substring(inx+1));
|
1062
|
path = path.substring(0, Math.abs(inx));
|
1063
|
} while ( inx > 0 );
|
1064
|
|
1065
|
// nested statements
|
1066
|
int i = 0;
|
1067
|
for (i = 0; i < nodes.size()-1; i++) {
|
1068
|
nestedStmts.append("(SELECT nodeid FROM xml_nodes" +
|
1069
|
" WHERE nodename LIKE '" +
|
1070
|
(String)nodes.elementAt(i) + "'" +
|
1071
|
" AND parentnodeid IN ");
|
1072
|
}
|
1073
|
// for the last statement: it is without " AND parentnodeid IN "
|
1074
|
nestedStmts.append("(SELECT nodeid FROM xml_nodes" +
|
1075
|
" WHERE nodename LIKE '" +
|
1076
|
(String)nodes.elementAt(i) + "'" );
|
1077
|
// node.size() number of closing brackets
|
1078
|
for (i = 0; i < nodes.size(); i++) {
|
1079
|
nestedStmts.append(")");
|
1080
|
}
|
1081
|
|
1082
|
|
1083
|
//System.out.println(nestedStmts.toString());
|
1084
|
return nestedStmts.toString();
|
1085
|
}
|
1086
|
|
1087
|
/**
|
1088
|
* create a String description of the query that this instance represents.
|
1089
|
* This should become a way to get the XML serialization of the query.
|
1090
|
*/
|
1091
|
public String toString() {
|
1092
|
|
1093
|
return this.printSQL(true);
|
1094
|
}
|
1095
|
}
|
1096
|
}
|