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
|
*
|
11
|
* '$Author: daigle $'
|
12
|
* '$Date: 2008-08-29 10:20:19 -0700 (Fri, 29 Aug 2008) $'
|
13
|
* '$Revision: 4335 $'
|
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 java.io.IOException;
|
33
|
import java.io.Reader;
|
34
|
import java.io.StringReader;
|
35
|
import java.util.Enumeration;
|
36
|
import java.util.Stack;
|
37
|
import java.util.Vector;
|
38
|
|
39
|
import edu.ucsb.nceas.metacat.util.MetaCatUtil;
|
40
|
|
41
|
import org.apache.log4j.Logger;
|
42
|
import org.xml.sax.Attributes;
|
43
|
import org.xml.sax.InputSource;
|
44
|
import org.xml.sax.SAXException;
|
45
|
import org.xml.sax.XMLReader;
|
46
|
import org.xml.sax.helpers.DefaultHandler;
|
47
|
import org.xml.sax.helpers.XMLReaderFactory;
|
48
|
import java.util.Iterator;
|
49
|
|
50
|
/**
|
51
|
* A Class that represents a structured query, and can be constructed from an
|
52
|
* XML serialization conforming to
|
53
|
*
|
54
|
* @see pathquery.dtd. The printSQL() method can be used to print a SQL
|
55
|
* serialization of the query.
|
56
|
*/
|
57
|
public class QuerySpecification extends DefaultHandler
|
58
|
{
|
59
|
|
60
|
/** flag determining whether extended query terms are present */
|
61
|
private boolean containsExtendedSQL = false;
|
62
|
|
63
|
/** flag determining whether predicates are present */
|
64
|
private boolean containsPredicates = false;
|
65
|
|
66
|
/** Identifier for this query document */
|
67
|
private String meta_file_id;
|
68
|
|
69
|
/** Title of this query */
|
70
|
private String queryTitle;
|
71
|
|
72
|
/** List of document types to be returned using package back tracing */
|
73
|
private Vector returnDocList;
|
74
|
|
75
|
/** List of document types to be searched */
|
76
|
private Vector filterDocList;
|
77
|
|
78
|
/** List of fields to be returned in result set */
|
79
|
private Vector returnFieldList;
|
80
|
|
81
|
/** List of fields with "[" and "]" in result set. This is a subset of returnFieldList.
|
82
|
* If some of return fields have [, those fields will be stored this vector (we have different query for those return fields */
|
83
|
private Vector returnFieldListWithPredicates;
|
84
|
|
85
|
/** List of users owning documents to be searched */
|
86
|
private Vector ownerList;
|
87
|
|
88
|
/** The root query group that contains the recursive query constraints */
|
89
|
private QueryGroup query = null;
|
90
|
|
91
|
/** A string buffer to stored normalized query (Sometimes, the query have
|
92
|
* a value like "&", it will cause problem in html transform). So we need a
|
93
|
* normalized query xml string.
|
94
|
*/
|
95
|
private StringBuffer xml = new StringBuffer();
|
96
|
|
97
|
// Query data structures used temporarily during XML parsing
|
98
|
private Stack elementStack;
|
99
|
|
100
|
private Stack queryStack;
|
101
|
|
102
|
private String currentValue;
|
103
|
|
104
|
private String currentPathexpr;
|
105
|
|
106
|
private String parserName = null;
|
107
|
|
108
|
private String accNumberSeparator = null;
|
109
|
|
110
|
private boolean percentageSearch = false;
|
111
|
|
112
|
private String userName = null;
|
113
|
|
114
|
private static final String PUBLIC = "public";
|
115
|
|
116
|
private String[] group = null;
|
117
|
|
118
|
public static final String ATTRIBUTESYMBOL = "@";
|
119
|
|
120
|
public static final char PREDICATE_START = '[';
|
121
|
|
122
|
public static final char PREDICATE_END = ']';
|
123
|
|
124
|
//private boolean hasAttributeReturnField = false;
|
125
|
|
126
|
//private Hashtable attributeReturnList = new Hashtable();
|
127
|
|
128
|
//private int countAttributeReturnField = 0;
|
129
|
|
130
|
private StringBuffer textBuffer = new StringBuffer();
|
131
|
|
132
|
|
133
|
private static Logger logMetacat = Logger.getLogger(QuerySpecification.class);
|
134
|
|
135
|
/**
|
136
|
* construct an instance of the QuerySpecification class
|
137
|
*
|
138
|
* @param queryspec
|
139
|
* the XML representation of the query (should conform to
|
140
|
* pathquery.dtd) as a Reader
|
141
|
* @param parserName
|
142
|
* the fully qualified name of a Java Class implementing the
|
143
|
* org.xml.sax.XMLReader interface
|
144
|
*/
|
145
|
public QuerySpecification(Reader queryspec, String parserName,
|
146
|
String accNumberSeparator) throws IOException
|
147
|
{
|
148
|
super();
|
149
|
|
150
|
// Initialize the class variables
|
151
|
returnDocList = new Vector();
|
152
|
filterDocList = new Vector();
|
153
|
elementStack = new Stack();
|
154
|
queryStack = new Stack();
|
155
|
returnFieldList = new Vector();
|
156
|
returnFieldListWithPredicates = new Vector();
|
157
|
ownerList = new Vector();
|
158
|
this.parserName = parserName;
|
159
|
this.accNumberSeparator = accNumberSeparator;
|
160
|
|
161
|
// Initialize the parser and read the queryspec
|
162
|
XMLReader parser = initializeParser();
|
163
|
if (parser == null) {
|
164
|
System.err.println("SAX parser not instantiated properly.");
|
165
|
}
|
166
|
try {
|
167
|
parser.parse(new InputSource(queryspec));
|
168
|
} catch (SAXException e) {
|
169
|
System.err.println("error parsing data in "
|
170
|
+ "QuerySpecification.QuerySpecification");
|
171
|
System.err.println(e.getMessage());
|
172
|
}
|
173
|
}
|
174
|
|
175
|
/**
|
176
|
* construct an instance of the QuerySpecification class
|
177
|
*
|
178
|
* @param queryspec
|
179
|
* the XML representation of the query (should conform to
|
180
|
* pathquery.dtd) as a String
|
181
|
* @param parserName
|
182
|
* the fully qualified name of a Java Class implementing the
|
183
|
* org.xml.sax.Parser interface
|
184
|
*/
|
185
|
public QuerySpecification(String queryspec, String parserName,
|
186
|
String accNumberSeparator) throws IOException
|
187
|
{
|
188
|
this(new StringReader(queryspec), parserName, accNumberSeparator);
|
189
|
}
|
190
|
|
191
|
/**
|
192
|
* construct an instance of the QuerySpecification class which don't need
|
193
|
* to parser a xml document
|
194
|
*
|
195
|
* @param accNumberSeparator
|
196
|
* the separator between doc version
|
197
|
*/
|
198
|
public QuerySpecification(String accNumberSeparator) throws IOException
|
199
|
{
|
200
|
// Initialize the class variables
|
201
|
returnDocList = new Vector();
|
202
|
filterDocList = new Vector();
|
203
|
elementStack = new Stack();
|
204
|
queryStack = new Stack();
|
205
|
returnFieldList = new Vector();
|
206
|
returnFieldListWithPredicates = new Vector();
|
207
|
ownerList = new Vector();
|
208
|
this.accNumberSeparator = accNumberSeparator;
|
209
|
}
|
210
|
|
211
|
/**
|
212
|
* Method to set user name
|
213
|
*
|
214
|
* @param myName
|
215
|
* the user name
|
216
|
*/
|
217
|
public void setUserName(String myName)
|
218
|
{
|
219
|
//to lower case
|
220
|
if (myName != null) {
|
221
|
this.userName = myName.toLowerCase();
|
222
|
} else {
|
223
|
this.userName = myName;
|
224
|
}
|
225
|
}
|
226
|
|
227
|
/**
|
228
|
* Method to set user group
|
229
|
*
|
230
|
* @param myGroup
|
231
|
* the user group
|
232
|
*/
|
233
|
public void setGroup(String[] myGroup)
|
234
|
{
|
235
|
this.group = myGroup;
|
236
|
}
|
237
|
|
238
|
/**
|
239
|
* Method to indicate this query is a percentage search
|
240
|
*/
|
241
|
public boolean isPercentageSearch()
|
242
|
{
|
243
|
return percentageSearch;
|
244
|
}
|
245
|
|
246
|
/*
|
247
|
* Method to get owner query. If it is owner it has all permission
|
248
|
*/
|
249
|
private String createOwerQuery()
|
250
|
{
|
251
|
String ownerQuery = null;
|
252
|
//if user is public, we don't need to run owner query
|
253
|
if (userName != null && !userName.equalsIgnoreCase(PUBLIC))
|
254
|
{
|
255
|
ownerQuery = "SELECT docid FROM xml_documents WHERE ";
|
256
|
if (userName != null && !userName.equals("")) {
|
257
|
ownerQuery = ownerQuery + "lower(user_owner) ='" + userName + "'";
|
258
|
}
|
259
|
}
|
260
|
logMetacat.info("OwnerQuery: " + ownerQuery);
|
261
|
return ownerQuery;
|
262
|
}
|
263
|
|
264
|
/*
|
265
|
* Method to create query for xml_access, this part is to get docid list
|
266
|
* which have a allow rule for a given user
|
267
|
*/
|
268
|
private String createAllowRuleQuery()
|
269
|
{
|
270
|
String allowQuery = null;
|
271
|
String allowString = constructAllowString();
|
272
|
allowQuery = "SELECT docid from xml_access WHERE( " + allowString;
|
273
|
allowQuery = allowQuery + ")";
|
274
|
logMetacat.info("allow query is: " + allowQuery);
|
275
|
return allowQuery;
|
276
|
|
277
|
}
|
278
|
|
279
|
/* Method to construct a allow rule string */
|
280
|
private String constructAllowString()
|
281
|
{
|
282
|
String allowQuery = "";
|
283
|
|
284
|
// add public
|
285
|
allowQuery = "(lower(principal_name) = '" + PUBLIC
|
286
|
+ "'";
|
287
|
|
288
|
// add user name
|
289
|
if (userName != null && !userName.equals("") && !userName.equalsIgnoreCase(PUBLIC)) {
|
290
|
allowQuery = allowQuery + "OR lower(principal_name) = '" + userName +"'";
|
291
|
|
292
|
}
|
293
|
// add group
|
294
|
if (group != null) {
|
295
|
for (int i = 0; i < group.length; i++) {
|
296
|
String groupUint = group[i];
|
297
|
if (groupUint != null && !groupUint.equals("")) {
|
298
|
groupUint = groupUint.toLowerCase();
|
299
|
allowQuery = allowQuery + " OR lower(principal_name) = '"
|
300
|
+ groupUint + "'";
|
301
|
}//if
|
302
|
}//for
|
303
|
}//if
|
304
|
// add allow rule
|
305
|
allowQuery = allowQuery + ") AND perm_type = 'allow'" + " AND permission > 3";
|
306
|
logMetacat.info("allow string is: " + allowQuery);
|
307
|
return allowQuery;
|
308
|
}
|
309
|
|
310
|
/*
|
311
|
* Method to create query for xml_access, this part is to get docid list
|
312
|
* which have a deny rule and perm_order is allowFirst for a given user.
|
313
|
* This means the user will be denied to read
|
314
|
*/
|
315
|
private String createDenyRuleQuery()
|
316
|
{
|
317
|
String denyQuery = null;
|
318
|
String denyString = constructDenyString();
|
319
|
denyQuery = "SELECT docid from xml_access WHERE( " + denyString;
|
320
|
denyQuery = denyQuery + ") ";
|
321
|
logMetacat.info("denyquery is: " + denyQuery);
|
322
|
return denyQuery;
|
323
|
|
324
|
}
|
325
|
|
326
|
/* Construct deny string */
|
327
|
private String constructDenyString()
|
328
|
{
|
329
|
String denyQuery = "";
|
330
|
|
331
|
// add public
|
332
|
denyQuery = "(lower(principal_name) = '" + PUBLIC
|
333
|
+ "'";
|
334
|
|
335
|
// add user name
|
336
|
if (userName != null && !userName.equals("") && !userName.equalsIgnoreCase(PUBLIC)) {
|
337
|
denyQuery = denyQuery + "OR lower(principal_name) = '" + userName +"'";
|
338
|
|
339
|
}
|
340
|
// add groups
|
341
|
if (group != null) {
|
342
|
for (int i = 0; i < group.length; i++) {
|
343
|
String groupUint = group[i];
|
344
|
if (groupUint != null && !groupUint.equals("")) {
|
345
|
groupUint = groupUint.toLowerCase();
|
346
|
denyQuery = denyQuery + " OR lower(principal_name) = '"
|
347
|
+ groupUint + "'";
|
348
|
}//if
|
349
|
}//for
|
350
|
}//if
|
351
|
// add deny rules
|
352
|
denyQuery = denyQuery + ") AND perm_type = 'deny'" + " AND perm_order ='allowFirst'" +" AND permission > 3";
|
353
|
logMetacat.info("allow string is: " + denyQuery);
|
354
|
return denyQuery;
|
355
|
|
356
|
}
|
357
|
|
358
|
/**
|
359
|
* Method to append a access control query to SQL. So in DBQuery class, we
|
360
|
* can get docid from both user specified query and access control query.
|
361
|
* We don't need to checking permission after we get the doclist. It will
|
362
|
* be good to performance
|
363
|
*
|
364
|
*/
|
365
|
public String getAccessQuery()
|
366
|
{
|
367
|
String accessQuery = null;
|
368
|
String onwer = createOwerQuery();
|
369
|
String allow = createAllowRuleQuery();
|
370
|
String deny = createDenyRuleQuery();
|
371
|
//logMetacat.warn("onwer " +onwer);
|
372
|
//logMetacat.warn("allow "+allow);
|
373
|
//logMetacat.warn("deny "+deny);
|
374
|
if (onwer != null)
|
375
|
{
|
376
|
accessQuery = " AND (docid IN(" + onwer + ")";
|
377
|
accessQuery = accessQuery + " OR (docid IN (" + allow + ")"
|
378
|
+ " AND docid NOT IN (" + deny + ")))";
|
379
|
}
|
380
|
else
|
381
|
{
|
382
|
accessQuery = " AND (docid IN (" + allow + ")"
|
383
|
+ " AND docid NOT IN (" + deny + "))";
|
384
|
}
|
385
|
logMetacat.warn("accessquery is: " + accessQuery);
|
386
|
return accessQuery;
|
387
|
}
|
388
|
|
389
|
/**
|
390
|
* Returns true if the parsed query contains and extended xml query (i.e.
|
391
|
* there is at least one <returnfield> in the pathquery document)
|
392
|
*/
|
393
|
public boolean containsExtendedSQL()
|
394
|
{
|
395
|
if (containsExtendedSQL) {
|
396
|
return true;
|
397
|
} else {
|
398
|
return false;
|
399
|
}
|
400
|
}
|
401
|
|
402
|
|
403
|
/**
|
404
|
* Accessor method to return the identifier of this Query
|
405
|
*/
|
406
|
public String getIdentifier()
|
407
|
{
|
408
|
return meta_file_id;
|
409
|
}
|
410
|
|
411
|
/**
|
412
|
* method to set the identifier of this query
|
413
|
*/
|
414
|
public void setIdentifier(String id)
|
415
|
{
|
416
|
this.meta_file_id = id;
|
417
|
}
|
418
|
|
419
|
/**
|
420
|
* Accessor method to return the title of this Query
|
421
|
*/
|
422
|
public String getQueryTitle()
|
423
|
{
|
424
|
return queryTitle;
|
425
|
}
|
426
|
|
427
|
/**
|
428
|
* method to set the title of this query
|
429
|
*/
|
430
|
public void setQueryTitle(String title)
|
431
|
{
|
432
|
this.queryTitle = title;
|
433
|
}
|
434
|
|
435
|
/**
|
436
|
* Accessor method to return a vector of the return document types as
|
437
|
* defined in the <returndoctype> tag in the pathquery dtd.
|
438
|
*/
|
439
|
public Vector getReturnDocList()
|
440
|
{
|
441
|
return this.returnDocList;
|
442
|
}
|
443
|
|
444
|
/**
|
445
|
* method to set the list of return docs of this query
|
446
|
*/
|
447
|
public void setReturnDocList(Vector returnDocList)
|
448
|
{
|
449
|
this.returnDocList = returnDocList;
|
450
|
}
|
451
|
|
452
|
/**
|
453
|
* Accessor method to return a vector of the filter doc types as defined in
|
454
|
* the <filterdoctype> tag in the pathquery dtd.
|
455
|
*/
|
456
|
public Vector getFilterDocList()
|
457
|
{
|
458
|
return this.filterDocList;
|
459
|
}
|
460
|
|
461
|
/**
|
462
|
* method to set the list of filter docs of this query
|
463
|
*/
|
464
|
public void setFilterDocList(Vector filterDocList)
|
465
|
{
|
466
|
this.filterDocList = filterDocList;
|
467
|
}
|
468
|
|
469
|
/**
|
470
|
* Accessor method to return a vector of the extended return fields as
|
471
|
* defined in the <returnfield> tag in the pathquery dtd.
|
472
|
*/
|
473
|
public Vector getReturnFieldList()
|
474
|
{
|
475
|
return this.returnFieldList;
|
476
|
}
|
477
|
|
478
|
/**
|
479
|
* method to set the list of fields to be returned by this query
|
480
|
*/
|
481
|
public void setReturnFieldList(Vector returnFieldList)
|
482
|
{
|
483
|
this.returnFieldList = returnFieldList;
|
484
|
}
|
485
|
|
486
|
/**
|
487
|
* Accessor method to return a vector of the owner fields as defined in the
|
488
|
* <owner> tag in the pathquery dtd.
|
489
|
*/
|
490
|
public Vector getOwnerList()
|
491
|
{
|
492
|
return this.ownerList;
|
493
|
}
|
494
|
|
495
|
/**
|
496
|
* method to set the list of owners used to constrain this query
|
497
|
*/
|
498
|
public void setOwnerList(Vector ownerList)
|
499
|
{
|
500
|
this.ownerList = ownerList;
|
501
|
}
|
502
|
|
503
|
/**
|
504
|
* get the QueryGroup used to express query constraints
|
505
|
*/
|
506
|
public QueryGroup getQueryGroup()
|
507
|
{
|
508
|
return query;
|
509
|
}
|
510
|
|
511
|
/**
|
512
|
* set the querygroup
|
513
|
*/
|
514
|
public void setQueryGroup(QueryGroup group)
|
515
|
{
|
516
|
query = group;
|
517
|
}
|
518
|
|
519
|
/**
|
520
|
* set if this query sepcification has extendQuery(has return doc type or
|
521
|
* not)
|
522
|
*/
|
523
|
public void setContainsExtenedSQL(boolean hasExtenedQuery)
|
524
|
{
|
525
|
containsExtendedSQL = hasExtenedQuery;
|
526
|
}
|
527
|
|
528
|
/**
|
529
|
* Set up the SAX parser for reading the XML serialized query
|
530
|
*/
|
531
|
private XMLReader initializeParser()
|
532
|
{
|
533
|
XMLReader parser = null;
|
534
|
|
535
|
// Set up the SAX document handlers for parsing
|
536
|
try {
|
537
|
|
538
|
// Get an instance of the parser
|
539
|
parser = XMLReaderFactory.createXMLReader(parserName);
|
540
|
|
541
|
// Set the ContentHandler to this instance
|
542
|
parser.setContentHandler(this);
|
543
|
|
544
|
// Set the error Handler to this instance
|
545
|
parser.setErrorHandler(this);
|
546
|
|
547
|
} catch (Exception e) {
|
548
|
System.err.println("Error in QuerySpcecification.initializeParser "
|
549
|
+ e.toString());
|
550
|
}
|
551
|
|
552
|
return parser;
|
553
|
}
|
554
|
|
555
|
/**
|
556
|
* callback method used by the SAX Parser when the start tag of an element
|
557
|
* is detected. Used in this context to parse and store the query
|
558
|
* information in class variables.
|
559
|
*/
|
560
|
public void startElement(String uri, String localName, String qName,
|
561
|
Attributes atts) throws SAXException
|
562
|
{
|
563
|
logMetacat.debug("start at startElement "+localName);
|
564
|
BasicNode currentNode = new BasicNode(localName);
|
565
|
//write element name into xml buffer.
|
566
|
xml.append("<");
|
567
|
xml.append(localName);
|
568
|
// add attributes to BasicNode here
|
569
|
if (atts != null) {
|
570
|
int len = atts.getLength();
|
571
|
for (int i = 0; i < len; i++) {
|
572
|
currentNode
|
573
|
.setAttribute(atts.getLocalName(i), atts.getValue(i));
|
574
|
xml.append(" ");
|
575
|
xml.append(atts.getLocalName(i));
|
576
|
xml.append("=\"");
|
577
|
xml.append(atts.getValue(i));
|
578
|
xml.append("\"");
|
579
|
}
|
580
|
}
|
581
|
xml.append(">");
|
582
|
|
583
|
elementStack.push(currentNode);
|
584
|
if (currentNode.getTagName().equals("querygroup")) {
|
585
|
QueryGroup currentGroup = new QueryGroup(currentNode
|
586
|
.getAttribute("operator"));
|
587
|
if (query == null) {
|
588
|
query = currentGroup;
|
589
|
} else {
|
590
|
QueryGroup parentGroup = (QueryGroup) queryStack.peek();
|
591
|
parentGroup.addChild(currentGroup);
|
592
|
}
|
593
|
queryStack.push(currentGroup);
|
594
|
}
|
595
|
logMetacat.debug("end in startElement "+localName);
|
596
|
}
|
597
|
|
598
|
/**
|
599
|
* callback method used by the SAX Parser when the end tag of an element is
|
600
|
* detected. Used in this context to parse and store the query information
|
601
|
* in class variables.
|
602
|
*/
|
603
|
public void endElement(String uri, String localName, String qName)
|
604
|
throws SAXException
|
605
|
{
|
606
|
logMetacat.debug("start in endElement "+localName);
|
607
|
BasicNode leaving = (BasicNode) elementStack.pop();
|
608
|
if (leaving.getTagName().equals("queryterm")) {
|
609
|
boolean isCaseSensitive = (new Boolean(leaving
|
610
|
.getAttribute("casesensitive"))).booleanValue();
|
611
|
QueryTerm currentTerm = null;
|
612
|
if (currentPathexpr == null) {
|
613
|
currentTerm = new QueryTerm(isCaseSensitive, leaving
|
614
|
.getAttribute("searchmode"), currentValue);
|
615
|
} else {
|
616
|
currentTerm = new QueryTerm(isCaseSensitive, leaving
|
617
|
.getAttribute("searchmode"), currentValue,
|
618
|
currentPathexpr);
|
619
|
}
|
620
|
QueryGroup currentGroup = (QueryGroup) queryStack.peek();
|
621
|
currentGroup.addChild(currentTerm);
|
622
|
currentValue = null;
|
623
|
currentPathexpr = null;
|
624
|
} else if (leaving.getTagName().equals("querygroup")) {
|
625
|
QueryGroup leavingGroup = (QueryGroup) queryStack.pop();
|
626
|
} else if (leaving.getTagName().equals("meta_file_id")) {
|
627
|
meta_file_id = textBuffer.toString().trim();
|
628
|
} else if (leaving.getTagName().equals("querytitle")) {
|
629
|
queryTitle = textBuffer.toString().trim();
|
630
|
} else if (leaving.getTagName().equals("value")) {
|
631
|
currentValue = textBuffer.toString().trim();
|
632
|
currentValue = MetaCatUtil.normalize(currentValue);
|
633
|
} else if (leaving.getTagName().equals("pathexpr")) {
|
634
|
currentPathexpr = textBuffer.toString().trim();
|
635
|
} else if (leaving.getTagName().equals("returndoctype")) {
|
636
|
returnDocList.add(textBuffer.toString().trim());
|
637
|
} else if (leaving.getTagName().equals("filterdoctype")) {
|
638
|
filterDocList.add(textBuffer.toString().trim());
|
639
|
} else if (leaving.getTagName().equals("returnfield")) {
|
640
|
handleReturnField(textBuffer.toString().trim());
|
641
|
} else if (leaving.getTagName().equals("filterdoctype")) {
|
642
|
filterDocList.add(textBuffer.toString().trim());
|
643
|
} else if (leaving.getTagName().equals("owner")) {
|
644
|
ownerList.add(textBuffer.toString().trim());
|
645
|
}
|
646
|
String normalizedXML = textBuffer.toString().trim();
|
647
|
logMetacat.debug("================before normailze "+normalizedXML);
|
648
|
normalizedXML = MetaCatUtil.normalize(normalizedXML);
|
649
|
logMetacat.debug("================after normailze "+normalizedXML);
|
650
|
xml.append(normalizedXML);
|
651
|
xml.append("</");
|
652
|
xml.append(localName);
|
653
|
xml.append(">");
|
654
|
//rest textBuffer
|
655
|
textBuffer = new StringBuffer();
|
656
|
|
657
|
}
|
658
|
|
659
|
/**
|
660
|
* Gets normailized query string in xml format, which can be transformed
|
661
|
* to html
|
662
|
*/
|
663
|
public String getNormalizedXMLQuery()
|
664
|
{
|
665
|
//System.out.println("normailized xml \n"+xml.toString());
|
666
|
return xml.toString();
|
667
|
}
|
668
|
|
669
|
|
670
|
/**
|
671
|
* callback method used by the SAX Parser when the text sequences of an xml
|
672
|
* stream are detected. Used in this context to parse and store the query
|
673
|
* information in class variables.
|
674
|
*/
|
675
|
public void characters(char ch[], int start, int length)
|
676
|
{
|
677
|
// buffer all text nodes for same element. This is for text was splited
|
678
|
// into different nodes
|
679
|
String text = new String(ch, start, length);
|
680
|
logMetacat.debug("the text in characters "+text);
|
681
|
textBuffer.append(text);
|
682
|
|
683
|
}
|
684
|
|
685
|
/**
|
686
|
* Method to handle return field. It will be callied in ecogrid part
|
687
|
* @param inputString
|
688
|
*/
|
689
|
public void handleReturnField(String inputString)
|
690
|
{
|
691
|
int attributePos = inputString.indexOf(ATTRIBUTESYMBOL);
|
692
|
int predicateStart = -1;
|
693
|
int predicateEnd;
|
694
|
boolean hasPredicate = false;
|
695
|
|
696
|
while (true)
|
697
|
{
|
698
|
predicateStart = inputString.indexOf(PREDICATE_START, predicateStart + 1);
|
699
|
|
700
|
if (attributePos == -1)
|
701
|
break;
|
702
|
|
703
|
if (predicateStart == -1)
|
704
|
break;
|
705
|
|
706
|
hasPredicate = true;
|
707
|
|
708
|
if (attributePos < predicateStart)
|
709
|
break;
|
710
|
|
711
|
predicateEnd = inputString.indexOf(PREDICATE_END, predicateStart);
|
712
|
|
713
|
if (predicateEnd == -1)
|
714
|
{
|
715
|
logMetacat.warn("handleReturnField(): ");
|
716
|
logMetacat.warn(" Invalid path: " + inputString);
|
717
|
return;
|
718
|
}
|
719
|
|
720
|
while (attributePos < predicateEnd)
|
721
|
{
|
722
|
attributePos = inputString.indexOf(ATTRIBUTESYMBOL, attributePos + 1);
|
723
|
|
724
|
if (attributePos == -1)
|
725
|
break;
|
726
|
}
|
727
|
}
|
728
|
|
729
|
if (hasPredicate)
|
730
|
{
|
731
|
containsPredicates = true;
|
732
|
returnFieldListWithPredicates.add(inputString);
|
733
|
}
|
734
|
|
735
|
containsExtendedSQL = true;
|
736
|
|
737
|
|
738
|
// no attribute value will be returned
|
739
|
logMetacat.info("QuerySpecification.handleReturnField(): " );
|
740
|
logMetacat.info(" there are no attributes in the XPATH statement" );
|
741
|
returnFieldList.add(inputString);
|
742
|
|
743
|
|
744
|
|
745
|
}
|
746
|
|
747
|
/**
|
748
|
* create a SQL serialization of the query that this instance represents
|
749
|
*/
|
750
|
public String printSQL(boolean useXMLIndex)
|
751
|
{
|
752
|
|
753
|
StringBuffer self = new StringBuffer();
|
754
|
StringBuffer queryString = new StringBuffer();
|
755
|
|
756
|
queryString.append("SELECT docid,docname,doctype,");
|
757
|
queryString.append("date_created, date_updated, rev ");
|
758
|
queryString.append("FROM xml_documents WHERE");
|
759
|
|
760
|
// Get the query from the QueryGroup and check
|
761
|
// if no query has been returned
|
762
|
String queryFromQueryGroup = query.printSQL(useXMLIndex);
|
763
|
logMetacat.info("Query from query in QuerySpec.printSQL: "
|
764
|
+ queryFromQueryGroup);
|
765
|
|
766
|
if(!queryFromQueryGroup.trim().equals("")){
|
767
|
self.append(" docid IN (");
|
768
|
self.append(queryFromQueryGroup);
|
769
|
self.append(") ");
|
770
|
}
|
771
|
|
772
|
// Add SQL to filter for doctypes requested in the query
|
773
|
// This is an implicit OR for the list of doctypes. Only doctypes in
|
774
|
// this
|
775
|
// list will be searched if the tag is present
|
776
|
if (!filterDocList.isEmpty()) {
|
777
|
boolean firstdoctype = true;
|
778
|
boolean emptyString = true;
|
779
|
|
780
|
if(!self.toString().equals("")){
|
781
|
self.append(" AND (");
|
782
|
emptyString = false;
|
783
|
}
|
784
|
|
785
|
Enumeration en = filterDocList.elements();
|
786
|
while (en.hasMoreElements()) {
|
787
|
String currentDoctype = (String) en.nextElement();
|
788
|
if (firstdoctype) {
|
789
|
firstdoctype = false;
|
790
|
self.append(" doctype = '" + currentDoctype + "'");
|
791
|
} else {
|
792
|
self.append(" OR doctype = '" + currentDoctype + "'");
|
793
|
}
|
794
|
}
|
795
|
|
796
|
if(!emptyString){
|
797
|
self.append(") ");
|
798
|
}
|
799
|
}
|
800
|
|
801
|
// Add SQL to filter for owners requested in the query
|
802
|
// This is an implicit OR for the list of owners
|
803
|
if (!ownerList.isEmpty()) {
|
804
|
boolean first = true;
|
805
|
boolean emptyString = true;
|
806
|
|
807
|
if(!self.toString().equals("")){
|
808
|
self.append(" AND (");
|
809
|
emptyString = false;
|
810
|
}
|
811
|
|
812
|
Enumeration en = ownerList.elements();
|
813
|
while (en.hasMoreElements()) {
|
814
|
String current = (String) en.nextElement();
|
815
|
if (current != null) {
|
816
|
current = current.toLowerCase();
|
817
|
}
|
818
|
if (first) {
|
819
|
first = false;
|
820
|
self.append(" lower(user_owner) = '" + current + "'");
|
821
|
} else {
|
822
|
self.append(" OR lower(user_owner) = '" + current + "'");
|
823
|
}
|
824
|
}
|
825
|
|
826
|
if(!emptyString){
|
827
|
self.append(") ");
|
828
|
}
|
829
|
}
|
830
|
|
831
|
// if there is only one percentage search item, this query is a
|
832
|
// percentage
|
833
|
// search query
|
834
|
logMetacat.info("percentage number: "
|
835
|
+ query.getPercentageSymbolCount());
|
836
|
if (query.getPercentageSymbolCount() == 1) {
|
837
|
logMetacat.info("It is a percentage search");
|
838
|
percentageSearch = true;
|
839
|
}
|
840
|
|
841
|
queryString.append(self.toString());
|
842
|
return queryString.toString();
|
843
|
}
|
844
|
|
845
|
|
846
|
|
847
|
/**
|
848
|
* This method prints sql based upon the <returnfield> tag in the
|
849
|
* pathquery document. This allows for customization of the returned fields.
|
850
|
* If the boolean useXMLIndex paramter is false, it uses a recursive query on
|
851
|
* xml_nodes to find the fields to be included by their path expression, and
|
852
|
* avoids the use of the xml_index table.
|
853
|
*
|
854
|
* @param doclist the list of document ids to search
|
855
|
* @param unaccessableNodePair the node pairs (start id and end id) which
|
856
|
* this user should not access
|
857
|
* @param useXMLIndex a boolean flag indicating whether to search using
|
858
|
* xml_index
|
859
|
*/
|
860
|
public String printExtendedSQL(String doclist, boolean useXMLIndex)
|
861
|
{
|
862
|
if (useXMLIndex && !containsPredicates)
|
863
|
{
|
864
|
return printExtendedSQL(doclist);
|
865
|
}
|
866
|
else
|
867
|
{
|
868
|
StringBuffer self = new StringBuffer();
|
869
|
boolean firstfield = true;
|
870
|
// first part comes from fields without predicates
|
871
|
String queryFromWithoutPrecidates = printExtendedSQL(doclist);
|
872
|
if (queryFromWithoutPrecidates != null)
|
873
|
{
|
874
|
// it has return fields without predicate
|
875
|
self.append(queryFromWithoutPrecidates);
|
876
|
firstfield = false;
|
877
|
}
|
878
|
//put the returnfields into the query
|
879
|
//the for loop allows for multiple fields
|
880
|
for (int i = 0; i < returnFieldListWithPredicates.size(); i++)
|
881
|
{
|
882
|
if (firstfield)
|
883
|
{
|
884
|
firstfield = false;
|
885
|
}
|
886
|
else
|
887
|
{
|
888
|
self.append(" UNION ");
|
889
|
}
|
890
|
String path = (String) returnFieldListWithPredicates.elementAt(i);
|
891
|
self.append("select xml_nodes.docid, ");
|
892
|
self.append("'"+ path.replaceAll("'", "''") + "' as path, ");
|
893
|
self.append("xml_nodes.nodedata, ");
|
894
|
self.append("xml_nodes.parentnodeid, ");
|
895
|
self.append("xml_nodes.nodetype ");
|
896
|
//self.append("from xml_nodes, xml_documents ");
|
897
|
self.append("from xml_nodes ");
|
898
|
self.append("where ");
|
899
|
self.append(QueryTerm.useNestedStatements(path));
|
900
|
|
901
|
self.append(" AND xml_nodes.docid in (");
|
902
|
self.append(doclist);
|
903
|
if (returnFieldIsAttribute(path))
|
904
|
{
|
905
|
self.append(")");
|
906
|
}
|
907
|
else
|
908
|
{
|
909
|
self.append(") AND xml_nodes.nodetype = 'TEXT'");
|
910
|
}
|
911
|
//self.append(" AND xml_nodes.rootnodeid = xml_documents.rootnodeid");
|
912
|
|
913
|
//addAccessRestrictionSQL(unaccessableNodePair, self);
|
914
|
}
|
915
|
|
916
|
return self.toString();
|
917
|
}
|
918
|
}
|
919
|
|
920
|
/*
|
921
|
* Determines the returnfield is an attribute of not.
|
922
|
* For given returnfield, this programm will cut the part of path after last slash.
|
923
|
* If no slash in the path, the original string will be considered as last part.
|
924
|
* If first character of last part is @ it will retrun true.
|
925
|
*/
|
926
|
private boolean returnFieldIsAttribute(String path)
|
927
|
{
|
928
|
boolean isAttribute = false;
|
929
|
if (path != null)
|
930
|
{
|
931
|
int slashIndex = path.lastIndexOf("/");
|
932
|
if (slashIndex !=-1)
|
933
|
{
|
934
|
// if there is slash in the path, path should be replace by the last part
|
935
|
path = path.substring(slashIndex+1);
|
936
|
}
|
937
|
logMetacat.debug("In QuerySpecification.returnFieldIsAttribute method, final path is "+path);
|
938
|
// if first of character of path is @, the path is attribute
|
939
|
if (path.charAt(0) == '@')
|
940
|
{
|
941
|
logMetacat.debug("it is attribute");
|
942
|
isAttribute = true;
|
943
|
}
|
944
|
}
|
945
|
return isAttribute;
|
946
|
}
|
947
|
|
948
|
/**
|
949
|
* This method prints sql based upon the <returnfield> tag in the
|
950
|
* pathquery document. This allows for customization of the returned fields.
|
951
|
* It uses the xml_index table and so assumes that this table has been
|
952
|
* built.
|
953
|
*
|
954
|
* @param doclist the list of document ids to search
|
955
|
* @param unaccessableNodePair the node pairs (start id and end id)
|
956
|
* which this user should not access
|
957
|
*/
|
958
|
private String printExtendedSQL(String doclist)
|
959
|
{
|
960
|
logMetacat.info("querySpecification.printExtendedSQL called\n");
|
961
|
StringBuffer self = new StringBuffer();
|
962
|
Vector elementVector = new Vector();
|
963
|
Vector attributeVector = new Vector();
|
964
|
|
965
|
boolean usePathIndex = true;
|
966
|
|
967
|
// test if the are elements in the return fields
|
968
|
if ( returnFieldList.size() == 0 ) {
|
969
|
return null;
|
970
|
}
|
971
|
|
972
|
for (int i = 0; i < returnFieldList.size(); i++) {
|
973
|
String path = (String)returnFieldList.elementAt(i);
|
974
|
// Since return fileds having preicates will be handle in another path,
|
975
|
// we should skip it.
|
976
|
if (returnFieldListWithPredicates.contains(path))
|
977
|
{
|
978
|
continue;
|
979
|
}
|
980
|
if (path != null && path.indexOf(ATTRIBUTESYMBOL) != -1)
|
981
|
{
|
982
|
attributeVector.add(path);
|
983
|
}
|
984
|
else
|
985
|
{
|
986
|
elementVector.add(path);
|
987
|
}
|
988
|
if(!MetaCatUtil.pathsForIndexing.contains(path)){
|
989
|
usePathIndex = false;
|
990
|
}
|
991
|
|
992
|
}
|
993
|
// check if has return field
|
994
|
if (elementVector.size() == 0 && attributeVector.size()==0)
|
995
|
{
|
996
|
return null;
|
997
|
}
|
998
|
|
999
|
if(usePathIndex){
|
1000
|
self.append("select docid, path, nodedata, parentnodeid, null as nodetype ");
|
1001
|
self.append("from xml_path_index where path in( '");
|
1002
|
|
1003
|
boolean firstfield = true;
|
1004
|
//put the returnfields into the query
|
1005
|
//the for loop allows for multiple fields
|
1006
|
for (int i = 0; i < returnFieldList.size(); i++) {
|
1007
|
if (firstfield) {
|
1008
|
firstfield = false;
|
1009
|
self.append( (String) returnFieldList.elementAt(i));
|
1010
|
self.append("' ");
|
1011
|
}
|
1012
|
else {
|
1013
|
self.append(", '");
|
1014
|
self.append( (String) returnFieldList.elementAt(i));
|
1015
|
self.append("' ");
|
1016
|
}
|
1017
|
}
|
1018
|
self.append(") AND docid in (");
|
1019
|
self.append(doclist);
|
1020
|
self.append(")");
|
1021
|
|
1022
|
} else {
|
1023
|
self.append("select xml_nodes.docid, xml_index.path, xml_nodes.nodedata, ");
|
1024
|
self.append("xml_nodes.parentnodeid, ");
|
1025
|
self.append("xml_nodes.nodetype ");
|
1026
|
self.append("FROM xml_index, xml_nodes WHERE (");
|
1027
|
|
1028
|
|
1029
|
boolean firstElement = true;
|
1030
|
boolean firstAttribute = true;
|
1031
|
//put the returnfields into the query
|
1032
|
//the for loop allows for multiple fields
|
1033
|
if (elementVector.size() != 0)
|
1034
|
{
|
1035
|
for (int i = 0; i < elementVector.size(); i++) {
|
1036
|
String path = (String) elementVector.elementAt(i);
|
1037
|
if (firstElement) {
|
1038
|
firstElement = false;
|
1039
|
self.append(" (xml_index.nodeid=xml_nodes.parentnodeid AND xml_index.path IN ('");
|
1040
|
self.append(path);
|
1041
|
self.append("'");
|
1042
|
}
|
1043
|
else
|
1044
|
{
|
1045
|
self.append(", '");
|
1046
|
self.append(path);
|
1047
|
self.append("' ");
|
1048
|
}
|
1049
|
}
|
1050
|
self.append(") AND xml_nodes.nodetype = 'TEXT')");
|
1051
|
}
|
1052
|
|
1053
|
if (attributeVector.size() != 0)
|
1054
|
{
|
1055
|
for (int j=0; j<attributeVector.size(); j++)
|
1056
|
{
|
1057
|
String path = (String) attributeVector.elementAt(j);
|
1058
|
if (firstAttribute)
|
1059
|
{
|
1060
|
firstAttribute = false;
|
1061
|
if (!firstElement)
|
1062
|
{
|
1063
|
self.append(" OR ");
|
1064
|
}
|
1065
|
self.append(" (xml_index.nodeid=xml_nodes.nodeid AND ( xml_index.path IN ( '");
|
1066
|
self.append(path);
|
1067
|
self.append("'");
|
1068
|
}
|
1069
|
else
|
1070
|
{
|
1071
|
self.append(", '");
|
1072
|
self.append(path);
|
1073
|
self.append("' ");
|
1074
|
}
|
1075
|
}
|
1076
|
self.append(") AND xml_nodes.nodetype = 'ATTRIBUTE'))");
|
1077
|
}
|
1078
|
|
1079
|
|
1080
|
self.append(") AND xml_nodes.docid in (");
|
1081
|
self.append(doclist);
|
1082
|
self.append(")");
|
1083
|
|
1084
|
}
|
1085
|
|
1086
|
return self.toString();
|
1087
|
}
|
1088
|
|
1089
|
|
1090
|
/**
|
1091
|
* Method to return a String generated after sorting the returnFieldList
|
1092
|
* Vector
|
1093
|
*/
|
1094
|
public String getSortedReturnFieldString(){
|
1095
|
String returnFields = "";
|
1096
|
|
1097
|
// Create a temporary vector and copy returnFieldList into it
|
1098
|
Vector tempVector = new Vector();
|
1099
|
|
1100
|
Iterator it = returnFieldList.iterator();
|
1101
|
while(it.hasNext()){
|
1102
|
tempVector.add(it.next());
|
1103
|
}
|
1104
|
|
1105
|
/*Enumeration attEnum = attributeReturnList.elements();
|
1106
|
while(attEnum.hasMoreElements()){
|
1107
|
Iterator tempIt = ((Vector)attEnum.nextElement()).iterator();
|
1108
|
String rfield = "";
|
1109
|
if(tempIt.hasNext()){
|
1110
|
String element = (String)tempIt.next();
|
1111
|
if(element != null) {
|
1112
|
rfield +=element;
|
1113
|
}
|
1114
|
}
|
1115
|
if(tempIt.hasNext()){
|
1116
|
String attribute = (String)tempIt.next();
|
1117
|
if(attribute != null) {
|
1118
|
rfield = rfield + "@" + attribute;
|
1119
|
}
|
1120
|
}
|
1121
|
tempVector.add(rfield);
|
1122
|
}*/
|
1123
|
|
1124
|
// Sort the temporary vector
|
1125
|
java.util.Collections.sort(tempVector);
|
1126
|
|
1127
|
// Generate the string and return it
|
1128
|
it = tempVector.iterator();
|
1129
|
while(it.hasNext()){
|
1130
|
returnFields = returnFields + it.next() + "|";
|
1131
|
}
|
1132
|
return returnFields;
|
1133
|
}
|
1134
|
|
1135
|
|
1136
|
|
1137
|
|
1138
|
|
1139
|
public static String printRelationSQL(String docid)
|
1140
|
{
|
1141
|
StringBuffer self = new StringBuffer();
|
1142
|
self.append("select subject, relationship, object, subdoctype, ");
|
1143
|
self.append("objdoctype from xml_relation ");
|
1144
|
self.append("where docid like '").append(docid).append("'");
|
1145
|
return self.toString();
|
1146
|
}
|
1147
|
|
1148
|
public static String printGetDocByDoctypeSQL(String docid)
|
1149
|
{
|
1150
|
StringBuffer self = new StringBuffer();
|
1151
|
|
1152
|
self.append("SELECT docid,docname,doctype,");
|
1153
|
self.append("date_created, date_updated ");
|
1154
|
self.append("FROM xml_documents WHERE docid IN (");
|
1155
|
self.append(docid).append(")");
|
1156
|
return self.toString();
|
1157
|
}
|
1158
|
|
1159
|
/**
|
1160
|
* create a String description of the query that this instance represents.
|
1161
|
* This should become a way to get the XML serialization of the query.
|
1162
|
*/
|
1163
|
public String toString()
|
1164
|
{
|
1165
|
return "meta_file_id=" + meta_file_id + "\n" + query;
|
1166
|
//DOCTITLE attr cleared from the db
|
1167
|
//return "meta_file_id=" + meta_file_id + "\n" +
|
1168
|
//"querytitle=" + querytitle + "\n" + query;
|
1169
|
}
|
1170
|
|
1171
|
/** A method to get rid of attribute part in path expression */
|
1172
|
public static String newPathExpressionWithOutAttribute(String pathExpression)
|
1173
|
{
|
1174
|
if (pathExpression == null) { return null; }
|
1175
|
int index = pathExpression.lastIndexOf(ATTRIBUTESYMBOL);
|
1176
|
String newExpression = null;
|
1177
|
if (index != 0) {
|
1178
|
newExpression = pathExpression.substring(0, index - 1);
|
1179
|
}
|
1180
|
logMetacat.info("The path expression without attributes: "
|
1181
|
+ newExpression);
|
1182
|
return newExpression;
|
1183
|
}
|
1184
|
|
1185
|
/** A method to get attribute name from path */
|
1186
|
public static String getAttributeName(String path)
|
1187
|
{
|
1188
|
if (path == null) { return null; }
|
1189
|
int index = path.lastIndexOf(ATTRIBUTESYMBOL);
|
1190
|
int size = path.length();
|
1191
|
String attributeName = null;
|
1192
|
if (index != 1) {
|
1193
|
attributeName = path.substring(index + 1, size);
|
1194
|
}
|
1195
|
logMetacat.info("The attirbute name from path: "
|
1196
|
+ attributeName);
|
1197
|
return attributeName;
|
1198
|
}
|
1199
|
|
1200
|
}
|