1
|
/**
|
2
|
* '$RCSfile$'
|
3
|
* Purpose: A Class that implements session utility methods
|
4
|
* Copyright: 2008 Regents of the University of California and the
|
5
|
* National Center for Ecological Analysis and Synthesis
|
6
|
* Authors: Michael Daigle
|
7
|
*
|
8
|
* '$Author: daigle $'
|
9
|
* '$Date: 2008-08-22 16:23:38 -0700 (Fri, 22 Aug 2008) $'
|
10
|
* '$Revision: 4297 $'
|
11
|
*
|
12
|
* This program is free software; you can redistribute it and/or modify
|
13
|
* it under the terms of the GNU General Public License as published by
|
14
|
* the Free Software Foundation; either version 2 of the License, or
|
15
|
* (at your option) any later version.
|
16
|
*
|
17
|
* This program is distributed in the hope that it will be useful,
|
18
|
* but WITHOUT ANY WARRANTY; without even the implied warranty of
|
19
|
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
|
20
|
* GNU General Public License for more details.
|
21
|
*
|
22
|
* You should have received a copy of the GNU General Public License
|
23
|
* along with this program; if not, write to the Free Software
|
24
|
* Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
|
25
|
*/
|
26
|
|
27
|
package edu.ucsb.nceas.metacat.service;
|
28
|
|
29
|
import java.io.IOException;
|
30
|
import java.io.StringReader;
|
31
|
import java.sql.PreparedStatement;
|
32
|
import java.sql.ResultSet;
|
33
|
import java.sql.SQLException;
|
34
|
import java.util.Vector;
|
35
|
import java.util.regex.Matcher;
|
36
|
import java.util.regex.Pattern;
|
37
|
|
38
|
import org.apache.log4j.Logger;
|
39
|
|
40
|
import edu.ucsb.nceas.metacat.DBConnection;
|
41
|
import edu.ucsb.nceas.metacat.DBConnectionPool;
|
42
|
import edu.ucsb.nceas.metacat.DocumentImpl;
|
43
|
import edu.ucsb.nceas.metacat.MetaCatServlet;
|
44
|
import edu.ucsb.nceas.utilities.FileUtil;
|
45
|
import edu.ucsb.nceas.utilities.PropertyNotFoundException;
|
46
|
import edu.ucsb.nceas.utilities.StringUtil;
|
47
|
|
48
|
public class XMLSchemaService extends BaseService {
|
49
|
|
50
|
public static final String NAMESPACEKEYWORD = "xmlns";
|
51
|
|
52
|
public static final String SCHEMA_DIR = "/schema/";
|
53
|
|
54
|
private static XMLSchemaService xmlSchemaService = null;
|
55
|
|
56
|
private static Logger logMetacat = Logger.getLogger(XMLSchemaService.class);
|
57
|
|
58
|
private static boolean useFullSchemaValidation = false;
|
59
|
|
60
|
// private static String documentNamespace = null;
|
61
|
|
62
|
// all schema objects that represent schemas registered in the db that
|
63
|
// actually have files on disk.
|
64
|
private static Vector<XMLSchema> registeredSchemaList = new Vector<XMLSchema>();
|
65
|
|
66
|
// a convenience list that holds the names of registered namespaces.
|
67
|
private static Vector<String> nameSpaceList = new Vector<String>();
|
68
|
|
69
|
// a convenience string that holds all name spaces and locations in a space
|
70
|
// delimited format
|
71
|
private static String nameSpaceAndLocationString = "";
|
72
|
|
73
|
/**
|
74
|
* private constructor since this is a singleton
|
75
|
*/
|
76
|
private XMLSchemaService() {
|
77
|
try {
|
78
|
setUseFullSchemaValidation();
|
79
|
populateRegisteredSchemaList();
|
80
|
createRegisteredNameSpaceList();
|
81
|
createRegisteredNameSpaceAndLocationString();
|
82
|
|
83
|
} catch (PropertyNotFoundException pnfe) {
|
84
|
logMetacat.error("Could not find property: xml.useFullSchemaValidation. " +
|
85
|
"Setting to false.");
|
86
|
}
|
87
|
}
|
88
|
|
89
|
/**
|
90
|
* Get the single instance of XMLService.
|
91
|
*
|
92
|
* @return the single instance of XMLService
|
93
|
*/
|
94
|
public static XMLSchemaService getInstance() {
|
95
|
if (xmlSchemaService == null) {
|
96
|
xmlSchemaService = new XMLSchemaService();
|
97
|
}
|
98
|
return xmlSchemaService;
|
99
|
}
|
100
|
|
101
|
public boolean refreshable() {
|
102
|
return true;
|
103
|
}
|
104
|
|
105
|
/**
|
106
|
* refresh the persistant values in this service.
|
107
|
*/
|
108
|
protected void doRefresh() throws ServiceException {
|
109
|
try {
|
110
|
populateRegisteredSchemaList();
|
111
|
setUseFullSchemaValidation();
|
112
|
createRegisteredNameSpaceList();
|
113
|
createRegisteredNameSpaceAndLocationString();
|
114
|
} catch (PropertyNotFoundException pnfe) {
|
115
|
logMetacat.error("Could not find property: xml.useFullSchemaValidation. " +
|
116
|
"Setting to false.");
|
117
|
}
|
118
|
}
|
119
|
|
120
|
/**
|
121
|
* Gets the registered schema list. This list holds schemas that exist in
|
122
|
* the xml_catalog table that also have associated files in the schema
|
123
|
* directory.
|
124
|
*
|
125
|
* @return a list of XMLSchema objects holding registered schema information
|
126
|
*/
|
127
|
public static Vector<XMLSchema> getRegisteredSchemaList() {
|
128
|
return registeredSchemaList;
|
129
|
}
|
130
|
|
131
|
/**
|
132
|
* Gets the name space and location string. This is a convenience method.
|
133
|
* The string will have space delimited namespaces and locations that are
|
134
|
* held in the registered schema list.
|
135
|
*
|
136
|
* @return a string that holds space delimited registered namespaces and
|
137
|
* locations.
|
138
|
*/
|
139
|
public static String getNameSpaceAndLocationString() {
|
140
|
return nameSpaceAndLocationString;
|
141
|
}
|
142
|
|
143
|
/**
|
144
|
* Gets a list of name spaces. This is a convenience method. The list will
|
145
|
* have all namespaces that are held in the registered schema list.
|
146
|
*
|
147
|
* @return a list that holds registered namespaces.
|
148
|
*/
|
149
|
public static Vector<String> getNameSpaceList() {
|
150
|
return nameSpaceList;
|
151
|
}
|
152
|
|
153
|
/**
|
154
|
* Report whether xml parsing is set to use full schema parsing. If full
|
155
|
* schema parsing is true, new schemas will be validated before being
|
156
|
* entered into the database and file system.
|
157
|
*
|
158
|
* @return true if the xml.useFullSchemaValidation property is set to true,
|
159
|
* false otherwise.
|
160
|
*/
|
161
|
public static boolean useFullSchemaValidation() {
|
162
|
return useFullSchemaValidation;
|
163
|
}
|
164
|
|
165
|
/**
|
166
|
* sets the UseFullSchemaValidation variable. The only way this should be
|
167
|
* set is in the constructor or the refresh methods.
|
168
|
*/
|
169
|
private static void setUseFullSchemaValidation() throws PropertyNotFoundException {
|
170
|
String strUseFullSchemaValidation =
|
171
|
PropertyService.getProperty("xml.useFullSchemaValidation");
|
172
|
useFullSchemaValidation = Boolean.valueOf(strUseFullSchemaValidation);
|
173
|
}
|
174
|
|
175
|
/**
|
176
|
* Populate the list of registered schemas. This reads all schemas in the
|
177
|
* xml_catalog table and then makes sure the schema actually exists and is
|
178
|
* readable on disk.
|
179
|
*/
|
180
|
public static void populateRegisteredSchemaList() {
|
181
|
DBConnection conn = null;
|
182
|
int serialNumber = -1;
|
183
|
PreparedStatement pstmt = null;
|
184
|
ResultSet resultSet = null;
|
185
|
registeredSchemaList = new Vector<XMLSchema>();
|
186
|
|
187
|
// get the system id from the xml_catalog table for all schemas.
|
188
|
String sql = "SELECT public_id, system_id FROM xml_catalog where "
|
189
|
+ "entry_type ='" + DocumentImpl.SCHEMA + "'";
|
190
|
try {
|
191
|
// check out DBConnection
|
192
|
conn = DBConnectionPool
|
193
|
.getDBConnection("XMLService.populateRegisteredSchemaList");
|
194
|
serialNumber = conn.getCheckOutSerialNumber();
|
195
|
pstmt = conn.prepareStatement(sql);
|
196
|
logMetacat.debug("Selecting schemas: " + pstmt.toString());
|
197
|
pstmt.execute();
|
198
|
resultSet = pstmt.getResultSet();
|
199
|
|
200
|
// make sure the schema actually exists on the file system. If so,
|
201
|
// add it to the registered schema list.
|
202
|
while (resultSet.next()) {
|
203
|
String fileNamespace = resultSet.getString(1);
|
204
|
String fileLocation = resultSet.getString(2);
|
205
|
logMetacat.debug("Registering schema: " + fileNamespace + " " + fileLocation);
|
206
|
XMLSchema xmlSchema = new XMLSchema(fileNamespace);
|
207
|
xmlSchema.setFileName(fileLocation);
|
208
|
|
209
|
if (FileUtil.getFileStatus(xmlSchema.getLocalFileDir()) >= FileUtil.EXISTS_READABLE) {
|
210
|
registeredSchemaList.add(xmlSchema);
|
211
|
} else {
|
212
|
logMetacat.warn("Schema file: " + xmlSchema.getLocalFileDir() + " is registered "
|
213
|
+ " in the database but does not exist on the file system.");
|
214
|
}
|
215
|
}
|
216
|
} catch (SQLException e) {
|
217
|
logMetacat.error("XMLService.populateRegisteredSchemaList(): "
|
218
|
+ e.getMessage());
|
219
|
} finally {
|
220
|
try {
|
221
|
pstmt.close();
|
222
|
}// try
|
223
|
catch (SQLException sqlE) {
|
224
|
logMetacat.error("Error in XMLService.populateRegisteredSchemaList(): "
|
225
|
+ sqlE.getMessage());
|
226
|
}
|
227
|
DBConnectionPool.returnDBConnection(conn, serialNumber);
|
228
|
}
|
229
|
}
|
230
|
|
231
|
/**
|
232
|
* create a space delimited string of all namespaces and locations
|
233
|
* in the registered schema list.
|
234
|
*/
|
235
|
private static void createRegisteredNameSpaceAndLocationString() {
|
236
|
boolean firstRow = true;
|
237
|
nameSpaceAndLocationString = "";
|
238
|
|
239
|
for (XMLSchema xmlSchema : registeredSchemaList) {
|
240
|
if (!firstRow) {
|
241
|
nameSpaceAndLocationString += " ";
|
242
|
}
|
243
|
nameSpaceAndLocationString += xmlSchema.getFileNamespace() + " "
|
244
|
+ xmlSchema.getLocalFileUri();
|
245
|
firstRow = false;
|
246
|
}
|
247
|
}
|
248
|
|
249
|
/**
|
250
|
* create a lsit of all namespaces in the registered schema list.
|
251
|
*/
|
252
|
private static void createRegisteredNameSpaceList() {
|
253
|
nameSpaceList = new Vector<String>();
|
254
|
for (XMLSchema xmlSchema : registeredSchemaList) {
|
255
|
nameSpaceList.add(xmlSchema.getFileNamespace());
|
256
|
}
|
257
|
}
|
258
|
|
259
|
/**
|
260
|
* Checks to see that all schemas are registered. If a single one in the
|
261
|
* list is not, this will return false.
|
262
|
*
|
263
|
* @param schemaList
|
264
|
* a list of schemas as they appear in xml.
|
265
|
* @return true if all schemas are registered.
|
266
|
*/
|
267
|
public static boolean areAllSchemasRegistered(Vector<XMLSchema> schemaList) {
|
268
|
for (XMLSchema xmlSchema : schemaList) {
|
269
|
if ( ! isSchemaRegistered(xmlSchema)) {
|
270
|
return false;
|
271
|
}
|
272
|
}
|
273
|
return true;
|
274
|
}
|
275
|
|
276
|
/**
|
277
|
* Returns true if the schema is registered.
|
278
|
*
|
279
|
* @param schema
|
280
|
* a single schema as it appears in xml
|
281
|
* @return true if the schema is registered, false otherwise.
|
282
|
*/
|
283
|
public static boolean isSchemaRegistered(XMLSchema xmlSchema) {
|
284
|
for (XMLSchema registeredXmlSchema : registeredSchemaList) {
|
285
|
if (registeredXmlSchema.getLocalFileUri().equals(
|
286
|
xmlSchema.getLocalFileUri())
|
287
|
&& registeredXmlSchema.getFileNamespace().equals(
|
288
|
xmlSchema.getFileNamespace())) {
|
289
|
return true;
|
290
|
}
|
291
|
}
|
292
|
|
293
|
return false;
|
294
|
}
|
295
|
|
296
|
/**
|
297
|
* See if schemas have been specified in the xml:schemalocation attribute.
|
298
|
* If so, return a vector of the system ids.
|
299
|
*
|
300
|
* @param xml
|
301
|
* the document we want to look in for schema location
|
302
|
* @return a vector of XMLSchema objects, or an empty vector if none are
|
303
|
* found
|
304
|
*/
|
305
|
public static Vector<XMLSchema> findSchemasInXML(StringReader xml) throws IOException {
|
306
|
Logger logMetacat = Logger.getLogger(MetaCatServlet.class);
|
307
|
Vector<XMLSchema> schemaList = new Vector<XMLSchema>();
|
308
|
|
309
|
// no xml. return empty vector
|
310
|
if (xml == null) {
|
311
|
logMetacat.debug("Returning empty schemaList.");
|
312
|
return schemaList;
|
313
|
}
|
314
|
|
315
|
// Get the "second line" from the xml
|
316
|
String targetLine = getSchemaLine(xml);
|
317
|
|
318
|
// see if there is a match for xsi.schemaLocation. If so, extract the
|
319
|
// schemas.
|
320
|
if (targetLine != null) {
|
321
|
String regex = "(\\p{Graph}*):schemaLocation=\"(.*)\"";
|
322
|
Pattern pattern = Pattern.compile(regex, Pattern.CASE_INSENSITIVE
|
323
|
| Pattern.DOTALL);
|
324
|
Matcher matcher = pattern.matcher(targetLine);
|
325
|
int i = 0;
|
326
|
while (i < targetLine.length()) {
|
327
|
if (!matcher.find(i)) {
|
328
|
break;
|
329
|
}
|
330
|
|
331
|
String uri = matcher.group(2);
|
332
|
uri = StringUtil.replaceTabsNewLines(uri);
|
333
|
uri = StringUtil.replaceDuplicateSpaces(uri);
|
334
|
|
335
|
// each namespace could have several schema locations. parsedUri
|
336
|
// will hold a list of uri and files.
|
337
|
Vector<String> parsedUri = StringUtil.toVector(uri, ' ');
|
338
|
for (int j = 0; j < parsedUri.size(); j = j + 2) {
|
339
|
if (j + 1 >= parsedUri.size()) {
|
340
|
throw new IOException(
|
341
|
"Odd number of elements found when parsing schema location: "
|
342
|
+ targetLine
|
343
|
+ ". There should be an even number of uri/files in location.");
|
344
|
}
|
345
|
XMLSchema xmlSchema = new XMLSchema(parsedUri.get(j), parsedUri
|
346
|
.get(j + 1));
|
347
|
schemaList.add(xmlSchema);
|
348
|
}
|
349
|
i = matcher.end();
|
350
|
}
|
351
|
}
|
352
|
|
353
|
logMetacat.debug("Schemas for eml are " + schemaList.toString());
|
354
|
|
355
|
return schemaList;
|
356
|
}
|
357
|
|
358
|
/**
|
359
|
* Returns all the namespace for an xml document. This is done by getting
|
360
|
* the internal namespace declaration (prefix) and looking for xmlns:<prefix>
|
361
|
*
|
362
|
* @param xml
|
363
|
* the document to search
|
364
|
* @return a string holding the namespace
|
365
|
*/
|
366
|
public static String findDocumentNamespace(StringReader xml) throws IOException {
|
367
|
String namespace = null;
|
368
|
|
369
|
String eml2_0_0NameSpace = DocumentImpl.EML2_0_0NAMESPACE;
|
370
|
String eml2_0_1NameSpace = DocumentImpl.EML2_0_1NAMESPACE;
|
371
|
String eml2_1_0NameSpace = DocumentImpl.EML2_1_0NAMESPACE;
|
372
|
|
373
|
if (xml == null) {
|
374
|
logMetacat.debug("XML doc is null. There is no namespace.");
|
375
|
return namespace;
|
376
|
}
|
377
|
|
378
|
String targetLine = getSchemaLine(xml);
|
379
|
|
380
|
// the prefix is at the beginning of the doc
|
381
|
String prefix = null;
|
382
|
String regex1 = "^\\s*(\\p{Graph}+):\\p{Graph}* ";
|
383
|
Pattern pattern = Pattern.compile(regex1, Pattern.CASE_INSENSITIVE);
|
384
|
Matcher matcher = pattern.matcher(targetLine);
|
385
|
if (matcher.find()) {
|
386
|
prefix = matcher.group(1).trim();
|
387
|
}
|
388
|
|
389
|
// if a prefix was found, we are looking for xmlns:<prefix>="namespace"
|
390
|
// if no prefix was found, we grab the first namespace.
|
391
|
String regex2;
|
392
|
if (prefix != null) {
|
393
|
regex2 = "xmlns:" + prefix + "=\"(.*)\"";
|
394
|
} else {
|
395
|
regex2 = "xmlns:.*=\"(.*)\"";
|
396
|
}
|
397
|
Pattern pattern2 = Pattern.compile(regex2, Pattern.CASE_INSENSITIVE);
|
398
|
Matcher matcher2 = pattern2.matcher(targetLine);
|
399
|
if (matcher2.find()) {
|
400
|
namespace = matcher2.group(1);
|
401
|
|
402
|
if (namespace.indexOf(eml2_0_0NameSpace) != -1) {
|
403
|
namespace = eml2_0_0NameSpace;
|
404
|
} else if (namespace.indexOf(eml2_0_1NameSpace) != -1) {
|
405
|
namespace = eml2_0_1NameSpace;
|
406
|
} else if (namespace.indexOf(eml2_1_0NameSpace) != -1) {
|
407
|
namespace = eml2_1_0NameSpace;
|
408
|
}
|
409
|
}
|
410
|
|
411
|
return namespace;
|
412
|
}
|
413
|
|
414
|
/**
|
415
|
* Return the line from xml that holds the metadata like namespace and
|
416
|
* schema location
|
417
|
*
|
418
|
* @param xml
|
419
|
* the document to parse
|
420
|
* @return the "second" line of the document
|
421
|
*/
|
422
|
private static String getSchemaLine(StringReader xml) throws IOException {
|
423
|
Logger logMetacat = Logger.getLogger(MetaCatServlet.class);
|
424
|
// find the line
|
425
|
String secondLine = null;
|
426
|
int count = 0;
|
427
|
final int TARGETNUM = 1;
|
428
|
StringBuffer buffer = new StringBuffer();
|
429
|
boolean comment = false;
|
430
|
boolean processingInstruction = false;
|
431
|
char thirdPreviousCharacter = '?';
|
432
|
char secondPreviousCharacter = '?';
|
433
|
char previousCharacter = '?';
|
434
|
char currentCharacter = '?';
|
435
|
int tmp = xml.read();
|
436
|
while (tmp != -1) {
|
437
|
currentCharacter = (char)tmp;
|
438
|
//in a comment
|
439
|
if (currentCharacter == '-' && previousCharacter == '-'
|
440
|
&& secondPreviousCharacter == '!'
|
441
|
&& thirdPreviousCharacter == '<') {
|
442
|
comment = true;
|
443
|
}
|
444
|
//out of comment
|
445
|
if (comment && currentCharacter == '>' && previousCharacter == '-'
|
446
|
&& secondPreviousCharacter == '-') {
|
447
|
comment = false;
|
448
|
}
|
449
|
|
450
|
//in a processingInstruction
|
451
|
if (currentCharacter == '?' && previousCharacter == '<') {
|
452
|
processingInstruction = true;
|
453
|
}
|
454
|
|
455
|
//out of processingInstruction
|
456
|
if (processingInstruction && currentCharacter == '>'
|
457
|
&& previousCharacter == '?') {
|
458
|
processingInstruction = false;
|
459
|
}
|
460
|
|
461
|
//this is not comment or a processingInstruction
|
462
|
if (currentCharacter != '!' && previousCharacter == '<'
|
463
|
&& !comment && !processingInstruction) {
|
464
|
count++;
|
465
|
}
|
466
|
|
467
|
// get target line
|
468
|
if (count == TARGETNUM && currentCharacter != '>') {
|
469
|
buffer.append(currentCharacter);
|
470
|
}
|
471
|
if (count == TARGETNUM && currentCharacter == '>') {
|
472
|
break;
|
473
|
}
|
474
|
thirdPreviousCharacter = secondPreviousCharacter;
|
475
|
secondPreviousCharacter = previousCharacter;
|
476
|
previousCharacter = currentCharacter;
|
477
|
tmp = xml.read();
|
478
|
}
|
479
|
secondLine = buffer.toString();
|
480
|
logMetacat.debug("the second line string is: " + secondLine);
|
481
|
|
482
|
xml.reset();
|
483
|
return secondLine;
|
484
|
}
|
485
|
|
486
|
/**
|
487
|
* Get a schema file name from the schema uri.
|
488
|
*
|
489
|
* @param uri
|
490
|
* the uri from which to extract the file name
|
491
|
* @return a string holding the file name
|
492
|
*/
|
493
|
public static String getSchemaFileNameFromUri(String uri) {
|
494
|
// get filename from systemId
|
495
|
String filename = uri;
|
496
|
|
497
|
if (filename != null && !(filename.trim()).equals("")) {
|
498
|
int slash = Math.max(filename.lastIndexOf('/'), filename.lastIndexOf('\\'));
|
499
|
if (slash > -1) {
|
500
|
filename = filename.substring(slash + 1);
|
501
|
}
|
502
|
}
|
503
|
|
504
|
return filename;
|
505
|
}
|
506
|
}
|