1
|
/**
|
2
|
* '$RCSfile$'
|
3
|
* Purpose: A Class for upgrading the database to version 1.5
|
4
|
* Copyright: 2000 Regents of the University of California and the
|
5
|
* National Center for Ecological Analysis and Synthesis
|
6
|
* Authors: Saurabh Garg
|
7
|
*
|
8
|
* '$Author: tao $'
|
9
|
* '$Date: 2014-11-20 09:00:01 -0800 (Thu, 20 Nov 2014) $'
|
10
|
* '$Revision: 8991 $'
|
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
|
package edu.ucsb.nceas.metacat.dataone;
|
27
|
|
28
|
import java.io.File;
|
29
|
import java.io.IOException;
|
30
|
import java.io.InputStream;
|
31
|
import java.math.BigInteger;
|
32
|
import java.net.URL;
|
33
|
import java.net.URLConnection;
|
34
|
import java.security.NoSuchAlgorithmException;
|
35
|
import java.sql.SQLException;
|
36
|
import java.util.ArrayList;
|
37
|
import java.util.Collections;
|
38
|
import java.util.Date;
|
39
|
import java.util.HashMap;
|
40
|
import java.util.Hashtable;
|
41
|
import java.util.List;
|
42
|
import java.util.Map;
|
43
|
import java.util.Vector;
|
44
|
|
45
|
import javax.xml.parsers.ParserConfigurationException;
|
46
|
import javax.xml.xpath.XPathExpressionException;
|
47
|
|
48
|
import org.apache.commons.beanutils.BeanUtils;
|
49
|
import org.apache.commons.io.IOUtils;
|
50
|
import org.apache.log4j.Logger;
|
51
|
import org.apache.wicket.protocol.http.mock.MockHttpServletRequest;
|
52
|
import org.dataone.client.v2.formats.ObjectFormatCache;
|
53
|
import org.dataone.eml.DataoneEMLParser;
|
54
|
import org.dataone.eml.EMLDocument;
|
55
|
import org.dataone.eml.EMLDocument.DistributionMetadata;
|
56
|
import org.dataone.ore.ResourceMapFactory;
|
57
|
import org.dataone.service.exceptions.BaseException;
|
58
|
import org.dataone.service.exceptions.NotFound;
|
59
|
import org.dataone.service.types.v1.AccessPolicy;
|
60
|
import org.dataone.service.types.v1.AccessRule;
|
61
|
import org.dataone.service.types.v1.Checksum;
|
62
|
import org.dataone.service.types.v1.Identifier;
|
63
|
import org.dataone.service.types.v1.NodeReference;
|
64
|
import org.dataone.service.types.v1.ObjectFormatIdentifier;
|
65
|
import org.dataone.service.types.v1.ReplicationPolicy;
|
66
|
import org.dataone.service.types.v1.Session;
|
67
|
import org.dataone.service.types.v1.Subject;
|
68
|
import org.dataone.service.types.v2.SystemMetadata;
|
69
|
import org.dataone.service.types.v1.util.ChecksumUtil;
|
70
|
import org.dataone.service.util.DateTimeMarshaller;
|
71
|
import org.dspace.foresite.ResourceMap;
|
72
|
import org.jibx.runtime.JiBXException;
|
73
|
import org.xml.sax.SAXException;
|
74
|
|
75
|
import java.util.Calendar;
|
76
|
|
77
|
import edu.ucsb.nceas.metacat.AccessionNumber;
|
78
|
import edu.ucsb.nceas.metacat.AccessionNumberException;
|
79
|
import edu.ucsb.nceas.metacat.DBUtil;
|
80
|
import edu.ucsb.nceas.metacat.DocumentImpl;
|
81
|
import edu.ucsb.nceas.metacat.IdentifierManager;
|
82
|
import edu.ucsb.nceas.metacat.McdbDocNotFoundException;
|
83
|
import edu.ucsb.nceas.metacat.McdbException;
|
84
|
import edu.ucsb.nceas.metacat.MetaCatServlet;
|
85
|
import edu.ucsb.nceas.metacat.MetacatHandler;
|
86
|
import edu.ucsb.nceas.metacat.accesscontrol.AccessControlException;
|
87
|
import edu.ucsb.nceas.metacat.client.InsufficientKarmaException;
|
88
|
import edu.ucsb.nceas.metacat.dataone.hazelcast.HazelcastService;
|
89
|
import edu.ucsb.nceas.metacat.index.MetacatSolrIndex;
|
90
|
import edu.ucsb.nceas.metacat.properties.PropertyService;
|
91
|
import edu.ucsb.nceas.metacat.replication.ReplicationService;
|
92
|
import edu.ucsb.nceas.metacat.shared.AccessException;
|
93
|
import edu.ucsb.nceas.metacat.shared.HandlerException;
|
94
|
import edu.ucsb.nceas.metacat.util.DocumentUtil;
|
95
|
import edu.ucsb.nceas.utilities.ParseLSIDException;
|
96
|
import edu.ucsb.nceas.utilities.PropertyNotFoundException;
|
97
|
|
98
|
public class SystemMetadataFactory {
|
99
|
|
100
|
public static final String RESOURCE_MAP_PREFIX = "resourceMap_";
|
101
|
private static Logger logMetacat = Logger.getLogger(SystemMetadataFactory.class);
|
102
|
/**
|
103
|
* use this flag if you want to update any existing system metadata values with generated content
|
104
|
*/
|
105
|
private static boolean updateExisting = true;
|
106
|
|
107
|
|
108
|
|
109
|
/**
|
110
|
* Create a system metadata object for insertion into metacat
|
111
|
* @param localId
|
112
|
* @param includeORE
|
113
|
* @param downloadData
|
114
|
* @return
|
115
|
* @throws McdbException
|
116
|
* @throws McdbDocNotFoundException
|
117
|
* @throws SQLException
|
118
|
* @throws IOException
|
119
|
* @throws AccessionNumberException
|
120
|
* @throws ClassNotFoundException
|
121
|
* @throws InsufficientKarmaException
|
122
|
* @throws ParseLSIDException
|
123
|
* @throws PropertyNotFoundException
|
124
|
* @throws BaseException
|
125
|
* @throws NoSuchAlgorithmException
|
126
|
* @throws JiBXException
|
127
|
* @throws AccessControlException
|
128
|
* @throws HandlerException
|
129
|
* @throws SAXException
|
130
|
* @throws AccessException
|
131
|
*/
|
132
|
public static SystemMetadata createSystemMetadata(String localId, boolean includeORE, boolean downloadData)
|
133
|
throws McdbException, McdbDocNotFoundException, SQLException,
|
134
|
IOException, AccessionNumberException, ClassNotFoundException,
|
135
|
InsufficientKarmaException, ParseLSIDException,
|
136
|
PropertyNotFoundException, BaseException, NoSuchAlgorithmException,
|
137
|
JiBXException, AccessControlException, HandlerException, SAXException, AccessException {
|
138
|
boolean indexDataFile = false;
|
139
|
return createSystemMetadata(indexDataFile, localId, includeORE, downloadData);
|
140
|
}
|
141
|
/**
|
142
|
* Creates a system metadata object for insertion into metacat
|
143
|
* @param indexDataFile
|
144
|
* Indicate if we need to index data file.
|
145
|
*
|
146
|
* @param localId
|
147
|
* The local document identifier
|
148
|
* @param user
|
149
|
* The user submitting the system metadata document
|
150
|
* @param groups
|
151
|
* The groups the user belongs to
|
152
|
*
|
153
|
* @return sysMeta The system metadata object created
|
154
|
* @throws SAXException
|
155
|
* @throws HandlerException
|
156
|
* @throws AccessControlException
|
157
|
* @throws AccessException
|
158
|
*/
|
159
|
public static SystemMetadata createSystemMetadata(boolean indexDataFile, String localId, boolean includeORE, boolean downloadData)
|
160
|
throws McdbException, McdbDocNotFoundException, SQLException,
|
161
|
IOException, AccessionNumberException, ClassNotFoundException,
|
162
|
InsufficientKarmaException, ParseLSIDException,
|
163
|
PropertyNotFoundException, BaseException, NoSuchAlgorithmException,
|
164
|
JiBXException, AccessControlException, HandlerException, SAXException, AccessException {
|
165
|
|
166
|
logMetacat.debug("createSystemMetadata() called for localId " + localId);
|
167
|
|
168
|
// check for system metadata
|
169
|
SystemMetadata sysMeta = null;
|
170
|
|
171
|
AccessionNumber accNum = new AccessionNumber(localId, "NONE");
|
172
|
int rev = Integer.valueOf(accNum.getRev());
|
173
|
|
174
|
// get/make the guid
|
175
|
String guid = null;
|
176
|
try {
|
177
|
// get the guid if it exists
|
178
|
guid = IdentifierManager.getInstance().getGUID(accNum.getDocid(), rev);
|
179
|
} catch (McdbDocNotFoundException dnfe) {
|
180
|
// otherwise create the mapping
|
181
|
logMetacat.debug("No guid found in the identifier table. Creating mapping for " + localId);
|
182
|
IdentifierManager.getInstance().createMapping(localId, localId);
|
183
|
guid = IdentifierManager.getInstance().getGUID(accNum.getDocid(), rev);
|
184
|
}
|
185
|
|
186
|
// look up existing system metadata if it exists
|
187
|
Identifier identifier = new Identifier();
|
188
|
identifier.setValue(guid);
|
189
|
try {
|
190
|
logMetacat.debug("Using hazelcast to get system metadata");
|
191
|
sysMeta = HazelcastService.getInstance().getSystemMetadataMap().get(identifier);
|
192
|
// TODO: if this is the case, we could return here -- what else do we gain?
|
193
|
if (!updateExisting ) {
|
194
|
return sysMeta;
|
195
|
}
|
196
|
} catch (Exception e) {
|
197
|
logMetacat.debug("No system metadata found in hz: " + e.getMessage());
|
198
|
|
199
|
}
|
200
|
|
201
|
if (sysMeta == null) {
|
202
|
// create system metadata
|
203
|
sysMeta = new SystemMetadata();
|
204
|
sysMeta.setIdentifier(identifier);
|
205
|
sysMeta.setSerialVersion(BigInteger.valueOf(1));
|
206
|
sysMeta.setArchived(false);
|
207
|
}
|
208
|
|
209
|
// get additional docinfo
|
210
|
Hashtable<String, String> docInfo = ReplicationService.getDocumentInfoMap(localId);
|
211
|
// set the default object format
|
212
|
String doctype = docInfo.get("doctype");
|
213
|
ObjectFormatIdentifier fmtid = null;
|
214
|
|
215
|
// set the object format, fall back to defaults
|
216
|
if (doctype.trim().equals("BIN")) {
|
217
|
// we don't know much about this file (yet)
|
218
|
fmtid = ObjectFormatCache.getInstance().getFormat("application/octet-stream").getFormatId();
|
219
|
} else if (doctype.trim().equals("metadata")) {
|
220
|
// special ESRI FGDC format
|
221
|
fmtid = ObjectFormatCache.getInstance().getFormat("FGDC-STD-001-1998").getFormatId();
|
222
|
} else {
|
223
|
try {
|
224
|
// do we know the given format?
|
225
|
fmtid = ObjectFormatCache.getInstance().getFormat(doctype).getFormatId();
|
226
|
} catch (NotFound nfe) {
|
227
|
// format is not registered, use default
|
228
|
fmtid = ObjectFormatCache.getInstance().getFormat("text/plain").getFormatId();
|
229
|
}
|
230
|
}
|
231
|
|
232
|
sysMeta.setFormatId(fmtid);
|
233
|
logMetacat.debug("The ObjectFormat for " + localId + " is " + fmtid.getValue());
|
234
|
|
235
|
// for retrieving the actual object
|
236
|
InputStream inputStream = null;
|
237
|
inputStream = MetacatHandler.read(localId);
|
238
|
|
239
|
// create the checksum
|
240
|
String algorithm = PropertyService.getProperty("dataone.checksumAlgorithm.default");
|
241
|
Checksum checksum = ChecksumUtil.checksum(inputStream, algorithm);
|
242
|
logMetacat.debug("The checksum for " + localId + " is " + checksum.getValue());
|
243
|
sysMeta.setChecksum(checksum);
|
244
|
|
245
|
// set the size from file on disk, don't read bytes again
|
246
|
File fileOnDisk = getFileOnDisk(localId);
|
247
|
long fileSize = 0;
|
248
|
if (fileOnDisk.exists()) {
|
249
|
fileSize = fileOnDisk.length();
|
250
|
}
|
251
|
sysMeta.setSize(BigInteger.valueOf(fileSize));
|
252
|
|
253
|
// submitter
|
254
|
Subject submitter = new Subject();
|
255
|
submitter.setValue(docInfo.get("user_updated"));
|
256
|
sysMeta.setSubmitter(submitter);
|
257
|
|
258
|
// rights holder
|
259
|
Subject owner = new Subject();
|
260
|
owner.setValue(docInfo.get("user_owner"));
|
261
|
sysMeta.setRightsHolder(owner);
|
262
|
|
263
|
// dates
|
264
|
String createdDateString = docInfo.get("date_created");
|
265
|
String updatedDateString = docInfo.get("date_updated");
|
266
|
Date createdDate = DateTimeMarshaller.deserializeDateToUTC(createdDateString);
|
267
|
Date updatedDate = DateTimeMarshaller.deserializeDateToUTC(updatedDateString);
|
268
|
sysMeta.setDateUploaded(createdDate);
|
269
|
//sysMeta.setDateSysMetadataModified(updatedDate);
|
270
|
// use current datetime
|
271
|
sysMeta.setDateSysMetadataModified(Calendar.getInstance().getTime());
|
272
|
|
273
|
// set the revision history
|
274
|
String docidWithoutRev = accNum.getDocid();
|
275
|
Identifier obsoletedBy = null;
|
276
|
Identifier obsoletes = null;
|
277
|
Vector<Integer> revisions = DBUtil.getRevListFromRevisionTable(docidWithoutRev);
|
278
|
// ensure this ordering since processing depends on it
|
279
|
Collections.sort(revisions);
|
280
|
for (int existingRev: revisions) {
|
281
|
// use the docid+rev as the guid
|
282
|
String existingPid = docidWithoutRev + "." + existingRev;
|
283
|
try {
|
284
|
existingPid = IdentifierManager.getInstance().getGUID(docidWithoutRev, existingRev);
|
285
|
} catch (McdbDocNotFoundException mdfe) {
|
286
|
// we'll be defaulting to the local id
|
287
|
logMetacat.warn("could not locate guid when processing revision history for localId: " + localId);
|
288
|
}
|
289
|
if (existingRev < rev) {
|
290
|
// it's the old docid, until it's not
|
291
|
obsoletes = new Identifier();
|
292
|
obsoletes.setValue(existingPid);
|
293
|
}
|
294
|
if (existingRev > rev) {
|
295
|
// it's the newer docid
|
296
|
obsoletedBy = new Identifier();
|
297
|
obsoletedBy.setValue(existingPid);
|
298
|
// only want the version just after it
|
299
|
break;
|
300
|
}
|
301
|
}
|
302
|
// set them on our object
|
303
|
sysMeta.setObsoletedBy(obsoletedBy);
|
304
|
sysMeta.setObsoletes(obsoletes);
|
305
|
|
306
|
// update the system metadata for the object[s] we are revising
|
307
|
if (obsoletedBy != null) {
|
308
|
SystemMetadata obsoletedBySysMeta = null;
|
309
|
try {
|
310
|
//obsoletedBySysMeta = HazelcastService.getInstance().getSystemMetadataMap().get(obsoletedBy);
|
311
|
obsoletedBySysMeta = IdentifierManager.getInstance().getSystemMetadata(obsoletedBy.getValue());
|
312
|
} catch (McdbDocNotFoundException e) {
|
313
|
// ignore
|
314
|
}
|
315
|
if (obsoletedBySysMeta != null) {
|
316
|
obsoletedBySysMeta.setObsoletes(identifier);
|
317
|
obsoletedBySysMeta.setDateSysMetadataModified(Calendar.getInstance().getTime());
|
318
|
HazelcastService.getInstance().getSystemMetadataMap().put(obsoletedBy, obsoletedBySysMeta);
|
319
|
}
|
320
|
}
|
321
|
if (obsoletes != null) {
|
322
|
SystemMetadata obsoletesSysMeta = null;
|
323
|
try {
|
324
|
//obsoletesSysMeta = HazelcastService.getInstance().getSystemMetadataMap().get(obsoletes);
|
325
|
obsoletesSysMeta = IdentifierManager.getInstance().getSystemMetadata(obsoletes.getValue());
|
326
|
} catch (McdbDocNotFoundException e) {
|
327
|
// ignore
|
328
|
}
|
329
|
if (obsoletesSysMeta != null) {
|
330
|
obsoletesSysMeta.setObsoletedBy(identifier);
|
331
|
// DO NOT set archived to true -- it will have unintended consequences if the CN sees this.
|
332
|
//obsoletesSysMeta.setArchived(true);
|
333
|
obsoletesSysMeta.setDateSysMetadataModified(Calendar.getInstance().getTime());
|
334
|
HazelcastService.getInstance().getSystemMetadataMap().put(obsoletes, obsoletesSysMeta);
|
335
|
}
|
336
|
}
|
337
|
|
338
|
// look up the access control policy we have in metacat
|
339
|
AccessPolicy accessPolicy = IdentifierManager.getInstance().getAccessPolicy(guid);
|
340
|
try {
|
341
|
List<AccessRule> allowList = accessPolicy.getAllowList();
|
342
|
int listSize = allowList.size();
|
343
|
sysMeta.setAccessPolicy(accessPolicy);
|
344
|
|
345
|
} catch (NullPointerException npe) {
|
346
|
logMetacat.info("The allow list is empty, can't include an empty " +
|
347
|
"access policy in the system metadata for " + guid);
|
348
|
|
349
|
}
|
350
|
|
351
|
// authoritative node
|
352
|
NodeReference nr = new NodeReference();
|
353
|
nr.setValue(PropertyService.getProperty("dataone.nodeId"));
|
354
|
sysMeta.setOriginMemberNode(nr);
|
355
|
sysMeta.setAuthoritativeMemberNode(nr);
|
356
|
|
357
|
// Set a default replication policy
|
358
|
ReplicationPolicy rp = getDefaultReplicationPolicy();
|
359
|
if (rp != null) {
|
360
|
sysMeta.setReplicationPolicy(rp);
|
361
|
}
|
362
|
|
363
|
// further parse EML documents to get data object format,
|
364
|
// describes and describedBy information
|
365
|
if (fmtid == ObjectFormatCache.getInstance().getFormat(
|
366
|
"eml://ecoinformatics.org/eml-2.0.0").getFormatId()
|
367
|
|| fmtid == ObjectFormatCache.getInstance().getFormat(
|
368
|
"eml://ecoinformatics.org/eml-2.0.1").getFormatId()
|
369
|
|| fmtid == ObjectFormatCache.getInstance().getFormat(
|
370
|
"eml://ecoinformatics.org/eml-2.1.0").getFormatId()
|
371
|
|| fmtid == ObjectFormatCache.getInstance().getFormat(
|
372
|
"eml://ecoinformatics.org/eml-2.1.1").getFormatId()) {
|
373
|
|
374
|
try {
|
375
|
|
376
|
// get it again to parse the document
|
377
|
logMetacat.debug("Re-reading document inputStream");
|
378
|
inputStream = MetacatHandler.read(localId);
|
379
|
|
380
|
DataoneEMLParser emlParser = DataoneEMLParser.getInstance();
|
381
|
EMLDocument emlDocument = emlParser.parseDocument(inputStream);
|
382
|
|
383
|
// iterate through the data objects in the EML doc and add sysmeta
|
384
|
logMetacat.debug("In createSystemMetadata() the number of data "
|
385
|
+ "entities is: "
|
386
|
+ emlDocument.distributionMetadata);
|
387
|
|
388
|
// for generating the ORE map
|
389
|
Map<Identifier, List<Identifier>> idMap = new HashMap<Identifier, List<Identifier>>();
|
390
|
List<Identifier> dataIds = new ArrayList<Identifier>();
|
391
|
|
392
|
// iterate through data objects described by the EML
|
393
|
if (emlDocument.distributionMetadata != null) {
|
394
|
for (int j = 0; j < emlDocument.distributionMetadata.size(); j++) {
|
395
|
|
396
|
DistributionMetadata distMetadata = emlDocument.distributionMetadata.elementAt(j);
|
397
|
String dataDocUrl = distMetadata.url;
|
398
|
String dataDocMimeType = distMetadata.mimeType;
|
399
|
// default to binary
|
400
|
if (dataDocMimeType == null) {
|
401
|
dataDocMimeType = "application/octet-stream";
|
402
|
}
|
403
|
|
404
|
// process the data
|
405
|
boolean remoteData = false;
|
406
|
String dataDocLocalId = null;
|
407
|
Identifier dataGuid = new Identifier();
|
408
|
|
409
|
// handle ecogrid, or downloadable data
|
410
|
String ecogridPrefix = "ecogrid://knb/";
|
411
|
if (dataDocUrl.trim().startsWith(ecogridPrefix)) {
|
412
|
dataDocLocalId = dataDocUrl.substring(dataDocUrl.indexOf(ecogridPrefix) + ecogridPrefix.length());
|
413
|
} else {
|
414
|
// should we try downloading the remote data?
|
415
|
if (downloadData) {
|
416
|
InputStream dataObject = null;
|
417
|
try {
|
418
|
// download the data from the URL
|
419
|
URL dataURL = new URL(dataDocUrl);
|
420
|
URLConnection dataConnection = dataURL.openConnection();
|
421
|
|
422
|
// default is to download the data
|
423
|
dataObject = dataConnection.getInputStream();
|
424
|
|
425
|
String detectedContentType = dataConnection.getContentType();
|
426
|
logMetacat.info("Detected content type: " + detectedContentType);
|
427
|
|
428
|
if (detectedContentType != null) {
|
429
|
// seems to be HTML from the remote location
|
430
|
if (detectedContentType.contains("html")) {
|
431
|
// if we are not expecting it, we skip it
|
432
|
if (!dataDocMimeType.contains("html")) {
|
433
|
// set to null so we don't download it
|
434
|
dataObject = null;
|
435
|
logMetacat.warn("Skipping remote resource, unexpected HTML content type at: " + dataDocUrl);
|
436
|
}
|
437
|
}
|
438
|
|
439
|
} else {
|
440
|
// if we don't know what it is, should we skip it?
|
441
|
dataObject = null;
|
442
|
logMetacat.warn("Skipping remote resource, unknown content type at: " + dataDocUrl);
|
443
|
}
|
444
|
|
445
|
} catch (Exception e) {
|
446
|
// error with the download
|
447
|
logMetacat.warn("Error downloading remote data. " + e.getMessage());
|
448
|
}
|
449
|
|
450
|
if (dataObject != null) {
|
451
|
// create the local version of it
|
452
|
dataDocLocalId = DocumentUtil.generateDocumentId(1);
|
453
|
IdentifierManager.getInstance().createMapping(dataDocLocalId, dataDocLocalId);
|
454
|
dataGuid.setValue(dataDocLocalId);
|
455
|
|
456
|
// save it locally
|
457
|
Session session = new Session();
|
458
|
session.setSubject(submitter);
|
459
|
MockHttpServletRequest request = new MockHttpServletRequest(null, null, null);
|
460
|
MNodeService.getInstance(request).insertDataObject(dataObject, dataGuid, session);
|
461
|
|
462
|
remoteData = true;
|
463
|
}
|
464
|
}
|
465
|
|
466
|
}
|
467
|
|
468
|
logMetacat.debug("Data local ID: " + dataDocLocalId);
|
469
|
logMetacat.debug("Data URL : " + dataDocUrl);
|
470
|
logMetacat.debug("Data mime : " + dataDocMimeType);
|
471
|
|
472
|
// check for valid docid.rev
|
473
|
String dataDocid = null;
|
474
|
int dataRev = 0;
|
475
|
if (dataDocLocalId != null) {
|
476
|
// look up the guid for the data
|
477
|
try {
|
478
|
dataDocid = DocumentUtil.getSmartDocId(dataDocLocalId);
|
479
|
dataRev = DocumentUtil.getRevisionFromAccessionNumber(dataDocLocalId);
|
480
|
} catch (Exception e) {
|
481
|
logMetacat.warn(e.getClass().getName() + " - Problem parsing accession number for: " + dataDocLocalId + ". Message: " + e.getMessage());
|
482
|
dataDocLocalId = null;
|
483
|
}
|
484
|
}
|
485
|
|
486
|
// now we have a local id for the data
|
487
|
if (dataDocLocalId != null) {
|
488
|
|
489
|
// check if data system metadata exists already
|
490
|
SystemMetadata dataSysMeta = null;
|
491
|
String dataGuidString = null;
|
492
|
try {
|
493
|
// look for the identifier
|
494
|
dataGuidString = IdentifierManager.getInstance().getGUID(dataDocid, dataRev);
|
495
|
// set it
|
496
|
dataGuid.setValue(dataGuidString);
|
497
|
// look up the system metadata
|
498
|
try {
|
499
|
dataSysMeta = HazelcastService.getInstance().getSystemMetadataMap().get(dataGuid);
|
500
|
} catch (Exception e) {
|
501
|
// probably not in the system
|
502
|
dataSysMeta = null;
|
503
|
}
|
504
|
//dataSysMeta = IdentifierManager.getInstance().getSystemMetadata(dataGuidString);
|
505
|
} catch (McdbDocNotFoundException nf) {
|
506
|
// we didn't find it
|
507
|
dataSysMeta = null;
|
508
|
}
|
509
|
|
510
|
// we'll have to generate it
|
511
|
if (dataSysMeta == null) {
|
512
|
// System metadata for data doesn't exist yet, so create it
|
513
|
logMetacat.debug("No exisiting SystemMetdata found, creating for: " + dataDocLocalId);
|
514
|
dataSysMeta = createSystemMetadata(dataDocLocalId, includeORE, false);
|
515
|
|
516
|
// now look it up again
|
517
|
dataGuidString = IdentifierManager.getInstance().getGUID(dataDocid, dataRev);
|
518
|
|
519
|
// set the guid
|
520
|
dataGuid.setValue(dataGuidString);
|
521
|
|
522
|
// inherit access rules from metadata, if we don't have our own
|
523
|
if (remoteData) {
|
524
|
dataSysMeta.setAccessPolicy(sysMeta.getAccessPolicy());
|
525
|
// TODO: use access rules defined in EML, per data file
|
526
|
}
|
527
|
|
528
|
}
|
529
|
|
530
|
// set object format for the data file
|
531
|
logMetacat.debug("Updating system metadata for " + dataGuid.getValue() + " to " + dataDocMimeType);
|
532
|
ObjectFormatIdentifier fmt = null;
|
533
|
try {
|
534
|
fmt = ObjectFormatCache.getInstance().getFormat(dataDocMimeType).getFormatId();
|
535
|
} catch (NotFound nfe) {
|
536
|
logMetacat.debug("Couldn't find format identifier for: "
|
537
|
+ dataDocMimeType
|
538
|
+ ". Setting it to application/octet-stream.");
|
539
|
fmt = new ObjectFormatIdentifier();
|
540
|
fmt.setValue("application/octet-stream");
|
541
|
}
|
542
|
dataSysMeta.setFormatId(fmt);
|
543
|
|
544
|
// update the values
|
545
|
HazelcastService.getInstance().getSystemMetadataMap().put(dataSysMeta.getIdentifier(), dataSysMeta);
|
546
|
|
547
|
// reindex data file if need it.
|
548
|
logMetacat.debug("do we need to reindex guid "+dataGuid.getValue()+"~~~~~~~~~~~~~~~~~~~~~~~~~~~~~?"+indexDataFile);
|
549
|
if(indexDataFile) {
|
550
|
reindexDataFile(dataSysMeta.getIdentifier(), dataSysMeta);
|
551
|
}
|
552
|
|
553
|
// include as part of the ORE package
|
554
|
dataIds.add(dataGuid);
|
555
|
|
556
|
} // end if (EML package)
|
557
|
|
558
|
} // end for (data entities)
|
559
|
|
560
|
} // data entities not null
|
561
|
|
562
|
// ORE map
|
563
|
if (includeORE) {
|
564
|
// can we generate them?
|
565
|
if (!dataIds.isEmpty()) {
|
566
|
// it doesn't exist in the system?
|
567
|
if (!oreExistsFor(sysMeta.getIdentifier())) {
|
568
|
|
569
|
// generate the ORE map for this datapackage
|
570
|
Identifier resourceMapId = new Identifier();
|
571
|
// use the local id, not the guid in case we have DOIs for them already
|
572
|
resourceMapId.setValue(RESOURCE_MAP_PREFIX + localId);
|
573
|
idMap.put(sysMeta.getIdentifier(), dataIds);
|
574
|
ResourceMap rm = ResourceMapFactory.getInstance().createResourceMap(resourceMapId, idMap);
|
575
|
String resourceMapXML = ResourceMapFactory.getInstance().serializeResourceMap(rm);
|
576
|
// copy most of the same system metadata as the packaging metadata
|
577
|
SystemMetadata resourceMapSysMeta = new SystemMetadata();
|
578
|
BeanUtils.copyProperties(resourceMapSysMeta, sysMeta);
|
579
|
resourceMapSysMeta.setIdentifier(resourceMapId);
|
580
|
Checksum oreChecksum = ChecksumUtil.checksum(IOUtils.toInputStream(resourceMapXML, MetaCatServlet.DEFAULT_ENCODING), algorithm);
|
581
|
resourceMapSysMeta.setChecksum(oreChecksum);
|
582
|
ObjectFormatIdentifier formatId = ObjectFormatCache.getInstance().getFormat("http://www.openarchives.org/ore/terms").getFormatId();
|
583
|
resourceMapSysMeta.setFormatId(formatId);
|
584
|
resourceMapSysMeta.setSize(BigInteger.valueOf(sizeOfStream(IOUtils.toInputStream(resourceMapXML, MetaCatServlet.DEFAULT_ENCODING))));
|
585
|
|
586
|
// set the revision graph
|
587
|
resourceMapSysMeta.setObsoletes(null);
|
588
|
resourceMapSysMeta.setObsoletedBy(null);
|
589
|
// look up the resource map that this one obsoletes
|
590
|
if (sysMeta.getObsoletes() != null) {
|
591
|
// use the localId in case we have a DOI
|
592
|
String obsoletesLocalId = IdentifierManager.getInstance().getLocalId(sysMeta.getObsoletes().getValue());
|
593
|
Identifier resourceMapObsoletes = new Identifier();
|
594
|
resourceMapObsoletes.setValue(RESOURCE_MAP_PREFIX + obsoletesLocalId );
|
595
|
resourceMapSysMeta.setObsoletes(resourceMapObsoletes);
|
596
|
SystemMetadata resourceMapObsoletesSystemMetadata = HazelcastService.getInstance().getSystemMetadataMap().get(resourceMapObsoletes);
|
597
|
if (resourceMapObsoletesSystemMetadata != null) {
|
598
|
resourceMapObsoletesSystemMetadata.setObsoletedBy(resourceMapId);
|
599
|
resourceMapObsoletesSystemMetadata.setArchived(true);
|
600
|
HazelcastService.getInstance().getSystemMetadataMap().put(resourceMapObsoletes, resourceMapObsoletesSystemMetadata);
|
601
|
}
|
602
|
}
|
603
|
// look up the resource map that this one is obsoletedBy
|
604
|
if (sysMeta.getObsoletedBy() != null) {
|
605
|
// use the localId in case we have a DOI
|
606
|
String obsoletedByLocalId = IdentifierManager.getInstance().getLocalId(sysMeta.getObsoletedBy().getValue());
|
607
|
Identifier resourceMapObsoletedBy = new Identifier();
|
608
|
resourceMapObsoletedBy.setValue(RESOURCE_MAP_PREFIX + obsoletedByLocalId);
|
609
|
resourceMapSysMeta.setObsoletedBy(resourceMapObsoletedBy);
|
610
|
resourceMapSysMeta.setArchived(true);
|
611
|
SystemMetadata resourceMapObsoletedBySystemMetadata = HazelcastService.getInstance().getSystemMetadataMap().get(resourceMapObsoletedBy);
|
612
|
if (resourceMapObsoletedBySystemMetadata != null) {
|
613
|
resourceMapObsoletedBySystemMetadata.setObsoletes(resourceMapId);
|
614
|
HazelcastService.getInstance().getSystemMetadataMap().put(resourceMapObsoletedBy, resourceMapObsoletedBySystemMetadata);
|
615
|
}
|
616
|
}
|
617
|
|
618
|
// save it locally, if it doesn't already exist
|
619
|
if (!IdentifierManager.getInstance().identifierExists(resourceMapId.getValue())) {
|
620
|
Session session = new Session();
|
621
|
session.setSubject(submitter);
|
622
|
MockHttpServletRequest request = new MockHttpServletRequest(null, null, null);
|
623
|
MNodeService.getInstance(request).insertDataObject(IOUtils.toInputStream(resourceMapXML, MetaCatServlet.DEFAULT_ENCODING), resourceMapId, session);
|
624
|
MNodeService.getInstance(request).insertSystemMetadata(resourceMapSysMeta);
|
625
|
logMetacat.info("Inserted ORE package: " + resourceMapId.getValue());
|
626
|
}
|
627
|
}
|
628
|
}
|
629
|
}
|
630
|
|
631
|
} catch (ParserConfigurationException pce) {
|
632
|
logMetacat.debug("There was a problem parsing the EML document. "
|
633
|
+ "The error message was: " + pce.getMessage());
|
634
|
|
635
|
} catch (SAXException saxe) {
|
636
|
logMetacat.debug("There was a problem traversing the EML document. "
|
637
|
+ "The error message was: " + saxe.getMessage());
|
638
|
|
639
|
} catch (XPathExpressionException xpee) {
|
640
|
logMetacat.debug("There was a problem searching the EML document. "
|
641
|
+ "The error message was: " + xpee.getMessage());
|
642
|
} catch (Exception e) {
|
643
|
logMetacat.debug("There was a problem creating System Metadata. "
|
644
|
+ "The error message was: " + e.getMessage());
|
645
|
e.printStackTrace();
|
646
|
} // end try()
|
647
|
|
648
|
} // end if()
|
649
|
|
650
|
return sysMeta;
|
651
|
}
|
652
|
|
653
|
/*
|
654
|
* Re-index the data file since the access rule was changed during the inserting of the eml document.
|
655
|
* (During first time to index the data file in Metacat API, the eml hasn't been inserted)
|
656
|
*/
|
657
|
private static void reindexDataFile(Identifier id, SystemMetadata sysmeta) {
|
658
|
try {
|
659
|
logMetacat.debug("@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@ reindex"+id.getValue());
|
660
|
if(sysmeta != null) {
|
661
|
if(!sysmeta.getArchived()) {
|
662
|
//set the archive to true to remove index.
|
663
|
sysmeta.setArchived(true);
|
664
|
MetacatSolrIndex.getInstance().submit(id, sysmeta, null, true);
|
665
|
//re-insert the index
|
666
|
sysmeta.setArchived(false);
|
667
|
MetacatSolrIndex.getInstance().submit(id, sysmeta, null, true);
|
668
|
} else {
|
669
|
MetacatSolrIndex.getInstance().submit(id, sysmeta, null, true);
|
670
|
}
|
671
|
}
|
672
|
|
673
|
} catch (Exception e) {
|
674
|
// TODO Auto-generated catch block
|
675
|
logMetacat.warn("Can't reindex the data object "+id.getValue()+" since "+e.getMessage());
|
676
|
//e.printStackTrace();
|
677
|
}
|
678
|
}
|
679
|
|
680
|
/**
|
681
|
* Checks for potential ORE object existence
|
682
|
* @param identifier
|
683
|
* @return
|
684
|
*/
|
685
|
public static boolean oreExistsFor(Identifier identifier) {
|
686
|
MockHttpServletRequest request = new MockHttpServletRequest(null, null, null);
|
687
|
List<Identifier> ids = MNodeService.getInstance(request).lookupOreFor(identifier, true);
|
688
|
return (ids != null && ids.size() > 0);
|
689
|
}
|
690
|
|
691
|
/**
|
692
|
* Generate SystemMetadata for any object in the object store that does
|
693
|
* not already have it. SystemMetadata documents themselves, are, of course,
|
694
|
* exempt. This is a utility method for migration of existing object
|
695
|
* stores to DataONE where SystemMetadata is required for all objects.
|
696
|
* @param idList
|
697
|
* @param includeOre
|
698
|
* @param downloadData
|
699
|
* @throws PropertyNotFoundException
|
700
|
* @throws NoSuchAlgorithmException
|
701
|
* @throws AccessionNumberException
|
702
|
* @throws SQLException
|
703
|
* @throws SAXException
|
704
|
* @throws HandlerException
|
705
|
* @throws JiBXException
|
706
|
* @throws BaseException
|
707
|
* @throws ParseLSIDException
|
708
|
* @throws InsufficientKarmaException
|
709
|
* @throws ClassNotFoundException
|
710
|
* @throws IOException
|
711
|
* @throws McdbException
|
712
|
* @throws AccessException
|
713
|
* @throws AccessControlException
|
714
|
*/
|
715
|
public static void generateSystemMetadata(List<String> idList, boolean includeOre, boolean downloadData)
|
716
|
throws PropertyNotFoundException, NoSuchAlgorithmException, AccessionNumberException, SQLException, AccessControlException, AccessException, McdbException, IOException, ClassNotFoundException, InsufficientKarmaException, ParseLSIDException, BaseException, JiBXException, HandlerException, SAXException
|
717
|
{
|
718
|
|
719
|
for (String localId : idList) {
|
720
|
logMetacat.debug("Creating SystemMetadata for localId " + localId);
|
721
|
logMetacat.trace("METRICS:\tGENERATE_SYSTEM_METADATA:\tBEGIN:\tLOCALID:\t" + localId);
|
722
|
|
723
|
SystemMetadata sm = null;
|
724
|
|
725
|
//generate required system metadata fields from the document
|
726
|
try {
|
727
|
sm = SystemMetadataFactory.createSystemMetadata(localId, includeOre, downloadData);
|
728
|
} catch (Exception e) {
|
729
|
logMetacat.error("Could not create/process system metadata for docid: " + localId, e);
|
730
|
continue;
|
731
|
}
|
732
|
|
733
|
//insert the systemmetadata object or just update it as needed
|
734
|
IdentifierManager.getInstance().insertOrUpdateSystemMetadata(sm);
|
735
|
logMetacat.info("Generated or Updated SystemMetadata for " + localId);
|
736
|
|
737
|
logMetacat.trace("METRICS:\tGENERATE_SYSTEM_METADATA:\tEND:\tLOCALID:\t" + localId);
|
738
|
|
739
|
}
|
740
|
logMetacat.info("done generating system metadata for given list");
|
741
|
}
|
742
|
|
743
|
/**
|
744
|
* Find the size (in bytes) of a stream. Note: This needs to refactored out
|
745
|
* of MetacatHandler and into a utility when stream i/o in Metacat is
|
746
|
* evaluated.
|
747
|
*
|
748
|
* @param is The InputStream of bytes
|
749
|
*
|
750
|
* @return size The size in bytes of the input stream as a long
|
751
|
*
|
752
|
* @throws IOException
|
753
|
*/
|
754
|
public static long sizeOfStream(InputStream is) throws IOException {
|
755
|
|
756
|
long size = 0;
|
757
|
byte[] b = new byte[1024];
|
758
|
int numread = is.read(b, 0, 1024);
|
759
|
while (numread != -1) {
|
760
|
size += numread;
|
761
|
numread = is.read(b, 0, 1024);
|
762
|
}
|
763
|
return size;
|
764
|
|
765
|
}
|
766
|
|
767
|
private static File getFileOnDisk(String docid) throws McdbException, PropertyNotFoundException {
|
768
|
|
769
|
DocumentImpl doc = new DocumentImpl(docid, false);
|
770
|
String filepath = null;
|
771
|
String filename = null;
|
772
|
|
773
|
// deal with data or metadata cases
|
774
|
if (doc.getRootNodeID() == 0) {
|
775
|
// this is a data file
|
776
|
filepath = PropertyService.getProperty("application.datafilepath");
|
777
|
} else {
|
778
|
filepath = PropertyService.getProperty("application.documentfilepath");
|
779
|
}
|
780
|
// ensure it is a directory path
|
781
|
if (!(filepath.endsWith("/"))) {
|
782
|
filepath += "/";
|
783
|
}
|
784
|
filename = filepath + docid;
|
785
|
File documentFile = new File(filename);
|
786
|
|
787
|
return documentFile;
|
788
|
}
|
789
|
|
790
|
/**
|
791
|
* Create a default ReplicationPolicy by reading properties from metacat's configuration
|
792
|
* and using those defaults. If the numReplicas property is not found, malformed, or less
|
793
|
* than or equal to zero, no policy needs to be set, so return null.
|
794
|
* @return ReplicationPolicy, or null if no replication policy is needed
|
795
|
*/
|
796
|
private static ReplicationPolicy getDefaultReplicationPolicy() {
|
797
|
ReplicationPolicy rp = null;
|
798
|
int numReplicas = -1;
|
799
|
try {
|
800
|
numReplicas = new Integer(PropertyService.getProperty("dataone.replicationpolicy.default.numreplicas"));
|
801
|
} catch (NumberFormatException e) {
|
802
|
// The property is not a valid integer, so return a null policy
|
803
|
return null;
|
804
|
} catch (PropertyNotFoundException e) {
|
805
|
// The property is not found, so return a null policy
|
806
|
return null;
|
807
|
}
|
808
|
|
809
|
if (numReplicas > 0) {
|
810
|
rp = new ReplicationPolicy();
|
811
|
rp.setReplicationAllowed(true);
|
812
|
rp.setNumberReplicas(numReplicas);
|
813
|
try {
|
814
|
String preferredNodeList = PropertyService.getProperty("dataone.replicationpolicy.default.preferredNodeList");
|
815
|
if (preferredNodeList != null) {
|
816
|
List<NodeReference> pNodes = extractNodeReferences(preferredNodeList);
|
817
|
if (pNodes != null && !pNodes.isEmpty()) {
|
818
|
rp.setPreferredMemberNodeList(pNodes);
|
819
|
}
|
820
|
}
|
821
|
} catch (PropertyNotFoundException e) {
|
822
|
// No preferred list found in properties, so just ignore it; no action needed
|
823
|
}
|
824
|
try {
|
825
|
String blockedNodeList = PropertyService.getProperty("dataone.replicationpolicy.default.blockedNodeList");
|
826
|
if (blockedNodeList != null) {
|
827
|
List<NodeReference> bNodes = extractNodeReferences(blockedNodeList);
|
828
|
if (bNodes != null && !bNodes.isEmpty()) {
|
829
|
rp.setBlockedMemberNodeList(bNodes);
|
830
|
}
|
831
|
}
|
832
|
} catch (PropertyNotFoundException e) {
|
833
|
// No blocked list found in properties, so just ignore it; no action needed
|
834
|
}
|
835
|
}
|
836
|
return rp;
|
837
|
}
|
838
|
|
839
|
/**
|
840
|
* Extract a List of NodeReferences from a String listing the node identifiers where
|
841
|
* each identifier is separated by whitespace, comma, or semicolon characters.
|
842
|
* @param nodeString the string containing the list of nodes
|
843
|
* @return the List of NodeReference objects parsed from the input string
|
844
|
*/
|
845
|
private static List<NodeReference> extractNodeReferences(String nodeString) {
|
846
|
List<NodeReference> nodeList = new ArrayList<NodeReference>();
|
847
|
String[] result = nodeString.split("[,;\\s]");
|
848
|
for (String r : result) {
|
849
|
if (r != null && r.length() > 0) {
|
850
|
NodeReference noderef = new NodeReference();
|
851
|
noderef.setValue(r);
|
852
|
nodeList.add(noderef);
|
853
|
}
|
854
|
}
|
855
|
return nodeList;
|
856
|
}
|
857
|
}
|