Project

General

Profile

metacat / README @ 3523

1
#
2
#  '$RCSfile$'
3
#   '$Author: tao $'
4
# '$Revision: 3376 $'
5
#     '$Date: 2007-08-27 18:02:39 -0700 (Mon, 27 Aug 2007) $'
6
#
7

    
8
Metacat: XML Metadata and Data Management System
9
------------------------------------------------
10

    
11
Version: 1.7.1 Release
12

    
13
Send feedback and bugs to: metacat-dev@ecoinformatics.org
14
                           http://bugzilla.ecoinformatics.org
15

    
16
Contributors: Matt Jones (jones@nceas.ucsb.edu)
17
              Chad Berkley (berkley@nceas.ucsb.edu)
18
              Jing Tao (tao@nceas.ucsb.edu)
19
              Jivka Bojilova (bojilova@nceas.ucsb.edu)
20
              Dan Higgins (higgins@nceas.ucsb.edu)
21
              Saurabh Garg (sgarg@nceas.ucsb.edu)
22
              Duane Costa (dcosta@lternet.edu)
23
              Veronique Connolly (connolly@nceas.ucsb.edu)
24
              John Harris (harris@nceas.ucsb.edu)
25
              Callie Bowdish (bowdish@ecoinformatics.org)
26
              Will Tyburczy (tyburczy@ecoinformatics.org)
27
              Matthew Perry (perry@nceas.ucsb.edu)
28
              Chad Burt (underbluewaters@gmail.com)
29

    
30
Patch contributors:
31
              Andrea Chadden (chadden@nceas.ucsb.edu)
32
              Chris Jones (cjones@msi.ucsb.edu)
33
              Johnoel Ancheta (johnoel@hawaii.edu)
34
              Owen Jones (owen.jones@imperial.ac.uk)
35

    
36
Metacat is a flexible database storage system for XML formatted
37
documents and associated binary files. Metacat models the structure of the
38
XML document, therefore allowing storage of XML documents with arbitrary
39
schemas. Metacat supports several databases on the back-end, including Oracle,
40
PostgreSQL and SQL Server (limited).  The server supports XML document
41
storage, retrieval, query, validation, and transformation (using the XSLT
42
stylesheet language).  In addition, there is a mechanism for replicating the
43
database to other metacat instances for backup and load balancing purposes.
44
Metacat currently uses an LDAP database for authenticating users, but was
45
written to accomodate other authentication services as well.
46

    
47
Installation
48
------------
49
See the file "docs/user/metacatinstall.html" for detailed instructions.
50

    
51
Release Notes for 1.7.1:
52
------------------------
53
The major focus of this release is the improvement of query performance. The 
54
selection query and access control query were modified to make SQL query executed 
55
more efficiently.  In order to get better performance, query cache mechanism was 
56
introduced to Metacat. Currently cached query mechanism only supports public user.
57

    
58
Bug Fixes:
59
* Metacat Performace: Rewrite the xml_nodes queries (Bug 2155)
60
* Metacat Performace: Rewrite the xml_access part of the query (Bug 2557)
61
* Query cache mechanism (Bug 2905)
62
* Metacat Performance: updates from Morpho of data packages are taking 
63
   longer than five minutes (Bug 2805)
64
 * Metacat Performance: Optimize Postgres and Tomcat (Bug 2157)
65
* Metacat Performace: Upgrading hardware setup (Bug 2175)
66
* Metacat Performace: Add/drop indices on key columns (Bug 2153)
67

    
68
Compatibility Notices:
69
* Tomcat 3 and 4 are no longer tested or supported. Users are 
70
  highly encouraged to upgrade to Tomcat 5.5
71

    
72
Release Notes for 1.7.0:
73
------------------------
74
This release features only a few minor changes to the core Metacat engine. 
75
The major focus of this release is the addition of the spatial functionality.
76
The geographic coverage of metadata documents can now be cached in a GIS-
77
accessible format and published via web mapping services and an online
78
interactive map. 
79

    
80
New Features:
81
* Three new open source libraries have been integrated with Metacat:
82
  - Geoserver:  A standards-compliant web map server to publish
83
                geographic data to the web.
84
                (http://docs.codehaus.org/display/GEOS/Home)
85
  - GeoTools:   A java-based GIS toolkit to programatically
86
                manipulate spatial datasets.
87
                (http://geotools.codehaus.org/)
88
  - Mapbuilder: A javascript application to provide an interactive
89
                web map front-end to Geoserver.
90
                (http://docs.codehaus.org/display/MAP/Home)
91
* Spatial Caching/Indexing mechanism; documents with geographic
92
  coverages will be added to the spatial cache.
93
* Spatial Query action in metacat servlet. 
94
* Interactive Web map to visualize and query the spatial 
95
  distribution of metadata documents.
96
* Experimental python, ruby and php client libraries to interface with metacat.
97
* Preliminary process in place for web based configuration of metacat.
98

    
99
Compatibility Notices:
100
* Tomcat 3 and 4 are no longer tested or supported. Users are 
101
  highly encouraged to upgrade to Tomcat 5.5
102

    
103
Bug Fixes:
104
* Metacat administrators are able to modify and delete documents 
105
  created by other users.
106
* Fixed problems with indexing XML attributes, see bug 2469
107

    
108

    
109
Release Notes for 1.6.0:
110
------------------------
111

    
112
This release comes with more performance improvements, initial support for 
113
LSIDs (Life Sciences Identifiers) and FGDC standard, more controls for 
114
Metacat administrators and bug fixes. Below is the list of bug fixes and
115
performance improvements.
116

    
117
Bug Fixes: 
118
* Improved the indexing algorithm. Earlier indexing failed for some documents 
119
  randomly and used to take a lot of time to complete for others. The 
120
  code for indexing was changed to take care of both these problems.
121
* Earlier, xmlns prefix was used to determine the namespace of the document. 
122
  Now schemaLocation is used instead xmlns prefix to determine the namespace 
123
  of the document as that is a better indicator of document type. 
124
* INSTR was used in some SQL statements and it is not supported by Postgres. 
125
  So SQL statements were modified so that INSTR is not used.
126
* Replication was changed to include replication of deleted documents also.
127
* Fixed bug in squery which didnt allow use of not-contains 
128

    
129
New Features:
130
* Added FGDC schema and XSLT so that Metacat can handle FGDC standard
131
* Added support for LSID. This includes modifying the stylesheets to 
132
  display the LSIDs, modifying the build to include a new target 
133
  'install-ldap'.
134
* Added following access control levels to Metacat:
135
    1. Administrators - users who can do the following:
136
       a. Run replication
137
       b. Delete any document
138
       c. Modify any document
139
       d. run indexing for any document
140
       e. check the logs
141
    2. allowedUsers - User who are allowed to submit
142
    3. deniedUsers - Users not allowed to submit
143
* Added a new skin for Ecological Society of America. 
144
* Created an Advanced search servlet which can be used from the web. 
145
* Various connections have been modified to be secure. e.g. connection between 
146
ldaps is made secure now, replication is done over secure channels.  
147

    
148
Performance:
149
* Reduced size of xml_nodes by creating a new table for holding nodes from 
150
  deleted documents and old revisions of the documents. 
151
* Provided a new variable in metacat.properties where more frequently searched 
152
  for paths can be specified. Seperate indexes are created for these paths 
153
  - so that search is faster.
154
* Included log4j for logging and debuging to prevent wasting of time in printing log statements. 
155

    
156
Release Notes for 1.5.0:
157
------------------------
158

    
159
This release comes with performace improvement, user interface enhancements,
160
and bug fixes. Below is the list of bug fixes and performance improvements. 
161

    
162
Bug Fixes:
163
* Modify property values of harvester registration servlets to match the
164
  servlet-mapping URL values in web.xml. The old values used the servlet
165
  class names. This worked in Tomcat 4 but seems to break in Tomcat 5 on
166
  Windows. The new values use the servlet-mapping URL values. This should
167
  work in both Tomcat 4 and Tomcat 5. (Bug 162)
168
* Re-implement logic to prune old log entries from the HARVEST_LOG and
169
  HARVEST_DETAIL_LOG tables. The old logic caused integrity constraint
170
  violations in the database because it tried to delete parent records from
171
  HARVEST_LOG prior to deleting child records from
172
  HARVEST_DETAIL_LOG. (Bug 162)
173
* In data registry, minor bugs fixed like the error page shows the link back
174
  to the form when error page was generated because of a document read
175
  request from search results page,  the successful delete button leading
176
  to a blank screen and the old ACL overwritten by the registry.
177
  (Bugs 1307, 1310, 1322, 1344, 1551)
178
* Changed the code to support insertion of following characters: ', `,
179
  charcters in the Numeric Character Reference format (&#xyz;) and
180
  characters like µ. Characters like µ will be converted to and stored in
181
  Numeric Character Reference format(&#xyz;). They will also be returned
182
  in Numeric Character Reference and it is upto the client to convert them
183
  and display them accordingly in the application. Partial fix provided by
184
  Johnoel Ancheta. (Bug 1538,1711)
185
* Spatial search failed on Oracle because of invalid entries made by
186
  some documents in the nodedata column of xml_nodes for paths involving
187
  'eastBoundingCoordinate', 'westBoundingCoordinate',
188
  'northBoundingCoordinate', 'southBoundingCoordinate'. A new column
189
  nodedatanumerical was added to xml_nodes table such that all the numerical
190
  nodedata is stored in this column and this column is now used for spatial
191
  search. (Bug 1703, 1718)
192
* In the default skin, clicking on the keywords on the home page did not
193
  result is a search being done. Fix provided by Owen Jones. (Bug 1768)
194
* Metacat generated a success message even when a document which didnt
195
  exsist was requested for deletion. This has been fixed. (Bug 1850)
196
* "packagetype" in xml-relation table was entered as eml200 namespace when
197
  an eml201 document was inserted. This is fixed now so that "packagetype"
198
  in xml-relation table points to eml201 namespace. (Bug 1979)
199
* System_id in xml_catalog pointed to http://knb.ecoinformatics.org/knb/
200
  earlier. Fixed such that it points to the locally installed metacat url. 
201
  (Bug 1986)
202
* Changes to the Query subsystem fix bugs that prevented attributes from 
203
  being expressed solely in the xpath query statement and the returnfield 
204
  values. For instance, a query URL may now include search strings like 
205
  '@packageId=sbclter%25' and return strings like 'returnfield=@packageId'.
206
  Previously, the attribute had to be appended to an element: 
207
  '/eml/@packageId=sbclter%25'. These fixes change DBQuery.java, 
208
  QuerySpecification.java, and QueryTerm.java (Bug 2052)
209
* Search query failed in case of Oracle if number of documents in the result
210
  were more than 1000. This has been fixed. (Bug 2000)
211
* Removed any occurence of enum which is now a keyword in Java 1.5
212

    
213

    
214
Enhancements:
215
* Major improvements done in user interface for the data registeries and
216
  various other skins. Improvements done based on suggestions from Mark
217
  Stromberg, Laura Downey and others. Improvements also done in resultset.xsl,
218
  the default skin and ldap templates. (Bug 1948)
219
* Coded added such that administrator can delete documents irrespective of
220
  who the document belongs to.
221
* Performance impovement done in searching of datasets. A part of the search
222
  time was taken up because of generation of resultset after the search had be
223
  done. So this was the sequence of events in earlier metacat
224
  1) Search for the given text
225
  2) A list of docids generated
226
  3) For each of the docids
227
     3a) Use xml_index and xml_nodes table to generate the string descibing
228
         the document including the returnfields requested in the search
229
  4) Add all the strings from step 3a to send back the resultant
230
     document. Here a decent amount of time was being taken by step 3a. 
231
  The algorithm is now modified by addition of two tables xml_queryresult and
232
  xml_returnfields and a user defined parameter xml_returnfield_count. The
233
  new algorithm works as follows:
234
  1) Search for the given text
235
  2) A list of docids is generated
236
  3) For the given set of return fields generate a unique string and check
237
     if that string exists in xml_returnfields
238
     3a) If string does not exist in xml_returnfield, then enter a new
239
         record usage_count as 1 and returnfield_string as the unique string
240
         generated above.
241
     3b) Else if the string exists, increment usage_count
242
  4) Given the docids from step 2 and the id of the returnfield record from
243
     step 3, query for any docids that already exist in xml_queryresult.
244
     For the docids that do exist, get the queryresult_string.
245
  5) For rest of the docids from step2 which were not found in step 4, do
246
     the following for each of the documents:
247
     5a) Use xml_index and xml_nodes table to generate the string descibing
248
         the document including the returnfields requested in the search
249
     5b) If usage_count of the record from step is more than
250
         xml_returnfield_count set in metacat.properties, then store the
251
         string in xml_queryresult as a record which has the returnfield_id
252
         representing the set of returnfields, docid representing the
253
         document and the string as queryresult_string.
254
  6) Add all the strings from step 4 and step 5a to send back the resultant
255
     document
256
  So the results from step 3a in previous algorithm are effectively cached
257
  and hence same computation is not done again and again for each search.
258
  When a document is deleted, all the entries for that document in
259
  xml_queryresult table are also deleted. When a document is updated, all
260
  the entries for that document in xml_queryresult table are deleted. This
261
  works fine because those entries will be generated and cached again the
262
  next time the document is part of a search is requested.
263
* Performance impovement done for % search.
264
* Following new functions added to metacat client API: a method to set access
265
  on an xml document and a method to get newest version for a given document.
266
* Implement a new HarvesterServlet for running Harvester as a servlet. This
267
  eliminates the need to run Harvester in a terminal window. By default, the
268
  HarvesterServlet is commented out in lib/web.xml.tomcat(3,4,5). The user
269
  documentation will be modified to instruct Harvester administrators to
270
  uncomment the HarvesterServlet entry.
271
* Minor enhancement to support multiple email addresses for harvester
272
  administrator and site contact. Each address is separated by a comma or
273
  semicolon.
274
* Increase number of rows in Harvest List Editor from 300 to 1200.
275
* Changed default maxHarvests value to 0. Added logic to ignore maxHarvests
276
  value when it is set to 0 or a negative number. This allows Harvester to
277
  run indefinitely without shutting down after reaching a maximum number of
278
  harvests. The previous default value of 30 would cause Harvester to
279
  terminate after 30 harvests.
280

    
281
Details of all changes can be found in the bug database here:
282
http://bugzilla.ecoinformatics.org/buglist.cgi?product=Metacat&target_milestone=1.5
283

    
284

    
285
Release Notes for 1.4.0:
286
------------------------
287

    
288
This is a major release, and is the first to introduce many new features to
289
Metacat.  Some of the major new features in this version of Metacat include:
290

    
291
* Added a new 'Harvester' that allows documents to be periodically pulled
292
  from distributed sources into metacat, useful for interfacing with other
293
  system types. See the Metacat Tour for details.
294
* Added new 'skins' capability to allow the GUI for the metacat web interface
295
  to be more easily customized to fit into site needs.  The skin is based on
296
  a flexible mix of XSLT, CSS, and Javascript.
297
* Added a web-based metadata entry form for submitting limited EML documents
298
  as part of a data registry.  The GUI for the registry is customizable and 
299
  is integrated with the 'skins' system for the main metacat interface.  The
300
  registry subsystem is written in Perl, and can simultaneously be used to 
301
  present multiple 'registry' interfaces.
302
* A new 'client API' has been developed and implemented in Java and in Perl 
303
  so that metacat functions can be accessed from any program in those
304
  languages.  The client API supports the major metacat servlet actions,
305
  including login(), logout(), query(), insert(), update(), delete(), and 
306
  upload(). See the Metacat Tour for details.
307
* Added a new 'getlog' action that produces a usage history for all major
308
  events.  So now an XML report can be generated for document reads, inserts,
309
  deletes, etc.  See the new section in the Metacat Tour.
310
* Added a new 'buildindex' action that can rebuild the index entries for any
311
  document in the database.
312
* The configuration file for metacat (metacat.properties) has been moved out
313
  of the jar file and into the WEB-INF directory, allowing far easier changes
314
  to the configuration parameters.
315
* Results in default web interface are sorted by title in the XSLT 
316
  (configurable)
317
* Many bugs were fixed.
318

    
319
Details of all changes can be found in the bug database here:
320
http://bugzilla.ecoinformatics.org/buglist.cgi?product=Metacat&target_milestone=1.4
321

    
322
Release Notes for 1.3.1:
323
------------------------
324

    
325
This is a simple interim bug fix. No major functionality changes. Bugs fixed 
326
include:
327
 1) Metacat 1.3.0 doesn't work in Java 1.3 because a subtle API differnce in 
328
    Java 1.3 and Java 1.4. Currently, Metacat 1.3.1 will work in both Java 1.3 
329
    and Java 1.4.
330
 2) Distribution package size was reduced.
331

    
332

    
333
Release Notes for 1.3.0
334
------------------------
335

    
336
In 1.3.0 release, the structure of the xml_access table was changed and a new
337
table, xml_accesssubtree was added. If you try to update a previously 
338
installed Metacat, you should run a script file to updated the table structure 
339
before installation.
340
For Oracle user: At the SQLPLUS prompt type the following -
341
@src/reviseformetacat13.sql
342
For Postgresql user: At install directory prompt type the following -
343
psql exp < src/reviseformetacat13_postgres.sql
344
(where 'exp' is the database name).
345

    
346
After installation, user should run "ant schemasql" command to register EML2
347
schema in xml_catalog table.
348

    
349
If you are a new Metacat user, this step is unneeded.
350

    
351
Note: 1) We recommend to use Tomcat 4 and JAVA 1.4 to run Metacat. Otherwise, 
352
         it will cause potential replication issues.
353
      2) Delete the xercesImpl.jar and xmlParserAPIs.jar files 
354
         which are in $CATALINA_HOME/common/endorsed. They are old version and
355
         don't support XML schema validation.
356

    
357
New Features in 1.3.0
358
  1) Partialy support EML2 document. User can store, query, read and write
359
     EML2 documents. However, Metacat only support access control in resource
360
     level. The subtree level access control will be ignored.
361
  2) Support other xml document base on namespace/schema.
362
  3) Support query for attribute value
363
  4) Assign MIME type to data file base on metadata when user try to read it.
364
  5) Owner can assign access rules to a document which does not have access 
365
     document to apply it.
366
  6) Support exporting single file, not only whole package
367
  7) Resupport Microsoft SQL Server.
368

    
369
Fixes in 1.3.0:
370
  1) Couldn't finish delta T replication for large set of documents.
371
  2) Couldn't create access control during delta T replication.
372
  3) Eorr will be written to a seperated log file if some documents
373
     were failed in replication.
374
  4) Decrease the time to create access rules during insert or update 
375
     a package.
376
     
377
Documentation
378
-------------
379
See the docs directory for detailed documentation and installation
380
instructions.
381

    
382
Details of the Metacat architecture can be found on the website for
383
the Knowledge Network for Biocomplexity (KNB):
384

    
385
  http://knb.ecoinformatics.org/software/metacat/
386

    
387
Contributions to this work are welcome.  Please see the above web site
388
for details on how to contribute.
389

    
390
Major Known Bugs or Feature Requests (see http://bugzilla.ecoinformatics.org)
391
-----------------------------------------------------------------------------
392
If you discover a bug please report it, either by email (above) or by using
393
our bug tracking system (http://bugzilla.ecoinformatics.org). There is a
394
list of currently unimplemented features in Bugzilla that we are working on
395
for the next release.
396

    
397
Legalese
398
--------
399
This software is copyrighted by The Regents of the University of California
400
and the National Center for Ecological Analysis and Synthesis
401
and licensed under the GNU GPL; see the 'LICENSE' file for
402
details.
403

    
404
This material is based upon work supported by the
405
National Science Foundation under Grant No. DEB99-80154, DBI99-04777, and
406
0225676 for SEEK.  Any opinions, findings and conclusions or recomendations 
407
expressed in this material are those of the author(s) and do not necessarily 
408
reflect the views of the National Science Foundation (NSF).
409

    
410
This software is partially supported by a grant from the Andrew W. 
411
Mellon Foundation.
412

    
413
This product includes software developed by the Apache Software
414
Foundation (http://www.apache.org/). See the LICENSE file in lib/apache
415
for details.
416

    
417
The source code, object code, and documentation in the com.oreilly.servlet
418
package is copyright and owned by Jason Hunter. See the cos-license.html file
419
for details of the license.  Licensor retains title to and ownership of the
420
Software and all enhancements, modifications, and updates to the Software.
421

    
422
This software includes the JDBC driver for PostgreSQL.  See the 
423
postgresql-license.txt file for details.