Project

General

Profile

Revision 2475

Added by sgarg over 16 years ago

Changes made for release 1.5

View differences:

README
36 36
------------
37 37
See the file "docs/user/metacatinstall.html" for detailed instructions.
38 38

  
39
Release Notes for 1.5.0:
40
------------------------
41

  
42
This release comes with performace improvement, user interface enhancements
43
and bug fixes. Below is the list of bug fixes and performance improvements. 
44

  
45
Bug Fixes:
46
* Modify property values of harvester registration servlets to match the
47
  servlet-mapping URL values in web.xml. The old values used the servlet
48
  class names. This worked in Tomcat 4 but seems to break in Tomcat 5 on
49
  Windows. The new values use the servlet-mapping URL values. This should
50
  work in both Tomcat 4 and Tomcat 5. (Bug 162)
51
* Re-implement logic to prune old log entries from the HARVEST_LOG and
52
  HARVEST_DETAIL_LOG tables. The old logic caused integrity constraint
53
  violations in the database because it tried to delete parent records from
54
  HARVEST_LOG prior to deleting child records from
55
  HARVEST_DETAIL_LOG. (Bug 162)
56
* In data registry, minor bugs fixed like the error page shows the link back
57
  to the form when error page was generated because of a document read
58
  request from search results page,  the successful delete button leading
59
  to a blank screen and the old ACL overwritten by the registry.
60
  (Bugs 1307, 1310, 1322, 1344, 1551)
61
* Changed the code to support insertion of following characters: ', `,
62
  charcters in the Numeric Character Reference format (&#xyz;) and
63
  characters like µ. Characters like µ will be converted to and stored in
64
  Numeric Character Reference format(&#xyz;). They will also be returned
65
  in Numeric Character Reference and it is upto the client to convert them
66
  and display them accordingly in the application. Partial fix provided by
67
  Johnoel Ancheta. (Bug 1538,1711)
68
* Spatial search failed on Oracle because of invalid entries made by
69
  some documents in the nodedata column of xml_nodes for paths involving
70
  'eastBoundingCoordinate', 'westBoundingCoordinate',
71
  'northBoundingCoordinate', 'southBoundingCoordinate'. A new column
72
  nodedatanumerical was added to xml_nodes table such that all the numerical
73
  nodedata is stored in this column and this column is now used for spatial
74
  search. (Bug 1703, 1718)
75
* In the default skin, clicking on the keywords on the home page did not
76
  result is a search being done. Fix provided by Owen Jones. (Bug 1768)
77
* Metacat generated a success message even when a document which didnt
78
  exsist was requested for deletion. This has been fixed. (Bug 1850)
79
* "packagetype" in xml-relation table was entered as eml200 namespace when
80
  an eml201 document was inserted. This is fixed now so that "packagetype"
81
  in xml-relation table points to eml201 namespace. (Bug 1979)
82
* System_id in xml_catalog pointed to http://knb.ecoinformatics.org/knb/
83
  earlier. Fixed such that it points to the locally installed metacat url. 
84
  (Bug 1986)
85
* Changes to the Query subsystem fix bugs that prevented attributes from 
86
  being expressed solely in the xpath query statement and the returnfield 
87
  values. For instance, a query URL may now include search strings like 
88
  '@packageId=sbclter%25' and return strings like 'returnfield=@packageId'.
89
  Previously, the attribute had to be appended to an element: 
90
  '/eml/@packageId=sbclter%25'. These fixes change DBQuery.java, 
91
  QuerySpecification.java, and QueryTerm.java (Bug 2052)
92
* Search query failed in case of Oracle if number of documents in the result
93
  were more than 1000. This has been fixed. (Bug 2000)
94
* Removed any occurence of enum which is now a keyword in Java 1.5
95

  
96

  
97
Enhancements:
98
* Major improvements done in user interface for the data registeries and
99
  various other skins. Improvements done based on suggestions from Mark
100
  Stromberg, Laura Downey and others. Improvements also done in resultset.xsl,
101
  the default skin and ldap templates. (Bug 1948)
102
* Coded added such that administrator can delete documents irrespective of
103
  who the document belongs to.
104
* Performance impovement done in searching of datasets. A part of the search
105
  time was taken up because of generation of resultset after the search had be
106
  done. So this was the sequence of events in earlier metacat
107
  1) Search for the given text
108
  2) A list of docids generated
109
  3) For each of the docids
110
     3a) Use xml_index and xml_nodes table to generate the string descibing
111
         the document including the returnfields requested in the search
112
  4) Add all the strings from step 3a to send back the resultant document.                                                                 Here a decent amount of time was being taken by step 3a. The algorithm is
113
  now modified by addition of two tables xml_queryresult and xml_returnfields
114
  and a user defined parameter xml_returnfield_count. The new algorithm works
115
  as follows
116
  1) Search for the given text
117
  2) A list of docids generated
118
  3) For the given set of return fields generate a unique string and check
119
     if that string exists in xml_returnfields
120
     3a) If string does not exist in xml_returnfield, then enter a new
121
         record usage_count as 1 and returnfield_string as the unique string
122
         generated above.
123
     3b) Else if the string exists, increment usage_count
124
  4) Given the docids from step 2 and the id of the returnfield record from
125
     step 3, query for any docids that already exist in xml_queryresult.
126
     For the docids that do exist, get the queryresult_string.
127
  5) For rest of the docids from step2 which were not found in step 4, do
128
     the following for each of the documents:
129
     5a) Use xml_index and xml_nodes table to generate the string descibing
130
         the document including the returnfields requested in the search
131
     5b) If usage_count of the record from step is more than
132
         xml_returnfield_count set in metacat.properties, then store the
133
         string in xml_queryresult as a record which has the returnfield_id
134
         representing the set of returnfields, docid representing the
135
         document and the string as queryresult_string.
136
  6) Add all the strings from step 4 and step 5a to send back the resultant
137
     document
138
  So the results from step 3a in previous algorithm are effectively cached
139
  and hence same computation is not done again and again for each search.
140
  When a document is deleted, all the entries for that document in
141
  xml_queryresult table are also deleted. When a document is updated, all
142
  the entries for that document in xml_queryresult table are deleted. This
143
  works fine because those entries will be generated and cached again the
144
  next time the document is part of a search is requested.
145
* Performance impovement done for % search.
146
* Following new functions added to metacat client API: a method to set access
147
  on an xml document and a method to get newest version for a given document.
148
* Implement a new HarvesterServlet for running Harvester as a servlet. This
149
  eliminates the need to run Harvester in a terminal window. By default, the
150
  HarvesterServlet is commented out in lib/web.xml.tomcat(3,4,5). The user
151
  documentation will be modified to instruct Harvester administrators to
152
  uncomment the HarvesterServlet entry.
153
* Minor enhancement to support multiple email addresses for harvester
154
  administrator and site contact. Each address is separated by a comma or
155
  semicolon.
156
* Increase number of rows in Harvest List Editor from 300 to 1200.
157
* Changed default maxHarvests value to 0. Added logic to ignore maxHarvests
158
  value when it is set to 0 or a negative number. This allows Harvester to
159
  run indefinitely without shutting down after reaching a maximum number of
160
  harvests. The previous default value of 30 would cause Harvester to
161
  terminate after 30 harvests.
162

  
163
Details of all changes can be found in the bug database here:
164
http://bugzilla.ecoinformatics.org/buglist.cgi?product=Metacat&target_milestone=1.5
165

  
166

  
167

  
39 168
Release Notes for 1.4.0:
40 169
------------------------
41 170

  

Also available in: Unified diff