Change log

From WandoraWiki
(Difference between revisions)
Jump to: navigation, search
Line 1: Line 1:
 
This page documents only Wandora application changes. See [[Special:Recentchanges|recent changes]] for a list of general Wiki changes.
 
This page documents only Wandora application changes. See [[Special:Recentchanges|recent changes]] for a list of general Wiki changes.
 +
 +
== 2011-02-25 ==
 +
 +
* Fixes a [http://www.wandora.net/wandora/forum/viewtopic.php?t=63 bug] related to [[Facebook Graph extractor|Wandora's Facebook extractor]].
  
 
== 2011-02-18 ==
 
== 2011-02-18 ==

Revision as of 11:08, 25 February 2011

This page documents only Wandora application changes. See recent changes for a list of general Wiki changes.

2011-02-25

2011-02-18

2010-12-17

Release contains some minor feature enhancements related to Email extractor, Simple document extractor, Simple directory structure extractor, Simple PDF extractor and Last.fm extractors. Enhanced extractors are more robust and generate cleaner topic maps compared to older versions.

2010-11-22

  • Maiana integration enables Wandora user to download topic maps from Maiana, Topic Maps Lab's service to host, explore, and share topic maps. Also, Wandora user can upload topic maps to Maiana.
  • Asset weight is a measure of a topic developed by Petra Haluzová of Department of Informatics and Telecommunication, Czech Technical University, Prague. Asset weight measure is described in detail in her TMRA'10 conference paper Evaluation of Instances Asset in a Topic Maps Ontology. Now Wandora user may calculate asset weights for any selectable topic collection.
  • Yahoo! YQL term extractor is a text classifier used to create topic and associations out of given text, an occurrence for example.
  • Tagthe extractor is a similar text classifier helping Wandora user to extract terms out of given text.
  • Yahoo! BOSS extractor is a search engine extractor using Yahoo's BOSS search service.
  • Changed AlchemyAPI extractors. Now extractors use POST method (instead of GET) while sending a given text to the web service.
  • Now Wandora user can refine a large collection of occurrences by right clicking a topic selection and choosing a context menu option Topic > Occurrences > Refine > .... There is eight different and well-known extractors available for occurrence refinement. Available occurrence refiners include OpenCalais classifier, AlchemyAPI extractors, Yahoo! YQL term extractor and Geonames extractors. The batch refine was developed and used to experiment with Afghanistan War Diary topic map.
  • Various bug fixes.

2010-08-19

2010-04-19

  • Export adjacency matrix. This feature exports a text matrix where a matrix cell represents a number of associations between column and row topics. Note that feature doesn't export all topic map structures. It is very lossy. Addition to a whole topic map export, you can export only current topic selection.
  • Export incidence matrix. This feature exports a text matrix where a matrix cell represents a number of row topics playing a role in column association. Note that feature doesn't export all topic map structures. It is very lossy. Addition to a whole topic map export, you can export only current topic selection.
  • Export topic map or selected topics as a GXL graph.
  • Add missing display scope to variant names. This feature is a result of an observation that some topic maps may have valid variant names but variant name scopes contain only language topic. Wandora doesn't view such variant names in Traditional topic panel. This new feature is used to fix such variant names. It adds standard display scope topic to variant names that haven't display nor sort scope topic.
  • Add missing language scope to variant names. This feature is a result of an observation that some topic maps may contain variant names that are missing language topic in their scope. Wandora doesn't view such variant names in Traditional topic panel. Feature adds user addressed language topic to given variant names that have no language topic in their scope.
  • Stands4 word describer
  • Past Wandora versions could export topic maps as a whole only. New Wandora version can export any selectable topic collection as XTM1, XTM2, LTM, JTM, RDF N3, graph, and matrix formats. This feature is a result of practical requirement to merge some parts of previously created topic map to another topic map. Now Wandora user can simply select required topics and then export them as a XTM serialization, and then merge exported topic map with another topic map. Selection export starts with context menu options Topics > Export.
  • Feature removing all topics except selected ones.

2010-01-04

2009-11-23

  • Now you can download and check not only subject locators but also URL occurrences.
  • Now 'XML transform and import' feature accepts also URL resources addition to local files.
  • Fixed custom topic tree tab feature.
  • Fixed YouTube search extractor.
  • Fixed GEDCOM extractor scheme.

2009-10-30

  • Wandora's embedded HTTP server feature has been refactored. New embedded HTTP server has now several parallel output modules. Wandora user can now easily activate and create new output modules. For example, Wandora user could define one output module for a dynamic and browsable HTML site, and other output module for an RSS feed containing topic's instances. As a practical example we have created several Drupal plugins that enable Wandora user to publish Wandora content with Drupal open source content management platform. Drupal plugins are available in Wandora's source code distribution package or as separate downloads. See Extras section on a download page.
  • Along the refactored HTTP server, Wandora's web service feature has been updated to work with Apache Axis2 and the web service feature is integrated with Wandora application. Now you don't need separate Wandora-Piccolo installation to run a topic map web service.
  • New topic map type, Shared Web service topic map, allows Wandora user to share his/her topic maps with other Wandora users. This feature enables a kind of peer-to-peer topic mapping.
  • GEDCOM (GEnealogical Data COMmunication) extractor converting GEDCOM files to topic maps format. Although GEDCOM conversion doesn't support all GEDCOM features, the converter is very useful. We have enclosed Wandora samples folder a Topic Maps conversion of European Royals GEDCOM composed by Denis R. Reid. Extractor is available at File > Extract > Other > GEDCOM extractor.
  • DBpedia extractor. Extractor reads RDF feed from DBpedia and converts it to a Topic Map format. Extractor is available at File > Extract > Subjects > DBpedia extractor.
  • Alchemy Entity Extractor takes a text fragment and extracts named entities out of text. Entities are then converted to topics and associated to the topic carrying the original text. Effect is very similar to OpenCalais classifier. Wandora's Alchemy Entity Extractor is available at File > Classification > Alchemy entity extractor. You need a valid Alchemy apikey to access the extractor.
  • Fixed YouTube extractor and Delicious extractors. YouTube's API libraries had changed. Now the API libraries are updated. Wandora used wrong Delicious web service endpoint. Wandora's Delicious endpoint is now fixed.
  • Fixed really embarrassing bug in Select Topic Button causing nullPointerExceptions.

2009-08-26

  • Important bug fixes:
    • Wandora used to halt irregularly in Windows XP and Vista while opening file dialog. Actual bug was never resolved but Wandora was changed to use same file dialog object again and again. This seems to prevent the halt.
    • Wandora's topic tree used to halt irregularly while browsing moderate large topics maps. Halt was due to lazy tree update. Tree update method has been altered and it looks like the tree halts no more.
  • Wandora specific and non-standard query language interface. Now you can execute queries in Search dialog opened with CTRL+F.
  • Customizable buttons. Now on you can attach you favorite tools to Wandora's top bar beside Open and arrow buttons. Or you could create new tools button sets with Tool manager.
  • Experimental MusicBrainz extractor. Extractor is used to retrieve artists, releases and tracks from enormous MusicBrainz database.
  • Enhanced drag'n'drop feature:
    • You can drop topics to layer stack. As an effect topic stub with only one subject identifier is created to the target layer.
    • You can drop topics to empty Class and Instance tables. As a result topic is attached to the inspected topic as a class or instance.
    • You can drop topics to Associations table title. If your topic map contains suitable schema specification, dropped topic is associated to the inspected topic with schema defined association type and roles. Feature speeds up the association creation.
  • Wandora's browser plug-in has been updated to work with Firefox 3.5.
  • SemanticHacker classifier has been updated to work with current SemanticHacker API.

2009-05-25

  • Added JTM (JSON Topic Maps) support. Now Wandora imports and exports JTM Topic Maps. To export JTM topic map, start Export topic map just like before and use file extension jtm.
  • Gellish ontology extractor. Try File > Extract > Other > Gellish ontology extractor...
  • Subj3ct record extractor. Try File > Extract > Subjects > Subj3ct record extractor...
  • Fixed right mouse click problem on Macs. Now Mac users can access context menus with CTRL+Mouse button.
  • Sleeping features Moby thesaurus extractor and VerbOcean extractor have been activated and are now available at Extract menu.
  • Edge generator has been renamed as Adjacency list import and has been moved to Import menu.
  • Antialized look
  • A lot of code refactoring

2009-03-13

Release contains also many small bug fixes.

2009-01-30

  • String similarity search
  • Enhanced occurrence text editor. Occurrence text editor is opened when you click existing occurrence text in Traditional topic panel for example. New occurrence text editor features include
    • Direct occurrence classifiers (OpenCalais and SemanticHacker). Now you can start the classifier in the occurrence editor using menu option Refine > Classify > Classify with OpenCalais and Refine > Classify > Classify with SemanticHacker.
    • Find topics with similar occurrences
    • Tidy HTML, Strip HTML
    • HTML entity conversion
    • Find topic references in occurrences. This feature scans through all available topics and compares topic names to words in occurrence. If topic name is found in occurrence text, topic is associated with the occurrence carrier topic.
    • Make topics out of occurrence selections. Now you can point out important words (word collections) in occurrence and order Wandora to create a topic for the pointed words. Optionally Wandora associates created topic with the occurrence carrier.
  • WWW site export templates has been refactored and simplified.
  • Sample Web Application has been refactored to work with Apache Tomcat 6.x.
    • HTML templates have also been updated.
  • Current Save project has been split to Save project and Save project as. First save option remembers current project file name while latter always asks file name. See documentation at How to save and load project.
  • Revert project feature to restore previously loaded Wandora project.

2008-12-10

2008-10-24

Wandora's base ontologies in resources/conf are now XTM 2.0. Old XTM 1.0 base ontologies are still in distribution package but they have been renamed with file extension xtm1.

2008-08-28

2008-05-30

2008-04-29

2008-03-14

2008-03-04

  • Wandora's license changed to GNU GPL!
  • Source code distribution package available!
  • Major refactoring! Base class path changed to org.wandora.
  • New Compare topic maps tool introduced!
  • Some minor bug fixes!

Wandora's topic map schema has also changed! Subject identifier base of schema topics is now http://www.wandora.org. It used to be http://www.wandora.net in previous Wandora releases! This change may cause problems while working with older Wandora projects as Wandora doesn't find Wandora class, Wandora language, and Wandora language name version topics. To fix the problem replace subject identifier base http://www.wandora.net to http://www.wandora.org in your old project.

2008-02-13

2008-01-22

2008-01-15

Minor update

  • Removed references to javax.swing.GroupLayout.

2007-11-05

  • Enhanced Drag and Drop support
    • Create topics and associations dragging text (base name) lists to Wandora
    • Extractors accept partial HTML snippets dragged from WWW browser and dropped to Drag and Drop Extractor.
  • Property Table Extractor
  • Simple Text Document Extractor
  • RSS 2.0 Extractor
  • Bibtex extractor schema changed. Creates now more topics and associations instead of occurrences.
  • Subject Locator preview for HTML pages
    • Extract directly from HTML preview
    • Create topic names and occurrences from HTML preview
  • Subject identifier GUI enhanced in Topic Panels
  • Some bug fixes

2007-10-17

  • Shell scripts to launch the application on Linux and Unix systems including Mac.
  • Some bug fixes on Linux-like systems.
  • Bibtex extractor

2007-09-24

This version contains some minor bug fixes. For example:

  • Topic map graph is not anymore drawn outside the topic panel.
  • User can interrupt topic map and project save operations.
  • Adding shortcut of a topic with multiple subject identifiers uses now fancier dialog.

2007-08-30

  • Added Graph Topic Panel enabling graph visualizations of topic maps.
  • Fixed minor bugs.

2007-07-31

  • XTM export fixed. InstanceOf elements contain now only one topic reference.
  • Menu structures have been updated.
  • Opens Wandora project file or XTM topic map file given as a command line parameter.
  • New layer operation doesn't automatically import minimal Wandora ontology.
  • Fixed minor bugs.

2007-06-04

  • New icons. More stylish look.
  • New statistical tools to analyze topic maps: Layer connections, Topic map diameter, and Clustering coefficient.
  • New tool category of Topic map generators including Random graph generator, Fully connected graph generator, Tree graph generator, and Linear list generator.
  • Fixed some minor bugs.

2007-05-22

  • Added SOAP web api option for Piccolo published topic map. Api may be used to access topics and associations directly from Flash applications for example.
  • Added new sample: Flash application using the SOAP api. Actionscript source classes are included.
  • Added a set of Audioscrobber extractors enabling to convert Last.fm's XML feeds to topic maps.

2007-05-02

  • Compiled and tested with Java 6.0.
  • Added MediaWiki extractor. Extractor converts MediaWiki's XML exports to topic map.
  • Fixed some minor bugs.

2007-03-08

  • Fixed HSQL database layer locking bug. Previous versions of Wandora locked HSQL database layer after layer creation blocking topic map access.
  • Fixed Tool manager's tool scanner. Previous versions of Wandora did not found tool classes with class string in their classname.
  • Added FreeDB extractor. Extractor converts freedb database files to a topic map.
  • Added FreeDB sample demonstrating Wandora's FreeDB extractor and simple modification layer used to merge typo-topics.
  • Fixed some minor bugs.

2007-02-27

  • Fix: Now XTM export filters out all characters that may break the XML document. Previous version of Wandora exported base names, text datas, and variant names to the XTM document without changes and broke the file sometimes when you had used (usually extracted) exotic characters in topic elements.
  • Generic SQL database import has been developed to allow you to select whether a table column is converted to a topic or an occurrence.

2007-02-09

  • LTM import robustness improved.
  • Added Copy all variant names tool.
  • Added Remove all empty variant names tool.
  • Added Remove variant name... tool.
  • Fixed Lucene index builder.
  • Speeding XTM import (Assuming the XTM topic map does not contain duplicate associations. I think next update will contain Paranoid XTM import forcing the removal of duplicate associations).
  • Fixed some minor bugs.

2007-01-19

2006-12-15

  • Fixed bug in Configure Topic Tree Types editor.
  • Added Merge project feature.
  • Fixed some minor bugs.

2006-11-15

  • SetClasspath.bat missed path settings for mstor/mstor.jar and activation.jar causing Email extractor to fail with MBOX email repositories. Fixed.
  • Added Lucene index builder tool to exporters. With this tool you can build Lucene search index from any Wandora topic map. Current version of Wandora does not use search index and the tool is mainly for external usage.

2006-11-06

  • Fixed schema association editor's bug causing creation of one player associations instead of multiple player associations in systems with multiple layers.


Retrieved from "http://wandora.org/w/index.php?title=Change_log&oldid=8414"