2.5.3

TopBraid Composer - Standard Edition

- Bugfix: TopBraid had to be restarted after entering the Google Maps key

- Bugfix: Various Sesame 2 fixes:
          - Removed username/password options from Sesame (until they are
            properly supported by Sesame 2.0
		  - Create new Sesame 2 remote repository attempted to open the DB
            even if it was read-only.
		  - Create new Sesame 2 native repository did not set the default
            namespace (based on the entered base URI).
          - Connections were not closed correctly at the end of a session
          - The action names for Sesame 2 are now more consistent with 
            action names for Sesame.
          
- Bugfix: Values of range rdfs:Datatype could not be entered on form

- Bugfix: Auto-completion in SPARQL editor inserted ':' for objects of default
          namespace even if the user had already entered ':' at the beginning.
          
- Bugfix: Sometimes Oracle databases could not be created with inferencing on.

- Bugfix: Caching of database triples was often inefficient

- Bugfix: Importing large URLs from the Web can now be canceled

- Bugfix: Form layout editor now has a vertical scroll bar if the preview is too high

- Bugfix: Popup forms (often used from Graph and SPARQLMotion editor) did not
          display vertical scrollbars correctly.  
          
- Bugfix: tops:constructName did replace characters such as / from URI templates

- Bugfix: The class tree display mode (number of instances) will be reset to
          direct instance count if a database project is opened.  Otherwise it
          may take too long to open.
          
- Bugfix: All TopBraid SPARQL property functions now use the graph used in the context 
		  of the query instead of the currently selected ontology.
		  
- Bugfix: For TopBraid library ontologies, predefined prefixes such as "dc" and "rss"
          are used for prefix generation during ontology import. 

- Update: RDFa importer now uses latest XSLT

- Update: RSS/Atom Feeds importer is significantly improved for RDF generation: 
		  1-) Depending on the type of feed, rss:guid, atom:id or the 
		  localname of referencing properties and instances are used to construct 
		  URIs for anonymous resources. 
	      2-) The resources with no types are given custom types. The URIs of the  
	      types are constructed using the localnames of properties that reference
	      the resources.
	      3-) Bag and Seq data structures are removed for simplicity.

- Feature: Added "Link with DBpedia..." menu item and batch operation (basket).
           This feature suggests matching DBpedia pages for resources in the
           current ontology.

- Feature: Nested forms now display rdfs:label and rdf:type first

- Feature: "Add existing..." dialog now also lists instances of subclasses if
           the "Show indirect instances" option is selected in the Instances View.
           
- Feature: Added extension point org.topbraid.sparql.propertyFunctions, enabling
           plugin developers to register their SPARQL property functions cleanly.
		   
- Feature: The spreadsheet importer can now import .txt files that are encoded in 
		   one of the text encodings that are supported by JRE. The importer dialog 
		   has a new combo box for the encoding. The list of supported encodings can be 
		   found here: http://java.sun.com/j2se/1.5.0/docs/guide/intl/encoding.doc.html	   

- Performance: Instances View, SPARQL Results view and Select Instance dialog have
               been made more scalable and now do lazy loading when the user scrolls.


TopBraid Composer - Maestro Edition

- Bugfix: Execute SPARQLMotion module and open temp file did not work when
          executed the first time (and the file wasn't there yet).
          
- Bugfix: Semantic XML loaded was not able to resolve relative DTD paths
          that pointed to files in the same directory
          
- Bugfix: The Maestro edition trial dialog was frozen at some UI operations.

- Feature: Added sparql:fetch to Semantic JSP tag library

- Feature: Semantic XML now supports additional ordering of XML elements with 
		   respect to element's attribute values during RDF to XML conversion. 
		   This ordering is defined using composite:child property on properties 
		   defining attributes. For example, the property, "idProperty", has the 
		   statement 'idProperty sxml:attribute "ID"'. In order to do ordering 
		   on the "ID" attribute, the following statement is used: 
		   'composite:index composite:child idProperty'. If composite:child 
		   is used transitively, i.e. secondary, tertiary or further ordering is 
		   desired or if only secondary ordering will be used, then the root 
		   subject should always be composite:index. 

- Changes in SPARQLMotion 1.1.0
  - Generalized input variables framework.  Now any configuration property in a
    module can be populated from input parameters.  For example if a module has
    a property sml:url and another module upstream of it binds a variable with
    the property's local name (url) then the module will use the variable value
    unless another value has been explicitly set by the module author.
  - As a result of this, the sparqlmotion core ontology has been simplified
    further (e.g., removed sm:Variable, sm:inputVariable, sm:name)
  - The graphical Modules editor now displays input variables much smarter and
    will also consider variables mentioned in string templates as well as those
    used in SPARQL queries. 

- Changes in SPARQLMotion Library 1.2.0
  - Added sml:ImportRDFa
  - Added sml:ImportGRDDL
  - Added sml:encoding option to spreadsheet converter