EDG 6.4 RELEASE
TopBraid EDG version 6.4 completes transition to the new user experience delivered in 6.3 for editing and viewing asset collections. With this, “old”, pre-6.3 editors are no longer available as an option. To make this transition possible we added various improvements to the editor including improving Problem and Suggestions panel, SPARQL query panel, and Change History panel.
Version 6.4 delivers a number of new capabilities including new Import and Data Cataloging options, support for GeoSPARQL query and a Map panel to display the results, various Collaboration and Personalization features, ability to store Document Attachments in S3. These and other key new features are described in more detail below. Additional smaller improvements, many of which are results of direct customer requests, and their associated ticket numbers can be found in the change-log to the right.
Key Improvements of 6.4
New Editor Application Features
- Two new Associations panels – for all data (root configured in the Manage tab or selected by users) and, specifically, for a selected resource. These panels show tree views of resources based on any relationships.
- The new Map Explorer panel to display results of searches and queries that have geo location information. Geo location information must use W3C vocabulary for representing latitude, longitude and altitude information in the WGS84 geodetic reference datum (namespace http://www.w3.org/2003/01/geo/wgs84_pos). To enable the panel, Google Map API needs to be entered in the Server Configuration page.
- For ontologies, a new option to Derive property shapes from instances – available under Modify menu of a class/node shape. Similarly, a new option to Derive missing classes – available for the Home resource (ontology itself) to suggest missing class/shape definitions..
- For ontologies, simpler one-click creation of inverse path property shape.
- Ability to define parametrized SPARQL queries. Use dollar sign ($) to indicate parameter variables. Once saved, users can run these from the SPARQL Query Library panel. A dialog will let them enter required parameters. We plan to provide in the future a number of useful queries e.g., to find terms with labels in one language, but no labels in another language.
- Additional configuration of asset visibility. New dash:hidden property when set to true will result in resources hidden in the tree views.
- Users can now formulate GeoSPARQL queries.
- For supported functions see https://jena.apache.org/documentation/geosparql/ under FILTER Functions and Property Functions.
- GeoSPARQL queries need to be enabled on an asset collection basis. EDG will build and maintain Geo indexes.
- Users can now create Announcements in the Governance collection with start and expiration date. These will be displayed on the home page.
- New Home Page widget for “What’s happening” – to make users aware of where work is happening
- New Home Page “Asset Collection Summary” widget to list the number of available asset collections by type.
- Users can now personalize selection of widgets available on their home page. This can also be done at the server level.
- Ability to Endorse assets and see Endorsements made by other users.
- Users can now add information about themselves: set avatar image, enter name and contact phone number.
Enhancements to RDF* Support
- Improved syntax.
- GraphQL queries for reified triples.
- Reification-based ordered lists.
- Reification-based sh:severity and sh:message per constraint.
New Import Options
- Ability to import TriG file – this will create multiple asset collections.
- Ability to import compressed files.
- New dataset importer – available for data asset collections to support cataloging of spreadsheet files. It creates instances of edg:DataSet class for each spreadsheet. Data in spreadsheets is profiled and all metadata is captured.
- EDG can now store document attachments for any asset.
- Attachments are stored in S3 and require administrator to configure S3 buckets.
- Each asset collection can use its own bucket configured on the Manage tab.
Export of Multiple Asset Collections
- A new basket operation enabling users to export several asset collections at a time.
- This will kick off a scheduled job to export each collection as a file and store the files in S3.
- Users will be notified upon /completion via email.
- Users can select the bucket for exports in the EDG Configuration Parameters page globally and override on the Manage tab.
Non Functional Requirements
- Improvements to compliance with accessibility standards.
- Upgrade to Apache Jena 3.15.0.
- User data moved to system level collection in order to be synced by Data Platform.
- Library upgrades to address CVE-2020-7662 and CVE-2019-0205, CVE-2019-0210. No known issues or concerns with EDG.
- Performance optimizations in the SHACL engine.
- Removed support for “includes based on subject area” – due to the lack of use and to make it possible to include an asset collection into multiple subject areas. Please see migration notes if using this feature.
- Removed old SWP editors as mentioned above.
- Removed Problems and Suggestions and Change History reports from the Reports tab – these were duplicative features. The same functionality is available as dedicated panels directly in the editors.
- Removed MultiTes importer for taxonomies.
Known issues with 6.4.
- The new avatar feature will be unavailable in 6.4.0 with the Data Platform option. The fix for allowing this feature is in Jena 3.16.0. EDG is currently using Jena 3.15.0. This will be fixed in 6.4.2.
- S3 export will fail silently if using an ELB. This will be fixed in 6.4.2.
- The following messages in the Tomcat log can be ignored. These are not an issue and should be resolved in an upcoming release.
- SLF4J: Failed to load class “org.slf4j.impl.StaticLoggerBinder”.
SLF4J: Defaulting to no-operation (NOP) logger implementation
SLF4J: See http://www.slf4j.org/codes.html#StaticLoggerBinder for further details
- SLF4J: Failed to load class “org.slf4j.impl.StaticLoggerBinder”.
Improvements and bug fixes for 6.4.1
- Reverted a change in 6.3 that caused some SPARQL queries to run in different execution order.
- Added Explore > Show in Change History to quickly review recent changes about an asset.
- Add regex option to search table filters
- CVE-2020-13822 with elliptic library (hosted from webpack)
- Lodash to version 4.17.19 to address CVE-2020-8203
- Prevent shapes marked sh:hidden from being indexed by Search the EDG
- Some mutations executed from the EDG UI (such as SPARQL UPDATEs) were doing an unnecessary, slow validation step
- Special characters in contains filters no longer cause an exception
- Columns and Filters quicksearch is now case insensitive
- Null pointers regressions in several search filters have been fixed
- Fixed various bugs on Problems and Suggestions panel if results were filtered by confidence value
- Fixed various issues with manage options being available on collections they should not apply to
- Fixed Read-only status of fields in UI is ignored when creating concept/label through search panel
- See change log link above for more
Scheduled for deprecation
- Taxonomy spell checker will be removed in the next major release.