Release Notes - PoolParty 2022 R3 (9.2.0)

A broken access control vulnerability in the user management allowing authenticated users to see information from other users has been fixed.

The PoolParty project SPARQL endpoint supports the CSV (comma separated values) and TSV (tab separated values) content types as output format.

Data Validator automatically triggers a snapshot of the project data to facilitate a rollback of changes.

Extractor sorting has been improved to increase reproducibility. Concept scores returned by the "extractor/api/extract" method have discrete (integer) scores with a range of 1 - 100. Sorting just by those scores is not completely deterministic because different concepts can have the same score. The preferred label was added as a search criteria to make the order of results easier reproducible and more stable.

Handling of language locales has been made more robust in the PoolParty API. The parameter is now treated in a case-insensitive manner and variations such as "en-gb" and "en-GB" are interchangeable. This affects methods like the sub-tree API for retrieving partial trees of a taxonomy or the Extractor API.

Data sanitation of user-uploaded content to corpus analysis or semantic classifier has been improved. It is now less likely for longer documents to be truncated or rejected.

Custom ontologies containing sub-classes or disjoint class definitions can be used again in projects without causing errors in the concept scheme view.

Matching properties (such as skos:exactMatch, skos:relatedMatch, and others) imported via tabular import are now shown correctly on the user interface.

In the GraphEditor the graphs can be again set to "editable".

Highlighting of the default language in the project creation dialog has been restored.

Synchronizing and de-synchronizing of customer schemes and ontologies work again.

Tomcat: 9.0.84

Keycloak: 22.0.5

Elasticsearch: 7.17.11

Solr: 8.11.1