* Avoid lock on graph uri update
* Run in VitroBackground named thread. Avoid creating threads if a thread to update graph uris has been already created.
---------
Co-authored-by: Georgy Litvinov <georgy.litvinov@tib.eu>
* Refactor blank-nodes-as-variables behavior from RDFServiceJena to JenaModelUtils; use this logic when removing triples from the TBox reasoner's internal assertions model. Add unit test.
* Remove complication of datasets and remove blank nodes directly from models
* Remove temporary debugging logging
* Simplify original blank node tree logic
* Remove unneeded imports
This is only used in a handful of places.
The core java libraries provide the same functionality but with a less exchaustive locale database.
The IBM ICU is more up to date than the Java version but is otherwise identical.
see: https://unicode-org.github.io/icu/userguide/icu4j/#introduction-to-icu4j
* Faux data properties
* Use not disjointed classes for faux properties domain and range options
* Fix faux object property shadowing real property
List faux data properties
List faux properties which base properties domain don't match current subject
* Write exception to log in VClassDaoJena
* filter faux properties returned in possiblePropInstForIndividual
* safety checks added
* extracted getBaseLabel method, fixed base labels for data properties
* fix for prev commit
* Formatting fixes
* More formatting fixes
* More formatting fixes
* More formatting fixes
* More formatting fixes
* More formatting fixes
Co-authored-by: Georgy Litvinov <georgy.litvinov@tib.eu>
* created RDFServiceBulkUnionUpdater
* fix and indents
* use bulk updater for RDFServiceGraph in blankNodeFilteringGraph
* use removeAll to optimize removal all triples from TDB modeles
* avoid additional serialization/deserialization cycle of removing model chunk
* fixed VitroModelFactory tests
* fixes for BulkUpdating models
* Created custom ModelCom, OntModelImpl, BulkGraphMem to avoid triple by triple transactions on add/remove model
Load n3 files from home directory into in-memory model to use bulk
loading.
* refact: simplified BulkGraphMem checks
* removed unused import
* renamed I18nBundle
* added I18nBundle interface
* Added translation provider
* prototype of TranslationConverter
* convert all properties
* fixes
* added caching
* Removed obsolete code
* Improved logging
* fixed getting already existing label
* Fix to get RDF Service for configuration models
* fix translation request query
* added INTERFACE_I18N_FIRSTTIME_BACKUP model
* converter test added
* formatting fixes
* Translation provider tests added
* cleanups, added cache test for translation provider
* fix: get theme info from web app dao factory as sparql queries on both content and configuration models not supported
