- Added a DictionaryAnnotator
- with support for word list files
- with support for Solr synonym files
- with support for loading files from disk
- with support for loading files from a web API
- Fixed a problem where the handling of chunk nodes did not take it’s ancestry into account, which resulted in incorrect determination of the chunk’s base language and if the chunk needed to be checked.
- Fixed a problem where annotations were not always correctly cleaned up on certain mutations in the content.
- Fixed incorrect documentation URL's in the analysis XSD and FDT FCQ module.
- Fixed inconsistency in the order in which annotation request results were handled during the initial document check when simultaneously making changes to the document.
- The required metadata in requests and responses for the HttpApiAnnotator are now actually handled as being required as per documentation.
- Listen to document capability changes in order to only check documents which have the editable capability (cap/editable).
- Resolved Microsoft Security Advisory CVE-2018-0875.
- Resolved Microsoft Security Advisory CVE-2018-0787.
- Changed the default
- Improve chunking logic in cases where an new chunk element is inserted which has sibling chunks to not unnecessarily re-check the sibling chunks which have not changed.
- Skip analysis configuration files that can't be parsed as XML documents nor have the appropriate analysis schema configured.
Was this page helpful?