…ularChangeBus which is now used by the repository.
…IPLE and PRIMARY. Corrected existing CNDs.
…tial problems or issues in the code. Also removed lots of meaningless JavaDoc
…l issues caused by the upgrade.
…warnings, and removed quite a few JavaDoc comments that are inherited via @Override.
…sequencers. Also added a mixin that is now being added to the nodes representing java packages.
… order that contain FK references
…on and to also process TIFF images. However, only the IFD0 part is searched & read.
…exposed Removed the separate 'JcrUtils' interface from the public API (along with the implementation), and moved the 'encode' and 'decode' methods to the ModeShape-specific 'Session' interface.
MODE-1958 DDL Sequencer should use Jsr283Encoder rather than its own custom code Added a getter in Session to return a class of utilities. Removed the code from DDL sequencer framework dealing with converting JCR-illegal characters and replaced with calls to the new utilities class.
Removed the unused constant. Also removed unused constants in the other DDL lexicon classes.
… MSOffice Sequencer and Tika Extractor, which were disabled as a result of https://issues.jboss.org/browse/MODE-1934.
Added the recommended changes to the DdlParsers API. One slight change was that the result "getScore" method returns and "int" and not an "Object."
…ws tests or test cases to be skipped altogether on certain OSes. Also, moved the SkipTestRule to the common.test package.
…t datatype Modified Teiid datatype parser to allow length on a type of "OBJECT." This is a new Teiid change.
…ences for in-line FK constraints when referenced table has not been parsed yet - added postProcess() method in DdlParser and subclasses - added postProcess() method in StatementParser and subclasses
… to inherit non-mixin types.
Handle undefined cells in Excel documents
…th A Name Containing An Allowed Symbol Changed Oracle DDL parsing of names to allow the valid Oracle name character symbols.
Corrected OracleDdlParser.areNextTokensCreateTableOptions(DdlTokenStrem) method to also look ahead and *not* process the tokens if they match a DDL statement start keywords. Added the one test I commented out in last commit back in. Also added a test to make sure the Oracle parser is handling the unterminated statements correctly. So the DDL parser framework still allows DDL files having statements that are not terminated.