For Jena and Commons RDF, see: https://issues.apache.org/jira/browse/COMMONSRDF-33
--- A. Soroka The University of Virginia Library > On Jul 29, 2016, at 6:49 AM, Andy Seaborne <[email protected]> wrote: > > On 26/07/16 21:05, Raphael W. Majeed wrote: >> Hello there, >> >> I was working with Jena dependencies in my projects for some time and was >> always wondering why the org.apache.jena.util.iterator.ClosableIterator >> didn't extend the java.lang.AutoCloseable interface. > > The API predates AutoCloseable. > > That means there is applciation and user code that already exists and works > perfectly well. > > Making changes > >> It appears to me like a simple change that would allow other developers >> (and me) to use these iterators in try-with-resource statements. This would >> simplify working with Jena significantly. >> >> Is there a specific reason why the ClosableIterator doesn't extend the >> AutoClosable interface? > > A way forward to consider is to develop a new API under a new package - > whether that's incremental change to the existing one, or a new oen with a > different focus (maybe designed for bulk processing from the ground up; maybe > a mor functional style, maybe ....). > > Jena can support multiple APIs over the same core (SPI) of Graph/Triple/Node. > > See also CommonsRDF although that is half way between the Graph/Triple/Node > level (which is what RDF 1.1 calls "generalised RDF") > and the Jena Model/Statement/Resource (which has an entity aspect > Resource.getProperty etc) > > CommonsRDF is typed (subjects can't be literal) but does not have the entity > aspect. > > Andy > > >> >> Regards, >> Raphael >> >
