I guess the fact that Lucene closes it is a legacy we may be stuck with - but I think it encourages bad practice code. There is plenty of room for an exception to be thrown before the Reader is sure to be closed in a finally block.
Part of why its standard convention to close your own I think, rather then somewhere in the method you pass it into. Best practice is close it in a finally block yourself, with the try starting right after you make it. Not a big deal though. If it was, we'd know about it by now ;) Chris Hostetter wrote: > : So in 2.9, the Reader is correctly closed, if the TokenStream chain is > : correctly set up, passing all close() calls to the delegate. > > Thanks for digging into that Uwe. > > So Daniel: The ball is in your court here: what analyzer / > tokenizer+tokenfilters is your app using in the cases where you see > Readers not getting closed by Lucene -- if they involve your own custom > Tokenizers then that may be where the problem is, but if all the Analysis > pieces you are using come out of hte box with Lucene please let us know so > we can check them. > > > -Hoss > > > --------------------------------------------------------------------- > To unsubscribe, e-mail: java-dev-unsubscr...@lucene.apache.org > For additional commands, e-mail: java-dev-h...@lucene.apache.org > > -- - Mark http://www.lucidimagination.com --------------------------------------------------------------------- To unsubscribe, e-mail: java-dev-unsubscr...@lucene.apache.org For additional commands, e-mail: java-dev-h...@lucene.apache.org