On Oct 19, 2010, at 12:20 PM, Robert Muir wrote: > On Tue, Oct 19, 2010 at 12:17 PM, DM Smith <dmsmith...@gmail.com> wrote: > >> I'd be surprised if there are use cases for non-reuse. >> >> IIRC: When we started down the reuse path, the goal was reuse only, not just >> reuse by default. But in order to bridge the past to the future, there was >> the possibility of continued non-reuse. In a sense non-reuse was deprecated, >> but I'm not sure that @deprecated as a mechanism was able to clearly >> indicate that. >> > > Exactly: i don't think theres a clear way to detect that your > tokenStream() method is "reuse-safe" and deprecate it: e.g. you have > to implement reset() correctly in your tokenstreams. > > But lets think about this: for non-experts, making Analyzer "reusable > by default" by removing reusableTokenStream() and reusing > tokenStream() would probably be the single largest indexing > performance improvement we could make... the API is so confusing that > I think many people probably have analyzers that aren't reusing today. > > I think its worth considering a backwards break, especially since as > Mike mentioned, for the very special (possibly even only theoretical!) > non-reuse case, there are ways they could still index: but the "fast > way" should be the "easy/default way".
To me, the backwards break is merely a code break. I can't see how it would break an index. -- DM --------------------------------------------------------------------- To unsubscribe, e-mail: dev-unsubscr...@lucene.apache.org For additional commands, e-mail: dev-h...@lucene.apache.org
