Hi Alan, As I mentioned in my previous email, our intention is to profile the library and reduce its size and signature as much as possible. We can do wonders just by removing ref-counting when its not really necessary, and hopefully we'll find more bottlenecks. We are interested in having an easy upgrade path too, so we'll keep to your structure, modifying only what isn't breaking that purpose.
Ben and Veit are on board with me on this, and I'm glad to read you are too. I'm still studying your code (from Ben's branch), and will report back as soon as I'm ready to begin. Itamar. On 04/04/2011 13:17, Alan Wright wrote: > Hi Itamar, > > Thanks for taking time to look at Lucene++ > > The intention behind Lucene++ has always been to closely reflect the java > lucene code with a view to being able to keep up with the pace of development. > > The good news with this approach means that I can quickly integrate bug > fix releases - in fact it only took a matter of days to integrate changes from > 3.0.0 -> ... -> 3.0.3. > > The disadvantages with this approach are that the library is quite large > (~10MB) because of the close modelling of the java code (there's a lot of > it!). > Also, I make use of boost::shared_ptr to represent pointers - this means > complex memory handling is very much reduced, at the expense of > performance. > > What I'm currently working on is taking Ben's changes an merging them in > the master (http://github.com/luceneplusplus/LucenePlusPlus), this should: > > 1) Reduce the number of symbols exported - reducing the link time and > library size > > 2) Add support for CMake. > > Looking forward, there's a new fairly substantial feature release of java > lucene > (3.1) that will require porting. I also look forward to working with you (and > anyone else!) in order to continually improve Lucene++/CLucene performance. > > cheers, > Alan. > > > On 31 March 2011 01:25, Itamar Syn-Hershko<ita...@code972.com> wrote: >> Hi all, >> >> >> CLucene has grown quite nicely the last few years, but yet we were >> unable to keep up with the high pace of Java Lucene's. Our goal has >> always been to have Lucene on steroids, and have it more maintainable >> and up to speed with Java Lucene. >> >> >> As Ben mentioned here last week, now there's Lucene++ written by Alan >> Wright which is easier to maintain, and therefore it allows to keep up >> with Java Lucene more easily - not to mention making bindings and >> porting of contribs more easily. However, Lucene++ is currently >> significantly heavier than CLucene, and is having several downsides we >> currently don't have (or don't want to have) in CLucene. >> >> >> Ben and I have decided to try and modify Lucene++ so it is lighter and >> works faster, and when we are done we plan on making the result the main >> branch of CLucene. This should allow us to concentrate on performance >> tuning, and on making more tools accessible to the community. Being more >> recent and with bindings to other languages should make the project more >> known and the community bigger. >> >> >> However, that means serious API change, and having a larger and heavier >> library, as mentioned in a previous thread. It also means we will not be >> able to support the 2_3_2 branch for long. The main downside here is if >> someone finds the current implementation perfect and is afraid of having >> a heavier library. >> >> >> We'd appreciate your opinions on the matter, or any help pulling this off. >> >> >> Itamar. >> >> >> ------------------------------------------------------------------------------ >> Create and publish websites with WebMatrix >> Use the most popular FREE web apps or write code yourself; >> WebMatrix provides all the features you need to develop and >> publish your website. http://p.sf.net/sfu/ms-webmatrix-sf >> _______________________________________________ >> CLucene-developers mailing list >> CLucene-developers@lists.sourceforge.net >> https://lists.sourceforge.net/lists/listinfo/clucene-developers > ------------------------------------------------------------------------------ Xperia(TM) PLAY It's a major breakthrough. An authentic gaming smartphone on the nation's most reliable network. And it wants your games. http://p.sf.net/sfu/verizon-sfdev _______________________________________________ CLucene-developers mailing list CLucene-developers@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/clucene-developers
