-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On 4/5/2011 10:52 AM, Robert Collins wrote: > So I made some time today, and I've got most of a rewritten > BatchNavigator that will let us use indexable DB constraints rather > than large OFFSET values to navigate batches. This has come up in > previous performance centric mails on this list, and Jeroen had a > really clear description of the semantic problems we have with > batching that he presented during the Epic - this work should go a > fair way towards addressing both of those things (at the cost of some > fuzziness around what 'page 3' actually means). > > The core grungy is roughly done, I have 'merely' integration glue - to > make sure that the existing URL's wont break and that the generation > of URLs adjacent to the current batch is done cheaply (e.g. without > materialising the adjacent batches). > > I'll skip the details for now, as its not done and is subject to > change without notice, but the code is at > lp:~lifeless/lazr.batchnavigator/keyoffset if you're interested in > playing around. > > Cheers, > Rob >
Do you have any numbers for real performance? John =:-> -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (Cygwin) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/ iEUEARECAAYFAk2a3ioACgkQJdeBCYSNAANMgACgghORiK9yU/X0iApGBfd1IONv NCcAmM5FFvgemLSlDzZMxOov888KdHQ= =9eo6 -----END PGP SIGNATURE----- _______________________________________________ Mailing list: https://launchpad.net/~launchpad-dev Post to : launchpad-dev@lists.launchpad.net Unsubscribe : https://launchpad.net/~launchpad-dev More help : https://help.launchpad.net/ListHelp
