Re: [HACKERS] KNNGIST next step: adjusting indexAM API

2010-12-01 Thread Dimitri Fontaine
Tom Lane t...@sss.pgh.pa.us writes: Lastly, I'm pretty un-thrilled with the way that the KNNGIST patch implements the interface to the opclass-specific hook functions. Seems like it would be cleaner to leave the Consistent function alone and invent a new, separate hook function for processing

Re: [HACKERS] KNNGIST next step: adjusting indexAM API

2010-12-01 Thread Robert Haas
On Wed, Dec 1, 2010 at 5:22 AM, Dimitri Fontaine dimi...@2ndquadrant.fr wrote: Tom Lane t...@sss.pgh.pa.us writes: Lastly, I'm pretty un-thrilled with the way that the KNNGIST patch implements the interface to the opclass-specific hook functions. Seems like it would be cleaner to leave the

Re: [HACKERS] KNNGIST next step: adjusting indexAM API

2010-12-01 Thread Tom Lane
Robert Haas robertmh...@gmail.com writes: On Wed, Dec 1, 2010 at 5:22 AM, Dimitri Fontaine dimi...@2ndquadrant.fr wrote: IIRC, the goal here was to be able to benefit from KNN GiST from existing GiST indexes as soon as you restart the server with the new code compiled in. I'm not sure it's

Re: [HACKERS] KNNGIST next step: adjusting indexAM API

2010-12-01 Thread Dimitri Fontaine
Tom Lane t...@sss.pgh.pa.us writes: Right, AFAIK there is nothing in KNNGIST that would involve an on-disk data change. Nice, that matches my Royal Oak memories. But any external module relying on GiST will have to provide for the new function you're thinking about, right? Updating was already

Re: [HACKERS] KNNGIST next step: adjusting indexAM API

2010-12-01 Thread Tom Lane
Dimitri Fontaine dimi...@2ndquadrant.fr writes: Tom Lane t...@sss.pgh.pa.us writes: Right, AFAIK there is nothing in KNNGIST that would involve an on-disk data change. But any external module relying on GiST will have to provide for the new function you're thinking about, right? Updating was

[HACKERS] KNNGIST next step: adjusting indexAM API

2010-11-30 Thread Tom Lane
In the current KNNGIST patch, the indexable ORDER BY clauses are transmitted to the executor by cramming them in with the index qual conditions (the IndexScan plan node's indexqual list), from whence they become part of the ScanKey array passed to the index AM. Robert complained that this was an

Re: [HACKERS] KNNGIST next step: adjusting indexAM API

2010-11-30 Thread Robert Haas
On Tue, Nov 30, 2010 at 2:50 PM, Tom Lane t...@sss.pgh.pa.us wrote: In the current KNNGIST patch, the indexable ORDER BY clauses are transmitted to the executor by cramming them in with the index qual conditions (the IndexScan plan node's indexqual list), from whence they become part of the