On Tue, Sep 20, 2005 at 05:05:05PM -0400, Bruce Momjian wrote: > Jim C. Nasby wrote: > > On Mon, Sep 19, 2005 at 10:16:36PM -0400, Bruce Momjian wrote: > > > > > > Added to TODO: > > > > > > * Allow DISTINCT to use hashing like GROUP BY > > > > 3 lines above we have... > > Consider using hash buckets to do DISTINCT, rather than sorting > > This would be beneficial when there are few distinct values. > > OK, I have merged these items into one. > > > > > Can you add > > http://archives.postgresql.org/pgsql-hackers/2005-09/msg00810.php? All I > > could find on the other TODO was > > http://archives.postgresql.org/pgsql-committers/2004-09/msg00028.php, > > which doesn't help much... > > What do these URL's have that the current TODO does not? > > * Consider using hash buckets to do DISTINCT, rather than sorting > > This would be beneficial when there are few distinct values. This is > already used by GROUP BY.
Maybe it's just me, but the recent run-through of the TODO list indicated that there's a fair number of items that people look at and don't really knowh what they mean. Providing the context (ie: email thread) that spawned an idea seems extremely valuable in being able to explain the idea behind a TODO item. They also usually contain valuable tips about how a TODO could be implemented. In this example, having quick reference to the discussion about hashagg and first()/last() would probably prove useful. -- Jim C. Nasby, Sr. Engineering Consultant [EMAIL PROTECTED] Pervasive Software http://pervasive.com work: 512-231-6117 vcard: http://jim.nasby.net/pervasive.vcf cell: 512-569-9461 ---------------------------(end of broadcast)--------------------------- TIP 9: In versions below 8.0, the planner will ignore your desire to choose an index scan if your joining column's datatypes do not match
