I've updated the GIT patch at http://community.enterprisedb.com/git/.
Bitrot caused by the findinsertloc-patch has been fixed, making that
part of the GIT patch a little bit smaller and cleaner. I also did some
refactoring, and minor cleanup and commenting.
Any comments on the design or patch? For your convenience, I copied the
same text I added to access/nbtree/README to
Should we start playing the name game at this point? I've been thinking
we should call this feature just Clustered Indexes, even though it's not
exactly the same thing as clustered indexes in other DBMSs. From user
point of view, they behave similarly enough that it may be best to use
the existing term.
As a next step, I'm hoping to get the indexam API changes from the
bitmap index patch committed soon, and in a way that supports GIT as well.
pgsql-hackers by date
|Next:||From: Abrie||Date: 2007-03-07 11:06:56|
|Previous:||From: Simon Riggs||Date: 2007-03-07 10:26:40|
|Subject: Re: Auto creation of Partitions|