From: Chris Gamache <cgg007@yahoo.com>
To: Tom Lane <tgl@sss.pgh.pa.us>
Cc: pgsql-sql@postgresql.org
Subject: Re: Schema + User-Defined Data Type Indexing problems...
Date: Thu, 10 Jun 2004 06:20:50 -0700 (PDT)
Message-ID: <20040610132050.94570.qmail@web13801.mail.yahoo.com> (raw)
In-Reply-To: <19454.1086841261@sss.pgh.pa.us>
--- Tom Lane <tgl@sss.pgh.pa.us> wrote:
> Chris Gamache <cgg007@yahoo.com> writes:
> > I'm having a heck of a time, and it seems like in my thrashing about
> > to find a solution to this problem I have ruined the uniqueidentifier
> > datatype in the schema...
>
> > CREATE INDEX mt_uuid_idx
> > ON my_schema.my_table USING btree (my_uuid);
>
> > ERROR: data type my_schema.uniqueidentifier has no default operator class
> for
> > access method "btree"
> > HINT: You must specify an operator class for the index or define a default
> > operator class for the data type.
>
> > I can look at the operator classes and see that there is an operator class
> for
> > btree for my_schema.uniqueidentifier.
>
> IIRC, the opclass has to be in a schema that is in your schema search
> path to be found by CREATE INDEX by default. If it isn't, you could
> specify it explicitly:
>
> CREATE INDEX mt_uuid_idx
> ON my_schema.my_table USING btree (my_uuid USING my_schema.uuidopclass);
>
> It's possible that we could think of a more convenient behavior for
> default opclasses, but I don't want to do something that would foreclose
> having similarly-named datatypes in different schemas. You have any
> suggestions?
That /is/ important to be able to have similarly named datatypes in different
schemas. I'll give the explicit opclass a go. Indeed, if I place the schema in
my search path the index creation and index scans seem to work perfectly. I had
wanted to have to specify the schema whenever I referenced objects in it
instead of putting it in my search path. I had no concept of exactly how truly
separated schemas are. The only idea that I can think of (and, again, I may be
underestimating the level of separation that needs to exist between schema) is
that object creation could implicitly looks to the current schema for a usable
index/opclass/whatever first before checking the search path. A SELECT could
look first to the schema of the table before checking the search path for a
usable index. Is it even possible to create an index that lives in a different
schema from the table it is indexing?
CG
__________________________________
Do you Yahoo!?
Friends. Fun. Try the all-new Yahoo! Messenger.
http://messenger.yahoo.com/
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Reply to all the recipients using the --to and --cc options:
reply via email
To: pgsql-sql@postgresql.org
Cc: cgg007@yahoo.com, tgl@sss.pgh.pa.us
Subject: Re: Schema + User-Defined Data Type Indexing problems...
In-Reply-To: <20040610132050.94570.qmail@web13801.mail.yahoo.com>
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
This inbox is served by DDX for PostgreSQL; see mirroring instructions
for how to clone and mirror all data and code used for this inbox