Postgres Pro
Downloads
Home   >   mailing lists

Thread: Gin index on array of uuid

Gin index on array of uuid

From
Enrique MailingLists
Date:

Currently creating an index on an array of UUID involves defining an operator class. I was wondering if this would be a valid request to add as part of the uuid-ossp extension? This seems like a reasonable operator to support as a default for UUIDs. Any downsides to adding this as a default?


This is the definition from the stack overflow reference:

CREATE OPERATOR CLASS _uuid_ops DEFAULT   FOR TYPE _uuid USING gin AS   OPERATOR 1 &&(anyarray, anyarray),   OPERATOR 2 @>(anyarray, anyarray),   OPERATOR 3 <@(anyarray, anyarray),   OPERATOR 4 =(anyarray, anyarray),   FUNCTION 1 uuid_cmp(uuid, uuid),   FUNCTION 2 ginarrayextract(anyarray, internal, internal),   FUNCTION 3 ginqueryarrayextract(anyarray, internal, smallint, internal, internal, internal, internal),   FUNCTION 4 ginarrayconsistent(internal, smallint, anyarray, integer, internal, internal, internal, internal),   STORAGE uuid;

This would be helpful for people trying to use arrays of UUIDs in cloud environments which limit root access.

Thank you,
Enrique

Re: Gin index on array of uuid

From
Peter Geoghegan
Date:
On Tue, Jun 21, 2016 at 11:42 AM, Enrique MailingLists
<enrique.mailing.lists@gmail.com> wrote:
> This would be helpful for people trying to use arrays of UUIDs in cloud
> environments which limit root access.

I have personally seen numerous requests for this from users of Heroku
Postgres. So, I agree that there is a demand for this.

-- 
Peter Geoghegan

Re: Gin index on array of uuid

From
Tom Lane
Date:
Enrique MailingLists <enrique.mailing.lists@gmail.com> writes:
> Currently creating an index on an array of UUID involves defining an
> operator class. I was wondering if this would be a valid request to add as
> part of the uuid-ossp extension? This seems like a reasonable operator to
> support as a default for UUIDs.

This makes me itch, really, because if we do this then we should logically
do it for every other add-on type.

It seems like we are not that far from being able to have just one GIN
opclass on "anyarray".  The only parts of this declaration that are
UUID-specific are the comparator function and the storage type, both of
which could be gotten without that much trouble, one would think.

> Any downsides to adding this as a default?

Well, it'd likely break things at dump/reload time for people who had
already created a competing "default for _uuid" opclass manually.  I'm not
entirely sure, but possibly replacing the core opclasses with a single one
that is "default for anyarray" could avoid such failures.  We'd have to
figure out ambiguity resolution rules.
        regards, tom lane

Re: Gin index on array of uuid

From
M Enrique
Date:
What's a good source code entry point to review how this is working for anyarray currently? I am new to the postgres code. I spend some time looking for it but all I found is the following (which I have not been able to decipher yet).

pasted1

Thank you,
Enrique



On Tue, Jun 21, 2016 at 12:20 PM Tom Lane <tgl@sss.pgh.pa.us> wrote:
Enrique MailingLists <enrique.mailing.lists@gmail.com> writes:
> Currently creating an index on an array of UUID involves defining an
> operator class. I was wondering if this would be a valid request to add as
> part of the uuid-ossp extension? This seems like a reasonable operator to
> support as a default for UUIDs.

This makes me itch, really, because if we do this then we should logically
do it for every other add-on type.

It seems like we are not that far from being able to have just one GIN
opclass on "anyarray".  The only parts of this declaration that are
UUID-specific are the comparator function and the storage type, both of
which could be gotten without that much trouble, one would think.

> Any downsides to adding this as a default?

Well, it'd likely break things at dump/reload time for people who had
already created a competing "default for _uuid" opclass manually.  I'm not
entirely sure, but possibly replacing the core opclasses with a single one
that is "default for anyarray" could avoid such failures.  We'd have to
figure out ambiguity resolution rules.

                        regards, tom lane
Attachment

Re: Gin index on array of uuid

From
Oleg Bartunov
Date:


On Wed, Jun 29, 2016 at 6:17 AM, M Enrique <enrique.mailing.lists@gmail.com> wrote:
What's a good source code entry point to review how this is working for anyarray currently? I am new to the postgres code. I spend some time looking for it but all I found is the following (which I have not been able to decipher yet).

 

pasted1

Thank you,
Enrique



On Tue, Jun 21, 2016 at 12:20 PM Tom Lane <tgl@sss.pgh.pa.us> wrote:
Enrique MailingLists <enrique.mailing.lists@gmail.com> writes:
> Currently creating an index on an array of UUID involves defining an
> operator class. I was wondering if this would be a valid request to add as
> part of the uuid-ossp extension? This seems like a reasonable operator to
> support as a default for UUIDs.

This makes me itch, really, because if we do this then we should logically
do it for every other add-on type.

It seems like we are not that far from being able to have just one GIN
opclass on "anyarray".  The only parts of this declaration that are
UUID-specific are the comparator function and the storage type, both of
which could be gotten without that much trouble, one would think.

> Any downsides to adding this as a default?

Well, it'd likely break things at dump/reload time for people who had
already created a competing "default for _uuid" opclass manually.  I'm not
entirely sure, but possibly replacing the core opclasses with a single one
that is "default for anyarray" could avoid such failures.  We'd have to
figure out ambiguity resolution rules.

                        regards, tom lane

Attachment

Re: Gin index on array of uuid

From
M Enrique
Date:
Thank you.

On Tue, Jun 28, 2016 at 11:06 PM Oleg Bartunov <obartunov@gmail.com> wrote:
On Wed, Jun 29, 2016 at 6:17 AM, M Enrique <enrique.mailing.lists@gmail.com> wrote:
What's a good source code entry point to review how this is working for anyarray currently? I am new to the postgres code. I spend some time looking for it but all I found is the following (which I have not been able to decipher yet).

 

pasted1

Thank you,
Enrique



On Tue, Jun 21, 2016 at 12:20 PM Tom Lane <tgl@sss.pgh.pa.us> wrote:
Enrique MailingLists <enrique.mailing.lists@gmail.com> writes:
> Currently creating an index on an array of UUID involves defining an
> operator class. I was wondering if this would be a valid request to add as
> part of the uuid-ossp extension? This seems like a reasonable operator to
> support as a default for UUIDs.

This makes me itch, really, because if we do this then we should logically
do it for every other add-on type.

It seems like we are not that far from being able to have just one GIN
opclass on "anyarray".  The only parts of this declaration that are
UUID-specific are the comparator function and the storage type, both of
which could be gotten without that much trouble, one would think.

> Any downsides to adding this as a default?

Well, it'd likely break things at dump/reload time for people who had
already created a competing "default for _uuid" opclass manually.  I'm not
entirely sure, but possibly replacing the core opclasses with a single one
that is "default for anyarray" could avoid such failures.  We'd have to
figure out ambiguity resolution rules.

                        regards, tom lane
Attachment

Re: Gin index on array of uuid

From
M Enrique
Date:
Ok, I think this is close as you point out.

I can see the GIN (2742 pg_am oid - access method) op classes defined explicitly for a lot of the core types in pg_opclass.h.

I can also see various GIN operator functions defined in pg_proc ("select proname, proargtypes  from pg_proc where proname like '%extract%' " and many comparison functions "select proname, proargtypes  from pg_proc where proname like '%cmp%' ").

It seems one way to do this is to add the missing remaining built in types such as UUID to pg_opclass.h.... but I am missing how this "DATA(insert(...))" operator class definition links to the four GIN functions (&&, @>, <@ and =). Is there a separate table or definition responsible for this linking? 

Taking _timestamp_ops as an example:
DATA(insert ( 2742 _timestamp_ops PGNSP PGUID 2745 1115 t 1114 ));
How does postgresql determine the GIN function bindings? Where does this function pointer resolution occur?

Thank you,
Enrique

pasted2

pasted3

pasted1



On Tue, Jun 21, 2016 at 12:20 PM Tom Lane <tgl@sss.pgh.pa.us> wrote:
Enrique MailingLists <enrique.mailing.lists@gmail.com> writes:
> Currently creating an index on an array of UUID involves defining an
> operator class. I was wondering if this would be a valid request to add as
> part of the uuid-ossp extension? This seems like a reasonable operator to
> support as a default for UUIDs.

This makes me itch, really, because if we do this then we should logically
do it for every other add-on type.

It seems like we are not that far from being able to have just one GIN
opclass on "anyarray".  The only parts of this declaration that are
UUID-specific are the comparator function and the storage type, both of
which could be gotten without that much trouble, one would think.

> Any downsides to adding this as a default?

Well, it'd likely break things at dump/reload time for people who had
already created a competing "default for _uuid" opclass manually.  I'm not
entirely sure, but possibly replacing the core opclasses with a single one
that is "default for anyarray" could avoid such failures.  We'd have to
figure out ambiguity resolution rules.

                        regards, tom lane
Attachment