Re: Custom table AMs need to include heapam.h because ofBulkInsertState - Mailing list pgsql-hackers

From Andres Freund
Subject Re: Custom table AMs need to include heapam.h because ofBulkInsertState
Date
Msg-id 20190601191943.zarudpbynmac62m7@alap3.anarazel.de
Whole thread Raw
In response to Custom table AMs need to include heapam.h because of BulkInsertState  (Michael Paquier <michael@paquier.xyz>)
Responses Re: Custom table AMs need to include heapam.h because ofBulkInsertState  (Michael Paquier <michael@paquier.xyz>)
Re: Custom table AMs need to include heapam.h because of BulkInsertState  (Robert Haas <robertmhaas@gmail.com>)
List pgsql-hackers
Hi,

On 2019-06-01 15:09:24 -0400, Michael Paquier wrote:
> I have been playing lately with the table AM API to do some stuff, and
> I got surprised that in the minimum set of headers which needs to be
> included for a table AM we have a hard dependency with heapam.h for
> BulkInsertState and vacuum.h for VacuumParams.

I've noted this before as a future todo.


> I'd like to think that the best way to deal with that and reduce the
> confusion would be to move anything related to bulk inserts into their
> own header/file, meaning the following set:
> - ReleaseBulkInsertStatePin
> - GetBulkInsertState
> - FreeBulkInsertState
> There is the argument that we could also move that part into tableam.h
> itself though as some of the rather generic table-related callbacks,
> but that seems grotty.  So I think that we could just move that stuff
> as backend/access/common/bulkinsert.c.

Yea, I think we should do that at some point. But I'm not sure this is
the right design. Bulk insert probably needs to rather be something
that's allocated inside the AM.

Greetings,

Andres Freund



pgsql-hackers by date:

Previous
From: Michael Paquier
Date:
Subject: psql completion bugs with access methods
Next
From: Andres Freund
Date:
Subject: Re: Table AM callbacks referring to heap in declarations (+typos)