Re: Allowing extensions to find out the OIDs of their member objects - Mailing list pgsql-hackers

From Chapman Flack
Subject Re: Allowing extensions to find out the OIDs of their member objects
Date
Msg-id 5C468D73.70902@anastigmatix.net
Whole thread Raw
In response to Re: Allowing extensions to find out the OIDs of their member objects  (Tom Lane <tgl@sss.pgh.pa.us>)
List pgsql-hackers
On 01/21/19 21:45, Tom Lane wrote:
> are concerned, but for cross-extension references it seems a
> lot better to be looking for "postgis / function_foo_int_int"
> than for "postgis / 3".

I wonder if postgis might offer a .h file with FUNCTION_POSTGIS_FOO_INT_INT
defined as 3, which extensions intending to use foo could be built against.

Any that aren't could still search by name and signature.

In the case of calls from core to some pluggable extension, of course
the .h file would be in core, with the implementing extensions expected
to build against it: in thy extension shalt thou provide XMLCAST at index 1,
XMLITERATE at index 2, etc.

Regards,
-Chap


pgsql-hackers by date:

Previous
From: Bruce Momjian
Date:
Subject: Re: Record last password change
Next
From: Andreas Karlsson
Date:
Subject: Re: Early WIP/PoC for inlining CTEs