Re: tid_blockno() and tid_offset() accessor functions

Hello

On 2026-Apr-05, Andres Freund wrote:

> I was reminded of this thread while in the process of doing some performance
> validation that required getting all the block numbers from an index scan,
> which required me to remember (ctid::text::point)[0], which I find to be very
> ugly.

Thanks for this patch!  I agree that the statu quo was sort of embarrasing :-)

Regards

-- 
Álvaro Herrera               48°01'N 7°57'E  —  https://www.EnterpriseDB.com/