On Thu, Mar 5, 2015 at 10:28 PM, Andres Freund <andres@2ndquadrant.com> wrote:
> On 2015-03-05 12:14:04 +0000, Syed, Rahila wrote:
>> Please find attached a patch. As discussed, flag to denote
>> compression and presence of hole in block image has been added in
>> XLogRecordImageHeader rather than block header.
>
> FWIW, I personally won't commit it with things done that way. I think
> it's going the wrong way, leading to a harder to interpret and less
> flexible format. I'm not going to further protest if Fujii or Heikki
> commit it this way though.
I'm pretty sure that we can discuss the *better* WAL format even after
committing this patch.
Regards,
--
Fujii Masao