Re: WAL format and API changes (9.5)

On 11/05/2014 11:30 AM, Andres Freund wrote:
> On 2014-11-04 18:33:34 +0200, Heikki Linnakangas wrote:
>> On 10/30/2014 06:02 PM, Andres Freund wrote:
>>> On 2014-10-29 10:24:20 +0200, Heikki Linnakangas wrote:
>>>> On 10/06/2014 02:29 PM, Andres Freund wrote:
>>>>> I've not yet really looked,
>>>>> but on a quick readthrough XLogInsertRecData() staying in xlog.c doesn't
>>>>> make me happy...
>>>>
>>>> Ok.. Can you elaborate?
>>>
>>> To me the split between xloginsert.c doing some of the record assembly,
>>> and xlog.c doing the lower level part of the assembly is just wrong.
>>
>> I moved the checks for bootstrap mode and xl_len == 0, from the current
>> XLogInsert to the new XLogInsert() function. I don't imagine that to be
>> enough address your feelings about the split between xlog.c and
>> xloginsert.c, but makes it a tiny bit clearer IMHO. I don't know what else
>> to do about it, as it feels very sensible to me as it is now. So unless you
>> object more loudly, I'm going to just go ahead with this and commit,
>> probably tomorrow.
>
> I'm still not particularly happy with the split. But I think this patch
> is enough of a win anyhow. So go ahead.

Committed. Now, to the main patch...

- Heikki