Re: subxcnt defined as signed integer in SnapshotData and SerializeSnapshotData

On Sat, May 9, 2015 at 8:48 AM, Michael Paquier
<michael.paquier@gmail.com> wrote:
> On Fri, May 8, 2015 at 10:27 PM, Simon Riggs wrote:
>> On 8 May 2015 at 13:02, Michael Paquier wrote:
>>> I think that we should redefine subxcnt as uint32 for consistency with
>>> xcnt, and remove the two assertions that 924bcf4 has introduced. I
>>> could get a patch quickly done FWIW.
>>
>> (uint32) +1
>
> Attached is the patch. This has finished by being far simpler than
> what I thought first.

I'm just going to remove the useless assertion for now.  What you're
proposing here may (or may not) be worth doing, but it carries a
non-zero risk of breaking something somewhere, if anyone is relying on
the signed-ness of that type.  Removing the assertion is definitely
safe.

-- 
Robert Haas
EnterpriseDB: http://www.enterprisedb.com
The Enterprise PostgreSQL Company