Tom Lane wrote:
> KaiGai Kohei <kaigai@ak.jp.nec.com> writes:
> > I'll try your approash in first, as follows:
>
> This seems a vast amount of uglification to avoid adding an argument to
> CreateTemplateTupleDesc. We do that kind of thing all the time --- it
> is a simple and reliable change to make.
>
> When designing a patch, you should generally try to make the code look
> like the patch has been there all along. Contorting logic to avoid
> a simple API change is not good.
Just to chime in, I agree with Simon's direction of making the security
specification for the table match WITH OIDS, and agree with Tom that the
implementation should follow the WITH OIDS API for clarity, not trying
to reduce the change footprint. Basically, if WITH OIDS and security
definer behave the same in the API, there is little additional code
_complexity_, even if the patch is now larger.
-- Bruce Momjian <bruce@momjian.us> http://momjian.us EnterpriseDB
http://enterprisedb.com
+ If your life is a hard drive, Christ can be your backup. +
