Robert Haas <robertmhaas@gmail.com> writes:
> On Thu, Jan 20, 2011 at 4:24 PM, Tom Lane <tgl@sss.pgh.pa.us> wrote:
>> If you're willing to substitute an incompatible table, it's not clear
>> why you don't just do
>>
>> begin;
>> drop table t;
>> alter table t_new rename to t;
>> commit;
> Because the whole source of this problem is dependency hell.
Well, if you want to preserve dependencies, you can *not* just blindly
substitute an incompatible table. You must ensure that views and
foreign keys referencing the table are still valid. So I'm not sure
where anybody got the idea that an implementation that fails to check
all that is even worth presenting.
regards, tom lane