On 04/02/2016 01:20 PM, Tom Lane wrote:
> Andrew Dunstan <andrew@dunslane.net> writes:
>> Looking at this briefly. It looks like the check should be called from
>> enum_in() and enum_recv(). What error should be raised if the enum row's
>> xmin isn't committed? ERRCODE_FEATURE_NOT_SUPPORTED? or maybe
>> ERRCODE_DATA_EXCEPTION? I don't see anything that fits very well.
> ERRCODE_OBJECT_NOT_IN_PREREQUISITE_STATE is something we use in some
> other places where the meaning is "just wait awhile, dude". Or you
> could invent a new ERRCODE.
>
>
OK, did that. Here is a patch that is undocumented but I think is
otherwise complete. It's been tested a bit and we haven't been able to
break it. Comments welcome.
cheers
andrew