Neil Conway <neilc(at)samurai(dot)com> writes:
>> (b) it can share infrastructure with the other RI triggers.
> Such as? I don't see anything it allows us to share.
Note the ri_BuildQueryKeyFull call, and the arguments thereto.
As to the notational convenience issue, I think it's good that
RI_FKey_keyequal_upd looks almost exactly like the other RI trigger
functions. Having it look different in order to save a couple lines
in trigger.c doesn't seem like a win to me.
regards, tom lane
In response to
pgsql-hackers by date
|Next:||From: Tom Lane||Date: 2005-05-27 15:12:06|
|Subject: Re: A 2 phase commit weirdness |
|Previous:||From: Mark Cave-Ayland||Date: 2005-05-27 15:01:00|
|Subject: Re: Cost of XLogInsert CRC calculations|