David Fetter wrote:
Folks,A co-worker pointed out to me that MySQL has a feature that, properly implemented and maybe extended, could be handy, namely what MySQL calls a "timestamp" field, so here's a proposal: 1. Create a generic (possibly overloaded) trigger function, bundled with PostgreSQL, which sets a field to some value. For example, a timestamptz version might set the field to now(). 2. Have some kind of pre-processing of CREATE and ALTER statements on tables which would attach the above function to the field at hand, something like: CREATE TABLE foo( last_updated TIMESTAMPTZ_UPDATED(), ... ); which would turn last_updated into a TIMESTAMPTZ with the expected behavior on UPDATEs. What do folks think of this idea?
Having the pre defined triggers at hand could be useful, especially for people not writing triggers so often to get used to it but I'm really not happy with the idea of magic preprocessing. I guess this is commonly used with timestamp fields so why not include a receipe to the docs under examples for timestamp which shows how to create and use a trigger? I may be wrong but my feeling is, not to much weirdness in the core please :) (I guess mysql had it because of lacking triggers and stuff for a long time?) T.
smime.p7s
Description: S/MIME Cryptographic Signature
