Re: [HACKERS] rule/notify bug?

From: Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
To: Brett McCormickS <brett(at)abraxas(dot)scene(dot)com>
Cc: pgsql-hackers(at)hub(dot)org
Subject: Re: [HACKERS] rule/notify bug?
Date: 1998-05-01 22:23:23
Message-ID: 28730.894061403@sss.pgh.pa.us
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

Brett McCormickS <brett(at)abraxas(dot)scene(dot)com> writes:
> create rule radius1 as on update to user where (current.usrppp <> new.usrppp) do notify radius;

> update user set usrname = 'Brett McCormick' where usrid = 'brett';
> NOTIFY

> the notification comes through on the radius relation
> (which, interestingly, doesn't exist)

Yup, that's what you told it to do: "notify radius". Listen/notify
names are really arbitrary identifiers, not relation names. This
is a good thing: you can signal conditions that aren't tightly
tied to a single table.

> should I be filling out a bug report? ;)

It's not a bug, it's a feature ;-)

The listen/notify documentation could be clearer about this, though.
Maybe it's a documentation bug.

regards, tom lane

Browse pgsql-hackers by date

  From Date Subject
Next Message Thomas G. Lockhart 1998-05-02 05:08:45 Re: [HACKERS] actual patch for close_ps() [file: geo_ops.c]
Previous Message Brett McCormickS 1998-05-01 20:45:32 rule/notify bug?