Re: rewriter in updateable views

From: Bernd Helmle <mailings(at)oopsware(dot)de>
To: Alvaro Herrera <alvherre(at)dcc(dot)uchile(dot)cl>, Bernd Helmle <mailings(at)oopsware(dot)de>
Cc: Simon Riggs <simon(at)2ndquadrant(dot)com>, Jaime Casanova <systemguards(at)gmail(dot)com>, pgsql-hackers(at)postgresql(dot)org
Subject: Re: rewriter in updateable views
Date: 2005-03-30 16:58:16
Message-ID: FC37604891A658BC5F1AFB35@sparkey.oopsware.intra
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

--On Mittwoch, März 30, 2005 11:35:05 -0400 Alvaro Herrera
<alvherre(at)dcc(dot)uchile(dot)cl> wrote:

> On Tue, Mar 29, 2005 at 07:30:13PM +0200, Bernd Helmle wrote:
>
>> Well, the patch is far away from being ready for -patches, but if you
>> want to "preview" you can get our latest patch against HEAD at
>>
>> http://www.oopsware.de/pgsql_viewupdate.html
>
> Well, that description there says what the patch doesn't do, but it
> would be helpful to say what does it do.

Implemented is:

- Create rules for SQL92-compliant updateable views. This creates implicit
rules
for DELETE, UPDATE and INSERT actions named
_DELETE,
_DELETE_NOTHING,
_UPDATE,
_UPDATE_NOTHING,
_INSERT and
_INSERT_NOTHING

and marks them with a new column in pg_rewrite as an impicit created update
rule.
The patch supports system columns and functions in the view's target list
rudimentary and allows CASCADED updates to base tables, since rules are
created in cascaded manner on the underlying relations (which is changed,
when we cover LOCAL | CASCADED CHECK OPTION). The WITH CHECK OPTION is
implemented and creates all mentioned update rules with a rule condition
added, which calls a new system function that elogs() an error if the
passed view condition is not evaluated to true.
I'm sure not all corner cases are covered for this yet, but it's still a
work in progress....

--

Bernd

In response to

Browse pgsql-hackers by date

  From Date Subject
Next Message Andrew Dunstan 2005-03-30 18:24:48 Re: minor windows & cygwin regression failures on stable
Previous Message Alvaro Herrera 2005-03-30 15:35:05 Re: rewriter in updateable views