From: | John McKown <john(dot)archie(dot)mckown(at)gmail(dot)com> |
---|---|
To: | Adrian Klaver <adrian(dot)klaver(at)aklaver(dot)com> |
Cc: | PostgreSQL General <pgsql-general(at)postgresql(dot)org> |
Subject: | Re: Enforce primary key on every table during dev? |
Date: | 2018-02-28 14:03:21 |
Message-ID: | CAAJSdjjvX9Lt4pAyANC-o9vcxLz-TKxEP6=y-b2AWc7ZB-KniA@mail.gmail.com |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-general |
On Wed, Feb 28, 2018 at 7:57 AM, Adrian Klaver <adrian(dot)klaver(at)aklaver(dot)com>
wrote:
> On 02/28/2018 05:52 AM, John McKown wrote:
>
>> On Wed, Feb 28, 2018 at 7:34 AM, Jeremy Finzel <finzelj(at)gmail(dot)com
>> <mailto:finzelj(at)gmail(dot)com>>wrote:
>>
>> We want to enforce a policy, partly just to protect those who might
>> forget, for every table in a particular schema to have a primary
>> key. This can't be done with event triggers as far as I can see,
>> because it is quite legitimate to do:
>>
>> BEGIN;
>> CREATE TABLE foo (id int);
>> ALTER TABLE foo ADD PRIMARY KEY (id);
>> COMMIT;
>>
>> It would be nice to have some kind of "deferrable event trigger" or
>> some way to enforce that no transaction commits which added a table
>> without a primary key.
>>
>> Any ideas?
>>
>> Thanks,
>> Jeremy
>>
>>
>>
>> What stops somebody from doing:
>>
>> CREATE TABLE foo (filler text primary key default null, realcol1 int,
>> realcol2 text);
>>
>> And then just never bother to ever insert anything into the column
>> FILLER? It fulfills your stated requirement of every table having a
>>
>
> Then you would get this:
>
> test=# CREATE TABLE foo (filler text primary key default null, realcol1
> int, realcol2 text);
> CREATE TABLE
> test=# insert into foo (realcol1, realcol2) values (1, 'test');
> ERROR: null value in column "filler" violates not-null constraint
> DETAIL: Failing row contains (null, 1, test).
>
Hum, it's been so long, I totally forgot. Which makes me wonder why the
parser doesn't "know" that a default of NULL for a primary key is going to
fail anyway and flag it at CREATE time. Oh, well. Thanks.
>
> primary key. Of course, you could amend the policy to say a "non-NULL
>> primary key".
>>
>>
>>
>> --
>> I have a theory that it's impossible to prove anything, but I can't prove
>> it.
>>
>> Maranatha! <><
>> John McKown
>>
>
>
> --
> Adrian Klaver
> adrian(dot)klaver(at)aklaver(dot)com
>
--
I have a theory that it's impossible to prove anything, but I can't prove
it.
Maranatha! <><
John McKown
From | Date | Subject | |
---|---|---|---|
Next Message | David G. Johnston | 2018-02-28 14:14:32 | Re: Enforce primary key on every table during dev? |
Previous Message | Melvin Davidson | 2018-02-28 14:01:34 | Re: Enforce primary key on every table during dev? |