Re: FW: Re: FW: Re: Shouldn;t this trigger be called?

On 9/16/19 12:55 PM, stan wrote:
>
> On Mon, Sep 16, 2019 at 12:44:49PM -0700, Adrian Klaver wrote:
>> On 9/16/19 11:53 AM, stan wrote:
>>> On Sun, Sep 15, 2019 at 09:16:35PM -0700, Adrian Klaver wrote:
>>>> On 9/15/19 6:04 PM, stan wrote:
>>>>> On Sun, Sep 15, 2019 at 12:27:14PM -0700, Adrian Klaver wrote:
>>>>>> On 9/15/19 10:46 AM, stan wrote:
>>>>>>> Forgot to cc the list again. Have to look at settings in mutt.
>>>>>>>
>>>>>>>>>>
>>>>
>>>>>> What validity check?
>>>>>>
>>>>>
>>>>> The check to see if it is the type enum.
>>>>>
>>>>
>>>> This would get solved a lot quicker if full information was provided:
>>>>
>>>> 1) Schema of the table.
>>>> Including associated triggers
>>>>
>>>> 2) The actual check code.
>>>>
>>>
>>> OK, please let me know if what I put in my reply to Tom Lane is not sufficient.
>>>
>>
>>
>> It was not sufficient, you did not include the table schema or the check
>> code.
>
> OK, understood here is the table:
>
> /* Contains one record for each customer */
>

> status activity_status NOT NULL DEFAULT 'ACTIVE',
> modtime timestamptz NOT NULL DEFAULT current_timestamp
> );
>
> I am not certain what you mean by the check. As you can see, there is nor
> specific check clause. I was referring to the built in check of data being
> entered versus the legal values for the user defined type. Make sense?
>

To be clear you are seeing an error because a value of say 'active' is
being rejected before your trigger has a chance to upper case it, correct?

Also this happens whether you use \copy or manually INSERT the values?

--
Adrian Klaver
adrian(dot)klaver(at)aklaver(dot)com