Re: Regression in COPY FROM caused by 9f8377f7a2

From: Andrew Dunstan <andrew(at)dunslane(dot)net>
To: Laurenz Albe <laurenz(dot)albe(at)cybertec(dot)at>, pgsql-hackers(at)lists(dot)postgresql(dot)org
Subject: Re: Regression in COPY FROM caused by 9f8377f7a2
Date: 2023-09-25 15:06:58
Message-ID: b815e4b1-8d8c-1822-ddd0-0a3c8b56c768@dunslane.net
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers


On 2023-09-25 Mo 04:59, Laurenz Albe wrote:
> On Mon, 2023-09-25 at 09:54 +0200, Laurenz Albe wrote:
>> In v16 and later, the following fails:
>>
>> CREATE TABLE boom (t character varying(5) DEFAULT 'a long string');
>>
>> COPY boom FROM STDIN;
>> ERROR:  value too long for type character varying(5)
>>
>> In PostgreSQL v15 and earlier, the COPY statement succeeds.
>>
>> The error is thrown in BeginCopyFrom in line 1578 (HEAD)
>>
>>   defexpr = expression_planner(defexpr);
>>
>> Bisecting shows that the regression was introduced by commit 9f8377f7a2,
>> which introduced DEFAULT values for COPY FROM.

Oops :-(

> I suggest the attached fix, which evaluates default values only if
> the DEFAULT option was specified or if the column does not appear in
> the column list of COPY.
>

Patch looks reasonable, haven't tested yet.

cheers

andrew

--
Andrew Dunstan
EDB:https://www.enterprisedb.com

In response to

Responses

Browse pgsql-hackers by date

  From Date Subject
Next Message Andrew Dunstan 2023-09-25 15:07:57 Re: Doc: vcregress .bat commands list lacks "taptest"
Previous Message jacktby jacktby 2023-09-25 15:04:18 Why need a lock?