Subject: Re: PostgreSQL CHARACTER VARYING vs CHARACTER VARYING (Length)
To: pgsql-admin@lists.postgresql.org
References: 
 <CAOasRJYCUwMYdd=BL29TupW_iCbvM3YRM01YTB_WYSUEays3CA@mail.gmail.com>
 <CAExHW5vcOh-X-cz_Z=pWxqCi1Bm3FmGZz_UtxT+DFCg1PPaGmA@mail.gmail.com>
 <c0149b4b-60d6-a1da-b068-05719d74a8dc@jakobs.com>
From: Ron <ronljohnsonjr@gmail.com>
Message-ID: <8d6eface-4bde-b165-34f1-eeaefe88a6aa@gmail.com>
Date: Tue, 28 Apr 2020 11:39:52 -0500
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:68.0) Gecko/20100101
 Thunderbird/68.4.1
MIME-Version: 1.0
In-Reply-To: <c0149b4b-60d6-a1da-b068-05719d74a8dc@jakobs.com>
Content-Type: text/plain; charset=utf-8; format=flowed
Content-Transfer-Encoding: 8bit
Content-Language: en-US
Precedence: bulk

An example:

test=# create table bargle (f1 varchar(10));
CREATE TABLE
test=# insert into bargle values ('01234567890123');
ERROR:  value too long for type character varying(10)


On 4/28/20 10:10 AM, Holger Jakobs wrote:
> Truncation will NEVER happen. PostgreSQL throws an ERROR on any attempt of 
> saving more characters (not bytes!) into a VARCHAR(50) column.
>
> There is some other well-known system which silently truncates, but we all 
> know why we would never use that.
>
> Am 28.04.20 um 13:46 schrieb Ashutosh Bapat:
>> On Tue, Apr 28, 2020 at 2:53 PM Rajin Raj <rajin.raj@opsveda.com> wrote:
>>> Is there any impact of using the character varying without providing the 
>>> length while creating tables?
>>> I have created two tables and inserted 1M records. But I don't see any 
>>> difference in pg_class. (size, relpage)
>>>
>>> create table test_1(name varchar);
>>> create table test_2(name varchar(50));
>> I don't think there's a difference in the way these two are stored
>> on-disk. But if you know that your strings will be at most 50
>> characters long, better set that limit so that server takes
>> appropriate action (i.e. truncates the strings to 50).
>>

-- 
Angular momentum makes the world go 'round.