Mailing-List: contact derby-dev-help@db.apache.org; run by ezmlm
Precedence: bulk
Reply-To: "Derby Development" <derby-dev@db.apache.org>
Received-SPF: neutral (hermes.apache.org: local policy)
Message-ID: <42235F98.5010204@apache.org>
Date: Mon, 28 Feb 2005 10:14:48 -0800
From: Jeremy Boynes <jboynes@apache.org>
User-Agent: Mozilla Thunderbird 0.8 (Windows/20040913)
MIME-Version: 1.0
To: Derby Development <derby-dev@db.apache.org>
Subject: Re: INTERSECT and EXCEPT Design
References: <421FBEC8.5020803@sbcglobal.net>
 <001801c51ba6$bc060780$1401a8c0@rpws002> <42235E21.1020606@sbcglobal.net>
In-Reply-To: <42235E21.1020606@sbcglobal.net>
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: 7bit

Jack Klebanoff wrote:
> RPost wrote:
> 
>> >Jack Klebanoff wrote:
>>  
>> >The syntax is (roughly):
>>
>> />queryExpression/ INTERSECT [ALL] /queryExpression/
>> />queryExpression/ EXCEPT [ALL] /queryExpression/
>>
>> Although DISTINCT is implicit if ALL is not specified the SQL standard 
>> also allows the DISTINCT operator to be specified explicitly. Would it 
>> make sense to do that in Derby also?
>>
>>      INTERSECT [ALL | DISTINCT] and EXCEPT [ALL | DISTINCT]
>>
> Other relational databases do not all support the "DISTINCT" keyword, so 
> applications that use it will not be portable.
> 

It is part of the spec so there is no reason not to support it; people 
seeking greater portability between (non-compliant) databases simply 
don't need to use it.

> That being said, I think that adding the "DISTINCT" keyword would be a 
> good project for a neophyte who wants to get his or her feet wet in the 
> Derby parser.
> 

I'll take it on as a first real commit (vs. patch).

--
Jeremy