Mailing-List: contact commons-user-help@jakarta.apache.org; run by ezmlm
Precedence: bulk
Reply-To: "Jakarta Commons Users List" <commons-user@jakarta.apache.org>
Received-SPF: neutral (hermes.apache.org: local policy)
Message-ID: <41586E90.5070708@apache.org>
Date: Mon, 27 Sep 2004 12:48:32 -0700
From: "matthew.hawthorne" <matth@apache.org>
User-Agent: Mozilla Thunderbird 0.8 (Windows/20040913)
MIME-Version: 1.0
To: Jakarta Commons Users List <commons-user@jakarta.apache.org>
Subject: Re: [Pool] Use of Exception in API
References: <32935.62.251.120.66.1095713399.squirrel@webmail.bibit.com>
 <20040927113231.G46335@minotaur.apache.org>
In-Reply-To: <20040927113231.G46335@minotaur.apache.org>
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: 7bit

Rodney Waldhoff wrote:
> Fundamentally this is a drawback to a strongly typed, checked exception 
> system.  I don't see how "replacing Exception with more specific 
> exceptions" could possibly work in the general case.


It at least would allow clients to avoid catching NullPointerExceptions, 
IllegalArgumentExceptions, and other RuntimeExceptions.

I've always disliked 'throws Exception' for this reason, and I think 
that creating a PoolException or something similar would be a good solution.

Unfortunately, this would change the public API and is probably 
impossible.  I've also barely used [pool] so I may be missing some more 
serious issues.

---------------------------------------------------------------------
To unsubscribe, e-mail: commons-user-unsubscribe@jakarta.apache.org
For additional commands, e-mail: commons-user-help@jakarta.apache.org