Mailing-List: contact dev-help@cayenne.apache.org; run by ezmlm
Precedence: bulk
Reply-To: dev@cayenne.apache.org
Received-SPF: pass (athena.apache.org: local policy)
Mime-Version: 1.0 (Apple Message framework v752.3)
In-Reply-To: <C357A9EE.FBFA%kmenard@servprise.com>
References: <C357A9EE.FBFA%kmenard@servprise.com>
Content-Type: text/plain; charset=US-ASCII; delsp=yes; format=flowed
Message-Id: <6C5249E3-1E73-46D6-BAE5-9E5709B53E85@objectstyle.org>
Content-Transfer-Encoding: 7bit
From: Andrus Adamchik <andrus@objectstyle.org>
Subject: Re: sorting and null in the path
Date: Thu, 8 Nov 2007 00:51:58 +0200
To: dev@cayenne.apache.org

My other purely semantic suggestion would be to move  
ObjPathResolvingException out of the root package. Maybe to the  
'reflect' package? And maybe call it something more user friendly  
(not sure what exactly ... NullPathException, ???)

Andrus

On Nov 8, 2007, at 12:44 AM, Kevin Menard wrote:

> Ahh, I hadn't even noticed the domain.  Cool.
>
> I generally like the idea.  Although, it would be nicer to have a  
> bit more
> control.  E.g., discard nulls or group them all together at the end.
>
> As a suggested code cleanup, the following:
>
> e.getCause() != null && e.getCause()
> instanceof org.apache.cayenne.ObjPathResolvingException
>
> Could be simplified to just the instanceof operation (it's null  
> safe and
> returns false if the argument is null).
>
> -- 
> Kevin
>
>
> On 11/7/07 5:28 PM, "Aristedes Maniatis" <ari@maniatis.org> wrote:
>
>>
>> On 08/11/2007, at 9:22 AM, Kevin Menard wrote:
>>
>>> Please submit the patch through JIRA.  By submitting through JIRA
>>> you can
>>> click a button that grants the ASF a non-exclusive license to use
>>> the code.
>>> This is a prerequisite for the code to get committed.
>>
>>
>> I'll take care of all that with Marcin and get him to sign a CLA. We
>> should be expecting more contributions from Marcin in the future.
>>
>> I'll also work with him to clean up the code, the naming and comment
>> it a bit, but what do you think of the concept? An extra switch in
>> Ordering which allows us to ignore paths that don't full resolve? Is
>> there a better way?
>>
>> Ari
>
>
>