Deprecating ChannelPipelineCoverage?

Mike McGrady mmcgrady at topiatechnology.com
Sat Nov 21 13:40:48 EST 2009 

I too think there is not really a reason to change the status quo and  
agree with these reasons,

Sent from my iPhone

On Nov 21, 2009, at 2:16 AM, jan de vries <jandevries145 at fastmail.fm>  
wrote:

> Not that i think i'm the best judge in this but for what its worth:
>
> 1) I think that might be a little better than  
> ChannelPipelineCoverage but i don't think
> ChannelPipelineCoverage was that unclear with the documentation.
> 2)It's a handy feature if you want to store extra info in the  
> handler and if you want
> to go for minimal memory usage you can easily use shared handlers.  
> As far as
> difficulty is concerned,  the documentation is very clear. Although  
> it is a feature
> which you cant grasp by just looking at the name  alone, its still  
> very usefull and forces
> the user a little to think about which data is stored in the  
> handler. The idea of the feature
> is good, it fits in the scope of the  framework and with the  
> documentation its more
> then clear enough.
>
> 1b)  Since i dont think Shared is way better then  
> ChannelPipelineCoverage, i tend
> to prefer to keep the things as they were, but either way wont be  
> much of an issue.
>
> PS
>
> I have been away for a while but i'm glad to see netty is still  
> going very
> strong. The project i worked on with netty (involving a flash game  
> server) sort
> of collapsed on itself purely because of personal arguments between  
> team
> members. I was to busy developing while others were busy arguing :(.  
> That kinda
> totally sucked since it worked very well technically. Anyway, i  
> gathered
> the old work and i'm giving a second live to it. I hope  it will be  
> more successful then
> the first attempt.  I just want to send my special thanks to Trustin  
> who made netty
> possible and of course the posters and especially the loyal ones who  
> help Trustin
> and all of us by testing and sharing ideas. I think netty is one of  
> the gems in the
> java-libraries.
>
> regards,
> jan
>
> Trustin Lee (이희승) wrote:
>>
>> Before making the final decision ...
>>
>> 1) (@Shared and @Unshared) or (@SharedHandler vs @UnsharedHandler)?
>> 2) Would it be a better idea to just get rid of
>> @ChannelPipelineCoverage without replacement?
>>
>> -- Trustin Lee, http://gleamynode.net/
>>
>> On Mon, Nov 9, 2009 at 8:12 PM, Christian Migowski <chrismfwrd at gmail.com 
>> > wrote:
>>
>>> 2009/11/9 Trustin Lee (이희승) <trustin at gmail.com>:
>>>
>>>> Hey Iain,
>>>>
>>>>
>>>>
>>>> On Mon, Nov 9, 2009 at 7:45 PM, Iain McGinniss  
>>>> <iainmcgin at gmail.com> wrote:
>>>>
>>>>> How about @SharedHandler and @UnsharedHandler? @Shared describes
>>>>> Mike's usage, where he has a handler used by multiple pipelines  
>>>>> but
>>>>> within which he has implemented the appropriate logic to avoid
>>>>> threading issues.
>>>>>
>>>> I love this idea.
>>>>
>>> I also think this is a good suggestion, better then Stateful and  
>>> Stateless.
>>>
>>> regards,
>>> christian!
>>>
>>>
>>>>
>>>>> A side question: does netty actually check these annotations at
>>>>> runtime, to make sure an unshared handler instance is not used by
>>>>> multiple pipelines?
>>>>>
>>>> Currently only for documentation purpose, but I'd like to add some
>>>> runtime check.
>>>>
>>>> Thanks!
>>>>
>>>>
>>>>> Iain
>>>>>
>>>>> On 9 Nov 2009, at 00:12, Trustin Lee (이희승) wrote:
>>>>>
>>>>>
>>>>>> Good point.  Then, do you have some suggestion for a better  
>>>>>> name than
>>>>>> ChannelPipelineCoverage, or do you think it's just fine?  I think
>>>>>> there might be a better name that makes it easier to understand.
>>>>>>
>>>>>> Thanks
>>>>>>
>>>>>> -- Trustin Lee, http://gleamynode.net/
>>>>>>
>>>>>>
>>>>>>
>>>>>> On Mon, Nov 9, 2009 at 7:00 AM, Mike McGrady
>>>>>> <mmcgrady at topiatechnology.com> wrote:
>>>>>>
>>>>>>> Hi, Trustin,
>>>>>>>
>>>>>>> I think that the present annotations are more explicit.  The  
>>>>>>> present
>>>>>>> annotations say what is actually happening.  The stateful and
>>>>>>> stateless annotations would not actually fit what is happening  
>>>>>>> with
>>>>>>> my
>>>>>>> code.  I use stateful code with "all" and this is not a  
>>>>>>> problem.  If
>>>>>>> the annotation were changed I would have "stateless" as the
>>>>>>> annotation
>>>>>>> but have a stateful handler.
>>>>>>>
>>>>>>> Mike
>>>>>>>
>>>>>>> On Nov 8, 2009, at 1:24 PM, Trustin Lee (이희승) wrote:
>>>>>>>
>>>>>>>
>>>>>>>> What do you think about replacing ChannelPipelineCoverage  
>>>>>>>> with more
>>>>>>>> explicit annotations such as @Stateful and @Stateless?
>>>>>>>>
>>>>>>>> ChannelPipelineCoverage has a confusing name (at least for a
>>>>>>>> beginner), and its value cannot be validated on compile time  
>>>>>>>> because
>>>>>>>> it's string.
>>>>>>>>
>>>>>>>> -- Trustin Lee, http://gleamynode.net/
>>>>>>>>
>>>>>>>> _______________________________________________
>>>>>>>> netty-dev mailing list
>>>>>>>> netty-dev at lists.jboss.org
>>>>>>>> https://lists.jboss.org/mailman/listinfo/netty-dev
>>>>>>>>
>>>>>>> Mike McGrady
>>>>>>> Principal Investigator AF081-028 AFRL SBIR
>>>>>>> Senior Engineer
>>>>>>> Topia Technology, Inc.
>>>>>>> 1.253.720.3365
>>>>>>> mmcgrady at topiatechnology.com
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>> _______________________________________________
>>>>>>> netty-dev mailing list
>>>>>>> netty-dev at lists.jboss.org
>>>>>>> https://lists.jboss.org/mailman/listinfo/netty-dev
>>>>>>>
>>>>>>>
>>>>>> _______________________________________________
>>>>>> netty-dev mailing list
>>>>>> netty-dev at lists.jboss.org
>>>>>> https://lists.jboss.org/mailman/listinfo/netty-dev
>>>>>>
>>>>> _______________________________________________
>>>>> netty-dev mailing list
>>>>> netty-dev at lists.jboss.org
>>>>> https://lists.jboss.org/mailman/listinfo/netty-dev
>>>>>
>>>>>
>>>> -- Trustin Lee, http://gleamynode.net/
>>>>
>>>> _______________________________________________
>>>> netty-dev mailing list
>>>> netty-dev at lists.jboss.org
>>>> https://lists.jboss.org/mailman/listinfo/netty-dev
>>>>
>>>>
>>> _______________________________________________
>>> netty-dev mailing list
>>> netty-dev at lists.jboss.org
>>> https://lists.jboss.org/mailman/listinfo/netty-dev
>>>
>>>
>>
>> _______________________________________________
>> netty-dev mailing list
>> netty-dev at lists.jboss.org
>> https://lists.jboss.org/mailman/listinfo/netty-dev
>>
>
> _______________________________________________
> netty-dev mailing list
> netty-dev at lists.jboss.org
> https://lists.jboss.org/mailman/listinfo/netty-dev
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.jboss.org/pipermail/netty-dev/attachments/20091121/5d02c9cb/attachment.html

More information about the netty-dev mailing list