Thanks a lot Bongjae!
Can I also ask you to add unit test? :)
Thanks.
WBR,
Alexey.
On 01/06/2012 03:41 AM, Bongjae Chang wrote:
> Hi Alexey,
>
> I committed the changes. It is issue #1169.
>
> HeapBuffer.java was also changed a little bit. ex) HeapBuffer#reset()
> will throw InvalidMarkException according to javadoc. I think the
> change doesn't have side-effect because I could find the usage case of
> Buffer#reset().
>
> Thanks.
>
> Regards,
> Bongjae Chang
>
> From: Oleksiy Stashok <oleksiy.stashok_at_oracle.com
> <mailto:oleksiy.stashok_at_oracle.com>>
> Reply-To: <dev_at_grizzly.java.net <mailto:dev_at_grizzly.java.net>>
> Date: Thu, 05 Jan 2012 15:10:57 +0100
> To: <dev_at_grizzly.java.net <mailto:dev_at_grizzly.java.net>>
> Subject: Re: About BuffersBuffer#mark() and reset()
>
> On 01/05/2012 03:00 PM, Bongjae Chang wrote:
>> Hi,
>>
>> BuffersBuffer#mark() and BuffersBuffer#reset() throw currently
>> UnsupportedOperationException.
>>
>> Could I know why they are not supported?
>>
>> When I parsed packets, sometimes the input buffer of my filter could
>> become to be BuffersBuffer type if there was the remainder.
>>
>> At that time, I got the exception if I used mark() and reset() methods.
>>
>> So I stored the position of BuffersBuffer when I needed mark() and
>> changed current position into the stored position when I needed
>> reset() temporarily(It is the similar behavior to the
>> HeapBuffer#mark() and reset()).
>>
>> If there were any issues about BuffersBuffer#mark() and reset(), I
>> think that I should also consider them again when I parses packets of
>> BuffersBuffer type.
>>
>> So I would like to know the reason.
> AFAIR there was no specific reason.
> You can implement it similar way it's implemented in HeapBuffer and
> commit the change.
>
> Thanks.
>
> WBR,
> Alexey.
>
>>
>> Thanks.
>>
>> Regards,
>> Bongjae Chang
>>
>>
>