TextLineDecoder bug?

[James Im <im...@hotmail.com>]

I need your help to determine if I have found a bug or not.

I use a slightly modified version of TextLineDecoder and I got an error
in decodeNormal() at line 272 (the line number correspond to the version
on the trunk):

in.limit( in.limit() - matchCount + oldMatchCount );


After investigating a bit I found that I was trying to set the limit to
8193 which was greater than the capacity (8192). Thus the error.

I've investigated some more and I saw that it happened when the
matchCount passed to decodeNormal() was 1 instead of 0.

I've investigated some more and I think that oldMatchCount should be
reset to zero when you have found a match.

Thus, after line 261 I have added:
oldMatchCount=0;


After I did this change the problem was solved. The problem did not
occurred 100% of the time on my server and thus I have no specific test
case to reproduce the problem.
As it is hard to analyze buffer, position, limit, remaining and capacity
in loops and I therefore ask you to do a mental effort to verify if I
am right or wrong.

_________________________________________________________________
Del dine store filer uden problemer pĺ MSN Messenger:  
http://messenger.msn.dk/

Re: TextLineDecoder bug?

[Trustin Lee <tr...@gmail.com>]

Hi James,

On 1/23/07, James Im <im...@hotmail.com> wrote:
>
> I need your help to determine if I have found a bug or not.
>
> I use a slightly modified version of TextLineDecoder and I got an error
> in decodeNormal() at line 272 (the line number correspond to the version
> on the trunk):
>
> in.limit( in.limit() - matchCount + oldMatchCount );
>
>
> After investigating a bit I found that I was trying to set the limit to
> 8193 which was greater than the capacity (8192). Thus the error.
>
> I've investigated some more and I saw that it happened when the
> matchCount passed to decodeNormal() was 1 instead of 0.
>
> I've investigated some more and I think that oldMatchCount should be
> reset to zero when you have found a match.
>
> Thus, after line 261 I have added:
> oldMatchCount=0;


Thank you for reporting a bug.  I agree with you that it's a bug.  Don't we
need to reset oldMatchCount whenever we reset matchCount, instead of adding
one sentence in line 261?

It would also be the best if you can create a JIRA issue for us, and attach
a patch file.

Thanks,
Trustin
-- 
what we call human nature is actually human habit
--
http://gleamynode.net/
--
PGP key fingerprints:
* E167 E6AF E73A CBCE EE41  4A29 544D DE48 FE95 4E7E
* B693 628E 6047 4F8F CFA4  455E 1C62 A7DC 0255 ECA6