Hi,

2010/3/28 NAKAMURA, Hiroshi <nakahiro / gmail.com>:
> On Sun, Mar 28, 2010 at 12:44, Yusuke ENDOH <mame / tsg.ne.jp> wrote:
>>> It would be better off waiting for Christian's response.
>>
>> When you revert the patch, he can complain if he is not satisfied
>> with the revert.
>> But I believe Christian will not complain about how to fix, unless
>> SEGV occurs.
>
> I'm not waiting for the confirmation. I'm asking Christian for
> guessing how the "fix" you committed is correct.

I cannot get your point.
I just commited the patch written by Christian himself.  Why
do we need his guessing?


>>> Would you please reopen this ticket, Yusuke?
>>
>> In terms of release management, I don't want to leave bug ticket
>> open just because of such a confirmation.
>
> Closing a ticket too soon generally is not good for release management
> I think. But I'm not a member of 1.9.1 development so it's your
> business.

Currently, there is a convention that a ticket is closed when
a commit is done.
In fact, redmine even has an automatic mechanism that closes
ticket when any committer writes ML reference to ChangeLog in
trunk.  (The mechanism does not work sometimes, though.)

It is good to discuss the convention, but I prefer the current
one.

Indeed, it is better to ask original poster to confirm whether
his/her issue is actually solved or not before closing tickets.
But it requires human-resource to manage tickets.  We cannot
afford to do so.


> I don't think it should be done but you can close these tickets as well:
>  http://redmine.ruby-lang.org/issues/show/3000
>  http://redmine.ruby-lang.org/issues/show/2596
>  http://redmine.ruby-lang.org/issues/show/1678
>  http://redmine.ruby-lang.org/issues/show/1142
>  http://redmine.ruby-lang.org/issues/show/405

Thank you.  I couldn't identify some of these tickets.

- #3000 and #1678 are already closed.
- I'm not sure whether #2596 is really the same problem.
  I'll close it later unless there is objection.
- #1142 needs backport to ruby_1_9_1.
- #405 needs backport to ruby_1_8_7.

-- 
Yusuke ENDOH <mame / tsg.ne.jp>