phobos commit, revision 2047


user: andrei

msg:
interim fix for bug 941

http://www.dsource.org/projects/phobos/changeset/2047

On 9/25/2010 1:38 AM, dsource.org wrote:
> phobos commit, revision 2047
> 
> 
> user: andrei
> 
> msg:
> interim fix for bug 941
> 
> http://www.dsource.org/projects/phobos/changeset/2047

Either leave this bug open or file another report to track the real fix. Otherwise it'll just get lost.

I consider it fixed.

Andrei

On 9/25/10 4:12 CDT, Brad Roberts wrote:
> On 9/25/2010 1:38 AM, dsource.org wrote:
>> phobos commit, revision 2047
>>
>>
>> user: andrei
>>
>> msg:
>> interim fix for bug 941
>>
>> http://www.dsource.org/projects/phobos/changeset/2047
>
> Either leave this bug open or file another report to track the real fix.
> Otherwise it'll just get lost.
> _______________________________________________
> phobos mailing list
> phobos at puremagic.com
> http://lists.puremagic.com/mailman/listinfo/phobos

Really?  Disabling optimization, calling it an interim fix, and suggesting that Walter can re-enable the optimization?  Those are all signs of not-fixed to me.

On 9/25/2010 6:16 AM, Andrei Alexandrescu wrote:
> I consider it fixed.
> 
> Andrei
> 
> On 9/25/10 4:12 CDT, Brad Roberts wrote:
>> On 9/25/2010 1:38 AM, dsource.org wrote:
>>> phobos commit, revision 2047
>>>
>>>
>>> user: andrei
>>>
>>> msg:
>>> interim fix for bug 941
>>>
>>> http://www.dsource.org/projects/phobos/changeset/2047
>>
>> Either leave this bug open or file another report to track the real fix. Otherwise it'll just get lost.

On 9/25/10 12:25 CDT, Brad Roberts wrote:
> Really?  Disabling optimization, calling it an interim fix, and suggesting that Walter can re-enable the optimization?  Those are all signs of not-fixed to me.

Brad, it is a fix in the sense that it makes the test work as expected and has no regression. I pored over the code for a couple of hours and I couldn't get into it. The bug is more than two years old, and I did the best I could to fix it. It now is fixed.

If you feel that regex has a performance issue, feel free to submit a new report - or better yet, figure out how to fix optimize().


Andrei

On Sat, 25 Sep 2010, Andrei Alexandrescu wrote:

> On 9/25/10 12:25 CDT, Brad Roberts wrote:
> > Really?  Disabling optimization, calling it an interim fix, and suggesting
> > that
> > Walter can re-enable the optimization?  Those are all signs of not-fixed to
> > me.
> 
> Brad, it is a fix in the sense that it makes the test work as expected and has no regression. I pored over the code for a couple of hours and I couldn't get into it. The bug is more than two years old, and I did the best I could to fix it. It now is fixed.
> 
> If you feel that regex has a performance issue, feel free to submit a new report - or better yet, figure out how to fix optimize().
> 
> Andrei

Calm down. I don't understand the defensiveness.  I saw a checkin that asked for follow up and suggested adding a ticket to track it.  The submit does work around the reported bug, which is all good by itself.  I'm not personally affected either way and don't have a vested interest in the bug.  I do have a desire to help keep us from loosing track of things and incrementally improve the d ecosystem.

Sigh,
Brad