Set with nil is fixed and waiting for a nice reviewer

> On Sun, 14 Mar 2010, Stéphane Ducasse wrote:
>
>> during the sprint niko schwarz and oscar nierstrasz fixed the Set in nil entry.
>> Thanks.
>> http://code.google.com/p/pharo/issues/detail?id=1907
>
> Why did you choose Igor's negative tally hack instead of the SetElement version?
>
>
> Levente
>
>>
>> Now does somebody with knowledge could have a look because this is a sensitive
>> part of the system and more eyes does not hurt.
>>
>> Stef
>> _______________________________________________
>> Pharo-project mailing list
>> [hidden email]
>> http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project
> _______________________________________________
> Pharo-project mailing list
> [hidden email]
> http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project

> On Sun, 14 Mar 2010, Stéphane Ducasse wrote:
>
>> during the sprint niko schwarz and oscar nierstrasz fixed the Set in nil
>> entry.
>> Thanks.
>> http://code.google.com/p/pharo/issues/detail?id=1907
>
> Why did you choose Igor's negative tally hack instead of the SetElement
> version?
>

>
> Levente
>
>>
>> Now does somebody with knowledge could have a look because this is a
>> sensitive
>> part of the system and more eyes does not hurt.
>>
>> Stef
>> _______________________________________________
>> Pharo-project mailing list
>> [hidden email]
>> http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project
>
> _______________________________________________
> Pharo-project mailing list
> [hidden email]
> http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project
>

> 2010/3/14 Levente Uzonyi<[hidden email]>:
>> On Sun, 14 Mar 2010, Stéphane Ducasse wrote:
>>
>>> during the sprint niko schwarz and oscar nierstrasz fixed the Set in nil
>>> entry.
>>> Thanks.
>>> http://code.google.com/p/pharo/issues/detail?id=1907
>> Why did you choose Igor's negative tally hack instead of the SetElement
>> version?
>>
> I'd prefer to use SetElement, unless there was a decision made, to use
> negative tally.
> We had a quite big discussion around it on squeak-dev, and most
> developers said that SetElement
> is the way to go, because it is clean, OO-oriented, even if somewhat
> slower comparing to other approaches.
>

> On 15.03.2010 00:18, Igor Stasenko wrote:
>> 2010/3/14 Levente Uzonyi<[hidden email]>:
>>> On Sun, 14 Mar 2010, Stéphane Ducasse wrote:
>>>
>>>> during the sprint niko schwarz and oscar nierstrasz fixed the Set in nil
>>>> entry.
>>>> Thanks.
>>>> http://code.google.com/p/pharo/issues/detail?id=1907
>>> Why did you choose Igor's negative tally hack instead of the SetElement
>>> version?
>>>
>> I'd prefer to use SetElement, unless there was a decision made, to use
>> negative tally.
>> We had a quite big discussion around it on squeak-dev, and most
>> developers said that SetElement
>> is the way to go, because it is clean, OO-oriented, even if somewhat
>> slower comparing to other approaches.
>>
> +1
>
> _______________________________________________
> Pharo-project mailing list
> [hidden email]
> http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project

> On Mon, 15 Mar 2010, Stéphane Ducasse wrote:
>
>> so may be the bug entry should have stated that because oscar and nico spend a day on it.
>
> The good news is that the tests are ready which is probably the larger part. The bad news is that integrating the tally hack is easier, than the SetElement version (which is also Igor's code, but that may not be clear from my previous mail).
>
>
> Levente
>
>>
>> Stef
>>
>> On Mar 15, 2010, at 12:56 AM, Henrik Sperre Johansen wrote:
>>
>>> On 15.03.2010 00:18, Igor Stasenko wrote:
>>>> 2010/3/14 Levente Uzonyi<[hidden email]>:
>>>>> On Sun, 14 Mar 2010, Stéphane Ducasse wrote:
>>>>>
>>>>>> during the sprint niko schwarz and oscar nierstrasz fixed the Set in nil
>>>>>> entry.
>>>>>> Thanks.
>>>>>> http://code.google.com/p/pharo/issues/detail?id=1907
>>>>> Why did you choose Igor's negative tally hack instead of the SetElement
>>>>> version?
>>>>>
>>>> I'd prefer to use SetElement, unless there was a decision made, to use
>>>> negative tally.
>>>> We had a quite big discussion around it on squeak-dev, and most
>>>> developers said that SetElement
>>>> is the way to go, because it is clean, OO-oriented, even if somewhat
>>>> slower comparing to other approaches.
>>>>
>>> +1
>>>
>>> _______________________________________________
>>> Pharo-project mailing list
>>> [hidden email]
>>> http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project
>>
>>
>> _______________________________________________
>> Pharo-project mailing list
>> [hidden email]
>> http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project
> _______________________________________________
> Pharo-project mailing list
> [hidden email]
> http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project

> On Mon, 15 Mar 2010, Stéphane Ducasse wrote:
>
>> so may be the bug entry should have stated that because oscar and nico spend a day on it.
>
> The good news is that the tests are ready which is probably the larger part. The bad news is that integrating the tally hack is easier, than the SetElement version (which is also Igor's code, but that may not be clear from my previous mail).
>
>
> Levente
>
>>
>> Stef
>>
>> On Mar 15, 2010, at 12:56 AM, Henrik Sperre Johansen wrote:
>>
>>> On 15.03.2010 00:18, Igor Stasenko wrote:
>>>> 2010/3/14 Levente Uzonyi<[hidden email]>:
>>>>> On Sun, 14 Mar 2010, Stéphane Ducasse wrote:
>>>>>
>>>>>> during the sprint niko schwarz and oscar nierstrasz fixed the Set in nil
>>>>>> entry.
>>>>>> Thanks.
>>>>>> http://code.google.com/p/pharo/issues/detail?id=1907
>>>>> Why did you choose Igor's negative tally hack instead of the SetElement
>>>>> version?
>>>>>
>>>> I'd prefer to use SetElement, unless there was a decision made, to use
>>>> negative tally.
>>>> We had a quite big discussion around it on squeak-dev, and most
>>>> developers said that SetElement
>>>> is the way to go, because it is clean, OO-oriented, even if somewhat
>>>> slower comparing to other approaches.
>>>>
>>> +1
>>>
>>> _______________________________________________
>>> Pharo-project mailing list
>>> [hidden email]
>>> http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project
>>
>>
>> _______________________________________________
>> Pharo-project mailing list
>> [hidden email]
>> http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project
> _______________________________________________
> Pharo-project mailing list
> [hidden email]
> http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project

>
> Yes, we focused on getting a working version with all the tests running.
>
> Frankly we could not load the SetElement version, so we focused on the other approach, even if it is hackier.  If we had had time, we would have then invested some effort into understanding how to get the other approach working without freezing the image, as it is certainly cleaner.
>
> There is still some work to do ...
>
> - on
>
>
> On 15 Mar 2010, at 07:08, Levente Uzonyi wrote:
>
>> On Mon, 15 Mar 2010, Stéphane Ducasse wrote:
>>
>>> so may be the bug entry should have stated that because oscar and nico spend a day on it.
>>
>> The good news is that the tests are ready which is probably the larger part. The bad news is that integrating the tally hack is easier, than the SetElement version (which is also Igor's code, but that may not be clear from my previous mail).
>>
>>
>> Levente
>>
>>>
>>> Stef
>>>
>>> On Mar 15, 2010, at 12:56 AM, Henrik Sperre Johansen wrote:
>>>
>>>> On 15.03.2010 00:18, Igor Stasenko wrote:
>>>>> 2010/3/14 Levente Uzonyi<[hidden email]>:
>>>>>> On Sun, 14 Mar 2010, Stéphane Ducasse wrote:
>>>>>>
>>>>>>> during the sprint niko schwarz and oscar nierstrasz fixed the Set in nil
>>>>>>> entry.
>>>>>>> Thanks.
>>>>>>> http://code.google.com/p/pharo/issues/detail?id=1907
>>>>>> Why did you choose Igor's negative tally hack instead of the SetElement
>>>>>> version?
>>>>>>
>>>>> I'd prefer to use SetElement, unless there was a decision made, to use
>>>>> negative tally.
>>>>> We had a quite big discussion around it on squeak-dev, and most
>>>>> developers said that SetElement
>>>>> is the way to go, because it is clean, OO-oriented, even if somewhat
>>>>> slower comparing to other approaches.
>>>>>
>>>> +1
>>>>
>>>> _______________________________________________
>>>> Pharo-project mailing list
>>>> [hidden email]
>>>> http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project
>>>
>>>
>>> _______________________________________________
>>> Pharo-project mailing list
>>> [hidden email]
>>> http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project
>> _______________________________________________
>> Pharo-project mailing list
>> [hidden email]
>> http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project
>
>
> _______________________________________________
> Pharo-project mailing list
> [hidden email]
> http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project