About a real solution for analysing Pharo in Moose

Previous Topic Next Topic
 
classic Classic list List threaded Threaded
5 messages Options
Reply | Threaded
Open this post in threaded view
|

About a real solution for analysing Pharo in Moose

Stéphane Ducasse
Hi guys

I think that we should leverage the pharo infrastructure to be able analyze pharo unstable.
We should make sure that we can
        - load code in a separate system dictionary
        - perform analysis of code in that separate namespace

this way we can use Moose 4.1 in pharo 1.1 to load pharo 1.2 unstable.
It will be shaky the first time but it should work.

wat do you think?

Stef
_______________________________________________
Moose-dev mailing list
[hidden email]
https://www.iam.unibe.ch/mailman/listinfo/moose-dev
Reply | Threaded
Open this post in threaded view
|

Re: About a real solution for analysing Pharo in Moose

Simon Denier-3
That could be interesting. Meanwhile we can just try to load Moose in Pharo 1.2 and see how it works :)

Simon

On 16 sept. 2010, at 09:28, Stéphane Ducasse <[hidden email]> wrote:

> Hi guys
>
> I think that we should leverage the pharo infrastructure to be able analyze pharo unstable.
> We should make sure that we can
>    - load code in a separate system dictionary
>    - perform analysis of code in that separate namespace
>
> this way we can use Moose 4.1 in pharo 1.1 to load pharo 1.2 unstable.
> It will be shaky the first time but it should work.
>
> wat do you think?
>
> Stef
> _______________________________________________
> Moose-dev mailing list
> [hidden email]
> https://www.iam.unibe.ch/mailman/listinfo/moose-dev

_______________________________________________
Moose-dev mailing list
[hidden email]
https://www.iam.unibe.ch/mailman/listinfo/moose-dev
Reply | Threaded
Open this post in threaded view
|

Re: About a real solution for analysing Pharo in Moose

Stéphane Ducasse

On Sep 16, 2010, at 9:55 AM, Simon Denier wrote:

> That could be interesting. Meanwhile we can just try to load Moose in Pharo 1.2 and see how it works :)

try but we will break Moose.
So may be investing one week would solve that problem is better than getting in 1.2.

>
> Simon
>
> On 16 sept. 2010, at 09:28, Stéphane Ducasse <[hidden email]> wrote:
>
>> Hi guys
>>
>> I think that we should leverage the pharo infrastructure to be able analyze pharo unstable.
>> We should make sure that we can
>>   - load code in a separate system dictionary
>>   - perform analysis of code in that separate namespace
>>
>> this way we can use Moose 4.1 in pharo 1.1 to load pharo 1.2 unstable.
>> It will be shaky the first time but it should work.
>>
>> wat do you think?
>>
>> Stef
>> _______________________________________________
>> Moose-dev mailing list
>> [hidden email]
>> https://www.iam.unibe.ch/mailman/listinfo/moose-dev
>
> _______________________________________________
> Moose-dev mailing list
> [hidden email]
> https://www.iam.unibe.ch/mailman/listinfo/moose-dev


_______________________________________________
Moose-dev mailing list
[hidden email]
https://www.iam.unibe.ch/mailman/listinfo/moose-dev
Reply | Threaded
Open this post in threaded view
|

Re: About a real solution for analysing Pharo in Moose

jannik laval
An idea is to create a "small" configuration with only things needed for generating mse file of Pharo.
Two question about that:
- is it possible to separate the behaviors ? I think yes.
- is it possible to generate a mse of pharo ? I'm not sure.

This small configuration could be useful to import code from squeak.
As they want to do modular Squeak, it will be a good thing to study the changes between versions.

But, putting all Moose on Pharo1.2 is useful. So we should do both.

Jannik

On Sep 16, 2010, at 11:40 , Stéphane Ducasse wrote:

>
> On Sep 16, 2010, at 9:55 AM, Simon Denier wrote:
>
>> That could be interesting. Meanwhile we can just try to load Moose in Pharo 1.2 and see how it works :)
>
> try but we will break Moose.
> So may be investing one week would solve that problem is better than getting in 1.2.
>
>>
>> Simon
>>
>> On 16 sept. 2010, at 09:28, Stéphane Ducasse <[hidden email]> wrote:
>>
>>> Hi guys
>>>
>>> I think that we should leverage the pharo infrastructure to be able analyze pharo unstable.
>>> We should make sure that we can
>>>  - load code in a separate system dictionary
>>>  - perform analysis of code in that separate namespace
>>>
>>> this way we can use Moose 4.1 in pharo 1.1 to load pharo 1.2 unstable.
>>> It will be shaky the first time but it should work.
>>>
>>> wat do you think?
>>>
>>> Stef
>>> _______________________________________________
>>> Moose-dev mailing list
>>> [hidden email]
>>> https://www.iam.unibe.ch/mailman/listinfo/moose-dev
>>
>> _______________________________________________
>> Moose-dev mailing list
>> [hidden email]
>> https://www.iam.unibe.ch/mailman/listinfo/moose-dev
>
>
> _______________________________________________
> Moose-dev mailing list
> [hidden email]
> https://www.iam.unibe.ch/mailman/listinfo/moose-dev

---
Jannik Laval


_______________________________________________
Moose-dev mailing list
[hidden email]
https://www.iam.unibe.ch/mailman/listinfo/moose-dev
Reply | Threaded
Open this post in threaded view
|

Re: About a real solution for analysing Pharo in Moose

Simon Denier-3
We want both, but for completely different reasons:

1) 'Namespace' (and robust) importer to load any pharo code into a working Moose image.

But could we run the tests and the lint rules in such a namespace? It's also important to build a report.


2) Load moose into pharo 1.2 just because early migration is less painful than late migration where you have to change a lot. So as soon as Pharo 1.2 is stable enough, we can move the Moose #bleedingEdge (to use the forthcoming Metacello terminology) to Pharo 1.2

Cyrille already worked in this direction and Moose did not break so much if I remember (at least at load time).


On 16 sept. 2010, at 12:07, Laval Jannik wrote:

> An idea is to create a "small" configuration with only things needed for generating mse file of Pharo.
> Two question about that:
> - is it possible to separate the behaviors ? I think yes.
> - is it possible to generate a mse of pharo ? I'm not sure.
>
> This small configuration could be useful to import code from squeak.
> As they want to do modular Squeak, it will be a good thing to study the changes between versions.


That could be an interesting idea. We have to take care that the importer works well in Squeak in this case.


>
> But, putting all Moose on Pharo1.2 is useful. So we should do both.
>
> Jannik
>
> On Sep 16, 2010, at 11:40 , Stéphane Ducasse wrote:
>
>>
>> On Sep 16, 2010, at 9:55 AM, Simon Denier wrote:
>>
>>> That could be interesting. Meanwhile we can just try to load Moose in Pharo 1.2 and see how it works :)
>>
>> try but we will break Moose.
>> So may be investing one week would solve that problem is better than getting in 1.2.
>>
>>>
>>> Simon
>>>
>>> On 16 sept. 2010, at 09:28, Stéphane Ducasse <[hidden email]> wrote:
>>>
>>>> Hi guys
>>>>
>>>> I think that we should leverage the pharo infrastructure to be able analyze pharo unstable.
>>>> We should make sure that we can
>>>> - load code in a separate system dictionary
>>>> - perform analysis of code in that separate namespace
>>>>
>>>> this way we can use Moose 4.1 in pharo 1.1 to load pharo 1.2 unstable.
>>>> It will be shaky the first time but it should work.
>>>>
>>>> wat do you think?
>>>>
>>>> Stef
>>>> _______________________________________________
>>>> Moose-dev mailing list
>>>> [hidden email]
>>>> https://www.iam.unibe.ch/mailman/listinfo/moose-dev
>>>
>>> _______________________________________________
>>> Moose-dev mailing list
>>> [hidden email]
>>> https://www.iam.unibe.ch/mailman/listinfo/moose-dev
>>
>>
>> _______________________________________________
>> Moose-dev mailing list
>> [hidden email]
>> https://www.iam.unibe.ch/mailman/listinfo/moose-dev
>
> ---
> Jannik Laval
>
>
> _______________________________________________
> Moose-dev mailing list
> [hidden email]
> https://www.iam.unibe.ch/mailman/listinfo/moose-dev

--
 Simon




_______________________________________________
Moose-dev mailing list
[hidden email]
https://www.iam.unibe.ch/mailman/listinfo/moose-dev