major repackaging planned

> Hi,
>
> For the next major version of Moose, we need to rethink the packaging of Moose.
>
> To rethink this, we first need to have a common understanding of the structure of Moose. For this, please read the short description of Moose from here:
> http://www.themoosebook.org/book/introduction/nutshell
>
> In particular, I want to start distinguishing clearly between several types of components:
> . Importers:
> -- These have the responsibility of importing data into models
> -- Examples: SmalltalkImporter, MSE
> . Analyses
> -- they do something specific with data
> -- Examples: metrics, system complexity, dependency browser ...
> . Meta-Models:
> -- they define the structure of data
> -- Examples: FAMIX, Hismo, ...
> . Engines
> -- they are meta-tools with which we control the analysis flow
> -- Examples: Moose-Core, PetitParser, Fame, Mondrian, Glamour, Magritte
>
> Then of course, there will be various kinds of flavors. For example: MooseFinder is both an engine (because it is generic and can be contextualized in several ways) and an analysis (because it is built on top of an engine).
>
> After we settle on a larger picture, I want to have clearly named configurations for each subpart. Then it will be clearer where people should put the extensions (because currently we all define the extensions in the same package)
>
> Cheers,
> Doru
>
>
> --
> www.tudorgirba.com
>
> "Be rather willing to give than demanding to get."
>
>
>
>
> _______________________________________________
> Moose-dev mailing list
> [hidden email]
> https://www.iam.unibe.ch/mailman/listinfo/moose-dev

> BTW I will do a pass over the metacello cha[ter. I produced a memento with dale and I think that people will find it useful.
>
> On Aug 25, 2011, at 2:31 PM, Tudor Girba wrote:
>
>> Hi,
>>
>> For the next major version of Moose, we need to rethink the packaging of Moose.
>>
>> To rethink this, we first need to have a common understanding of the structure of Moose. For this, please read the short description of Moose from here:
>> http://www.themoosebook.org/book/introduction/nutshell
>>
>> In particular, I want to start distinguishing clearly between several types of components:
>> . Importers:
>> -- These have the responsibility of importing data into models
>> -- Examples: SmalltalkImporter, MSE
>> . Analyses
>> -- they do something specific with data
>> -- Examples: metrics, system complexity, dependency browser ...
>> . Meta-Models:
>> -- they define the structure of data
>> -- Examples: FAMIX, Hismo, ...
>> . Engines
>> -- they are meta-tools with which we control the analysis flow
>> -- Examples: Moose-Core, PetitParser, Fame, Mondrian, Glamour, Magritte
>>
>> Then of course, there will be various kinds of flavors. For example: MooseFinder is both an engine (because it is generic and can be contextualized in several ways) and an analysis (because it is built on top of an engine).
>>
>> After we settle on a larger picture, I want to have clearly named configurations for each subpart. Then it will be clearer where people should put the extensions (because currently we all define the extensions in the same package)
>>
>> Cheers,
>> Doru
>>
>>
>> --
>> www.tudorgirba.com
>>
>> "Be rather willing to give than demanding to get."
>>
>>
>>
>>
>> _______________________________________________
>> 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

>
> On Aug 25, 2011, at 11:11 PM, Tudor Girba wrote:
>
>> Hi,
>>
>> On 25 Aug 2011, at 22:49, Stéphane Ducasse wrote:
>>
>>> this is ok for me.
>>> Now I would like to make sure that we can reload moose.
>>> I still do not understand why this is not possible.
>>
>> The problem is that we have nested configurations. To reliably version something, we need to version everything recursively. I am not aware of the support for this.
>
>
> Did you check the metacelloToolBox: api?
> Did you check the MetacelloBrowser?

>
>> I have to do that.
>>
>> So if nested version recursively does not work
>> we should either release more often
>> or at least version at the level of certain package
>>
>> Not been able to reload is not good.
>
> Definitely. That is why I asked for help on this matter multiple times :)
>
> Doru
>
>> Stef
>
> --
> www.tudorgirba.com
>
> "In a world where everything is moving ever faster,
> one might have better chances to win by moving slower."
>
>
>
>
> _______________________________________________
> Moose-dev mailing list
> [hidden email]
> https://www.iam.unibe.ch/mailman/listinfo/moose-dev

> Metacello supports "nested configurations" ... Seaside/Pier/Magritte/etc. all have long chains of project dependencies ...
>
> I am in Edinburgh until tomorrow night so I am willing to sit down with someone who has an example of the "moose problem" on their laptop and we can characterize the problem ... I do not know if this is metacello bug, a configuration error or something in between .... but given the complexity of the moose configuration I need to be able to talk to someone about the intent of the configuration so that I can identify the bug/misunderstanding/configuration error...
>
> Dale
>
> ----- Original Message -----
> | From: "stephane ducasse" <[hidden email]>
> | To: "Moose-related development" <[hidden email]>
> | Cc: [hidden email]
> | Sent: Friday, August 26, 2011 12:06:45 AM
> | Subject: [Metacello] Re: [Moose-dev] Re: major repackaging planned
> |
> |
> | On Aug 25, 2011, at 11:11 PM, Tudor Girba wrote:
> |
> | > Hi,
> | >
> | > On 25 Aug 2011, at 22:49, Stéphane Ducasse wrote:
> | >
> | >> this is ok for me.
> | >> Now I would like to make sure that we can reload moose.
> | >> I still do not understand why this is not possible.
> | >
> | > The problem is that we have nested configurations. To reliably
> | > version something, we need to version everything recursively. I am
> | > not aware of the support for this.
> |
> |
> | Did you check the metacelloToolBox: api?
> | Did you check the MetacelloBrowser?
> | I have to do that.
> |
> | So if nested version recursively does not work
> | we should either release more often
> | or at least version at the level of certain package
> |
> | Not been able to reload is not good.
> |
> | Stef
>
> _______________________________________________
> Moose-dev mailing list
> [hidden email]
> https://www.iam.unibe.ch/mailman/listinfo/moose-dev

> I am with Dale working on the configuration of moose problem.
>
> Alexandre
>
>
> On 26 Aug 2011, at 10:21, Dale Henrichs wrote:
>
>> Metacello supports "nested configurations" ... Seaside/Pier/Magritte/etc. all have long chains of project dependencies ...
>>
>> I am in Edinburgh until tomorrow night so I am willing to sit down with someone who has an example of the "moose problem" on their laptop and we can characterize the problem ... I do not know if this is metacello bug, a configuration error or something in between .... but given the complexity of the moose configuration I need to be able to talk to someone about the intent of the configuration so that I can identify the bug/misunderstanding/configuration error...
>>
>> Dale
>>
>> ----- Original Message -----
>> | From: "stephane ducasse" <[hidden email]>
>> | To: "Moose-related development" <[hidden email]>
>> | Cc: [hidden email]
>> | Sent: Friday, August 26, 2011 12:06:45 AM
>> | Subject: [Metacello] Re: [Moose-dev] Re: major repackaging planned
>> |
>> |
>> | On Aug 25, 2011, at 11:11 PM, Tudor Girba wrote:
>> |
>> | > Hi,
>> | >
>> | > On 25 Aug 2011, at 22:49, Stéphane Ducasse wrote:
>> | >
>> | >> this is ok for me.
>> | >> Now I would like to make sure that we can reload moose.
>> | >> I still do not understand why this is not possible.
>> | >
>> | > The problem is that we have nested configurations. To reliably
>> | > version something, we need to version everything recursively. I am
>> | > not aware of the support for this.
>> |
>> |
>> | Did you check the metacelloToolBox: api?
>> | Did you check the MetacelloBrowser?
>> | I have to do that.
>> |
>> | So if nested version recursively does not work
>> | we should either release more often
>> | or at least version at the level of certain package
>> |
>> | Not been able to reload is not good.
>> |
>> | Stef
>>
>> _______________________________________________
>> Moose-dev mailing list
>> [hidden email]
>> https://www.iam.unibe.ch/mailman/listinfo/moose-dev
>
> --
> _,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:
> Alexandre Bergel  http://www.bergel.eu
> ^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;.
>
>
>
>
>

> Great. Any news?
>
> Doru
>
>
> On 26 Aug 2011, at 12:00, Alexandre Bergel wrote:
>
>> I am with Dale working on the configuration of moose problem.
>>
>> Alexandre
>>
>>
>> On 26 Aug 2011, at 10:21, Dale Henrichs wrote:
>>
>>> Metacello supports "nested configurations" ... Seaside/Pier/Magritte/etc. all have long chains of project dependencies ...
>>>
>>> I am in Edinburgh until tomorrow night so I am willing to sit down with someone who has an example of the "moose problem" on their laptop and we can characterize the problem ... I do not know if this is metacello bug, a configuration error or something in between .... but given the complexity of the moose configuration I need to be able to talk to someone about the intent of the configuration so that I can identify the bug/misunderstanding/configuration error...
>>>
>>> Dale
>>>
>>> ----- Original Message -----
>>> | From: "stephane ducasse" <[hidden email]>
>>> | To: "Moose-related development" <[hidden email]>
>>> | Cc: [hidden email]
>>> | Sent: Friday, August 26, 2011 12:06:45 AM
>>> | Subject: [Metacello] Re: [Moose-dev] Re: major repackaging planned
>>> |
>>> |
>>> | On Aug 25, 2011, at 11:11 PM, Tudor Girba wrote:
>>> |
>>> | > Hi,
>>> | >
>>> | > On 25 Aug 2011, at 22:49, Stéphane Ducasse wrote:
>>> | >
>>> | >> this is ok for me.
>>> | >> Now I would like to make sure that we can reload moose.
>>> | >> I still do not understand why this is not possible.
>>> | >
>>> | > The problem is that we have nested configurations. To reliably
>>> | > version something, we need to version everything recursively. I am
>>> | > not aware of the support for this.
>>> |
>>> |
>>> | Did you check the metacelloToolBox: api?
>>> | Did you check the MetacelloBrowser?
>>> | I have to do that.
>>> |
>>> | So if nested version recursively does not work
>>> | we should either release more often
>>> | or at least version at the level of certain package
>>> |
>>> | Not been able to reload is not good.
>>> |
>>> | Stef
>>>
>>> _______________________________________________
>>> Moose-dev mailing list
>>> [hidden email]
>>> https://www.iam.unibe.ch/mailman/listinfo/moose-dev
>>
>> --
>> _,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:
>> Alexandre Bergel  http://www.bergel.eu
>> ^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;.
>>
>>
>>
>>
>>
>
> --
> www.tudorgirba.com
>
> "Speaking louder won't make the point worthier."
>
>
> _______________________________________________
> Moose-dev mailing list
> [hidden email]
> https://www.iam.unibe.ch/mailman/listinfo/moose-dev

> I am just back home. Dale found a number of errors in many configuration. Since Squeaksource was down, we couldn't go as far as we wished.
>
> I saved the configuration and I am now trying to load version 4dale of Moose to see if it loads.
>
> Cheers,
> Alexandre
>
>
> On 26 Aug 2011, at 18:28, Tudor Girba wrote:
>
>> Great. Any news?
>>
>> Doru
>>
>>
>> On 26 Aug 2011, at 12:00, Alexandre Bergel wrote:
>>
>>> I am with Dale working on the configuration of moose problem.
>>>
>>> Alexandre
>>>
>>>
>>> On 26 Aug 2011, at 10:21, Dale Henrichs wrote:
>>>
>>>> Metacello supports "nested configurations" ... Seaside/Pier/Magritte/etc. all have long chains of project dependencies ...
>>>>
>>>> I am in Edinburgh until tomorrow night so I am willing to sit down with someone who has an example of the "moose problem" on their laptop and we can characterize the problem ... I do not know if this is metacello bug, a configuration error or something in between .... but given the complexity of the moose configuration I need to be able to talk to someone about the intent of the configuration so that I can identify the bug/misunderstanding/configuration error...
>>>>
>>>> Dale
>>>>
>>>> ----- Original Message -----
>>>> | From: "stephane ducasse" <[hidden email]>
>>>> | To: "Moose-related development" <[hidden email]>
>>>> | Cc: [hidden email]
>>>> | Sent: Friday, August 26, 2011 12:06:45 AM
>>>> | Subject: [Metacello] Re: [Moose-dev] Re: major repackaging planned
>>>> |
>>>> |
>>>> | On Aug 25, 2011, at 11:11 PM, Tudor Girba wrote:
>>>> |
>>>> | > Hi,
>>>> | >
>>>> | > On 25 Aug 2011, at 22:49, Stéphane Ducasse wrote:
>>>> | >
>>>> | >> this is ok for me.
>>>> | >> Now I would like to make sure that we can reload moose.
>>>> | >> I still do not understand why this is not possible.
>>>> | >
>>>> | > The problem is that we have nested configurations. To reliably
>>>> | > version something, we need to version everything recursively. I am
>>>> | > not aware of the support for this.
>>>> |
>>>> |
>>>> | Did you check the metacelloToolBox: api?
>>>> | Did you check the MetacelloBrowser?
>>>> | I have to do that.
>>>> |
>>>> | So if nested version recursively does not work
>>>> | we should either release more often
>>>> | or at least version at the level of certain package
>>>> |
>>>> | Not been able to reload is not good.
>>>> |
>>>> | Stef
>>>>
>>>> _______________________________________________
>>>> Moose-dev mailing list
>>>> [hidden email]
>>>> https://www.iam.unibe.ch/mailman/listinfo/moose-dev
>>>
>>> --
>>> _,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:
>>> Alexandre Bergel  http://www.bergel.eu
>>> ^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;.
>>>
>>>
>>>
>>>
>>>
>>
>> --
>> www.tudorgirba.com
>>
>> "Speaking louder won't make the point worthier."
>>
>>
>> _______________________________________________
>> Moose-dev mailing list
>> [hidden email]
>> https://www.iam.unibe.ch/mailman/listinfo/moose-dev
>
> --
> _,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:
> Alexandre Bergel  http://www.bergel.eu
> ^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;.
>
>
>
>
>

> Hi,
>
> Currently, trying to load the ConfigurationOfMoose default leads to some sort of an infinite loop.
>
> Now, here is the bummer. The new changes in the Moose-related configurations did *not* change the versions I am trying to load. And yet, I still cannot load the configuration I want.
>
> The reason is most likely due to the problem of reifying *all* versions before loading only one of them. Thus, if one version that is not related to anything is broken everything is broken.
>
> I raised this issue before, and I strongly believe this is not the way to go. Loading should be light in that it should only reify the version we are looking for.
>
> Cheers,
> Doru
>
>
> On 29 Aug 2011, at 20:57, Alexandre Bergel wrote:
>
>> I am just back home. Dale found a number of errors in many configuration. Since Squeaksource was down, we couldn't go as far as we wished.
>>
>> I saved the configuration and I am now trying to load version 4dale of Moose to see if it loads.
>>
>> Cheers,
>> Alexandre
>>
>>
>> On 26 Aug 2011, at 18:28, Tudor Girba wrote:
>>
>>> Great. Any news?
>>>
>>> Doru
>>>
>>>
>>> On 26 Aug 2011, at 12:00, Alexandre Bergel wrote:
>>>
>>>> I am with Dale working on the configuration of moose problem.
>>>>
>>>> Alexandre
>>>>
>>>>
>>>> On 26 Aug 2011, at 10:21, Dale Henrichs wrote:
>>>>
>>>>> Metacello supports "nested configurations" ... Seaside/Pier/Magritte/etc. all have long chains of project dependencies ...
>>>>>
>>>>> I am in Edinburgh until tomorrow night so I am willing to sit down with someone who has an example of the "moose problem" on their laptop and we can characterize the problem ... I do not know if this is metacello bug, a configuration error or something in between .... but given the complexity of the moose configuration I need to be able to talk to someone about the intent of the configuration so that I can identify the bug/misunderstanding/configuration error...
>>>>>
>>>>> Dale
>>>>>
>>>>> ----- Original Message -----
>>>>> | From: "stephane ducasse" <[hidden email]>
>>>>> | To: "Moose-related development" <[hidden email]>
>>>>> | Cc: [hidden email]
>>>>> | Sent: Friday, August 26, 2011 12:06:45 AM
>>>>> | Subject: [Metacello] Re: [Moose-dev] Re: major repackaging planned
>>>>> |
>>>>> |
>>>>> | On Aug 25, 2011, at 11:11 PM, Tudor Girba wrote:
>>>>> |
>>>>> | > Hi,
>>>>> | >
>>>>> | > On 25 Aug 2011, at 22:49, Stéphane Ducasse wrote:
>>>>> | >
>>>>> | >> this is ok for me.
>>>>> | >> Now I would like to make sure that we can reload moose.
>>>>> | >> I still do not understand why this is not possible.
>>>>> | >
>>>>> | > The problem is that we have nested configurations. To reliably
>>>>> | > version something, we need to version everything recursively. I am
>>>>> | > not aware of the support for this.
>>>>> |
>>>>> |
>>>>> | Did you check the metacelloToolBox: api?
>>>>> | Did you check the MetacelloBrowser?
>>>>> | I have to do that.
>>>>> |
>>>>> | So if nested version recursively does not work
>>>>> | we should either release more often
>>>>> | or at least version at the level of certain package
>>>>> |
>>>>> | Not been able to reload is not good.
>>>>> |
>>>>> | Stef
>>>>>
>>>>> _______________________________________________
>>>>> Moose-dev mailing list
>>>>> [hidden email]
>>>>> https://www.iam.unibe.ch/mailman/listinfo/moose-dev
>>>>
>>>> --
>>>> _,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:
>>>> Alexandre Bergel  http://www.bergel.eu
>>>> ^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;.
>>>>
>>>>
>>>>
>>>>
>>>>
>>>
>>> --
>>> www.tudorgirba.com
>>>
>>> "Speaking louder won't make the point worthier."
>>>
>>>
>>> _______________________________________________
>>> Moose-dev mailing list
>>> [hidden email]
>>> https://www.iam.unibe.ch/mailman/listinfo/moose-dev
>>
>> --
>> _,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:
>> Alexandre Bergel  http://www.bergel.eu
>> ^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;.
>>
>>
>>
>>
>>
>
> --
> www.tudorgirba.com
>
> "Every thing has its own flow."
>
>
>
>
>
> _______________________________________________
> Moose-dev mailing list
> [hidden email]
> https://www.iam.unibe.ch/mailman/listinfo/moose-dev

> what do you mean by reify?
> So that I can follow.
>
> Stef
>
> On Aug 29, 2011, at 11:18 PM, Tudor Girba wrote:
>
>> Hi,
>>
>> Currently, trying to load the ConfigurationOfMoose default leads to some sort of an infinite loop.
>>
>> Now, here is the bummer. The new changes in the Moose-related configurations did *not* change the versions I am trying to load. And yet, I still cannot load the configuration I want.
>>
>> The reason is most likely due to the problem of reifying *all* versions before loading only one of them. Thus, if one version that is not related to anything is broken everything is broken.
>>
>> I raised this issue before, and I strongly believe this is not the way to go. Loading should be light in that it should only reify the version we are looking for.
>>
>> Cheers,
>> Doru
>>
>>
>> On 29 Aug 2011, at 20:57, Alexandre Bergel wrote:
>>
>>> I am just back home. Dale found a number of errors in many configuration. Since Squeaksource was down, we couldn't go as far as we wished.
>>>
>>> I saved the configuration and I am now trying to load version 4dale of Moose to see if it loads.
>>>
>>> Cheers,
>>> Alexandre
>>>
>>>
>>> On 26 Aug 2011, at 18:28, Tudor Girba wrote:
>>>
>>>> Great. Any news?
>>>>
>>>> Doru
>>>>
>>>>
>>>> On 26 Aug 2011, at 12:00, Alexandre Bergel wrote:
>>>>
>>>>> I am with Dale working on the configuration of moose problem.
>>>>>
>>>>> Alexandre
>>>>>
>>>>>
>>>>> On 26 Aug 2011, at 10:21, Dale Henrichs wrote:
>>>>>
>>>>>> Metacello supports "nested configurations" ... Seaside/Pier/Magritte/etc. all have long chains of project dependencies ...
>>>>>>
>>>>>> I am in Edinburgh until tomorrow night so I am willing to sit down with someone who has an example of the "moose problem" on their laptop and we can characterize the problem ... I do not know if this is metacello bug, a configuration error or something in between .... but given the complexity of the moose configuration I need to be able to talk to someone about the intent of the configuration so that I can identify the bug/misunderstanding/configuration error...
>>>>>>
>>>>>> Dale
>>>>>>
>>>>>> ----- Original Message -----
>>>>>> | From: "stephane ducasse" <[hidden email]>
>>>>>> | To: "Moose-related development" <[hidden email]>
>>>>>> | Cc: [hidden email]
>>>>>> | Sent: Friday, August 26, 2011 12:06:45 AM
>>>>>> | Subject: [Metacello] Re: [Moose-dev] Re: major repackaging planned
>>>>>> |
>>>>>> |
>>>>>> | On Aug 25, 2011, at 11:11 PM, Tudor Girba wrote:
>>>>>> |
>>>>>> | > Hi,
>>>>>> | >
>>>>>> | > On 25 Aug 2011, at 22:49, Stéphane Ducasse wrote:
>>>>>> | >
>>>>>> | >> this is ok for me.
>>>>>> | >> Now I would like to make sure that we can reload moose.
>>>>>> | >> I still do not understand why this is not possible.
>>>>>> | >
>>>>>> | > The problem is that we have nested configurations. To reliably
>>>>>> | > version something, we need to version everything recursively. I am
>>>>>> | > not aware of the support for this.
>>>>>> |
>>>>>> |
>>>>>> | Did you check the metacelloToolBox: api?
>>>>>> | Did you check the MetacelloBrowser?
>>>>>> | I have to do that.
>>>>>> |
>>>>>> | So if nested version recursively does not work
>>>>>> | we should either release more often
>>>>>> | or at least version at the level of certain package
>>>>>> |
>>>>>> | Not been able to reload is not good.
>>>>>> |
>>>>>> | Stef
>>>>>>
>>>>>> _______________________________________________
>>>>>> Moose-dev mailing list
>>>>>> [hidden email]
>>>>>> https://www.iam.unibe.ch/mailman/listinfo/moose-dev
>>>>>
>>>>> --
>>>>> _,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:
>>>>> Alexandre Bergel  http://www.bergel.eu
>>>>> ^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;.
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>
>>>>
>>>> --
>>>> www.tudorgirba.com
>>>>
>>>> "Speaking louder won't make the point worthier."
>>>>
>>>>
>>>> _______________________________________________
>>>> Moose-dev mailing list
>>>> [hidden email]
>>>> https://www.iam.unibe.ch/mailman/listinfo/moose-dev
>>>
>>> --
>>> _,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:
>>> Alexandre Bergel  http://www.bergel.eu
>>> ^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;.
>>>
>>>
>>>
>>>
>>>
>>
>> --
>> www.tudorgirba.com
>>
>> "Every thing has its own flow."
>>
>>
>>
>>
>>
>> _______________________________________________
>> 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

> Hi,
>
> Currently, trying to load the ConfigurationOfMoose default leads to some sort of an infinite loop.
>
> Now, here is the bummer. The new changes in the Moose-related configurations did *not* change the versions I am trying to load. And yet, I still cannot load the configuration I want.
>
> The reason is most likely due to the problem of reifying *all* versions before loading only one of them. Thus, if one version that is not related to anything is broken everything is broken.
>
> I raised this issue before, and I strongly believe this is not the way to go. Loading should be light in that it should only reify the version we are looking for.
>
> Cheers,
> Doru
>
>
> On 29 Aug 2011, at 20:57, Alexandre Bergel wrote:
>
>> I am just back home. Dale found a number of errors in many configuration. Since Squeaksource was down, we couldn't go as far as we wished.
>>
>> I saved the configuration and I am now trying to load version 4dale of Moose to see if it loads.
>>
>> Cheers,
>> Alexandre
>>
>>
>> On 26 Aug 2011, at 18:28, Tudor Girba wrote:
>>
>>> Great. Any news?
>>>
>>> Doru
>>>
>>>
>>> On 26 Aug 2011, at 12:00, Alexandre Bergel wrote:
>>>
>>>> I am with Dale working on the configuration of moose problem.
>>>>
>>>> Alexandre
>>>>
>>>>
>>>> On 26 Aug 2011, at 10:21, Dale Henrichs wrote:
>>>>
>>>>> Metacello supports "nested configurations" ... Seaside/Pier/Magritte/etc. all have long chains of project dependencies ...
>>>>>
>>>>> I am in Edinburgh until tomorrow night so I am willing to sit down with someone who has an example of the "moose problem" on their laptop and we can characterize the problem ... I do not know if this is metacello bug, a configuration error or something in between .... but given the complexity of the moose configuration I need to be able to talk to someone about the intent of the configuration so that I can identify the bug/misunderstanding/configuration error...
>>>>>
>>>>> Dale
>>>>>
>>>>> ----- Original Message -----
>>>>> | From: "stephane ducasse" <[hidden email]>
>>>>> | To: "Moose-related development" <[hidden email]>
>>>>> | Cc: [hidden email]
>>>>> | Sent: Friday, August 26, 2011 12:06:45 AM
>>>>> | Subject: [Metacello] Re: [Moose-dev] Re: major repackaging planned
>>>>> |
>>>>> |
>>>>> | On Aug 25, 2011, at 11:11 PM, Tudor Girba wrote:
>>>>> |
>>>>> | > Hi,
>>>>> | >
>>>>> | > On 25 Aug 2011, at 22:49, Stéphane Ducasse wrote:
>>>>> | >
>>>>> | >> this is ok for me.
>>>>> | >> Now I would like to make sure that we can reload moose.
>>>>> | >> I still do not understand why this is not possible.
>>>>> | >
>>>>> | > The problem is that we have nested configurations. To reliably
>>>>> | > version something, we need to version everything recursively. I am
>>>>> | > not aware of the support for this.
>>>>> |
>>>>> |
>>>>> | Did you check the metacelloToolBox: api?
>>>>> | Did you check the MetacelloBrowser?
>>>>> | I have to do that.
>>>>> |
>>>>> | So if nested version recursively does not work
>>>>> |    we should either release more often
>>>>> |    or at least version at the level of certain package
>>>>> |
>>>>> | Not been able to reload is not good.
>>>>> |
>>>>> | Stef
>>>>>
>>>>> _______________________________________________
>>>>> Moose-dev mailing list
>>>>> [hidden email]
>>>>> https://www.iam.unibe.ch/mailman/listinfo/moose-dev
>>>>
>>>> --
>>>> _,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:
>>>> Alexandre Bergel  http://www.bergel.eu
>>>> ^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;.
>>>>
>>>>
>>>>
>>>>
>>>>
>>>
>>> --
>>> www.tudorgirba.com
>>>
>>> "Speaking louder won't make the point worthier."
>>>
>>>
>>> _______________________________________________
>>> Moose-dev mailing list
>>> [hidden email]
>>> https://www.iam.unibe.ch/mailman/listinfo/moose-dev
>>
>> --
>> _,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:
>> Alexandre Bergel  http://www.bergel.eu
>> ^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;.
>>
>>
>>
>>
>>
>
> --
> www.tudorgirba.com
>
> "Every thing has its own flow."
>
>
>
>
>
> _______________________________________________
> Moose-dev mailing list
> [hidden email]
> https://www.iam.unibe.ch/mailman/listinfo/moose-dev

> There is indeed a kind of loop. Dale confirmed it this Friday.
> I have the impression that all of this stems from the default baseline. We should load the last defined version instead. But that was nit easy in the pre-MetacelloBrowser era.
>
> How do we proceed?
>
> Alexandre
>
>
>
> Le 29 août 2011 à 17:18, Tudor Girba <[hidden email]> a écrit :
>
>> Hi,
>>
>> Currently, trying to load the ConfigurationOfMoose default leads to some sort of an infinite loop.
>>
>> Now, here is the bummer. The new changes in the Moose-related configurations did *not* change the versions I am trying to load. And yet, I still cannot load the configuration I want.
>>
>> The reason is most likely due to the problem of reifying *all* versions before loading only one of them. Thus, if one version that is not related to anything is broken everything is broken.
>>
>> I raised this issue before, and I strongly believe this is not the way to go. Loading should be light in that it should only reify the version we are looking for.
>>
>> Cheers,
>> Doru
>>
>>
>> On 29 Aug 2011, at 20:57, Alexandre Bergel wrote:
>>
>>> I am just back home. Dale found a number of errors in many configuration. Since Squeaksource was down, we couldn't go as far as we wished.
>>>
>>> I saved the configuration and I am now trying to load version 4dale of Moose to see if it loads.
>>>
>>> Cheers,
>>> Alexandre
>>>
>>>
>>> On 26 Aug 2011, at 18:28, Tudor Girba wrote:
>>>
>>>> Great. Any news?
>>>>
>>>> Doru
>>>>
>>>>
>>>> On 26 Aug 2011, at 12:00, Alexandre Bergel wrote:
>>>>
>>>>> I am with Dale working on the configuration of moose problem.
>>>>>
>>>>> Alexandre
>>>>>
>>>>>
>>>>> On 26 Aug 2011, at 10:21, Dale Henrichs wrote:
>>>>>
>>>>>> Metacello supports "nested configurations" ... Seaside/Pier/Magritte/etc. all have long chains of project dependencies ...
>>>>>>
>>>>>> I am in Edinburgh until tomorrow night so I am willing to sit down with someone who has an example of the "moose problem" on their laptop and we can characterize the problem ... I do not know if this is metacello bug, a configuration error or something in between .... but given the complexity of the moose configuration I need to be able to talk to someone about the intent of the configuration so that I can identify the bug/misunderstanding/configuration error...
>>>>>>
>>>>>> Dale
>>>>>>
>>>>>> ----- Original Message -----
>>>>>> | From: "stephane ducasse" <[hidden email]>
>>>>>> | To: "Moose-related development" <[hidden email]>
>>>>>> | Cc: [hidden email]
>>>>>> | Sent: Friday, August 26, 2011 12:06:45 AM
>>>>>> | Subject: [Metacello] Re: [Moose-dev] Re: major repackaging planned
>>>>>> |
>>>>>> |
>>>>>> | On Aug 25, 2011, at 11:11 PM, Tudor Girba wrote:
>>>>>> |
>>>>>> | > Hi,
>>>>>> | >
>>>>>> | > On 25 Aug 2011, at 22:49, Stéphane Ducasse wrote:
>>>>>> | >
>>>>>> | >> this is ok for me.
>>>>>> | >> Now I would like to make sure that we can reload moose.
>>>>>> | >> I still do not understand why this is not possible.
>>>>>> | >
>>>>>> | > The problem is that we have nested configurations. To reliably
>>>>>> | > version something, we need to version everything recursively. I am
>>>>>> | > not aware of the support for this.
>>>>>> |
>>>>>> |
>>>>>> | Did you check the metacelloToolBox: api?
>>>>>> | Did you check the MetacelloBrowser?
>>>>>> | I have to do that.
>>>>>> |
>>>>>> | So if nested version recursively does not work
>>>>>> |    we should either release more often
>>>>>> |    or at least version at the level of certain package
>>>>>> |
>>>>>> | Not been able to reload is not good.
>>>>>> |
>>>>>> | Stef
>>>>>>
>>>>>> _______________________________________________
>>>>>> Moose-dev mailing list
>>>>>> [hidden email]
>>>>>> https://www.iam.unibe.ch/mailman/listinfo/moose-dev
>>>>>
>>>>> --
>>>>> _,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:
>>>>> Alexandre Bergel  http://www.bergel.eu
>>>>> ^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;.
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>
>>>>
>>>> --
>>>> www.tudorgirba.com
>>>>
>>>> "Speaking louder won't make the point worthier."
>>>>
>>>>
>>>> _______________________________________________
>>>> Moose-dev mailing list
>>>> [hidden email]
>>>> https://www.iam.unibe.ch/mailman/listinfo/moose-dev
>>>
>>> --
>>> _,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:
>>> Alexandre Bergel  http://www.bergel.eu
>>> ^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;.
>>>
>>>
>>>
>>>
>>>
>>
>> --
>> www.tudorgirba.com
>>
>> "Every thing has its own flow."
>>
>>
>>
>>
>>
>> _______________________________________________
>> 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

> Dale suggested to have a kind of MooseCore package that will load all the Moose packages. ConfigurationOfMoose will then load ConfigurationOfMooseCore and all the dependent packages. Other configurations will then to be adjusted to depend on ConfigurationOfMooseCore instead of ConfigurationOfMoose (as it happens in ConfigurationOfDSM).
>
> This represents quite some engineering work.
>
> Cheers,
> Alexandre
>
>
> On 29 Aug 2011, at 20:19, Alexandre Bergel wrote:
>
>> There is indeed a kind of loop. Dale confirmed it this Friday.
>> I have the impression that all of this stems from the default baseline. We should load the last defined version instead. But that was nit easy in the pre-MetacelloBrowser era.
>>
>> How do we proceed?
>>
>> Alexandre
>>
>>
>>
>> Le 29 août 2011 à 17:18, Tudor Girba <[hidden email]> a écrit :
>>
>>> Hi,
>>>
>>> Currently, trying to load the ConfigurationOfMoose default leads to some sort of an infinite loop.
>>>
>>> Now, here is the bummer. The new changes in the Moose-related configurations did *not* change the versions I am trying to load. And yet, I still cannot load the configuration I want.
>>>
>>> The reason is most likely due to the problem of reifying *all* versions before loading only one of them. Thus, if one version that is not related to anything is broken everything is broken.
>>>
>>> I raised this issue before, and I strongly believe this is not the way to go. Loading should be light in that it should only reify the version we are looking for.
>>>
>>> Cheers,
>>> Doru
>>>
>>>
>>> On 29 Aug 2011, at 20:57, Alexandre Bergel wrote:
>>>
>>>> I am just back home. Dale found a number of errors in many configuration. Since Squeaksource was down, we couldn't go as far as we wished.
>>>>
>>>> I saved the configuration and I am now trying to load version 4dale of Moose to see if it loads.
>>>>
>>>> Cheers,
>>>> Alexandre
>>>>
>>>>
>>>> On 26 Aug 2011, at 18:28, Tudor Girba wrote:
>>>>
>>>>> Great. Any news?
>>>>>
>>>>> Doru
>>>>>
>>>>>
>>>>> On 26 Aug 2011, at 12:00, Alexandre Bergel wrote:
>>>>>
>>>>>> I am with Dale working on the configuration of moose problem.
>>>>>>
>>>>>> Alexandre
>>>>>>
>>>>>>
>>>>>> On 26 Aug 2011, at 10:21, Dale Henrichs wrote:
>>>>>>
>>>>>>> Metacello supports "nested configurations" ... Seaside/Pier/Magritte/etc. all have long chains of project dependencies ...
>>>>>>>
>>>>>>> I am in Edinburgh until tomorrow night so I am willing to sit down with someone who has an example of the "moose problem" on their laptop and we can characterize the problem ... I do not know if this is metacello bug, a configuration error or something in between .... but given the complexity of the moose configuration I need to be able to talk to someone about the intent of the configuration so that I can identify the bug/misunderstanding/configuration error...
>>>>>>>
>>>>>>> Dale
>>>>>>>
>>>>>>> ----- Original Message -----
>>>>>>> | From: "stephane ducasse" <[hidden email]>
>>>>>>> | To: "Moose-related development" <[hidden email]>
>>>>>>> | Cc: [hidden email]
>>>>>>> | Sent: Friday, August 26, 2011 12:06:45 AM
>>>>>>> | Subject: [Metacello] Re: [Moose-dev] Re: major repackaging planned
>>>>>>> |
>>>>>>> |
>>>>>>> | On Aug 25, 2011, at 11:11 PM, Tudor Girba wrote:
>>>>>>> |
>>>>>>> | > Hi,
>>>>>>> | >
>>>>>>> | > On 25 Aug 2011, at 22:49, Stéphane Ducasse wrote:
>>>>>>> | >
>>>>>>> | >> this is ok for me.
>>>>>>> | >> Now I would like to make sure that we can reload moose.
>>>>>>> | >> I still do not understand why this is not possible.
>>>>>>> | >
>>>>>>> | > The problem is that we have nested configurations. To reliably
>>>>>>> | > version something, we need to version everything recursively. I am
>>>>>>> | > not aware of the support for this.
>>>>>>> |
>>>>>>> |
>>>>>>> | Did you check the metacelloToolBox: api?
>>>>>>> | Did you check the MetacelloBrowser?
>>>>>>> | I have to do that.
>>>>>>> |
>>>>>>> | So if nested version recursively does not work
>>>>>>> |    we should either release more often
>>>>>>> |    or at least version at the level of certain package
>>>>>>> |
>>>>>>> | Not been able to reload is not good.
>>>>>>> |
>>>>>>> | Stef
>>>>>>>
>>>>>>> _______________________________________________
>>>>>>> Moose-dev mailing list
>>>>>>> [hidden email]
>>>>>>> https://www.iam.unibe.ch/mailman/listinfo/moose-dev
>>>>>>
>>>>>> --
>>>>>> _,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:
>>>>>> Alexandre Bergel  http://www.bergel.eu
>>>>>> ^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;.
>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>
>>>>> --
>>>>> www.tudorgirba.com
>>>>>
>>>>> "Speaking louder won't make the point worthier."
>>>>>
>>>>>
>>>>> _______________________________________________
>>>>> Moose-dev mailing list
>>>>> [hidden email]
>>>>> https://www.iam.unibe.ch/mailman/listinfo/moose-dev
>>>>
>>>> --
>>>> _,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:
>>>> Alexandre Bergel  http://www.bergel.eu
>>>> ^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;.
>>>>
>>>>
>>>>
>>>>
>>>>
>>>
>>> --
>>> www.tudorgirba.com
>>>
>>> "Every thing has its own flow."
>>>
>>>
>>>
>>>
>>>
>>> _______________________________________________
>>> 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
>
> --
> _,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:
> Alexandre Bergel  http://www.bergel.eu
> ^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;.
>
>
>
>
>

>> In the meantime, we have to get the configuration to load, so please remove the problematic versions. Or at least remove the pragmas from the methods.
>
> I commented out the pragmas of the versions and baselines we added.
> Can you restart Jenkins ?
>
> Cheers,
> Alexandre
>
> --
> _,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:
> Alexandre Bergel  http://www.bergel.eu
> ^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;.
>
>
>
>
>

>> In the meantime, we have to get the configuration to load, so please remove the problematic versions. Or at least remove the pragmas from the methods.
>
> I commented out the pragmas of the versions and baselines we added.
> Can you restart Jenkins ?
>
> Cheers,
> Alexandre
>
> --
> _,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:
> Alexandre Bergel  http://www.bergel.eu
> ^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;.
>
>
>
>
>