2 Jul
2016
2 Jul
'16
9:43 a.m.
Alex,
Extending the Famix meta model has 2 difficulties:
- put the right pragmas in the right place to extend the meta-model
There was a presentation done by Steph. and Andre Hora, I will send it
to you.
(basically, you must put a pragma on the class side to describe the
class itself
+ put pragmas in the getters of the properties that you want to be
recognized in the meta model)
- putting in place the mechanism so that you don't have to take care of
both ends of an association (eg if you say that a method has for parent
a given class, you don't need to say to the class that it owns this method)
This is where MultiLinks come in place.
For this to work, you typically need to put something in the initialize
(at both ends of the association), and in the setters of the properties
(at both ends of the association).
And because associations are known by the metamodel, you actually have
three classes to deal with (association + both ends of it, like method
and class in the example above) and four properties (instance
variables): two in the association and one in each class at the end of
the association.
This is indeed a bit cumbersome and error prone :-)
What I typically do for both problems (pragms, multilinks) is that I
look at something existing, I open several nautilus and I replicate
exactly hwat was done before.
As Guillaume explained, you have to take care of chossing the right
example: 1-to-* or *-to-* links.
nicolas
On 02/07/2016 00:16, Alexandre Bergel wrote:
Vincent, can you put this in the class comments?
It would be fantastic to have a Help on this. Googling before using Fame is really not
optimal...
Alexandre
> On Jul 1, 2016, at 12:10 PM, Blondeau Vincent <vincent.blondeau(a)worldline.com>
wrote:
>
> Hi,
>
> I am afraid that the documentation about fame (and few other moose things like the
moose algorithms) has vanished with the old Moose website… If someone can make it
available again, it will be nice.
>
> But, to give more usage examples:
>
> - FMMultivalueLink
> Used to create 1 to n links between famix elements. Example with inheritance where a
FAMIXInheritance has a link to 1 FAMIXType and where FAMIXType have 1 link to n
FAMIXInheritances:
>
> In FAMIXInheritance (subclass is an instance variable):
>
> FAMIXInheritance >>subclass
> <MSEProperty: #subclass type: #FAMIXType opposite:
#superInheritances>
> <MSEComment: 'Subclass linked to in this relationship. from-side of
the association'>
>
> ^subclass
>
> FAMIXInheritance >> subclass: aType
> subclass := FMMultivalueLink on: self
> update: #superInheritances
> from: self subclass
> to: aType
>
> In: FAMIXType (superInheritances is an instance variable)
> FAMIXType >>initialize
> superInheritances := FMNullMultivalueLink on: self opposite: #subclass:
selector: #superInheritances.
>
> FAMIXType >>superInheritances
> <MSEProperty: #superInheritances type: #FAMIXInheritance opposite:
#subclass> <multivalued> <derived>
> <MSEComment: 'Superinheritance relationships, i.e. known
superclasses of this type.'>
>
> ^superInheritances
>
> FAMIXType >> superInheritances: anInheritance
> superInheritances value: anInheritance
>
> - FMMultiMultivalueLink. To create n to n associations. Example:
> In FAMIXInvocations (candidates is an instance variable):
> FAMIXInvocations >> initialize
> candidates := FMMultiMultivalueLink on: self opposite:
#incomingInvocations.
>
> FAMIXInvocations >> candidates
> <MSEProperty: #candidates type: #FAMIXBehaviouralEntity opposite:
#incomingInvocations> <multivalued>
> <MSEComment: 'List of candidate behavioural entities for receiving
the invocation'>
>
> ^candidates
> FAMIXInvocations >> candidates: aCollection
> candidates value: aCollection
>
> In FAMIXBehaviouralEntity (incomingInvocations is an instance variable):
> FAMIXBehaviouralEntity >> initialize
> incomingInvocations := FMMultiMultivalueLink on: self opposite:
#candidates.
> FAMIXBehaviouralEntity >> incomingInvocations
> <MSEProperty: #incomingInvocations type: #FAMIXInvocation opposite:
#candidates> <multivalued> <derived>
> <MSEComment: 'Incoming invocations from other behaviours computed
by the candidate operator.'>
>
> ^incomingInvocations
>
> FAMIXBehaviouralEntity >> incomingInvocations: anInvocation
> incomingInvocations value: anInvocation
>
> - FMNullMultivalueLink: used to reduce memory consumption (see other answers)
> - <multivalued>: used to describe in the meta model that the return value is a
collection of entities and no only one entity.
> - <derived>: as said by Guillaume: “is for things that don't require to be
stored in the .mse for example because you can derive it (compute) from other data”
> Example:
> numberOfAccesses
> <MSEProperty: #numberOfAccesses type: #Number>
> <derived>
> <MSEComment: 'The number of accesses from a method'>
>
> ^ self
> lookUpPropertyNamed: #numberOfAccesses
> computedAs: [ self accesses size ]
>
> I know that the colors are flashy but it is to better understand the links between
the names.
>
> All these pragmas are mandatory to describe relations in Fame Meta Meta Model and
should be provided accurately.
>
> And you should put something like:
> annotation
> <MSEClass: #BehaviouralEntity super: #FAMIXContainerEntity>
> <package: #FAMIX>
> ^self
>
> On class side of your Moose classes to reference them in the meta model.
>
> After a change in the pragmas, do : MooseModel resetMeta It avoid a lots of problems
;)
>
> Don’t hesitate if you have more questions,
>
> Cheers,
> Vincent
>
>> -----Message d'origine-----
>> De : moose-dev-bounces(a)list.inf.unibe.ch [mailto:moose-dev-
>> bounces(a)list.inf.unibe.ch] De la part de Anne Etien
>> Envoyé : vendredi 1 juillet 2016 17:10
>> À : Moose-related development
>> Objet : [Moose-dev] Re: Documentation about Fame?
>>
>> Hi,
>>
>> We asked for an engineer to modify it. We do not know yet if we got it or not.
>>
>> What I do is that I use the example of Famix for ST to create my own FAMIX.
>> <mutivalued> means that the maximal multiplicity is *. In that case, you
need
>> FMMultiValueLink. But when you want to initiate one FMMultivalueLink to
>> null, you create a FMNullMultiValueLink.
>> I don’t really know when there is <derived> or not.
>>
>> Anne
>>
>>
>>
>> Le 1 juil. 2016 à 16:47, Alexandre Bergel <alexandre.bergel(a)me.com> a écrit
:
>>
>>> hi
>>>
>>> Is there at least a bit of documentation about Fame?
>>> I am trying to meta-describes some C elements, but it is very difficult.
>>> All these are very mysterious:
>>> FMNullMultivalueLink,
>>> FMMultivalueLink
>>> FMMultiMultivalueLink
>>> <multivalued>
>>> <derived>
>>>
>>> I feel Fame is very powerful, but very hard to understand.
>>>
>>> Alexandre
>>> --
>>> _,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:
>>> Alexandre Bergel http://www.bergel.eu
>>> ^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;.
>>>
>>>
>>>
>>> _______________________________________________
>>> Moose-dev mailing list
>>> Moose-dev(a)list.inf.unibe.ch
>>> https://www.list.inf.unibe.ch/listinfo/moose-dev
>> _______________________________________________
>> Moose-dev mailing list
>> Moose-dev(a)list.inf.unibe.ch
>> https://www.list.inf.unibe.ch/listinfo/moose-dev
>
!!!*************************************************************************************
> "Ce message et les pièces jointes sont confidentiels et réservés à l'usage
exclusif de ses destinataires. Il peut également être protégé par le secret professionnel.
Si vous recevez ce message par erreur, merci d'en avertir immédiatement
l'expéditeur et de le détruire. L'intégrité du message ne pouvant être assurée sur
Internet, la responsabilité de Worldline ne pourra être recherchée quant au contenu de ce
message. Bien que les meilleurs efforts soient faits pour maintenir cette transmission
exempte de tout virus, l'expéditeur ne donne aucune garantie à cet égard et sa
responsabilité ne saurait être recherchée pour tout dommage résultant d'un virus
transmis.
>
> This e-mail and the documents attached are confidential and intended solely for the
addressee; it may also be privileged. If you receive this e-mail in error, please notify
the sender immediately and destroy it. As its integrity cannot be secured on the Internet,
the Worldline liability cannot be triggered for the message content. Although the sender
endeavours to maintain a computer virus-free network, the sender does not warrant that
this transmission is virus-free and will not be liable for any damages resulting from any
virus transmitted.!!!"
> _______________________________________________
> Moose-dev mailing list
> Moose-dev(a)list.inf.unibe.ch
> https://www.list.inf.unibe.ch/listinfo/moose-dev
--
Nicolas Anquetil
RMod team -- Inria Lille