Some people suggested trying moose for analysis of SAS programs, and I'm trying to figure out where to start with the meta-modeling.
SAS is not object oriented, though parts of it could be imagined to be. The main SAS code contains blocks of statements that begin with DATA or PROC; each block could probably be thought of as a function--to be precise, the application of a function.
If I want to make some new models, where do I start? FM3? FAMIX? elsewhere? The 4.0 release announcement says FM3 and FAMIX3 are both implemented in Fame, so maybe start with Fame? I've read some documentation, but I can't tell.
My inspiration for the project is understanding how some SAS datasets were produced. I have a system that creates numerous datasets, which feed into later datasets, etc. This is split at least across a couple of program files and it's really too complicated to keep in the brain.
If I get ambitious it would also be useful to trace where particular variables came from, in the sense of both datasets and individual expressions.
The only thing resembling a SAS parser I've run into (except for SAS, which is closed source) is a Perl module, and it looked relatively primitive. I've been using PetitParser, and at the point I started thinking about what it would parse results into.
Thanks for any pointers. Ross Boylan
Hi Ross,
Some people suggested trying moose for analysis of SAS programs, and I'm trying to figure out where to start with the meta-modeling.
Welcome to Moose :).
SAS is not object oriented, though parts of it could be imagined to be. The main SAS code contains blocks of statements that begin with DATA or PROC; each block could probably be thought of as a function--to be precise, the application of a function.
If I want to make some new models, where do I start? FM3? FAMIX? elsewhere? The 4.0 release announcement says FM3 and FAMIX3 are both implemented in Fame, so maybe start with Fame? I've read some documentation, but I can't tell.
FAMIX is a meta-model, FM3 is the meta-meta-model. If a clarification of terms is needed, you can check: http://www.tudorgirba.com/blog/subject-model-meta-model
The core of FAMIX models in a language independent fashion object- oriented and procedural languages. You can see a diagrammatic overview here: http://www.themoosebook.org/book/internals/famix/core
I do not know SAS, but you should check to see if the concepts present in FAMIX are of use to you. If yes, you can directly instantiate them from your parser. If not, then you will want to code a new meta-model, or extend the current ones to accommodate your needs. You can then describe this new meta-model using FM3 (this is achieved via Smalltalk pragmas) and you get a browser and import-export in the MSE format for free.
Once you have your Smalltalk objects as instances of the meta-model, you can then use Mondrian and Glamour for building visual tools, or the MooseAlgos to express graph-specific algorithms.
Does this help?
Cheers, Doru
My inspiration for the project is understanding how some SAS datasets were produced. I have a system that creates numerous datasets, which feed into later datasets, etc. This is split at least across a couple of program files and it's really too complicated to keep in the brain.
If I get ambitious it would also be useful to trace where particular variables came from, in the sense of both datasets and individual expressions.
The only thing resembling a SAS parser I've run into (except for SAS, which is closed source) is a Perl module, and it looked relatively primitive. I've been using PetitParser, and at the point I started thinking about what it would parse results into. Thanks for any pointers. Ross Boylan
Moose-dev mailing list Moose-dev@iam.unibe.ch https://www.iam.unibe.ch/mailman/listinfo/moose-dev
-- www.tudorgirba.com
"Relationships are of two kinds: those we choose and those that happen. They both matter."
On Wed, 2010-06-16 at 02:17 +0200, Tudor Girba wrote:
I do not know SAS, but you should check to see if the concepts present in FAMIX are of use to you. If yes, you can directly instantiate them from your parser. If not, then you will want to code a new meta-model, or extend the current ones to accommodate your needs. You can then describe this new meta-model using FM3 (this is achieved via Smalltalk pragmas) and you get a browser and import-export in the MSE format for free.
Once you have your Smalltalk objects as instances of the meta-model, you can then use Mondrian and Glamour for building visual tools, or the MooseAlgos to express graph-specific algorithms.
Does this help?
Well, I was hoping that I would need to specify the meta-model or the meta-meta-model, but not both. That is, I was hoping the tools could build one out of the other. It sounds as if you're telling me I need to do both.
I hope I'm misunderstanding :)
Also, is there a guide to what the pragmas are?
Ross
Hi,
On 16 Jun 2010, at 02:27, Ross Boylan wrote:
On Wed, 2010-06-16 at 02:17 +0200, Tudor Girba wrote:
I do not know SAS, but you should check to see if the concepts present in FAMIX are of use to you. If yes, you can directly instantiate them from your parser. If not, then you will want to code a new meta-model, or extend the current ones to accommodate your needs. You can then describe this new meta-model using FM3 (this is achieved via Smalltalk pragmas) and you get a browser and import-export in the MSE format for free.
Once you have your Smalltalk objects as instances of the meta-model, you can then use Mondrian and Glamour for building visual tools, or the MooseAlgos to express graph-specific algorithms.
Does this help?
Well, I was hoping that I would need to specify the meta-model or the meta-meta-model, but not both. That is, I was hoping the tools could build one out of the other. It sounds as if you're telling me I need to do both.
You do not define the meta-meta-model, you use the existing one (FM3) by just describing in your meta-model elements (classes and attributes) using pragmas. That is, the meta-model holds the Smalltalk implementation, while the pragmas provide a higher level description for other kinds of manipulations.
I hope I'm misunderstanding :)
I think you are :)
Also, is there a guide to what the pragmas are?
We do not have much of a guide for the moment, but we have plenty of examples. For a class named "C" with one attribute "interestingAttribute" of type String, you have to define the followings: - On the class size: annotation <MSEClass: 'X'>
- On the instance side: interestingAttribute <MSEProperty: #interestingAttribute type: #String> ^ interestingAttribute
Cheers, Doru
Ross
Moose-dev mailing list Moose-dev@iam.unibe.ch https://www.iam.unibe.ch/mailman/listinfo/moose-dev
-- www.tudorgirba.com
"Not knowing how to do something is not an argument for how it cannot be done."
On Wed, 2010-06-16 at 02:43 +0200, Tudor Girba wrote:
Hi,
On 16 Jun 2010, at 02:27, Ross Boylan wrote:
On Wed, 2010-06-16 at 02:17 +0200, Tudor Girba wrote:
I do not know SAS, but you should check to see if the concepts present in FAMIX are of use to you. If yes, you can directly instantiate them from your parser. If not, then you will want to code a new meta-model, or extend the current ones to accommodate your needs. You can then describe this new meta-model using FM3 (this is achieved via Smalltalk pragmas) and you get a browser and import-export in the MSE format for free.
Once you have your Smalltalk objects as instances of the meta-model, you can then use Mondrian and Glamour for building visual tools, or the MooseAlgos to express graph-specific algorithms.
Does this help?
Well, I was hoping that I would need to specify the meta-model or the meta-meta-model, but not both. That is, I was hoping the tools could build one out of the other. It sounds as if you're telling me I need to do both.
You do not define the meta-meta-model, you use the existing one (FM3) by just describing in your meta-model elements (classes and attributes) using pragmas. That is, the meta-model holds the Smalltalk implementation, while the pragmas provide a higher level description for other kinds of manipulations.
I hope I'm misunderstanding :)
I think you are :)
Also, is there a guide to what the pragmas are?
We do not have much of a guide for the moment, but we have plenty of examples. For a class named "C" with one attribute "interestingAttribute" of type String, you have to define the followings:
- On the class size:
annotation <MSEClass: 'X'>
- On the instance side:
interestingAttribute <MSEProperty: #interestingAttribute type: #String> ^ interestingAttribute
Cheers, Doru
So, rephrasing the question slightly, is there a way to generate the smalltalk implementation directly from the pragmas? Or, even better, a graphical tool that allows you to specify the model and produces everything else from it?
On the pragmas, at least some of them are "documented" by the methods FMPragmaProcessor>>processCompiledMethod: MSEPragmaProcessor>>processClass:
There are some huge strings that are (meta-)*model specifications, e.g., FAMIXSpecifications>>famixCore30mse and their use by FAMIXSpecifications>>generateFamixCore30. What role do they play in the system?
Ross
Hi,
On 16 Jun 2010, at 05:33, Ross Boylan wrote:
On Wed, 2010-06-16 at 02:43 +0200, Tudor Girba wrote:
Hi,
On 16 Jun 2010, at 02:27, Ross Boylan wrote:
On Wed, 2010-06-16 at 02:17 +0200, Tudor Girba wrote:
I do not know SAS, but you should check to see if the concepts present in FAMIX are of use to you. If yes, you can directly instantiate them from your parser. If not, then you will want to code a new meta-model, or extend the current ones to accommodate your needs. You can then describe this new meta-model using FM3 (this is achieved via Smalltalk pragmas) and you get a browser and import-export in the MSE format for free.
Once you have your Smalltalk objects as instances of the meta- model, you can then use Mondrian and Glamour for building visual tools, or the MooseAlgos to express graph-specific algorithms.
Does this help?
Well, I was hoping that I would need to specify the meta-model or the meta-meta-model, but not both. That is, I was hoping the tools could build one out of the other. It sounds as if you're telling me I need to do both.
You do not define the meta-meta-model, you use the existing one (FM3) by just describing in your meta-model elements (classes and attributes) using pragmas. That is, the meta-model holds the Smalltalk implementation, while the pragmas provide a higher level description for other kinds of manipulations.
I hope I'm misunderstanding :)
I think you are :)
Also, is there a guide to what the pragmas are?
We do not have much of a guide for the moment, but we have plenty of examples. For a class named "C" with one attribute "interestingAttribute" of type String, you have to define the followings:
- On the class size:
annotation <MSEClass: 'X'>
- On the instance side:
interestingAttribute <MSEProperty: #interestingAttribute type: #String> ^ interestingAttribute
Cheers, Doru
So, rephrasing the question slightly, is there a way to generate the smalltalk implementation directly from the pragmas? Or, even better, a graphical tool that allows you to specify the model and produces everything else from it?
At this moment, there is no graphical tool to define the meta-model.
On the pragmas, at least some of them are "documented" by the methods FMPragmaProcessor>>processCompiledMethod: MSEPragmaProcessor>>processClass:
Those are all that are relevant for the meta-describing your meta- model part.
There are some huge strings that are (meta-)*model specifications, e.g., FAMIXSpecifications>>famixCore30mse and their use by FAMIXSpecifications>>generateFamixCore30. What role do they play in the system?
These strings are indeed the meta-model specifications written in the MSE format. From an MSE format you can generate the Smalltalk implementation and the other way around. For example, through the MSE format we can import meta-models developed in another language and generate the Smalltalk implementation.
As mentioned before, there is no support for graphical support for writing MSE files. The main reason here is that Smalltalk already provides a good enough solution for quickly building classes and methods, and annotating them is a practical solution.
However, we do have a tool for browsing an existing meta-model: http://www.themoosebook.org/book/internals/famix/meta-browser
Cheers, Doru
Ross
Moose-dev mailing list Moose-dev@iam.unibe.ch https://www.iam.unibe.ch/mailman/listinfo/moose-dev
-- www.tudorgirba.com
"Not knowing how to do something is not an argument for how it cannot be done."
On Wed, 2010-06-16 at 09:37 +0200, Tudor Girba wrote:
As mentioned before, there is no support for graphical support for writing MSE files. The main reason here is that Smalltalk already provides a good enough solution for quickly building classes and methods, and annotating them is a practical solution.
I gather there is no facility to generate the smalltalk out of the method annotations. Is there any reason, in principle, that it couldn't be done?
Ross
Hi Ross,
On 16 Jun 2010, at 21:17, Ross Boylan wrote:
On Wed, 2010-06-16 at 09:37 +0200, Tudor Girba wrote:
As mentioned before, there is no support for graphical support for writing MSE files. The main reason here is that Smalltalk already provides a good enough solution for quickly building classes and methods, and annotating them is a practical solution.
I gather there is no facility to generate the smalltalk out of the method annotations. Is there any reason, in principle, that it couldn't be done?
Well, if you have the method annotations, you already have the Smalltalk code :).
Right now, the only possibility is to generate Smalltalk code out of MSE code that represents your meta-model. This is like the famixCore string that you saw in the image.
Cheers, Doru
-- www.tudorgirba.com
"Not knowing how to do something is not an argument for how it cannot be done."
On Wed, 2010-06-16 at 22:14 +0200, Tudor Girba wrote:
Hi Ross,
On 16 Jun 2010, at 21:17, Ross Boylan wrote:
On Wed, 2010-06-16 at 09:37 +0200, Tudor Girba wrote:
As mentioned before, there is no support for graphical support for writing MSE files. The main reason here is that Smalltalk already provides a good enough solution for quickly building classes and methods, and annotating them is a practical solution.
I gather there is no facility to generate the smalltalk out of the method annotations. Is there any reason, in principle, that it couldn't be done?
Well, if you have the method annotations, you already have the Smalltalk code :).
Consider RPGDragon>>hoard <MSEProperty: #hoard type: #RPGTreasure opposite: #keeper> <multivalued> ^hoard ^hoard is smalltalk code. You also have to create the instance variable, define hoard: aTreasure hoard value: aTreasure and initialize super initialize. hoard := FMMultivalueLink on: self opposite: #keeper:. killedBy := FMMultiMultivalueLink on: self opposite: #kills.
It's the redundancy, along with the possibilities for error that would put the smalltalk and annotation out of sync, that concerns me (and that I'm trying to avoid).
I do not in some other classes the names in the annotations and the names of the methods and variables are not quite in sync, so maybe this is a feature. Ross
Right now, the only possibility is to generate Smalltalk code out of MSE code that represents your meta-model. This is like the famixCore string that you saw in the image.
Cheers, Doru
-- www.tudorgirba.com
"Not knowing how to do something is not an argument for how it cannot be done."
Moose-dev mailing list Moose-dev@iam.unibe.ch https://www.iam.unibe.ch/mailman/listinfo/moose-dev
Hi Ross
It's true that the metamodeling tools in Moose seem crude compared to what exist elsewhere, but that's because we spend much more time using metamodels in new tools than building metamodels :)
Here is the procedure when you need to define a completely new metamodel. You can still make mistake when defining the metamodel but it eliminates most redundancy. Once you have defined your metamodel, a set of lint rules can be run to check some aspects of the metamodel (see below)
The essential pragma documentation is here (should be integrated in the book) http://code.google.com/p/moose-technology/wiki/FameMetaDescriptions
Also, a good rule of thumb is to have MooseEntity as root instead of Object, it facilitates integration with Moose tools.
1) Metamodel declaration For the example below: - Create class RPGDragon from a class browser - add (class-side) RPGDragon class>>annotation <MSEClass: #Dragon super: #Object> <package: #RPG>
Notice I just define the method with pragmas, I don't add any code in the body for now.
For each attribute, I need to define the accessor (the getter, not the setter): - define RPGDragon>>hoard <MSEProperty: #hoard type: #RPGTreasure opposite: #keeper><multivalued>
(same thing, method decl + pragma, no code) No need to add the initialize, it will be defined by the generator.
2) To generate the implementation, do: MooseModel generateClassesFrom: {RPGDragon. RPGHero. RPGWall} "collection of classes defining the metamodel"
It should open a preview window, where you can see which method will be defined/overridden by the generator. You can always tap Cmd-x to remove the selected definition (in case you are doing some incremental metamodel definition). You can repeat the generation step as much as you want, provided you understand it will overwrite existing methods.
It works like a charm if you are doing your own metamodel. If you need to extend (as in class extension, not the Java extend for inheritance) some existing Famix classes with new attributes, you should be a bit more careful. Actually, when I need to extend an existing metamodel with a few attributes, I just define the attributes directly in the Smalltalk way (with pragmas AND method body). It's fast to do.
3) optionally, check that your metamodel is correct with lint rules (especially soundness of opposite relationships): can be done from the meta browser...
On 16 juin 2010, at 22:36, Ross Boylan wrote:
On Wed, 2010-06-16 at 22:14 +0200, Tudor Girba wrote:
Hi Ross,
On 16 Jun 2010, at 21:17, Ross Boylan wrote:
On Wed, 2010-06-16 at 09:37 +0200, Tudor Girba wrote:
As mentioned before, there is no support for graphical support for writing MSE files. The main reason here is that Smalltalk already provides a good enough solution for quickly building classes and methods, and annotating them is a practical solution.
I gather there is no facility to generate the smalltalk out of the method annotations. Is there any reason, in principle, that it couldn't be done?
Well, if you have the method annotations, you already have the Smalltalk code :).
Consider RPGDragon>>hoard <MSEProperty: #hoard type: #RPGTreasure opposite: #keeper>
<multivalued> ^hoard ^hoard is smalltalk code. You also have to create the instance variable, define hoard: aTreasure hoard value: aTreasure and initialize super initialize. hoard := FMMultivalueLink on: self opposite: #keeper:. killedBy := FMMultiMultivalueLink on: self opposite: #kills.
It's the redundancy, along with the possibilities for error that would put the smalltalk and annotation out of sync, that concerns me (and that I'm trying to avoid).
I do not in some other classes the names in the annotations and the names of the methods and variables are not quite in sync, so maybe this is a feature. Ross
Right now, the only possibility is to generate Smalltalk code out of MSE code that represents your meta-model. This is like the famixCore string that you saw in the image.
Cheers, Doru
-- www.tudorgirba.com
"Not knowing how to do something is not an argument for how it cannot be done."
Moose-dev mailing list Moose-dev@iam.unibe.ch https://www.iam.unibe.ch/mailman/listinfo/moose-dev
Moose-dev mailing list Moose-dev@iam.unibe.ch https://www.iam.unibe.ch/mailman/listinfo/moose-dev
-- Simon
Thanks simon for this documentation
Alexandre
On 24 Jun 2010, at 18:21, Simon Denier wrote:
Hi Ross
It's true that the metamodeling tools in Moose seem crude compared to what exist elsewhere, but that's because we spend much more time using metamodels in new tools than building metamodels :)
Here is the procedure when you need to define a completely new metamodel. You can still make mistake when defining the metamodel but it eliminates most redundancy. Once you have defined your metamodel, a set of lint rules can be run to check some aspects of the metamodel (see below)
The essential pragma documentation is here (should be integrated in the book) http://code.google.com/p/moose-technology/wiki/FameMetaDescriptions
Also, a good rule of thumb is to have MooseEntity as root instead of Object, it facilitates integration with Moose tools.
- Metamodel declaration
For the example below:
- Create class RPGDragon from a class browser
- add (class-side)
RPGDragon class>>annotation <MSEClass: #Dragon super: #Object> <package: #RPG>
Notice I just define the method with pragmas, I don't add any code in the body for now.
For each attribute, I need to define the accessor (the getter, not the setter):
- define RPGDragon>>hoard <MSEProperty: #hoard type: #RPGTreasure opposite: #keeper><multivalued>
(same thing, method decl + pragma, no code) No need to add the initialize, it will be defined by the generator.
- To generate the implementation, do:
MooseModel generateClassesFrom: {RPGDragon. RPGHero. RPGWall} "collection of classes defining the metamodel"
It should open a preview window, where you can see which method will be defined/overridden by the generator. You can always tap Cmd-x to remove the selected definition (in case you are doing some incremental metamodel definition). You can repeat the generation step as much as you want, provided you understand it will overwrite existing methods.
It works like a charm if you are doing your own metamodel. If you need to extend (as in class extension, not the Java extend for inheritance) some existing Famix classes with new attributes, you should be a bit more careful. Actually, when I need to extend an existing metamodel with a few attributes, I just define the attributes directly in the Smalltalk way (with pragmas AND method body). It's fast to do.
- optionally, check that your metamodel is correct with lint rules (especially soundness of opposite relationships): can be done from the meta browser...
On 16 juin 2010, at 22:36, Ross Boylan wrote:
On Wed, 2010-06-16 at 22:14 +0200, Tudor Girba wrote:
Hi Ross,
On 16 Jun 2010, at 21:17, Ross Boylan wrote:
On Wed, 2010-06-16 at 09:37 +0200, Tudor Girba wrote:
As mentioned before, there is no support for graphical support for writing MSE files. The main reason here is that Smalltalk already provides a good enough solution for quickly building classes and methods, and annotating them is a practical solution.
I gather there is no facility to generate the smalltalk out of the method annotations. Is there any reason, in principle, that it couldn't be done?
Well, if you have the method annotations, you already have the Smalltalk code :).
Consider RPGDragon>>hoard <MSEProperty: #hoard type: #RPGTreasure opposite: #keeper>
<multivalued> ^hoard ^hoard is smalltalk code. You also have to create the instance variable, define hoard: aTreasure hoard value: aTreasure and initialize super initialize. hoard := FMMultivalueLink on: self opposite: #keeper:. killedBy := FMMultiMultivalueLink on: self opposite: #kills.
It's the redundancy, along with the possibilities for error that would put the smalltalk and annotation out of sync, that concerns me (and that I'm trying to avoid).
I do not in some other classes the names in the annotations and the names of the methods and variables are not quite in sync, so maybe this is a feature. Ross
Right now, the only possibility is to generate Smalltalk code out of MSE code that represents your meta-model. This is like the famixCore string that you saw in the image.
Cheers, Doru
-- www.tudorgirba.com
"Not knowing how to do something is not an argument for how it cannot be done."
Moose-dev mailing list Moose-dev@iam.unibe.ch https://www.iam.unibe.ch/mailman/listinfo/moose-dev
Moose-dev mailing list Moose-dev@iam.unibe.ch https://www.iam.unibe.ch/mailman/listinfo/moose-dev
-- Simon
Moose-dev mailing list Moose-dev@iam.unibe.ch https://www.iam.unibe.ch/mailman/listinfo/moose-dev
There was some questions last week about extending the Famix metamodel. I just forward the following guide I wrote last year. It should be formatted and put online too.
Begin forwarded message:
From: Simon Denier Simon.Denier@inria.fr Date: 24 juin 2010 18:21:34 HAEC To: Related to the development of Moose and other related tools moose-dev@iam.unibe.ch Cc: Simon Denier Simon.Denier@inria.fr Subject: Re: [Moose-dev] Re: Moose for SAS
Hi Ross
It's true that the metamodeling tools in Moose seem crude compared to what exist elsewhere, but that's because we spend much more time using metamodels in new tools than building metamodels :)
Here is the procedure when you need to define a completely new metamodel. You can still make mistake when defining the metamodel but it eliminates most redundancy. Once you have defined your metamodel, a set of lint rules can be run to check some aspects of the metamodel (see below)
The essential pragma documentation is here (should be integrated in the book) http://code.google.com/p/moose-technology/wiki/FameMetaDescriptions
Also, a good rule of thumb is to have MooseEntity as root instead of Object, it facilitates integration with Moose tools.
- Metamodel declaration
For the example below:
- Create class RPGDragon from a class browser
- add (class-side)
RPGDragon class>>annotation <MSEClass: #Dragon super: #Object> <package: #RPG>
Notice I just define the method with pragmas, I don't add any code in the body for now.
For each attribute, I need to define the accessor (the getter, not the setter):
- define RPGDragon>>hoard <MSEProperty: #hoard type: #RPGTreasure opposite: #keeper><multivalued>
(same thing, method decl + pragma, no code) No need to add the initialize, it will be defined by the generator.
- To generate the implementation, do:
MooseModel generateClassesFrom: {RPGDragon. RPGHero. RPGWall} "collection of classes defining the metamodel"
It should open a preview window, where you can see which method will be defined/overridden by the generator. You can always tap Cmd-x to remove the selected definition (in case you are doing some incremental metamodel definition). You can repeat the generation step as much as you want, provided you understand it will overwrite existing methods.
It works like a charm if you are doing your own metamodel. If you need to extend (as in class extension, not the Java extend for inheritance) some existing Famix classes with new attributes, you should be a bit more careful. Actually, when I need to extend an existing metamodel with a few attributes, I just define the attributes directly in the Smalltalk way (with pragmas AND method body). It's fast to do.
- optionally, check that your metamodel is correct with lint rules (especially soundness of opposite relationships): can be done from the meta browser...
On 16 juin 2010, at 22:36, Ross Boylan wrote:
On Wed, 2010-06-16 at 22:14 +0200, Tudor Girba wrote:
Hi Ross,
On 16 Jun 2010, at 21:17, Ross Boylan wrote:
On Wed, 2010-06-16 at 09:37 +0200, Tudor Girba wrote:
As mentioned before, there is no support for graphical support for writing MSE files. The main reason here is that Smalltalk already provides a good enough solution for quickly building classes and methods, and annotating them is a practical solution.
I gather there is no facility to generate the smalltalk out of the method annotations. Is there any reason, in principle, that it couldn't be done?
Well, if you have the method annotations, you already have the Smalltalk code :).
Consider RPGDragon>>hoard <MSEProperty: #hoard type: #RPGTreasure opposite: #keeper>
<multivalued> ^hoard ^hoard is smalltalk code. You also have to create the instance variable, define hoard: aTreasure hoard value: aTreasure and initialize super initialize. hoard := FMMultivalueLink on: self opposite: #keeper:. killedBy := FMMultiMultivalueLink on: self opposite: #kills.
It's the redundancy, along with the possibilities for error that would put the smalltalk and annotation out of sync, that concerns me (and that I'm trying to avoid).
I do not in some other classes the names in the annotations and the names of the methods and variables are not quite in sync, so maybe this is a feature. Ross
Right now, the only possibility is to generate Smalltalk code out of MSE code that represents your meta-model. This is like the famixCore string that you saw in the image.
Cheers, Doru
-- www.tudorgirba.com
"Not knowing how to do something is not an argument for how it cannot be done."
Moose-dev mailing list Moose-dev@iam.unibe.ch https://www.iam.unibe.ch/mailman/listinfo/moose-dev
Moose-dev mailing list Moose-dev@iam.unibe.ch https://www.iam.unibe.ch/mailman/listinfo/moose-dev
-- Simon
-- Simon Denier
I was discussing with Usman yesterday about these emails to the list that are also documentation artefacts.
It would be nice to have a mean to flag such emails (manually) so that they can form collectively some documentation of Moose that would be easier to search in than the entire mailing list archive.
nicolas
----- Mail original -----
De: "Simon Denier" simon.denier@gmail.com À: "Moose-related development" moose-dev@iam.unibe.ch Envoyé: Mardi 7 Juin 2011 09:40:55 Objet: [Moose-dev] Fwd: Re: Moose for SAS
There was some questions last week about extending the Famix metamodel. I just forward the following guide I wrote last year. It should be formatted and put online too.
Begin forwarded message:
From: Simon Denier < Simon.Denier@inria.fr >
Date: 24 juin 2010 18:21:34 HAEC
To: Related to the development of Moose and other related tools < moose-dev@iam.unibe.ch >
Cc: Simon Denier < Simon.Denier@inria.fr >
Subject: Re: [Moose-dev] Re: Moose for SAS
Hi Ross
It's true that the metamodeling tools in Moose seem crude compared to what exist elsewhere, but that's because we spend much more time using metamodels in new tools than building metamodels :)
Here is the procedure when you need to define a completely new metamodel. You can still make mistake when defining the metamodel but it eliminates most redundancy. Once you have defined your metamodel, a set of lint rules can be run to check some aspects of the metamodel (see below)
The essential pragma documentation is here (should be integrated in the book) http://code.google.com/p/moose-technology/wiki/FameMetaDescriptions
Also, a good rule of thumb is to have MooseEntity as root instead of Object, it facilitates integration with Moose tools.
1) Metamodel declaration For the example below: - Create class RPGDragon from a class browser - add (class-side) RPGDragon class>>annotation <MSEClass: #Dragon super: #Object> <package: #RPG>
Notice I just define the method with pragmas, I don't add any code in the body for now.
For each attribute, I need to define the accessor (the getter, not the setter): - define RPGDragon>>hoard <MSEProperty: #hoard type: #RPGTreasure opposite: #keeper><multivalued>
(same thing, method decl + pragma, no code) No need to add the initialize, it will be defined by the generator.
2) To generate the implementation, do: MooseModel generateClassesFrom: {RPGDragon. RPGHero. RPGWall} "collection of classes defining the metamodel"
It should open a preview window, where you can see which method will be defined/overridden by the generator. You can always tap Cmd-x to remove the selected definition (in case you are doing some incremental metamodel definition). You can repeat the generation step as much as you want, provided you understand it will overwrite existing methods.
It works like a charm if you are doing your own metamodel. If you need to extend (as in class extension, not the Java extend for inheritance) some existing Famix classes with new attributes, you should be a bit more careful. Actually, when I need to extend an existing metamodel with a few attributes, I just define the attributes directly in the Smalltalk way (with pragmas AND method body). It's fast to do.
3) optionally, check that your metamodel is correct with lint rules (especially soundness of opposite relationships): can be done from the meta browser...
On 16 juin 2010, at 22:36, Ross Boylan wrote:
On Wed, 2010-06-16 at 22:14 +0200, Tudor Girba wrote:
Hi Ross,
On 16 Jun 2010, at 21:17, Ross Boylan wrote:
On Wed, 2010-06-16 at 09:37 +0200, Tudor Girba wrote:
As mentioned before, there is no support for graphical support for
writing MSE files. The main reason here is that Smalltalk already
provides a good enough solution for quickly building classes and
methods, and annotating them is a practical solution.
I gather there is no facility to generate the smalltalk out of the
method annotations. Is there any reason, in principle, that it
couldn't
be done?
Well, if you have the method annotations, you already have the
Smalltalk code :).
Consider RPGDragon>>hoard
<MSEProperty: #hoard type: #RPGTreasure opposite: #keeper>
<multivalued>
^hoard
^hoard is smalltalk code. You also have to create the instance
variable, define
hoard: aTreasure
hoard value: aTreasure
and
initialize
super initialize.
hoard := FMMultivalueLink on: self opposite: #keeper:.
killedBy := FMMultiMultivalueLink on: self opposite: #kills.
It's the redundancy, along with the possibilities for error that would
put the smalltalk and annotation out of sync, that concerns me (and that
I'm trying to avoid).
I do not in some other classes the names in the annotations and the
names of the methods and variables are not quite in sync, so maybe this
is a feature.
Ross
Right now, the only possibility is to generate Smalltalk code out of
MSE code that represents your meta-model. This is like the famixCore
string that you saw in the image.
Cheers,
Doru
On 7 juin 2011, at 09:53, Nicolas Anquetil wrote:
I was discussing with Usman yesterday about these emails to the list that are also documentation artefacts.
It would be nice to have a mean to flag such emails (manually) so that they can form collectively some documentation of Moose that would be easier to search in than the entire mailing list archive.
I think the simplest solution would be to copy-paste the email on the website, with just a bit of formatting. Is it ok?
nicolas
De: "Simon Denier" simon.denier@gmail.com À: "Moose-related development" moose-dev@iam.unibe.ch Envoyé: Mardi 7 Juin 2011 09:40:55 Objet: [Moose-dev] Fwd: Re: Moose for SAS
There was some questions last week about extending the Famix metamodel. I just forward the following guide I wrote last year. It should be formatted and put online too.
Begin forwarded message:
From: Simon Denier Simon.Denier@inria.fr Date: 24 juin 2010 18:21:34 HAEC To: Related to the development of Moose and other related tools moose-dev@iam.unibe.ch Cc: Simon Denier Simon.Denier@inria.fr Subject: Re: [Moose-dev] Re: Moose for SAS
Hi Ross
It's true that the metamodeling tools in Moose seem crude compared to what exist elsewhere, but that's because we spend much more time using metamodels in new tools than building metamodels :)
Here is the procedure when you need to define a completely new metamodel. You can still make mistake when defining the metamodel but it eliminates most redundancy. Once you have defined your metamodel, a set of lint rules can be run to check some aspects of the metamodel (see below)
The essential pragma documentation is here (should be integrated in the book) http://code.google.com/p/moose-technology/wiki/FameMetaDescriptions
Also, a good rule of thumb is to have MooseEntity as root instead of Object, it facilitates integration with Moose tools.
- Metamodel declaration
For the example below:
- Create class RPGDragon from a class browser
- add (class-side)
RPGDragon class>>annotation <MSEClass: #Dragon super: #Object> <package: #RPG>
Notice I just define the method with pragmas, I don't add any code in the body for now.
For each attribute, I need to define the accessor (the getter, not the setter):
- define RPGDragon>>hoard <MSEProperty: #hoard type: #RPGTreasure opposite: #keeper><multivalued>
(same thing, method decl + pragma, no code) No need to add the initialize, it will be defined by the generator.
- To generate the implementation, do:
MooseModel generateClassesFrom: {RPGDragon. RPGHero. RPGWall} "collection of classes defining the metamodel"
It should open a preview window, where you can see which method will be defined/overridden by the generator. You can always tap Cmd-x to remove the selected definition (in case you are doing some incremental metamodel definition). You can repeat the generation step as much as you want, provided you understand it will overwrite existing methods.
It works like a charm if you are doing your own metamodel. If you need to extend (as in class extension, not the Java extend for inheritance) some existing Famix classes with new attributes, you should be a bit more careful. Actually, when I need to extend an existing metamodel with a few attributes, I just define the attributes directly in the Smalltalk way (with pragmas AND method body). It's fast to do.
- optionally, check that your metamodel is correct with lint rules (especially soundness of opposite relationships): can be done from the meta browser...
On 16 juin 2010, at 22:36, Ross Boylan wrote:
On Wed, 2010-06-16 at 22:14 +0200, Tudor Girba wrote: Hi Ross,
On 16 Jun 2010, at 21:17, Ross Boylan wrote:
On Wed, 2010-06-16 at 09:37 +0200, Tudor Girba wrote: As mentioned before, there is no support for graphical support for writing MSE files. The main reason here is that Smalltalk already provides a good enough solution for quickly building classes and methods, and annotating them is a practical solution. I gather there is no facility to generate the smalltalk out of the method annotations. Is there any reason, in principle, that it couldn't be done?
Well, if you have the method annotations, you already have the Smalltalk code :). Consider RPGDragon>>hoard <MSEProperty: #hoard type: #RPGTreasure opposite: #keeper>
<multivalued> ^hoard ^hoard is smalltalk code. You also have to create the instance variable, define hoard: aTreasure hoard value: aTreasure and initialize super initialize. hoard := FMMultivalueLink on: self opposite: #keeper:. killedBy := FMMultiMultivalueLink on: self opposite: #kills.
It's the redundancy, along with the possibilities for error that would put the smalltalk and annotation out of sync, that concerns me (and that I'm trying to avoid).
I do not in some other classes the names in the annotations and the names of the methods and variables are not quite in sync, so maybe this is a feature. Ross
Right now, the only possibility is to generate Smalltalk code out of MSE code that represents your meta-model. This is like the famixCore string that you saw in the image.
Cheers, Doru
-- www.tudorgirba.com
"Not knowing how to do something is not an argument for how it cannot be done."
Moose-dev mailing list Moose-dev@iam.unibe.ch https://www.iam.unibe.ch/mailman/listinfo/moose-dev
Moose-dev mailing list Moose-dev@iam.unibe.ch https://www.iam.unibe.ch/mailman/listinfo/moose-dev
-- Simon
-- Simon Denier
Moose-dev mailing list Moose-dev@iam.unibe.ch https://www.iam.unibe.ch/mailman/listinfo/moose-dev
Moose-dev mailing list Moose-dev@iam.unibe.ch https://www.iam.unibe.ch/mailman/listinfo/moose-dev
-- Simon Denier
Yes :)
Doru
On 7 Jun 2011, at 11:20, Simon Denier wrote:
On 7 juin 2011, at 09:53, Nicolas Anquetil wrote:
I was discussing with Usman yesterday about these emails to the list that are also documentation artefacts.
It would be nice to have a mean to flag such emails (manually) so that they can form collectively some documentation of Moose that would be easier to search in than the entire mailing list archive.
I think the simplest solution would be to copy-paste the email on the website, with just a bit of formatting. Is it ok?
nicolas
De: "Simon Denier" simon.denier@gmail.com À: "Moose-related development" moose-dev@iam.unibe.ch Envoyé: Mardi 7 Juin 2011 09:40:55 Objet: [Moose-dev] Fwd: Re: Moose for SAS
There was some questions last week about extending the Famix metamodel. I just forward the following guide I wrote last year. It should be formatted and put online too.
Begin forwarded message:
From: Simon Denier Simon.Denier@inria.fr Date: 24 juin 2010 18:21:34 HAEC To: Related to the development of Moose and other related tools moose-dev@iam.unibe.ch Cc: Simon Denier Simon.Denier@inria.fr Subject: Re: [Moose-dev] Re: Moose for SAS
Hi Ross
It's true that the metamodeling tools in Moose seem crude compared to what exist elsewhere, but that's because we spend much more time using metamodels in new tools than building metamodels :)
Here is the procedure when you need to define a completely new metamodel. You can still make mistake when defining the metamodel but it eliminates most redundancy. Once you have defined your metamodel, a set of lint rules can be run to check some aspects of the metamodel (see below)
The essential pragma documentation is here (should be integrated in the book) http://code.google.com/p/moose-technology/wiki/FameMetaDescriptions
Also, a good rule of thumb is to have MooseEntity as root instead of Object, it facilitates integration with Moose tools.
- Metamodel declaration
For the example below:
- Create class RPGDragon from a class browser
- add (class-side)
RPGDragon class>>annotation <MSEClass: #Dragon super: #Object> <package: #RPG>
Notice I just define the method with pragmas, I don't add any code in the body for now.
For each attribute, I need to define the accessor (the getter, not the setter):
- define RPGDragon>>hoard <MSEProperty: #hoard type: #RPGTreasure opposite: #keeper><multivalued>
(same thing, method decl + pragma, no code) No need to add the initialize, it will be defined by the generator.
- To generate the implementation, do:
MooseModel generateClassesFrom: {RPGDragon. RPGHero. RPGWall} "collection of classes defining the metamodel"
It should open a preview window, where you can see which method will be defined/overridden by the generator. You can always tap Cmd-x to remove the selected definition (in case you are doing some incremental metamodel definition). You can repeat the generation step as much as you want, provided you understand it will overwrite existing methods.
It works like a charm if you are doing your own metamodel. If you need to extend (as in class extension, not the Java extend for inheritance) some existing Famix classes with new attributes, you should be a bit more careful. Actually, when I need to extend an existing metamodel with a few attributes, I just define the attributes directly in the Smalltalk way (with pragmas AND method body). It's fast to do.
- optionally, check that your metamodel is correct with lint rules (especially soundness of opposite relationships): can be done from the meta browser...
On 16 juin 2010, at 22:36, Ross Boylan wrote:
On Wed, 2010-06-16 at 22:14 +0200, Tudor Girba wrote: Hi Ross,
On 16 Jun 2010, at 21:17, Ross Boylan wrote:
On Wed, 2010-06-16 at 09:37 +0200, Tudor Girba wrote: As mentioned before, there is no support for graphical support for writing MSE files. The main reason here is that Smalltalk already provides a good enough solution for quickly building classes and methods, and annotating them is a practical solution. I gather there is no facility to generate the smalltalk out of the method annotations. Is there any reason, in principle, that it couldn't be done?
Well, if you have the method annotations, you already have the Smalltalk code :). Consider RPGDragon>>hoard <MSEProperty: #hoard type: #RPGTreasure opposite: #keeper>
<multivalued> ^hoard ^hoard is smalltalk code. You also have to create the instance variable, define hoard: aTreasure hoard value: aTreasure and initialize super initialize. hoard := FMMultivalueLink on: self opposite: #keeper:. killedBy := FMMultiMultivalueLink on: self opposite: #kills.
It's the redundancy, along with the possibilities for error that would put the smalltalk and annotation out of sync, that concerns me (and that I'm trying to avoid).
I do not in some other classes the names in the annotations and the names of the methods and variables are not quite in sync, so maybe this is a feature. Ross
Right now, the only possibility is to generate Smalltalk code out of MSE code that represents your meta-model. This is like the famixCore string that you saw in the image.
Cheers, Doru
-- www.tudorgirba.com
"Not knowing how to do something is not an argument for how it cannot be done."
Moose-dev mailing list Moose-dev@iam.unibe.ch https://www.iam.unibe.ch/mailman/listinfo/moose-dev
Moose-dev mailing list Moose-dev@iam.unibe.ch https://www.iam.unibe.ch/mailman/listinfo/moose-dev
-- Simon
-- Simon Denier
Moose-dev mailing list Moose-dev@iam.unibe.ch https://www.iam.unibe.ch/mailman/listinfo/moose-dev
Moose-dev mailing list Moose-dev@iam.unibe.ch https://www.iam.unibe.ch/mailman/listinfo/moose-dev
-- Simon Denier
Moose-dev mailing list Moose-dev@iam.unibe.ch https://www.iam.unibe.ch/mailman/listinfo/moose-dev
-- www.tudorgirba.com
"Problem solving efficiency grows with the abstractness level of problem understanding."
+1 on that. Having selected e-mails as a reference/documentation on-line is much better than having to search the mailing lists.
On 07 Jun 2011, at 05:26, Tudor Girba wrote:
Yes :)
Doru
On 7 Jun 2011, at 11:20, Simon Denier wrote:
On 7 juin 2011, at 09:53, Nicolas Anquetil wrote:
I was discussing with Usman yesterday about these emails to the list that are also documentation artefacts.
It would be nice to have a mean to flag such emails (manually) so that they can form collectively some documentation of Moose that would be easier to search in than the entire mailing list archive.
I think the simplest solution would be to copy-paste the email on the website, with just a bit of formatting. Is it ok?
nicolas
De: "Simon Denier" simon.denier@gmail.com À: "Moose-related development" moose-dev@iam.unibe.ch Envoyé: Mardi 7 Juin 2011 09:40:55 Objet: [Moose-dev] Fwd: Re: Moose for SAS
There was some questions last week about extending the Famix metamodel. I just forward the following guide I wrote last year. It should be formatted and put online too.
Begin forwarded message:
From: Simon Denier Simon.Denier@inria.fr Date: 24 juin 2010 18:21:34 HAEC To: Related to the development of Moose and other related tools moose-dev@iam.unibe.ch Cc: Simon Denier Simon.Denier@inria.fr Subject: Re: [Moose-dev] Re: Moose for SAS
Hi Ross
It's true that the metamodeling tools in Moose seem crude compared to what exist elsewhere, but that's because we spend much more time using metamodels in new tools than building metamodels :)
Here is the procedure when you need to define a completely new metamodel. You can still make mistake when defining the metamodel but it eliminates most redundancy. Once you have defined your metamodel, a set of lint rules can be run to check some aspects of the metamodel (see below)
The essential pragma documentation is here (should be integrated in the book) http://code.google.com/p/moose-technology/wiki/FameMetaDescriptions
Also, a good rule of thumb is to have MooseEntity as root instead of Object, it facilitates integration with Moose tools.
- Metamodel declaration
For the example below:
- Create class RPGDragon from a class browser
- add (class-side)
RPGDragon class>>annotation <MSEClass: #Dragon super: #Object> <package: #RPG>
Notice I just define the method with pragmas, I don't add any code in the body for now.
For each attribute, I need to define the accessor (the getter, not the setter):
- define RPGDragon>>hoard <MSEProperty: #hoard type: #RPGTreasure opposite: #keeper><multivalued>
(same thing, method decl + pragma, no code) No need to add the initialize, it will be defined by the generator.
- To generate the implementation, do:
MooseModel generateClassesFrom: {RPGDragon. RPGHero. RPGWall} "collection of classes defining the metamodel"
It should open a preview window, where you can see which method will be defined/overridden by the generator. You can always tap Cmd-x to remove the selected definition (in case you are doing some incremental metamodel definition). You can repeat the generation step as much as you want, provided you understand it will overwrite existing methods.
It works like a charm if you are doing your own metamodel. If you need to extend (as in class extension, not the Java extend for inheritance) some existing Famix classes with new attributes, you should be a bit more careful. Actually, when I need to extend an existing metamodel with a few attributes, I just define the attributes directly in the Smalltalk way (with pragmas AND method body). It's fast to do.
- optionally, check that your metamodel is correct with lint rules (especially soundness of opposite relationships): can be done from the meta browser...
On 16 juin 2010, at 22:36, Ross Boylan wrote:
On Wed, 2010-06-16 at 22:14 +0200, Tudor Girba wrote: Hi Ross,
On 16 Jun 2010, at 21:17, Ross Boylan wrote:
On Wed, 2010-06-16 at 09:37 +0200, Tudor Girba wrote: As mentioned before, there is no support for graphical support for writing MSE files. The main reason here is that Smalltalk already provides a good enough solution for quickly building classes and methods, and annotating them is a practical solution. I gather there is no facility to generate the smalltalk out of the method annotations. Is there any reason, in principle, that it couldn't be done?
Well, if you have the method annotations, you already have the Smalltalk code :). Consider RPGDragon>>hoard <MSEProperty: #hoard type: #RPGTreasure opposite: #keeper>
<multivalued> ^hoard ^hoard is smalltalk code. You also have to create the instance variable, define hoard: aTreasure hoard value: aTreasure and initialize super initialize. hoard := FMMultivalueLink on: self opposite: #keeper:. killedBy := FMMultiMultivalueLink on: self opposite: #kills.
It's the redundancy, along with the possibilities for error that would put the smalltalk and annotation out of sync, that concerns me (and that I'm trying to avoid).
I do not in some other classes the names in the annotations and the names of the methods and variables are not quite in sync, so maybe this is a feature. Ross
Right now, the only possibility is to generate Smalltalk code out of MSE code that represents your meta-model. This is like the famixCore string that you saw in the image.
Cheers, Doru
-- www.tudorgirba.com
"Not knowing how to do something is not an argument for how it cannot be done."
Moose-dev mailing list Moose-dev@iam.unibe.ch https://www.iam.unibe.ch/mailman/listinfo/moose-dev
Moose-dev mailing list Moose-dev@iam.unibe.ch https://www.iam.unibe.ch/mailman/listinfo/moose-dev
-- Simon
-- Simon Denier
Moose-dev mailing list Moose-dev@iam.unibe.ch https://www.iam.unibe.ch/mailman/listinfo/moose-dev
Moose-dev mailing list Moose-dev@iam.unibe.ch https://www.iam.unibe.ch/mailman/listinfo/moose-dev
-- Simon Denier
Moose-dev mailing list Moose-dev@iam.unibe.ch https://www.iam.unibe.ch/mailman/listinfo/moose-dev
-- www.tudorgirba.com
"Problem solving efficiency grows with the abstractness level of problem understanding."
Moose-dev mailing list Moose-dev@iam.unibe.ch https://www.iam.unibe.ch/mailman/listinfo/moose-dev
-- Johan Fabry jfabry@dcc.uchile.cl - http://dcc.uchile.cl/~jfabry PLEIAD Lab - Computer Science Department (DCC) - University of Chile
I posted a first version here, which is possibly incomplete http://www.moosetechnology.org/docs/famix/extendingFamix
It should also include http://code.google.com/p/moose-technology/wiki/FameMetaDescriptions - how to reset the metamodel - how to run lint rules - what are the lint rules for metamodel
On 7 juin 2011, at 16:18, Johan Fabry wrote:
+1 on that. Having selected e-mails as a reference/documentation on-line is much better than having to search the mailing lists.
On 07 Jun 2011, at 05:26, Tudor Girba wrote:
Yes :)
Doru
On 7 Jun 2011, at 11:20, Simon Denier wrote:
On 7 juin 2011, at 09:53, Nicolas Anquetil wrote:
I was discussing with Usman yesterday about these emails to the list that are also documentation artefacts.
It would be nice to have a mean to flag such emails (manually) so that they can form collectively some documentation of Moose that would be easier to search in than the entire mailing list archive.
I think the simplest solution would be to copy-paste the email on the website, with just a bit of formatting. Is it ok?
nicolas
De: "Simon Denier" simon.denier@gmail.com À: "Moose-related development" moose-dev@iam.unibe.ch Envoyé: Mardi 7 Juin 2011 09:40:55 Objet: [Moose-dev] Fwd: Re: Moose for SAS
There was some questions last week about extending the Famix metamodel. I just forward the following guide I wrote last year. It should be formatted and put online too.
Begin forwarded message:
From: Simon Denier Simon.Denier@inria.fr Date: 24 juin 2010 18:21:34 HAEC To: Related to the development of Moose and other related tools moose-dev@iam.unibe.ch Cc: Simon Denier Simon.Denier@inria.fr Subject: Re: [Moose-dev] Re: Moose for SAS
Hi Ross
It's true that the metamodeling tools in Moose seem crude compared to what exist elsewhere, but that's because we spend much more time using metamodels in new tools than building metamodels :)
Here is the procedure when you need to define a completely new metamodel. You can still make mistake when defining the metamodel but it eliminates most redundancy. Once you have defined your metamodel, a set of lint rules can be run to check some aspects of the metamodel (see below)
The essential pragma documentation is here (should be integrated in the book) http://code.google.com/p/moose-technology/wiki/FameMetaDescriptions
Also, a good rule of thumb is to have MooseEntity as root instead of Object, it facilitates integration with Moose tools.
- Metamodel declaration
For the example below:
- Create class RPGDragon from a class browser
- add (class-side)
RPGDragon class>>annotation <MSEClass: #Dragon super: #Object> <package: #RPG>
Notice I just define the method with pragmas, I don't add any code in the body for now.
For each attribute, I need to define the accessor (the getter, not the setter):
- define RPGDragon>>hoard <MSEProperty: #hoard type: #RPGTreasure opposite: #keeper><multivalued>
(same thing, method decl + pragma, no code) No need to add the initialize, it will be defined by the generator.
- To generate the implementation, do:
MooseModel generateClassesFrom: {RPGDragon. RPGHero. RPGWall} "collection of classes defining the metamodel"
It should open a preview window, where you can see which method will be defined/overridden by the generator. You can always tap Cmd-x to remove the selected definition (in case you are doing some incremental metamodel definition). You can repeat the generation step as much as you want, provided you understand it will overwrite existing methods.
It works like a charm if you are doing your own metamodel. If you need to extend (as in class extension, not the Java extend for inheritance) some existing Famix classes with new attributes, you should be a bit more careful. Actually, when I need to extend an existing metamodel with a few attributes, I just define the attributes directly in the Smalltalk way (with pragmas AND method body). It's fast to do.
- optionally, check that your metamodel is correct with lint rules (especially soundness of opposite relationships): can be done from the meta browser...
On 16 juin 2010, at 22:36, Ross Boylan wrote:
On Wed, 2010-06-16 at 22:14 +0200, Tudor Girba wrote: Hi Ross,
On 16 Jun 2010, at 21:17, Ross Boylan wrote:
On Wed, 2010-06-16 at 09:37 +0200, Tudor Girba wrote: As mentioned before, there is no support for graphical support for writing MSE files. The main reason here is that Smalltalk already provides a good enough solution for quickly building classes and methods, and annotating them is a practical solution. I gather there is no facility to generate the smalltalk out of the method annotations. Is there any reason, in principle, that it couldn't be done?
Well, if you have the method annotations, you already have the Smalltalk code :). Consider RPGDragon>>hoard <MSEProperty: #hoard type: #RPGTreasure opposite: #keeper>
<multivalued> ^hoard ^hoard is smalltalk code. You also have to create the instance variable, define hoard: aTreasure hoard value: aTreasure and initialize super initialize. hoard := FMMultivalueLink on: self opposite: #keeper:. killedBy := FMMultiMultivalueLink on: self opposite: #kills.
It's the redundancy, along with the possibilities for error that would put the smalltalk and annotation out of sync, that concerns me (and that I'm trying to avoid).
I do not in some other classes the names in the annotations and the names of the methods and variables are not quite in sync, so maybe this is a feature. Ross
Right now, the only possibility is to generate Smalltalk code out of MSE code that represents your meta-model. This is like the famixCore string that you saw in the image.
Cheers, Doru
-- www.tudorgirba.com
"Not knowing how to do something is not an argument for how it cannot be done."
Moose-dev mailing list Moose-dev@iam.unibe.ch https://www.iam.unibe.ch/mailman/listinfo/moose-dev
Moose-dev mailing list Moose-dev@iam.unibe.ch https://www.iam.unibe.ch/mailman/listinfo/moose-dev
-- Simon
-- Simon Denier
Moose-dev mailing list Moose-dev@iam.unibe.ch https://www.iam.unibe.ch/mailman/listinfo/moose-dev
Moose-dev mailing list Moose-dev@iam.unibe.ch https://www.iam.unibe.ch/mailman/listinfo/moose-dev
-- Simon Denier
Moose-dev mailing list Moose-dev@iam.unibe.ch https://www.iam.unibe.ch/mailman/listinfo/moose-dev
-- www.tudorgirba.com
"Problem solving efficiency grows with the abstractness level of problem understanding."
Moose-dev mailing list Moose-dev@iam.unibe.ch https://www.iam.unibe.ch/mailman/listinfo/moose-dev
-- Johan Fabry jfabry@dcc.uchile.cl - http://dcc.uchile.cl/~jfabry PLEIAD Lab - Computer Science Department (DCC) - University of Chile
Moose-dev mailing list Moose-dev@iam.unibe.ch https://www.iam.unibe.ch/mailman/listinfo/moose-dev
-- Simon Denier
Just to update people on my SAS work: I got caught in a maze of twisty passages with the SAS preprocessor and macro language, and solved my immediate SAS needs by hand.
One alternative would be to parse a SAS log instead of the SAS code, where the log is generated with SAS options like mprint to print the results of macro expansion (and some other options I don't recall for printing code from included files). This would replace the relatively tricky task of getting the SAS preprocessor right with the more straightforward one of extracting the expanded code from the log file. However, it would require old code to be rerun, which might be a problem.
Replicating the behavior of the SAS preprocessor is not an all or nothing activity; there are various tricky cases that are relatively unlikely to arise.
Ross
On Tue, 2011-06-07 at 09:40 +0200, Simon Denier wrote:
There was some questions last week about extending the Famix metamodel. I just forward the following guide I wrote last year. It should be formatted and put online too.
Begin forwarded message:
From: Simon Denier Simon.Denier@inria.fr
Date: 24 juin 2010 18:21:34 HAEC
To: Related to the development of Moose and other related tools moose-dev@iam.unibe.ch
Cc: Simon Denier Simon.Denier@inria.fr
Subject: Re: [Moose-dev] Re: Moose for SAS
Hi Ross
It's true that the metamodeling tools in Moose seem crude compared to what exist elsewhere, but that's because we spend much more time using metamodels in new tools than building metamodels :)
......
On Tue, 2010-06-15 at 17:27 -0700, Ross Boylan wrote:
On Wed, 2010-06-16 at 02:17 +0200, Tudor Girba wrote:
I do not know SAS, but you should check to see if the concepts present in FAMIX are of use to you. If yes, you can directly instantiate them from your parser. If not, then you will want to code a new meta-model, or extend the current ones to accommodate your needs. You can then describe this new meta-model using FM3 (this is achieved via Smalltalk pragmas) and you get a browser and import-export in the MSE format for free.
Once you have your Smalltalk objects as instances of the meta-model, you can then use Mondrian and Glamour for building visual tools, or the MooseAlgos to express graph-specific algorithms.
Does this help?
Well, I was hoping that I would need to specify the meta-model or the meta-meta-model, but not both.
Whoops: I think I meant model or meta-model.
That is, I was hoping the tools could build one out of the other. It sounds as if you're telling me I need to do both.
I hope I'm misunderstanding :)
Also, is there a guide to what the pragmas are?
I meant the FM3 pragmas.
Ross
-
Alexandre Bergel -
Johan Fabry -
Nicolas Anquetil -
Ross Boylan -
Simon Denier -
Simon Denier -
Tudor Girba -
Tudor Girba