PetitJava AST model status

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

PetitJava AST model status

cbc
So, it looks like everyone that wanted to speak up has spoken up.  From a strictly numeric standpoint it looks like the consensus is tied - 3 for strict grammar adherence, and 3 for allowing it to be looser (maybe significantly).  But, popular vote isn't really the best way to solve technical problems.

After reviewing the arguments again, the case for following the published grammar does make a lot of sense, especially since there is one.  For my purposes, I'll either subclass that grammar or figure out how to use it directly (probably by extended it in my image as necessary).  I still like a more focused AST, but given that I am not a Java expert, I definitely shouldn't be the one creating it - at least not the general one.

So, how/when should I/we go about converting the current AST over?  I will be interested in converting at least the parts that I want - which is more than there is now, but not the entire grammar.  Note that this will change any Visitors using the AST, meaning that the Fast build will probably continue to fail as work is progressing.

Might I suggest that a symbolic version be made against PetitJava that the Fast job uses - that uses a static version of PetitJava. This would stop the job from failing; the static version can be updated when you are ready to incorporate the changes in.

Thanks,
Chris

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

Re: PetitJava AST model status

Stephan Eggermont-3
Sounds like a good idea, if you have the time.

How do you build the AST incrementally and iteratively,
so you can show (end-user) results every day?

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