Re: [cdt-core-dev] Refactoring

[Douglas Schaefer <dschaefe@xxxxxxxxxx>]

I think we're almost ready on the AST front as well. We've done a lot of 
thinking and prototyping and I think we're now on a path towards the AST 
architecture we think will meet the most needs. I'll put together a 
document describing this by the end of this week (if not sooner). I have 
no firm opinion on how to proceed with writing out AST changes. The JDT 
rewriter attempts to be generic, but in the end, you need a lot of 
knowledge about the detailed structure of the AST to properly make 
changes. I'll take another look at your proposal and see how it meshes 
with our AST.

I think getting together would be a great idea. I'm sure the QNX folks 
here in Ottawa would also be interested (although maybe closer to the end 
of 2.1). We'll also set up a tele/web conference for others who are 
interested but can't make it out here to the "Great White North" (although 
it's not white here, yet...). When would be a good time for everyone?

Doug Schaefer, IBM's Technical Lead, Eclipse CDT
Ottawa Lab, IBM Canada, Rational Software Division



Markus.Schorn@xxxxxxxxxxxxx 
Sent by: cdt-core-dev-admin@xxxxxxxxxxx
11/15/2004 06:09 AM
Please respond to
cdt-core-dev


To
Douglas Schaefer/Ottawa/IBM@IBMCA
cc
cdt-core-dev@xxxxxxxxxxx
Subject
[cdt-core-dev] Refactoring






Hi Doug,
as you are about to release CDT 2.1 it might be a good point to
reengage in the attempt to put refactoring in place. As this
does require some colaboration between AST and the refactoring
code I cannot just go ahead without your help.
So I ask for getting towards a decision on how to interface with
the AST and I can see two approaches:

The first one is along the lines of JDT, making use of an AST-
rewriter that does not modify the AST. This approach is quite
complex, though promising in the long run.

I tend to follow an easier approach first, keeping the doors
open to bring in the rewriter at a later point. This has the
advantage that I can get to some working implementation faster.
I have summarized what I need to get out of an AST for this
simpler approach at:
https://bugs.eclipse.org/bugs/show_bug.cgi?id=75364
 
If it helps, I am also ready to make a visit in Canada to make
sure that our ideas are in line.

Markus.
_______________________________________________
cdt-core-dev mailing list
cdt-core-dev@xxxxxxxxxxx
http://dev.eclipse.org/mailman/listinfo/cdt-core-dev