[
Date Prev][
Date Next][
Thread Prev][
Thread Next][
Date Index][
Thread Index]
[
List Home]
|
RE: [cdt-dev] CModel vs. DOM
|
Great question, Markus. I guess the question to ask is, with these new
views, are users going to expect the same menus they get in the Outline and
C/C++ Project Views. I'd say yes, in which case you need to use ICElement
objects to keep the objectContributions working.
My vision for ICElement is to bring it closer to the DOM, and in fact, be
created from the DOM objects using the same parsing mode I use for the Fast
indexer. As a part of this, we need to ensure that we can easily navigate
between CModel and DOM (which we do BTW for the References and Declarations
search menu items).
Once I finish the PDOM work in 3.1.1, I will focus on this activity. My main
intention is to remove the old parser as the last in the clean up work I
started in 3.1.
In the meantime, if there are any changes to ICElement to make your work
easier, let me know and we can make changes.
Doug Schaefer
QNX Software Systems
Eclipse CDT Project Lead
http://cdtdoug.blogspot.com
-----Original Message-----
From: cdt-dev-bounces@xxxxxxxxxxx [mailto:cdt-dev-bounces@xxxxxxxxxxx] On
Behalf Of Schorn, Markus
Sent: Tuesday, August 01, 2006 4:57 AM
To: CDT General developers list.
Subject: [cdt-dev] CModel vs. DOM
Hi,
while implementing the call-hierarchy I have to make the decision how to
model
the functions/methods in the tree. I have the choice of using the CModel
(ICElement) or the DOM (IASTName, IBinding).
It is not a good idea to hold on to objects of type IASTName and
IBinding when they are generated by the parser. This would use up way
too much memory. The situation may be different with objects obtained
from the PDom. If it is ok to use those objects I will be heading this
way. Doug, is it ok?
Using ICElements causes the problem of bridging back to DOM objects when
doing
queries in the PDom. When we want to go this way, then I think the query
API to be created on top of the PDom has to work with ICElements itself.
Markus.
_______________________________________________
cdt-dev mailing list
cdt-dev@xxxxxxxxxxx
https://dev.eclipse.org/mailman/listinfo/cdt-dev
