Skip to main content

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index] [List Home]
RE: parser contributions soapbox (was Re: [cdt-dev] FW: [eclipse-dev] C#plugin for Eclipse)

Hi John,

 

Thanks for speaking up with a voice of experience and reason.

 

Regards,

Leo

 


From: cdt-dev-bounces@xxxxxxxxxxx [mailto:cdt-dev-bounces@xxxxxxxxxxx] On Behalf Of John Camelon
Sent: Tuesday, June 13, 2006 11:14 PM
To: CDT General developers list.
Subject: parser contributions soapbox (was Re: [cdt-dev] FW: [eclipse-dev] C#plugin for Eclipse)

 


Ryan Hapgood wrote on 06/13/2006 05:45:57 PM:
> It may also be a good idea to consider objective C when going through  
> any design steps. This would make things much easier when Bug 68083  
> (https://bugs.eclipse.org/bugs/show_bug.cgi?id=68083) gets more votes/
> support.
>
> Ryan.

Just wanted to tell it like it is, from beyond the CDT-grave ... I set
that defect to "Future" almost 2 years ago BTW. :-)

Speaking from my experience, it takes a considerable development effort
to implement, test & support these type of features for a varied cross
product of target languages, compilers, os's & architectures.  Unless
some company steps up to the table to fund development & testing of
Objective-C extensions, it will not happen, and I cannot see how
any number of bugzilla votes will make a real difference.  

In CDT 3.1, Doug has done a comparable amount of work to what the IBM
CDT team did in CDT 3.0.  While the effort and result is quite impressive,
I worry as to whether or not it is sustainable, particularly as we consider
new languages (Fortran, C#).  I have tried to chip in despite my current
work commitments, but my contributions have unfortunately been just a
drop in the bucket.

I do agree that the long-term goal of the CDT must include support for a
greater variety of compiled languages; I do not believe that we should fool
ourselves into believing that its just a matter of finding the right
architecture and that these parsers, doms & indexers will just build,
test and tune themselves.  I would consider this effort to be that of a
marathon, not a sprint, and I hope that those who keep bringing the topic
up are interested in going the full distance, because that is just how long
it may take to get there.

Thanks
JohnC


Back to the top