Re: [cdt-dev] Vote for new CDT features

["Alain Magloire" <alain@xxxxxxx>]

> 
> Hey all, now that we've extracted the builders out of the CDT "platform", 
> the CDT platform features become pretty useless on their own.  To help the 
> user who is simply downloading the CDT from the update site, I see two 
> alternatives to make their life easier (and one not):
> 
> 1) Create new CDT "product" features that include the CDT platform and the 
> two make builders. This would be one new feature/plugin combination for 
> each os.ws we currently have.
> 2) Include dependencies from the current platform features to the two make 
> builders and have those who want to remove the builders from their 
> products to hand edit the feature.xml files off stream.
> 3) Too bad, they'll just have to get used to downloading all the features 
> the want individually and we'll deal with the user mistakes in the 
> newgroups/bugzilla.
> 
> I'd like to see a vote by the various committers on this as to which 
> alternative they prefer.
> 
> I am +1 on option 2, -1 on option 3, and a zero (in more ways than one) on 
> option 1.
> 

I like:
(1) +1
(2) -1
(3) 0


The rational:
- CDT is also a framework that in itself does not do much except to provide
  a "rendez-vous" point for all the modules to work in cooperation.
  It provides basic/common C/C++ environment i.e. CEditor, C-Parser, views etc ...

- The full potential can be reach when "enhance" with the appropriate plugins.

- CDT should come with a complete implementation of those frameworks, GDB for the
  debugger and managed make for the builder etc ..

I can see the complete CDT product i.e. with gdb/mi, managed make, GNU make
etc ... as a brand(default) product.

And let other compagnies, TimeSys, Tensillica, QnX have there own product base on
the core.