Re: A guide on setting up C/C++ development environment for Emacs

[Óscar Fuentes]

David Engster <deng@randomsample.de> writes:

> Óscar Fuentes writes:
>> However, I don't see CEDET's team throwing away their Elisp parser and
>> start coding one on C/C++. They could use Clang's right away, or GCC's
>> once the modularization is completed, but they insist on their own's.
>
> Stop this, Óscar. I think you're very well aware that the clang binary
> in itself only provides smart completions, nothing more. 

I'm aware of the possibilities provided by libclang, which go *far*
beyond than whatever the executable does.

> I *have* added
> support for this in CEDET. I'm pretty sure you know that as well,
> because I've already written this to you the last time this was
> discussed.

Yes, there are a number of packages that interact with Clang. That is
unrelated to the topic of this sub-thread: the usability of CEDET's
built-in C++ parser.

> What you are talking about is walking the AST using libclang/libtooling,
> which is a whole different matter and nothing which one does "right
> away". Just look at projects like rtags which do that, and see what it
> entails.

That's about 0.001% of the work required for implementing a C++
parser/analyzer.

> This is just to set things straight for others which are reading this. I
> have absolutely no interest in further discussing this with you.

Fair enough. And as I mentioned on that other discussion, this is not
about you, David. Please don't take this personally.

What irks me is that, whenever Emacs is compared to modern IDEs and
features like smart code completion, refactoring, advanced code
navigation and querying are discussed, CEDET is the Great Hand-waving
Enabler. Since CEDET was regarded as the solution by the community,
years passed, C++11 came, then C++14 and fact is that CEDET still can't
handle C++ code that was ordinary circa 1995. The net effect is that
Emacs is no longer competitive on the C++ realm and I cannot recommend
it to my C++ colleagues and students. The only hope is that some third
party package fixes this outside of the core Emacs development community
(be it on top of CEDET or not).

Of course, Emacs can also consider C++ a lost cause. After all, C++ has
just a fraction of the user base of Java.