Re: Emacs as C++-IDE

[Jorgen Grahn <grahn+nntp@snipabacken.se>]

On Thu, 2011-05-05, Nicolas Neuss wrote:
> Hello,
>
> a question of interest [*] to the C++ programmers using Emacs.
>
> I asked a C++ developer who had written up something about using
> KDevelop as C++-IDE what were the advantages compared with Emacs, and he
> answered as follows:
>
>   http://www.dealii.org/~archiver/dealii/msg06854.html
>
> Is his description adequate?  Or has Emacs comparable functionality
> (maybe via TAGS, or others).

He describes some of the things IDE people like to do, but doesn't
really contrast it with what Emacs does.

I don't know ... TAGS and dabbrev-expand seem to fill my needs in the
areas he describes[1].  I don't feel a need for popups telling me
which blanks to fill in -- if I need to, I find-tag into another
buffer where the definitions are, and type+dabbrev-expand the code.

He also lists the "project" concept as a positive thing, but fails to
explain *why*.  I have no idea why people find that useful; to me it
just seems to get in the way.  We already have version control + file
system + build system + TAGS to tell us what the sources are; why add
yet another layer?

[1] One exception: I'm not at all happy with how find-tag finds "foo",
    but isn't aware that I'm interested in Foo::foo(int) rather than
    Bar::foo() or char* foo(Baz*) ... or any of a dozen things called
    "foo".  This is something that C++ IDEs probably do a *lot* better
    than the stock etags program.  If there's a more context-aware
    TAGS for C++, I'm very interested to hear about it.

/Jorgen

-- 
  // Jorgen Grahn <grahn@  Oo  o.   .  .
\X/     snipabacken.se>   O  o   .