Re: [CEDET-devel] EDE-compdb now available through MELPA
Brought to you by:
zappo
From: Alastair R. <ala...@gi...> - 2014-03-09 19:47:55
|
On Mar 9, 2014, at 3:41 AM, David Engster <de...@ra...> wrote: > Well, I think we should really merge this package into CEDET. Would you > be OK with that? Yes, absolutely. My motivation for making it available as a package on MELPA was merely to enable more feedback sooner. As a package, people can install it with minimal fuss, without upgrading to the latest CEDET or Emacs or both. Also having it in a separate repo lets me make changes with a lower risk of impacting anything else. In the long run I can see this would make sense to be part of CEDET, and I'm happy to spend time on doing this. > If so, would you like to maintain this package directly in the bzr > repository? Of course, you could still use GitHub as your primary > repository if you prefer, but if you could take care of merging your > commits to the CEDET repository, that would be really appreciated. The > only thing you need for this is a SourceForge account and Eric can then > add you to the CEDET project. Yeah no problems - I am "alastair" on sourceforge. > I know that some people don't like dealing with bzr, but given that > Emacs is switching to git, I have a hunch we'll do the same eventually, > so bzr is probably a temporary thing by now. Bzr doesn't worry me - I used it for many years at $WORK. > When ede-compdb is upstream, there would be two things left to do: > - Integrate your documentation into ede.texi > - Integrate your tests into our test suite I will take a look at this. >> Early draft is here: >> https://github.com/randomphrase/dotfiles/wiki/Creating-a-C---Development-Environment > > That's really great! Regarding clang completion, you might want to try > the semantic-clang package in CEDET's bzr repository. It overrides > semantic-analyze-possible-completions to use clang and uses EDE to get > the necessary compile information. You can try it by using 'M-x > semantic-clang-activate'. I'm not sure how well it works with > ede-compdb, but if there are problems, we can sure fix them. I believe there is an issue here - and I've been meaning to find more time to track it down. Basically when I tried it last, the compiler internal include directories were not being detected properly. I will investigate futher and if necessary follow up in another thread about this. |