On Wed, May 07, 2008 at 03:58:26PM +0000, David Nicol wrote: > the only thing > that an IDE provides that would be a nice-to-have in gvim is flexible > keyword completion > that knows about the valid symbols available at any point. Having > this prevents cycles > of compile/fix due to miscapitalizations, for instance. That feature > is really the I in IDE > in my opinion, as the editing tool must know the language being > written in well enough > to provide the correct subset of available symbols in any context.
I use one of emacs's expansion features to complete things. It cycles through all words in all open buffers to try to complete the word I'm curently typing. It's not perfect since it makes lots of suggestions that just don't make sense at that point and will miss method names that I haven't defined or used in open files, but it's nice. I think it will also use that TAGS or CTAGS business or whatever that is. I've obviously never played with it. I have used hooks in emacs to make it run the test suite when I M-x compile, and stuff like that. It's nice, but I typically switch windows and repeat the previous command instead. not to start an editor war, but I've found that emacs makes a good enough development environment for me. I suspect that's the same experience that the (g)vi(m) crowd has had. The editors seem to be "good enough" and making them better seems to be "real hard", so we're stuck here for a while. -Andy _______________________________________________ kc mailing list [email protected] http://mail.pm.org/mailman/listinfo/kc
