If semantic-analyse-proto-impl-toggle fails, jump to any implementation with the correct class, function name and signature.

In the huge (huge !) majority of cases there is only one such match ; and even if there are more than one and you pick the wrong one, it's not worse than returning an error...

To reduce the scope of my searches (and the number of possible mismatches), I also added a notion of project, which I define as a list of header files and a list of source files. For proto-impl-toggle, I only search inside the header's project.


On Thu, Oct 31, 2013 at 10:22 PM, David Engster <deng@randomsample.de> wrote:
Vincent Semeria writes:
> Haha, I hear poems about the proper ways of coding quite often, sometimes even
> in rhymes :)
> However I cannot change the way my functions are implemented. I work on 14000
> c++ files, most of which are 20 years old, and all of which are written like I
> showed you.

Yeah well, I already assumed it was something like this... :-)

> I already wrote my version of proto-impl-toggle in elisp which works pretty
> good. I can send it to you if you're interested.

If it's something that could be applied to CEDET as a patch, I'm
interested, of course (although if it's longer than ~15 lines, we would
need a copyright assignment from you...).