Bind C-c C-c to #'racket-run

[cpitclaudel]

This is in line with the practice of other major modes, like
python-mode. Some other modes (like haskell-mode) use this to map to a
mode-specific version of #'compile, but I'm not sure that this would be
too relevant here. What do you think?

[greghendershott]

Good point. I've wondered about this.

Racket does have raco make to create bytecode zo and dep files under compiled/. But in my experience usually this gets run as part of raco setup or raco pkg install -- not directly as a raco make of a specific file.

So if we were to bind C-c C-c to something "compilation related", where it saves people needing to go to a shell, maybe it should be raco setup, re issue #41 ? But I'm not sure....

[cpitclaudel]

I've added a comment there :) I don't think C-c C-c should be made redundant with a properly configured M-x compile. I also expect racket-run to be much more common that compiling.

I use C-c c to run compile, and that has been pretty convenient. That could be suggested in the docs somewhere.

[greghendershott]

1. Sorry: I confused things by mentioning issue Offer to help with bytecode mismatches? #41 in connection with this. When issue Offer to help with bytecode mismatches? #41 arises, it's almost always because some other file(s) need to be recompiled. Closing issue Offer to help with bytecode mismatches? #41 would entail doing something like what I mentioned in the first comment -- using https://github.com/rmculpepper/custom-load.
2. I do like your comment -- which I misled you to leave over there :) -- about setting Emacs compile-command to raco make in racket-mode buffers. I'll do that.
3. Once I do 2, is there anything left to-do on this issue? Should we leave it up to individual users to bind C-c C-c if/as they prefer? Or should we bind it to compile in racket-mode as a reasonable default? What do you think?

[cpitclaudel]

Cool for 2. For 3, I would actually vote for biding C-c C-c to the same as C-c C-k, as it matches other language modes like Python-mode. I wouldn't bind it to compile, because users often already have a binding for that, and because it would be slightly inconsistent with other language modes. Of course, that wouldn't prevent users from rebinding C-c C-c.

Anyhow, this is a minor thing; it just bit me when I first tried the mode :)

[greghendershott]

OK. Done. Thanks for the PR!