Throw warning for unsupported runtimes, e.g. Node < 6

[GeoffreyBooth]

Running coffee or cake in Node 4 or below results in an unhelpful stack trace (per here):

user@computer:~/some/path $ coffee -c code.coffee 
/usr/local/lib/node_modules/coffeescript/lib/coffeescript/command.js:23
  ({spawn, exec} = require('child_process'));
   ^

ReferenceError: Invalid left-hand side in assignment
    at exports.runInThisContext (vm.js:53:16)
    at Module._compile (module.js:374:25)
    at Object.Module._extensions..js (module.js:417:10)
    at Module.load (module.js:344:32)
    at Function.Module._load (module.js:301:12)
    at Module.require (module.js:354:17)
    at require (internal/module.js:12:17)
    at Object.<anonymous> (/usr/local/lib/node_modules/coffeescript/bin/coffee:15:5)
    at Module._compile (module.js:410:26)
    at Object.Module._extensions..js (module.js:417:10)

We can do better than this. This minor addition first checks that the runtime supports destructuring, and throws an informative error if your runtime doesn’t:

user@computer:~/some/path $ coffee -c code.coffee
Your JavaScript runtime does not support some features used by the coffee command.
Please use Node 6 or later.

Destructuring is really just a proxy for all the advanced ES2015 features that the CoffeeScript module uses, but it should be enough. We already use CI to test that CoffeeScript works in Node 6, and destructuring isn’t supported in Node 4 and below, so this should catch all the unsupported runtimes.

[GeoffreyBooth]

Can someone who’s more an expert in the coffee command and launch stuff please review this? It’s very small. @lydell?

[lydell]

Looks like a working solution to me. But I can't really know how safe it is.

Any particular reason you didn't use eval directly?

[GeoffreyBooth]

Any particular reason you didn’t use eval directly?

Well, mainly because it’s been drilled into me that eval is evil, so therefore anything else must be better. More practically, new Function creates its own scope whereas eval uses the scope it’s in, so this PR doesn’t end up creating a new a variable. I guess that’s as good a reason as any to prefer new Function, unless there’s some reason you think I should use eval.

[lydell]

Thanks for the explanation. Sounds good.