"during development" I agree with that. When doing a scientific simulation, we typically run the same script several times, but each times with different parameters.
On Mon, Apr 27, 2015 at 5:08 PM, Stefan Karpinski < [email protected]> wrote: > It will probably continue to be a warning when possible since redefining > constants is often handy for reloading code during development. > > > On Apr 27, 2015, at 9:14 AM, Zheng Wendell <[email protected]> wrote: > > Ah, OK. Currently, the constant-value global doesn't prevent the user to > change its value, it only gives a warning. This makes me think that it _is_ > a constant-type global. > Imaging that when the constant-type global finally comes out, the behavior > of constant-value global will also change -- it will give an error instead > of just a warning. > > On Mon, Apr 27, 2015 at 3:02 PM, Patrick O'Leary <[email protected] > > wrote: > >> On Monday, April 27, 2015 at 7:32:59 AM UTC-5, Sisyphuss wrote: >>> >>> In the documentation: >>> >Currently, type declarations cannot be used in global scope, e.g. in >>> the REPL, since Julia does not yet have constant-type globals. >>> But we already have >>> ``` >>> const pi = 3.14 >>> ``` >>> Isn't it a constant-type global? >>> >> >> It is not just a constant-type global, though. It is also a >> constant-value global. A type-declared global variable would need to relax >> that restriction, which is not something that is currently supported. See >> also https://github.com/JuliaLang/julia/issues/964 (it's already linked >> in #8870). >> > >
