[Date Prev][Date Next][Thread Prev][Thread Next]
[Date Index]
[Thread Index]
- Subject: Re: mathlib
- From: steve donovan <steve.j.donovan@...>
- Date: Tue, 8 Apr 2014 14:07:00 +0200
On Tue, Apr 8, 2014 at 12:51 PM, Christopher Berardi
<cberardi32@gmail.com> wrote:
> import module1
> local import module2
The first form could be implemented simply as a function, since it
just involves a loop at run-time copying key,value pairs from the
require'd module into _G.
The second form has to expand at compile-time, and in general this
isn't possible, because we don't know what the contents of 'module2'
are at compile-time.
> And the same for getting sub-modules
>
> import func1, func2 from module4
> local import func3, func4 from module5
I doubt there's any need for the first form here - the second form can
be implemented at compile-time (or macro expand time) because the
names are specified.
One way of evaluating a proposal is to think of documenting it, and I
think it's going to take a fair amount of text to explain this all....
It's matter of what's easy to type and what's easy to read; if the
typing is a problem, teach your editor to do the expansion, if reading
is a problem..... well, then I don't know.. Personally I don't see
the problem with reading explicit code. This feels like an attempt to
impose another language's semantics onto Lua - as well as syntax.
- References:
- Re: mathlib, Roberto Ierusalimschy
- Re: mathlib, Coroutines
- Re: mathlib, Luiz Henrique de Figueiredo
- Re: mathlib, Roberto Ierusalimschy
- Re: mathlib, Christopher Berardi
- Re: mathlib, steve donovan
- Re: mathlib, Christopher Berardi
- Re: mathlib, Sean Conner
- Re: mathlib, Coroutines
- Re: mathlib, Sean Conner
- Re: mathlib, Christopher Berardi