>  But if/cond accepts any expression, as is, no special cases. That’s the line 
> I don’t want to (and we won’t) blurry.

I see, I understand what you mean, maybe I still not fully understand all the 
reasoning behind it, but I trust you know what’s the best for it and I’m not 
seeing something you are.

> Then it has to be made a special form and that has even further implications, 
> so it is unlikely to happen either.

Not sure what are the further implications you mean, but I will study a bit 
more on the subject while trying to create that macro, maybe I will have those 
answers in the end.

> And if we don’t make it, then if starts to look inconsistent (which was my 
> point).

Yeah, I got it now, although to understand the kind of inconsistency you mean, 
one might actually know about some internals of the language that not too many 
people know. My point about the `for` and `with` inconsistency compared to `if` 
and `cond` was taking into account a totally lay person that might think `<-` 
is just another kind of "match operator”. But I see your point: it is not and 
it was never the intention for it to be.

Thinking about that actually just gave me an idea for the lib: a `left <- 
right` operator that works exactly like `=` but instead of raising when it 
doesn’t match, it returns `nil`. Will check how that goes and what are the 
implications of the conflict 

> If you can’t tell the complexity, you can trust the opinion of someone who 
> can (and is describing it) :)

Sure, yeah! I trust you and everyone else on the core team, you never let me 
down before, I am sure you have really good reasons for all the decisions you 
make :)

Just to be clear, I was just wondering about the feature, sorry about being 
stubborn, I know it’s very unlikely you would change a core feature like that 
too, will just shut up now and take advantage of the fact it is possible to do 
that on a lib.

Best,
Kelvin Stinghen
[email protected]

> On Oct 17, 2019, at 20:24, José Valim <[email protected]> wrote:
> 
> If you can’t tell the complexity, you can trust the opinion of someone who 
> can (and is describing it) :)

-- 
You received this message because you are subscribed to the Google Groups 
"elixir-lang-core" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
To view this discussion on the web visit 
https://groups.google.com/d/msgid/elixir-lang-core/9A786838-DAC0-4306-B540-4CD8FDDF4CCD%40gmail.com.

Reply via email to