Re: 'table' as fallback for tables

[Philipp Janda <siffiejoe@...>]

Am 01.07.2016 um 13:30 schröbte Martin:
> On 16-06-30 05:11 PM, Tim Hill wrote:
> > > What about {1, [3]=3, 2} ? {1, [2]=2, 2} ? etc
> >
> > 2 and 2. The largest implicit key. As I said, this is the *least* surprising, not that it has no surprises at all.
>
> Huh. And what length of sequence should be after
>
> local t = {1, [2]=2, 2}
> table.remove(t, 2)
> --print(t[2])
> ?

That's easy (well, at least it would be the same as it is now): `nil`. 
To quote from the manual:

> Finally, fields of the form exp are equivalent to [i] = exp, where i
> are consecutive integers starting with 1. Fields in the other formats
> do not affect this counting.

... so the `[2]=2` doesn't affect the index of the following `2` at all.


> I think, better approach is to explicitly separate array part from hash
> part. But then we need separate indexing method for array part. Maybe
> syntax like "a.2" ?

And `a.2` and `a[2]` could have different values?! That doesn't seem 
like a good idea ...


Philipp