brendan at mozilla.com
Wed May 12 13:17:24 PDT 2010
Agreed, we need a[-1] for a[a.length-1] syntactic sugar, or there's no
point. The Array slice method, among others, already does sensible and
Pythonic processing of negative arguments.
The way to provide syntax for negative indexes (if not Pythonic
slicing) is under a new version. But that would complicate the spec,
the various implementations, and the migration landscape for old code
(did this old code that I did not write, but that I'm paid to port,
use anything that evaluated to -1 as an array index?).
And such a strawman lacks a champion to write up the proposal and
carry it through prototyping and standardization. I've toyed with
slice syntax before, and may again, but I'm not that champion. So this
seems unlikely to make it past "straw" state.
On May 12, 2010, at 12:53 PM, Jordan Osete wrote:
> De : Dmitry A. Soshnikov <dmitry.soshnikov at gmail.com>
> Envoyé le : Mer 12 mai 2010, 9h 12min 01s
> > That's too bad concurrency for alternative and already invented
> syntactic sugar for that -- a bracket notation used in many
> languages. I should mention Python again and its very convenient
> > ...
> > So, don't think that JS -- a highly abstracted language, needs
> this getItem / setItems, even if for less abstracted languages (even
> for Assembler) there is bracket notation for addressing offset from
> the base of a structure.
> Well, I can agree of that, I would prefer to have that syntactic
> This kind of getItem / setItem would have been just temporary
> solutions for the current implementations until that sugar is ready
> to be used in a few years, just like filter and map are temporary
> replacements for array comprehensions.
> es-discuss mailing list
> es-discuss at mozilla.org
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the es-discuss