Array literal syntax
Brian Rice
water at tunes.org
Mon Dec 19 10:04:14 PST 2005
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
On Dec 19, 2005, at 3:03 AM, Tony Garnock-Jones wrote:
> Brian Rice wrote:
>> Smalltalk does #[] for that, but for us, that's compile-time
>> evaluation
>> of a block. Generally, we use #- prefix to denote compile-time
>> evaluation.
>
> How frequently is CTE used? Perhaps a single-char syntax for that is
> overkill - mightn't `atCompileTime be better?
Pretty often. And we do have `evaluate.
> {1. 'hi'. #something} `atCompileTime "oh ok it's #used a lot. oops"
>
> {my complexComputation. another trickyComputation} `atCompileTime
> {my complexComputation. another trickyComputation} `preEvaluated
>
> (Obviously, notions of phase and scoping rear their ugly heads...)
>
> Actually this is interesting - the #something above is using '#' to
> mean
> 'quote' (roughly). To properly resolve this would require a proper
> investigation into the nature of quoting and any relationship it
> has to
> staged evaluation in Slate. Tricky stuff. I retract the suggestion for
> now :-)
Yep. Let's consider byte array literal syntax unnecessary for now.
- --
- -Brian
http://tunes.org/~water/brice.vcf
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.1 (Darwin)
iD8DBQFDpvYjPVkAvLW1zf4RAmCHAJ4+K3aHlA6TM3GUpU+7F8G0TJJUbQCglVZB
EzMNYyy0R6S8j/gLIDTGSI8=
=Msag
-----END PGP SIGNATURE-----
More information about the Slate
mailing list