At 03:57 PM 9/12/00 +0100, Leon Brocard wrote:
>Perl6 RFC Librarian sent the following bits through the ether:
>
> > Allow Perl to create serialize both data and code from the core.
>
>Hmmm, would it be enough to emit and take in bytecode? Might there be
>versions of Perl 6 which omit this functionality?
Probably not. That's the one thing I can't see being left out, since most
of the cut-down versions will probably end up losing everything except
bytecode reading. Though emitting bytecode might be tossed for small
devices. (Like the palm, say)
>As well as binary code, a human-readable version might also be
>interesting: or do we expect bytecode -> source translators to do that
>for us?
Maybe. Whether that's possible depends on what the bytecode looks like.
>I'd strongly suggest that we drop local encodings and go for network
>encodings at all times, but this point may be mute.
As long as it's for data streams that are getting serialized and rebuilt,
sure. Internally perl should use the native encoding for speed reasons.
(Not that I think you're recommending the internal stuff be network
encoded, but just in case someone thinks this is a nifty idea...)
Dan
--------------------------------------"it's like this"-------------------
Dan Sugalski even samurai
[EMAIL PROTECTED] have teddy bears and even
teddy bears get drunk
