All --
FWIW, this stuff came up early on in Parrot's infancy. At one time we had
fingerprinting, then we removed it. At one time it was MD5, then we went
away from MD5. IM(NS)HO, the better approach overall is to remove the need
for fingerprinting altogether. I used to know roughly how to get from
where we were to that state, but when I just looked at the code, I
realized I'd have to catch up on a lot of history to say anything useful
about how to get there from where we are now.
If there is interest in the idea in general, I'll see if I can find
pointers to the prior discussion. Else, I'll just duck back into the
corner :) ...
On a related note, I'm working on a toy VM outside of Parrot to
demonstrate the technique I've proposed here in the past, but alas, time
is short. Anyway, if I can get something produced that is worth looking
at, I'll post here.
Regards,
-- Gregor
Josh Wilmes <[EMAIL PROTECTED]>
Sent by: [EMAIL PROTECTED]
11/01/2002 11:34 PM
Please respond to Josh Wilmes
To: "Brent Dax" <[EMAIL PROTECTED]>
cc: "'Andy Dougherty'" <[EMAIL PROTECTED]>, "'Perl6 Internals'"
<[EMAIL PROTECTED]>
Subject: Re: What to do if Digest::MD5 is unavailable?
I think this solution is the simplest... i'll go ahead and commit it.
--Josh
At 10:15 on 11/01/2002 PST, "Brent Dax" <[EMAIL PROTECTED]> wrote:
> Andy Dougherty:
> # At the moment, the bytecode "fingerprint" is built with
> # Digest::MD5. Alas, Digest::MD5 wasn't standard with perl
> # versions prior to 5.8.0. What should happen in those cases?
> # Anybody have any good ideas?
>
> Not sure if this qualifies as "good" :^), but we *could* package the
> pure-Perl implementation of Digest::MD5 with Parrot.
>
> --Brent Dax <[EMAIL PROTECTED]>
> @roles=map {"Parrot $_"} qw(embedding regexen Configure)
>
> Wire telegraph is a kind of a very, very long cat. You pull his tail in
> New York and his head is meowing in Los Angeles. And radio operates
> exactly the same way. The only difference is that there is no cat.
> --Albert Einstein (explaining radio)
