On Sat, May 06, 2000 at 01:27:17PM -0700, Bill Trost wrote:
> Oskar Sandberg writes:
>     I think it is fair to allow for zeros at the front of the value to
>     be cut off.
> 
> What business does a node have making modifications to the ID generated
> by a node?  All the UniqueID is being used for is a comparison value.
> There is no point in converting the ID to a number, only to convert it
> back to characters when the next packet is sent.  This is especially
> relevant to implementations that don't support numbers long enough to
> contain a UniqueID (like gcc on the i386).

But not allowing variations that don't actually change the value of the
UniqueID (such as using lowercase a-f instead of uppercase A-F or the
opposite) is just inviting crockishness.

> There's no reason to treat the UniqueID as anything other than a string
> of characters, and no reason for the protocol to arbitrarily place
> limits on or require conversions of that string.
> 
>     With the changes I'm making now and since we have to change the
>     protocol for crypto anyways, I'm going to make all numbers hex
>     strings (as was decided in the discussion before iirc).

Because the UniqueID is merely a hex value and not a full alphanumeric
string, I think that it should be handled as a number equivalent to the
hex string.

> Do you mean to say that "Depth" will be hex instead of decimal?  I
> didn't think that discussion had reached any real conclusion.  It sounds
> like a gratuitous change to the protocol to me.  Currently, the only
> data types are character strings (UniqueID's, addresses, and keys) and
> decimal integers (Depth and Hops).
> 
>     On Sat, 06 May 2000, Bill Trost wrote:
>     >         Recieved message 'e7cc6d5223fdc8a' expected '0e7cc6d5223fdc8a'
> 
> _______________________________________________
> Freenet-dev mailing list
> Freenet-dev at lists.sourceforge.net
> http://lists.sourceforge.net/mailman/listinfo/freenet-dev

-- 
Travis Bemann
Yes, I still haven't edited /etc/sendmail.cf.
My address should be bemann at execpc.com

_______________________________________________
Freenet-dev mailing list
Freenet-dev at lists.sourceforge.net
http://lists.sourceforge.net/mailman/listinfo/freenet-dev

Reply via email to