[PATCH] Fix 2 possible SEGVs in tftp client

Jason Schoon floydpink at gmail.com
Wed Jun 14 21:44:11 UTC 2006


On 6/14/06, Rob Landley <rob at landley.net> wrote:
>
> http://www.faqs.org/rfcs/rfc2348.html
>
> Does not contain the string "octect".  It does contain 17 occurrences of
> the
> string "octet", which it uses as a synonym for byte and which they refer
> to
> rfc1350 for where "octet" is defined as "raw 8 bit bytes", and the
> extended
> description goes off on a tangent about the DEC-20 (a 36-bit machine).
> Apparently, in 1992 there was still one of those in use somewhere.


Doh, you're right, I thought for sure I looked at that, having been
skeptical myself.  Octect is a ASN.1 type used widely in SNMP.  My mind is
apparently becoming cross-wired.

I continue top pretty strongly believe that if a constant that doesn't
> change
> is only ever used in one place, making a #define for that constant is the
> wrong thing to do, and instead a comment should be placed at the site of
> the
> one use of the thing specifying what the constant means and/or where it
> comes
> from.
>
>
I don't necessarily care if it is a #define or not.  Having a comment giving
the source of the magic number is probably sufficient also.  I do think you
have 2 contradictary statements here though:

"They're in the spec.  We must use these numbers."

vs.

"...it means that you have to stop what you're doing and look
something up in order to follow what's actually going on, and this happens
for no reason."
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.busybox.net/pipermail/busybox/attachments/20060614/41849778/attachment-0001.htm 


More information about the busybox mailing list