[CrackMonkey] Another nail in the Pine coffin

Paul J Collins sneakums at eircom.net
Sat Sep 30 20:16:49 PDT 2000


>>>>> "Bad" == Bad  <mr.bad at pigdog.org> writes:

>>>>> "PJC" == Paul J Collins <sneakums at eircom.net> writes:
    PJC> Your contrived example does not prove that every use of those
    PJC> functions in Pine is safe and not prone to buffer overflows.

    Bad> Well, did anyone say they were? He just said that careful use
    Bad> of the functions can avoid B.O.

His example wasn't even ``careful use''.  It was contrived junk that
no-one who thought in a straight line, or even in two or three
dimensions, would ever write.

    Bad> I find the idea that programs that use strcpy, strcat and
    Bad> sprintf are fundamentally insecure to be ABSURD.

I didn't agree with that view point.  However, functions that are easy
to use misuse and often are make it more difficult to determine
whether a program is easily compromised or not.

    PJC> By the way, use of the integer constant 0 for '\0' is not a
    PJC> good idea.  Say what you mean.

    Bad> Yeah, but it's also extremely widespread C usage. Sure, it's
    Bad> a pain for people doing new multi-byte character set ports,
    Bad> but fuck them. 7-bit character types were good enough for my
    Bad> father, and they're good enough for me!

char can be an 8-bit type.  It depends on your implementation.

-- 
Paul Collins <sneakums at eircom.net> - - - - - [ A&P,a&f ]
 GPG: 0A49 49A9 2932 0EE5 89B2  9EE0 3B65 7154 8131 1BCD
``Attention all MP3s: resistance is futile.
  You will be vorbized.''





More information about the Crackmonkey mailing list