On Wed, Jul 07, 1999 at 06:19:32PM -0400, Jeffrey Perry wrote:
> What is your objection to the idea of having a run-time interpretor. It
> sounds like a number of people like the idea and would find it useful.
> (myself included) You will NEVER cover ALL "common protocols", this makes
> it easier for people, including programmers like myself.
>
> Is your objection - startup time? Having to write a parser?
It's all of them. It's the mere amount of work which IMHO is better
spent elswhere. It makes ethereal take longer to startup and probably
slower dissecting the packets, then all these users have to hassle
around with additional files. Then it's nearly impossible to writa a
parser which covers every possible protocol, thus leading to
bug-reports/feature-requests which need even more work to deal with.
I think a simple parser which generates C source for ethreal for the
most common parts, which can be hand-edited to work perfectly would be
the most efficient way.
CU,
Sec
--
rfc0896: This 4000% overhead is annoying, but tolerable on
lightly loaded networks.