Re: none

[Michael Kirkham <mikek@muonics.com>]

On 21 May 2001, Frank Strauss wrote:

> BTW, could you tell us some details about your SMIng parser? Is it possible
> to make it freely available? Do you use lex/yacc? What is the background of
> your experiments.

I haven't decided yet whether or not it will be freely available, so it's
probably not appropriate for me to go into too much detail about the scope
of the project on an IETF list.  But if you'd like to know more I can
email you privately about it.  It's entirely possible I will make it part
of it freely available, at least for non-commerical use (perhaps an
SMIv1/v2 <--> SMIng converter, for example).

I'm not currently using lex/yacc.  I'm using a vaguely lex/yacc-like
add-on I'm writing for the tcLex Tcl extension
(http://www.multimania.com/fbonnet/Tcl/tcLex/index.en.htm) because the
rest of the project will be done in Tcl.  I may yet go true lex/yacc if
efficiency becomes a big issue.

The extent of my experiments with parsing SMIng have pretty much just been
creating a keyword/symbol table in my add-on to tcLex and running
draft-ietf-sming-01.txt through it, to make sure it was going to work okay
for SMIng as well as SMIv1/v2, before putting too much work into the
yacc-like parts of the add-on.  And that's where I noticed the quoting
problem: it seemed to be tokenizing it just fine until it got to that abnf
statement.

> Please note that the current sming spec I-Ds are just the current proposal.
> There are still a lot things being discussed in the requirements process
> that will lead to significant changes to the language.

Certainly, and good to hear, since I sent a couple suggestions to the list
recently :).  Just good to be prepared..  I'm not focussing on SMIng in my
project, but I wanted to implement things in a way that would be useful
for it as well.