Re: strings, Re: where is everybody? Re. etc..
On 2 Nov 2000, at 12:52, Irv wrote:
> On Thu, 02 Nov 2000, you wrote:
> > On 2 Nov 2000, at 10:17, Irv wrote:
> >
> > > (1), Hey, I think Moe, Larry and Curly are funny, too.
> > > (2) "We went to the zoo, and saw an elephant, a lion, and 23."
> > > "My boss is Mr. Smith, and my girlfriend is 19"
> >
> > "I am Maxwell Smart, and this is 99."
> >
> > Irv, if you are trying to do natural language parsing, the native support in
> > any puter language won't handle any of those. "Maxwell", "Smart", and "99"
> > are all tokens in a string, and what the parser does to those tokens is up
> > to the token definition table you gave the program. It's not something that
> > the programing language can deal with... well... maybe Ox or Py can deal
> > with it...
>
> Thanks, Kat, but no, this has nothing to do with natural language parsing.
> It's simply an illustration that data (in the real world) isn't all numbers.
> We also saw a {103,105,114,97,102,102,101} and an
> {97,108,108,105,103,97,116,111,114} at the zoo. Just doesn't cut it, does it?
Well, no, but the {*} and {*} in the debug or ex.err is just the raw dump of
memory locations. We've had such dumps in hex for 30 or more yrs, and i
welcome the ascii for text dumps. You could write a post-processor to
munge those files? Possibly Robert can give us some ex.err switches for
the dump? The truncated sequences in ex.err are a pain, and mostly less
than useful. So is not being able to drill down into a nested sequence in a
trace window.
Kat
|
Not Categorized, Please Help
|
|
