E2 to E1 SHTOLO test files

[Phil Spiby]

Ed,

I think we are in violent agreement!
My view of syntax is "a closed collection of schemas used to define a model
for a particular purpose".
I also came to the conclusion that the language version id is now looking
like a parser invocation option and could probably be removed.

If this is the case then we have to assume that all schemas will be parsed
against the latest version of EXPRESS.

Phil

PS. To go back to my CADDETC roots ...
I would like then to see a standard set of test files made publicly
available, which is used to test EXPRESS parsers, so that users can feel
confident about the EXPRESS toolkits they are using. I doubt if that will
happen though because the last time I tried to suggest that at EUG'95 there
was no interest from either the user community or the vendor community.

> -----Original Message-----
> From: Ed Barkmeyer [mailto:edbark at nist.gov]
> Sent: Wednesday, May 08, 2002 3:55 PM
> To: Phil.Spiby at eurostep.com
> Cc: Hans Karsten Dahl; wg11 at steptools.com; Hendrix, Thomas E; Arne Tøn
> Subject: Re: E2 to E1 SHTOLO test files
>
>
> Phil,
>
> You and I are not quite communicating.  I also have respect for Pascal's
> position.  But at the end of your message, you said:
>
> > My view is now that we should either have one language version
> identifier
> > for a syntax, or none at all and assume that everything obeys edition 2.
>
> The problem is, as Jochen and I have said, that we do not have a
> definition
> of "syntax".  By "syntax", do you mean "all the Express schemas you need"?
> I interpret any other definition to mean "every schema is (implicitly or
> explicitly) labeled with a language version id", which is Peter's
> position.
>
> To make my position clear, if your definition of "syntax" is "all
> the Express
> schemas you need", then I agree with your position as stated above.
>
> (What is not clear to me is how a syntax with that definition can be
> "labelled", which is, I think, implicit in Pascal's position.  Where would
> that language-version-id physically appear?  It can only appear at the
> beginning of the "syntax".  So if I cobble multiple Express files
> together
> to make a "syntax" for a parser run, at most one of them can have a
> language-version-id in it, and that one has to appear first.  By the time
> I have arranged that, the language-version-id has become a parser
> invocation option.  I only know where to put it for a long-form schema --
> the language-version-id labels the schema and the schema = the syntax.)
>
> -Ed
>
> --
> Edward J. Barkmeyer                       Email: edbark at nist.gov
> National Institute of Standards & Technology
> Manufacturing Systems Integration Division
> 100 Bureau Drive, Mail Stop 8260          Tel: +1 301-975-3528
> Gaithersburg, MD 20899-8260               FAX: +1 301-975-4482
>
> "The opinions expressed above do not reflect consensus of NIST,
> and have not been reviewed by any Government authority."
>