Hi,
> Example:
> var verdicttype a_verdict_var;
> a_verdict_var := error;
> setverdict(a_verdict_var);
yes, this usage of the 'error' verdict is something which has not
been considered. At the moment I don't know how to resolve this
without runtime check. Disallowing 'error' on the right side of
assignments or restricting the usage of 'error' to the control
part only seem not to be correct solutions.
Regards
Jens
Mariusz Kupiec schrieb:
>
> Hi,
>
> Thanks for good explanation and confirmation of Mr Syntax example.
> What about this example:
>
> Example:
> var verdicttype a_verdict_var;
> a_verdict_var := error;
> setverdict(a_verdict_var);
>
> An 'error' verdict was not produced by getverdict operation because it is
> impossible according 25.2.1. I hope it is legally produced, user made
> 'error' verdict, assigned in a similar way as it is allowed for 'none'
> verdict. It does not violate static semantics:
>
> 513. SetLocalVerdict ::= SetVerdictKeyword "(" SingleExpression ")"
>
> /* STATIC SEMANTICS -SingleExpression must resolve to a value of type
> verdict */
>
> /* STATIC SEMANTICS - the SetLocalVerdict shall not be used to assign the
> Value error */
>
> The only protection against this kind of gap in specification seems to be
> the sentence from 25.2.1: 'The error verdict is special in that it is set by
> the test system'.
> Is it enough? In the example I have only initialized the variable and put it
> as an argument for system function. It is the system, which does all the
> dirty job with setting what is in variable ;-)
> Should run-time error / dynamic test case error be generated in this
> example?
>
> BR,
> Mariusz Kupiec
>
>
Original Message
> From: "Jens Grabowski" <
This email address is being protected from spambots. You need JavaScript enabled to view it.>
> To: <
This email address is being protected from spambots. You need JavaScript enabled to view it.>
> Sent: Saturday, January 18, 2003 3:18 PM
> Subject: Re: Question on 'none' verdict
>
> Hi,
>
> the example of 'syntay' was the main reason for allowing
> 'setverdict(none)'.
>
> Whereas it is easy to check statically 'setverdict(none)',
> the value of a variable 'a_verdict_var' which is used in a
> 'setverdict(a_verdict_var)' can in general ony be checked
> dynamically.
>
> By disallowing 'setverdict(none)' we would also have to
> disallow 'setverdict(a_verdict_var)' if 'a_verdict_var == none'.
> This would require a lot runtime-checks.
>
> Since 'setverdict(none)' has no effect, we believed that it
> is more efficient to allow it instead of forcing further
> runtime-checks.
>
> Regards
> Jens
>
>
> Poczta nowych mozliwosci >>>
link.interia.pl/f16bc
--
======================================================================
Dr. Jens Grabowski
Institute for Telematics phone: +49 451 500 3723
University of Luebeck fax: +49 451 500 3722
Ratzeburger Allee 160 eMail:
This email address is being protected from spambots. You need JavaScript enabled to view it.
D-23538 Luebeck or
This email address is being protected from spambots. You need JavaScript enabled to view it.
(Germany) WWW:
www.itm.mu-luebeck.de
======================================================================