Re: Apex 3.0.0b can't handle dynamic positional agregates?

Re: Apex 3.0.0b can't handle dynamic positional agregates?

[Robert A Duff]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 481 bytes --]

"Dr. Joachim Schr�er" <joachim.schroeer@vs.dasa.de> writes:

> To be serious: By sovereignly ignoring language standards, Rational seems to
> intentionally bring some C++ habits to the Ada world.

That's not fair.  In Ada 83, there *was* no standard way to do unsigned
integers, so it's not surprising that Rational invented one.  Now, of
course, they have to keep supporting it, for upward compatibility.

- Bob
-- 
Change robert to bob to get my real email address.  Sorry.

Re: Apex 3.0.0b can't handle dynamic positional agregates?

[Samuel T. Harris]

"Dr. Joachim Schröer" wrote:
> 
> Robert Jordan schrieb in Nachricht <37FE1232.EB0AD5EE@boeing.com>...
> >I try to compile a statement like
> >
> >stuff : stuff_type := (first, second, third, Others => Too_Many);
> >
> >But Apex 3.0.0.b comes back with an error:
> >
> >[MID] Unimplemented feature: dynamic positional aggregate with Others.
> 
> Having unimplemented features in Ada95 is not too bad if you consider, that
> the Rational Ada83 compiler gives you some benifits of Ada95 e.g. modular
> types. If you look into the package spec. of system (system.1.ada)  in  the
> Ada83 model
> 
>    lrm.ss/sun4_solaris2.ada83.3.0.0.rel
> 
> you find some modular type declarations like
> 
>     type Address is mod 2 ** 32;

This is not such a great thing when running ASIS on code
and traversing to a ground type which is one of these modulo's
and then getting an ASIS error that modulo types are not
an Ada 83 feature! Have to work around that one many times.
However, all in all, Apex is a great environment.

> 
> Even better, the Ada83 compiler translates modular types in your application
> SW
> without complaint. Doesn't that compensate your unimplemented feature?
> 
> To be serious: By sovereignly ignoring language standards, Rational seems to
> intentionally bring some C++ habits to the Ada world.
> 
>     J. Schröer
> 
> >The LRM (4.3.3) actually shows an example of this being used.  Even when
> >I try to use the exact LRM example
> >
> >stuff : stuff_type := stuff_type'(first, second, Third, Others =>
> >Too_Many);
> >
> >I still get the same error.
> >
> >There's lots of ways around this problem, but it's going to make my code
> >hard to read.  I wanted the Others statement and the remaining elements
> >to be as compact as possible so the numerous tables I need to define are
> >easily maintained (One per line, columns lined up, etc...).
> >
> >Apex claims to be ADA95 compliant, so how can they have Unimplemented
> >features which are not related to the annexes?
> >
> >Robert

-- 
Samuel T. Harris, Principal Engineer
Raytheon, Scientific and Technical Systems
"If you can make it, We can fake it!"

Re: Apex 3.0.0b can't handle dynamic positional agregates?

[Martin Dowie]

which version of Apex 3.0.0b are you using? i'm using the version for
VxWorks and it is quite happy about this construct. perhaps time to call up
some support...

Robert Jordan wrote:

> I try to compile a statement like
>
> stuff : stuff_type := (first, second, third, Others => Too_Many);
>
> But Apex 3.0.0.b comes back with an error:
>
> [MID] Unimplemented feature: dynamic positional aggregate with Others.
>
> The LRM (4.3.3) actually shows an example of this being used.  Even when
> I try to use the exact LRM example
>
> stuff : stuff_type := stuff_type'(first, second, Third, Others =>
> Too_Many);
>
> I still get the same error.
>
> There's lots of ways around this problem, but it's going to make my code
> hard to read.  I wanted the Others statement and the remaining elements
> to be as compact as possible so the numerous tables I need to define are
> easily maintained (One per line, columns lined up, etc...).
>
> Apex claims to be ADA95 compliant, so how can they have Unimplemented
> features which are not related to the annexes?
>
> Robert

Re: Apex 3.0.0b can't handle dynamic positional agregates?

[Wes Groleau]

> stuff : stuff_type := (first, second, third, Others => Too_Many);
> But Apex 3.0.0.b comes back with an error:
> [MID] Unimplemented feature: dynamic positional aggregate with Others.

Send a report to support@rational.com

> Apex claims to be ADA95 compliant, so how can they have Unimplemented
> features which are not related to the annexes?

Is it mentioned on the validation report?