Tiny error in GNAT: wrong message about unknown discrimant

Tiny error in GNAT: wrong message about unknown discrimant

[Yannick Duchêne (Hibou57)]

Hello,

Create a type with an unknown discriminant in the public part of a package  
(and optionally a concrete discriminant in the full definition). Declare  
an instance of that type in a procedure outside of that package, and do so  
without any initialization, and instead, erroneously add a discriminant.  
GNAT will complain “invalid constraint: type has no discriminant”. This is  
inconsistent, as the type has a discriminated view; it should instead say  
“invalid constraint: type discriminant is unknown”.

Not sure it is worth a bug report, but at least, this is probably  
confusing enough for beginners.

-- 
Si les chats miaulent et font autant de vocalises bizarres, c’est pas pour  
les chiens.
“ c++; /* this makes c bigger but returns the old value */ ” [Anonymous]

Re: Tiny error in GNAT: wrong message about unknown discrimant

[Georg Bauhaus]

Yannick Duchêne (Hibou57 ) <yannick_duchene@yahoo.fr> wrote:

> 
> Not sure it is worth a bug report, but at least, this is probably  
> confusing enough for beginners.

The GCC bugzilla has a choice list for qualifying reports.
You can choose "enhancement".

Re: Tiny error in GNAT: wrong message about unknown discrimant

[Yannick Duchêne (Hibou57)]

Le Fri, 27 May 2011 10:45:41 +0200, Georg Bauhaus  
<rm-host.bauhaus@maps.arcor.de> a écrit:

> Yannick Duchêne (Hibou57 ) <yannick_duchene@yahoo.fr> wrote:
>
>>
>> Not sure it is worth a bug report, but at least, this is probably
>> confusing enough for beginners.
>
> The GCC bugzilla has a choice list for qualifying reports.
> You can choose "enhancement".
Thanks for the point (never noticed this entry).
Done: http://gcc.gnu.org/bugzilla/show_bug.cgi?id=49192


-- 
“Syntactic sugar causes cancer of the semi-colons.”  [Epigrams on  
Programming — Alan J. — P. Yale University]
“Structured Programming supports the law of the excluded muddle.” [Idem]
“c++; /* this makes c bigger but returns the old value */” [Anonymous]