comp.lang.ada
 help / color / mirror / Atom feed
From: "Randy Brukardt" <randy@rrsoftware.com>
Subject: Re: Why is not Stream_Access defined Ada.Streams ?
Date: Thu, 6 May 2010 21:24:49 -0500
Date: 2010-05-06T21:24:49-05:00	[thread overview]
Message-ID: <hrvtlm$92v$1@munin.nbi.dk> (raw)
In-Reply-To: op.vca852lexmjfy8@garhos

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 950 bytes --]

"Yannick Duch�ne (Hibou57)" <yannick_duchene@yahoo.fr> wrote in message 
news:op.vca852lexmjfy8@garhos...
> Hi all and Hi others as well,
>
> My dirty stuff of the day : why is Stream_Access not defined once a time 
> in Ada.Streams and is instead defined elsewhere multiple times, like in 
> Ada.Streams.Stream_IO, Ada.Text_IO.Text_Streams, 
> Ada.Wide_Text_IO.Text_Streams and Ada.Wide_Wide_Text_IO.Text_Streams ?
>
> Isn't it funny design ?

Ada 95 didn't have anonymous access returns, so a named type had to be 
defined. I suspect that there was no intent that it be used for anything 
other than defining the result of these functions -- and there was no intent 
that that result be stored: just directly dereferenced and passed to a 
stream attribute.

If we were writing it today, I'm pretty sure that no named type would be 
used at all. But making a change like that now would be incompatible.

                                Randy.





      reply	other threads:[~2010-05-07  2:24 UTC|newest]

Thread overview: 2+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2010-05-07  1:02 Why is not Stream_Access defined Ada.Streams ? Yannick Duchêne (Hibou57)
2010-05-07  2:24 ` Randy Brukardt [this message]
replies disabled

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox