From mboxrd@z Thu Jan  1 00:00:00 1970
X-Spam-Checker-Version: SpamAssassin 3.4.4 (2020-01-24) on polar.synack.me
X-Spam-Level: 
X-Spam-Status: No, score=-1.3 required=5.0 tests=BAYES_00,INVALID_MSGID
	autolearn=no autolearn_force=no version=3.4.4
X-Google-Language: ENGLISH,ASCII-7-bit
X-Google-Thread: 103376,f92fbb4a0420dd57
X-Google-Attributes: gid103376,public
From: pitre@n5160d.nrl.navy.mil (Richard Pitre)
Subject: Re: some questions re. Ada/GNAT from a C++/GCC user
Date: 1996/03/30
Message-ID: <4jjul6$637@ra.nrl.navy.mil>#1/1
X-Deja-AN: 145055100
references: <Dp1sMM.23q@world.std.com>
organization: Naval Research Laboratory
newsgroups: comp.lang.ada
Date: 1996-03-30T00:00:00+00:00
List-Id: <comp.lang.ada>

In article <Dp1sMM.23q@world.std.com> bobduff@world.std.com (Robert A Duff)  
writes:
> In article <leschkes.828053356@ferret>,
> Scott Leschke <leschkes@ferret.cig.mot.com> wrote:
> >My first question would be, why do you want redundant instantiations.
> 
> Usually you don't.  But you might declare a type Apple_Count, and not
> know whether anybody wants to do I/O on it.  Then one client does, and
> instantiates Text_IO.Integer_IO on Apple_Count.  And another client
> independently does the same.  One might like to share the code of the
> two instantiations.
> 
> But, as Bill Newman pointed out, the more important thing is when you
> have Apple_Count and Orange_Count, which are logically distinct, but
> happen to share the same hardware representation.  That's a more
> important case where you'd like shared generic code.
> 
> Shared generic code isn't easy to implement in general, but many Ada 83
> compilers support it, at least for some kinds of generics.  The problem
> is that you can't count on it in portable code, so you have to do it "by
> hand".  That's not so awful, I suppose -- at least the ugly parts are
> buried in one place.
> 
> >You can use a block statement.  This is different than C++ in the sense
> >that objects declared within the block are only in existence within the
> >block and are finalized at the end.
> 
> No, I think this is exactly the same in C++ and Ada.  See r.6.7 of The
> C++ Programming Language, Second Edition, by Bjarne Stroustrup.  That
> is, if you declare an object in C++ in a local sequence of statements,
> it will be finalized at the end of that sequence.  
> The only difference
> is that Ada requires some extra syntactic baggage.
> 
> >declare
> >   Obj : Pkg.SomeType;
> >begin
> >   Pkg.Operation (Object => Obj);
> >   -- Other stuff
> >exception
> >   when Pkg.Some_Exception =>
> >
> >      Do_Something;
> >end;
> 
> The exception handler makes this look reasonable, but in the usual case,
> where there's no need for any exception handling, and you just want to
> declare Obj, the extra 3 lines of code ("declare", "begin", "end") are
> just an annoyance.

I thing of "begin" and "end" are visual cues and I prefer them to eyeburner
baggage like "{" and "}".

richard