comp.std.c++ Google Group

amongst Global up

brion.kam...@suddenlink.net — Sat, 10 Oct 2008 09:31:02 UT

[link]

up to Nadia Moidu neighbouring

murray.sto...@suddenlink.net — Sat, 10 Oct 2008 05:07:27 UT

Dolf wrote:
[link]

raceing in part comp.std.c++

evan.schra...@msn.com — Fri, 10 Oct 2008 19:58:03 UT

Pat wrote on 17 Oct 2008 19:58:03 GMT:
[link]

Re: What's wrong with this picture?

p...@versatilecoding.com (Pete Becker) — Wed, 01 Jan 2008 17:11:57 UT

============================== ======= MODERATOR'S COMMENT:
Please take care when replying to ensure that this thread remains topical.
============================== ======= END OF MODERATOR'S COMMENT
On 2008-01-09 10:27:39 -0500, wal...@digitalmars-nospamm.com (Walter
Bright) said:
Our document test suites are never good enough, either. That's why so

Re: What's wrong with this picture?

wal...@digitalmars-nospamm.com (Walter Bright) — Wed, 01 Jan 2008 15:27:39 UT

The difference with Latex, however, is that the user can simply look at
the output of Latex and see if it is right or not. This is not true of
programming language libraries, because our test suites are never good
enough.
---
[ comp.std.c++ is moderated. To submit articles, try just posting with ]

Re: What's wrong with this picture?

na...@animats.com (John Nagle) — Wed, 01 Jan 2008 07:16:03 UT

I think it made it into the Ada discussion group documents, which
went out on microfiche, and somewhere in some obscure library cabinet
those probably still exist. Not that anybody cares at this late date.
The intention is not that these are all different types. They're all
type "int" with different attributes. The space used to store them is

Re: What's wrong with this picture?

seewebsiteforem...@erdani.org (Andrei Alexandrescu (See Website For Email)) — Tue, 01 Jan 2008 21:17:25 UT

A while ago, I've discussed with Howard Hinnant the scenario in which
functions could overload on value vs. reference, e.g.:
struct A {};
void foo(A &);
void foo(const A &);
void foo(A);
would describe three distinct overloads, which would be selected based
on the "true" constness and lvalueness of the passed-in argument. Then

Re: What's wrong with this picture?

seewebsiteforem...@erdani.org (Andrei Alexandrescu (See Website For Email)) — Tue, 01 Jan 2008 18:57:05 UT

Interesting. I couldn't find the paper, but you did a great job at
conveying the gist of it.
There is one element that could help your case in C++ (e.g. as a library
implementation) but a few that work against:
+ The advent of auto would allow people to let the compiler take care of
computing the result type of an expression. Otherwise it would be

Re: What's wrong with this picture?

lont...@gmail.com — Tue, 01 Jan 2008 14:44:42 UT

I am glad someone not only sees that something is wrong but actually
has a pretty good idea of what exactly is wrong. By the way, I have
looked at D language and I am impressed. Seems like a huge improvement
on C++ (for the most part). But until we can all switch to D [:-)], do
you think something can be done for C++ to at least mitigate this

Re: What's wrong with this picture?

na...@animats.com (John Nagle) — Tue, 01 Jan 2008 06:09:38 UT

A long time ago, back when Ada was being designed, I wrote a short paper
titled "Type Integer Considered Harmful", in which I addressed this issue.
The fundamental problem here is that typing addresses a numeric problem
by linguistic means.
The position I took was that the basic integer numeric concept should be

Re: why is operator+=() allowed to be non-member function also ?

pedro.lama...@gmail.com (Pedro Lamarão) — Mon, 01 Jan 2008 17:56:35 UT

Hum... Current GCC 4.3 compiles this without complaint:
using namespace std;
class Test
{
public:
Test(int arg = 10);
int val;
Test::Test(int arg) : val(arg)
{
cout << "one arg ctor called" << endl;
Test operator+=(Test&& lhs, Test const& rhs)
{
cout << "from operator+=" << endl;

Re: why is operator+=() allowed to be non-member function also ?

andreytarasev...@hotmail.com (Andrey Tarasevich) — Mon, 01 Jan 2008 10:51:19 UT

I think the way you see this is is a bit backwards, so to say. The proper way to
see it is that for the purpose of uniformity _all_ overloadable operators are
(and should be) allowed to have both member and non-member (standalone) forms,
unless there is a good reason to prohibit it. So, it is not about _allowing_

Re: What's wrong with this picture?

seewebsiteforem...@erdani.org ("Andrei Alexandrescu (See Website For Email)") — Mon, 01 Jan 2008 07:09:36 UT

I agree; this is perhaps a good example of the law of diminishing
returns at work. The min/max solution in the (rejected) proposal
painstakingly handles by hand all of the exceedingly uninteresting
corner cases and also all semantics-dependent static typing (e.g. having
min invoked against an unsigned short and a signed char yields a signed

Re: What's wrong with this picture?

albertobarb...@libero.it (Alberto Ganesh Barbati) — Mon, 01 Jan 2008 04:20:22 UT

lont...@gmail.com ha scritto:
In fact, that proposal has been rejected by the LWG, according to
[link]
Ganesh
---
[ comp.std.c++ is moderated. To submit articles, try just posting with ]
[ your news-reader. If that fails, use mailto:std-...@ncar.ucar.edu ]

Re: What's wrong with this picture?

lont...@gmail.com — Sun, 01 Jan 2008 23:44:07 UT

Thanks for the link, quite funny.
I was beginning to think I was the only one who saw this trend as a
problem.
---
[ comp.std.c++ is moderated. To submit articles, try just posting with ]
[ your news-reader. If that fails, use mailto:std-...@ncar.ucar.edu ]
[ --- Please see the FAQ before posting. --- ]