From: Sven Eschenberg <eschenb@cs.uni-frankfurt.de>
To: Mustafa4LP <mustafa.hussain@nechclst.in>,
GCC-help <gcc-help@gcc.gnu.org>
Subject: Re: Copy constructor not called.
Date: Wed, 12 Aug 2009 11:28:00 -0000 [thread overview]
Message-ID: <4A8264C2.5080509@cs.uni-frankfurt.de> (raw)
In-Reply-To: <24930356.post@talk.nabble.com>
As far as I can tell:
Your assignment Operator does not return a reference, that's why the
compiler complains about not finding a matching copy contructor. Afaik
semantically ( b = a ) would yield a copy constructor to create a
temporary for the RV, which is A and not A & in your case. I don't know
though, why a const in the copy constructor fixes the compile error as well.
And I want to quote from the man page:
-fno-elide-constructors
The C++ standard allows an implementation to omit creating a
tempo-
rary which is only used to initialize another object of the same
type. Specifying this option disables that optimization, and
forces G++ to call the copy constructor in all cases.
As I read this this only relates to temporarys used for initialization.
I don't see any initialization, only a RV temporary passed to the
assignment operator. Thus the RVO will still be done and not be disabled
by that option, as I read it.
Regards
-Sven
Mustafa4LP schrieb:
> Hi all,
>
> Please consider the following program:
>
> #include <iostream>
> using namespace std;
> class A
> {
> public:
> A ()
> {
> cout << "constructor" << endl;
> }
> A (A &a)
> {
> cout << "copy constructor" << endl;
> }
> A operator = (const A &a)
> {
> cout << "= operator" << endl;
> }
> };
> int main()
> {
> A a;
> A b;
> b = ( b = a );
> }
>
>
> Following error is issued while running the above test case on gcc 3.4.6
>
> test007.cpp: In function `int main()':
> test007.cpp:23: error: no matching function for call to `A::A(A)'
> test007.cpp:11: note: candidates are: A::A(A&)
>
> However, if I change the argument of copy constructor to "const A &", the
> program compiles finely.
> I could not understand why const is required in copy constructor for running
> the test case.
>
> Also note that in the output of the modified program (which compiled
> finely), no copy constructor is called.
> I am running the test case with --no-elide-constructors option to disable
> the RVO (so that I can get all copy constructor calls).
>
> Thanks in advance for any help.
>
> Regards,
> Mustafa
next prev parent reply other threads:[~2009-08-12 6:45 UTC|newest]
Thread overview: 4+ messages / expand[flat|nested] mbox.gz Atom feed top
2009-08-12 10:03 Mustafa4LP
2009-08-12 11:28 ` Sven Eschenberg [this message]
2009-08-12 20:04 ` John (Eljay) Love-Jensen
2009-08-13 9:28 ` Mustafa4LP
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=4A8264C2.5080509@cs.uni-frankfurt.de \
--to=eschenb@cs.uni-frankfurt.de \
--cc=gcc-help@gcc.gnu.org \
--cc=mustafa.hussain@nechclst.in \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).