220 40682 article Path: news.gmane.org!.POSTED!not-for-mail From: mihailnajdenov@gmail.com Newsgroups: gmane.comp.lang.c++.isocpp.proposals Subject: Re: Re: Use new keyword to indicate that a function should return it's argument allocated on the heap Date: Tue, 23 Oct 2018 00:16:18 -0700 (PDT) Lines: 300 Approved: news@gmane.org Message-ID: References: <269cd5f8-0201-431b-9683-52a945aa9c3d@isocpp.org> Reply-To: std-proposals@isocpp.org NNTP-Posting-Host: blaine.gmane.org Mime-Version: 1.0 Content-Type: multipart/mixed; boundary="----=_Part_4882_74888822.1540278979047" X-Trace: blaine.gmane.org 1540278855 19641 195.159.176.226 (23 Oct 2018 07:14:15 GMT) X-Complaints-To: usenet@blaine.gmane.org NNTP-Posting-Date: Tue, 23 Oct 2018 07:14:15 +0000 (UTC) Cc: sunsetburninginmysoul@gmail.com To: ISO C++ Standard - Future Proposals Original-X-From: std-proposals+bncBCUJ3A7GRAPRBQ4VXPPAKGQEAJIUGMY@isocpp.org Tue Oct 23 09:14:11 2018 Return-path: Envelope-to: gclcip-std-proposals@m.gmane.org Original-Received: from mail-yb1-f200.google.com ([209.85.219.200]) by blaine.gmane.org with esmtp (Exim 4.84_2) (envelope-from ) id 1gEqtG-0004yL-RF for gclcip-std-proposals@m.gmane.org; Tue, 23 Oct 2018 09:14:11 +0200 Original-Received: by mail-yb1-f200.google.com with SMTP id z8-v6sf192588ybo.17 for ; Tue, 23 Oct 2018 00:16:21 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=isocpp-org.20150623.gappssmtp.com; s=20150623; h=date:from:to:cc:message-id:in-reply-to:references:subject :mime-version:x-original-sender:reply-to:precedence:mailing-list :list-id:list-post:list-help:list-archive:list-subscribe :list-unsubscribe; bh=01HzCuaNIIrQcoKIBSJD44OGcjp4iAwpKLE3QhpJ2d4=; b=tCYSGNIs6KB+xzHB7kzZX89q1lvQyOS1wQJ3HaDewGTtbkkm7tc7yshjq1MmS7hTfj k1gjX25HFa3rRd1a8vCE751e6OOlxYMGJBP1uZjAoBOnuv3UBBfhrc2HdBfOCYe1wQWQ /zmtFydaqeRP8X8tucDMsw1rhNVta3GhYdy8fOsM8xqKYIS8XTgj9SftdxpzTeLdkCMX aRZdmzkp7mTmzZvKyBmD/ZNLKWkHgCsR1FsiZP1tZpWQq3WHX1JCqkyl0ODO2YIoqif6 lRrUy9CemrTK/ppPUkuFjY1rUAmlkqzinfwe7UoGZyFjcUbPfWXQGjVU4Vdn9YTezCdK Sxeg== DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20161025; h=date:from:to:cc:message-id:in-reply-to:references:subject :mime-version:x-original-sender:reply-to:precedence:mailing-list :list-id:list-post:list-help:list-archive:list-subscribe :list-unsubscribe; bh=01HzCuaNIIrQcoKIBSJD44OGcjp4iAwpKLE3QhpJ2d4=; b=EhrnEBW21S1jaTFLK1Be5hgivMCEVQDqyrk89AX+7gGC1IJHh3F6kT/uzyT4MpNymC U0sYwBHW4AQdc6MdY2iu1bHQJ1H5rGf0Lzq7H1+9PezS39bR2fC5JwcBfacMLglwQZoJ hJQTCRHLF3MXHSnGHgQLdbb5/TiKPyM1pR228uQ3YevkARlpXaQpH+RJNSWD9KEnQyzX mM5xtVuFjIQEshNBsgcUiV/1mdXsKDBDKJw4HdHWni1/Lp5qeu2MTiaR32VZkXbImFUv 2f11kl/0wSCw+ye4GXARZv/mu5crskIi3h8K7g3JW0lBW6BXcfc8zGWYrO0TxXwYI0xq ZQBw== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20161025; h=x-gm-message-state:date:from:to:cc:message-id:in-reply-to :references:subject:mime-version:x-original-sender:reply-to :precedence:mailing-list:list-id:x-spam-checked-in-group:list-post :list-help:list-archive:list-subscribe:list-unsubscribe; bh=01HzCuaNIIrQcoKIBSJD44OGcjp4iAwpKLE3QhpJ2d4=; b=Mx429H/1LvrpTxotDvs7BCCJj0L2oshwdItRKjCIL6czzuu3nwLP3A1Qr15bBHYXz8 QSpyFRlNOBhN/SBQeOJv3KyDlG7gDwrUTCNEBQ8s0Ice8wrS5UiAhcd6H6hIlCeXHgsq NRR5p6guVSD6NALl14IcRU5yvCzaZgQptE0Pn4qHlH1vmn0vmBpedKsDmG4MxhlAbriN uELFxkE14NGp6jfAADybaZo0ioZHuagy3wZ5EN+Zrini2C61jH7kj4t2RU/QukOZRaFx 8gc+faXYRlnK+S4hfjOkwvg22/8f+3WPHqCzVtDIPTwZ1nY6YAnYx641hX9j2sjkp76X 5mxg== X-Gm-Message-State: ABuFfohDBsQsYiOVWUn2731RqJ8H3CWO6191Tyf/BZfhodhVqxI1xI/Y YorCvFZgq4D0Z5VzZbGSBnFWgg== X-Google-Smtp-Source: ACcGV61j38cP+2m8+2DZYUUamAqb87iCuM385jMmLCxwuooqVSTLmIRiiTa7xHdXrK6dPKCqmTK8VQ== X-Received: by 2002:a0d:c841:: with SMTP id k62-v6mr28810711ywd.6.1540278981122; Tue, 23 Oct 2018 00:16:21 -0700 (PDT) X-BeenThere: std-proposals@isocpp.org Original-Received: by 2002:a25:bb8d:: with SMTP id y13-v6ls1078339ybg.4.gmail; Tue, 23 Oct 2018 00:16:19 -0700 (PDT) X-Received: by 2002:a25:daca:: with SMTP id n193-v6mr107403ybf.7.1540278979705; Tue, 23 Oct 2018 00:16:19 -0700 (PDT) In-Reply-To: X-Original-Sender: MihailNajdenov@gmail.com Precedence: list Mailing-list: list std-proposals@isocpp.org; contact std-proposals+owners@isocpp.org List-ID: X-Spam-Checked-In-Group: std-proposals@isocpp.org X-Google-Group-Id: 399137483710 List-Post: , List-Help: , List-Archive: List-Subscribe: , List-Unsubscribe: , Xref: news.gmane.org gmane.comp.lang.c++.isocpp.proposals:40682 Archived-At: ------=_Part_4882_74888822.1540278979047 Content-Type: multipart/alternative; boundary="----=_Part_4883_1926723071.1540278979048" ------=_Part_4883_1926723071.1540278979048 Content-Type: text/plain; charset="UTF-8" On Tuesday, October 23, 2018 at 5:30:52 AM UTC+3, Richard Smith wrote: > > On Mon, 22 Oct 2018 at 01:23, > wrote: > >> You must give better motivation. >> If you want to use a function that returns an object and have that object >> new-ed without copy/move, then you should explain what is stopping you >> creating the returned object via the new expression in the first place, >> especially now with template deduction guide. >> If you want a copy/move of the object, returned by the function you can >> do this today using new auto(create_object()); >> > > If create_object() returns by value (as make_tuple does, for instance), > this does not perform a copy or move. Rather, it constructs directly into > the allocated storage. That is, we already have the feature that is being > requested here: just use 'new'. > Great to hear! It didn't know the new RVO rules work here as well as it is not so obvious - we tend to think this is stack-only. Very interesting... I might as well update cppreference to note that it works here as well. > > My guess is you want the former, but some better examples would be nice. >> >> >> On Monday, October 22, 2018 at 7:43:42 AM UTC+3, sunsetburn...@gmail.com >> wrote: >>> >>> We could write a function to take something allocated on the stack and >>> return a copy of it on the heap: >>> >>> template T* as_new(T&& obj) { >>> return new T{ std::forward(obj)}; >>> } >>> >>> And this works in most cases while only incurring a performance hit. >>> However, it breaks down when using objects that either can't be copied or >>> moved. Providing overloads of functions that allocate something in a >>> specific provided space can be tedious, too. >>> >>> *Proposal: *putting the new keyword in parethesis before a function or >>> expression results in the output of that expression being allocated on the >>> heap, or in the provided buffer. >>> >>> Let's take the function make_tuple as an example: >>> >>> #include >>> int main() { >>> using std::tuple; >>> using std::make_tuple; >>> >>> /* Allocates tuple normally */ >>> tuple t1 = make_tuple(4, 0.3); >>> >>> /*Allocates tuple on heap, returns pointer */ >>> tuple* t2 = (new) make_tuple(4, 0.3); >>> >>> char buff[100]; >>> >>> /* Allocates tuple in buff, returns pointer */ >>> tuple* t3 = (new(buff)) make_tuple(4, 0.3); >>> } >>> >>> >>> if a function returns a reference, it's a compiler error to use this >>> construct, and if the function returns a pointer, then this construct >>> returns a pointer to that pointer. >>> >>> >>> -- >> You received this message because you are subscribed to the Google Groups >> "ISO C++ Standard - Future Proposals" group. >> To unsubscribe from this group and stop receiving emails from it, send an >> email to std-proposal...@isocpp.org . >> To post to this group, send email to std-pr...@isocpp.org . >> To view this discussion on the web visit >> https://groups.google.com/a/isocpp.org/d/msgid/std-proposals/c5e8d685-d72c-4163-bb6d-f9cee0fa190b%40isocpp.org >> >> . >> > -- You received this message because you are subscribed to the Google Groups "ISO C++ Standard - Future Proposals" group. To unsubscribe from this group and stop receiving emails from it, send an email to std-proposals+unsubscribe@isocpp.org. To post to this group, send email to std-proposals@isocpp.org. To view this discussion on the web visit https://groups.google.com/a/isocpp.org/d/msgid/std-proposals/bdbf2214-87c0-4d14-8831-c6ffd5da2eba%40isocpp.org. ------=_Part_4883_1926723071.1540278979048 Content-Type: text/html; charset="UTF-8" Content-Transfer-Encoding: quoted-printable


On Tuesday, October 23, 2018 at 5:30:52 AM UTC+3, = Richard Smith wrote:
On Mon, 22 Oct 2018 at= 01:23, <mihailn...@gmail.com> wrote:
You must give better motivation.=C2=A0
If you = want to use a function that returns an object and have that object new-ed w= ithout copy/move, then you should explain what is stopping you creating the= returned object via the new expression in the first place, especially now = with template deduction guide.
If you want a copy/move of the obj= ect, returned by the function you can do this today using new auto(create_object());

If create_object() returns by value (as make_tuple = does, for instance), this does not perform a copy or move. Rather, it const= ructs directly into the allocated storage. That is, we already have the fea= ture that is being requested here: just use 'new'.

Great to hear! It didn't know the new= RVO rules work here as well as it is not so obvious - we tend to think thi= s is stack-only. Very interesting...=C2=A0
I might as well update= cppreference to note that it works here as well.
=C2=A0

<= div>My guess is you want the former, but some better examples would be nice= ..


On Monday, October 22, 2018 at 7:43:42 AM UTC+3, sun= setburn...@gmail.com wrote:
We could write a function to take something allocated on the s= tack and return a copy of it on the heap:

template<class<= /span> T> T* as_new(T&&<= /span> obj) {<= span style=3D"color:#000">
=C2=A0 =C2=A0 return new T{ std:= :forward<decltype(obj= )>(<= /span>obj= )};
= }

<= /div>And this works in most cases while only incurring a performance hit. H= owever, it breaks down when using objects that either can't be copied o= r moved. Providing overloads of functions that allocate something in a spec= ific provided space can be tedious, too.=C2=A0

= Proposal: putting the new keyword in parethesis before a function or ex= pression results in the output of that expression being allocated on the he= ap, or in the provided buffer.=C2=A0

Let's tak= e the function make_tuple as an example:

#inc= lude <tuple>
int main() {
=C2=A0 =C2= =A0using std::tuple;
=C2=A0 =C2=A0using std::make_tuple;
= =C2=A0 =C2=A0 /* Allocates tuple normally */
=C2=A0 =C2=A0tuple<int, double> t1 =3D make_tuple(4, 0.3);
=C2=A0 =C2=A0 /*Allocates= tuple on heap, returns pointer */
=C2=A0 =C2=A0tuple<int, double>* t2 =3D (new) make_tuple(4, 0.3);

=C2=A0 =C2=A0 char buff[100];
=C2=A0 =C2=A0
=C2=A0 =C2=A0/* Allocates tuple in buff, returns = pointer */
=C2=A0 =C2= =A0tuple<= ;int= , double>* t3 =3D (<= span style=3D"color:#008">new(buff)) make_tuple(4= , 0.3);
}


if a function returns a reference, it's a compiler error to use th= is construct, and if the function returns a pointer, then this construct re= turns a pointer to that pointer.=C2=A0


<= /div>

--
You received this message because you are subscribed to the Google Groups &= quot;ISO C++ Standard - Future Proposals" group.
To unsubscribe from this group and stop receiving emails from it, send an e= mail to std-proposal...@isocpp.org.
To post to this group, send email to std-pr...@isocpp.org.
To view this discussion on the web visit https://groups.google.com= /a/isocpp.org/d/msgid/std-proposals/c5e8d685-d72c-4163-bb6d-= f9cee0fa190b%40isocpp.org.

--
You received this message because you are subscribed to the Google Groups &= quot;ISO C++ Standard - Future Proposals" group.
To unsubscribe from this group and stop receiving emails from it, send an e= mail to std-proposa= ls+unsubscribe@isocpp.org.
To post to this group, send email to std-proposals@isocpp.org.
To view this discussion on the web visit https://groups.google.= com/a/isocpp.org/d/msgid/std-proposals/bdbf2214-87c0-4d14-8831-c6ffd5da2eba= %40isocpp.org.
------=_Part_4883_1926723071.1540278979048-- ------=_Part_4882_74888822.1540278979047-- .