220 25750 <77c1b655-f29c-4836-aab9-334d6e77af73@isocpp.org> article Path: news.gmane.org!not-for-mail From: "'Walt Karas' via ISO C++ Standard - Future Proposals" Newsgroups: gmane.comp.lang.c++.isocpp.proposals Subject: Re: "Components" template with implicitly defined partial specializations for all classes Date: Mon, 2 May 2016 14:38:34 -0700 (PDT) Lines: 237 Approved: news@gmane.org Message-ID: <77c1b655-f29c-4836-aab9-334d6e77af73@isocpp.org> References: Reply-To: std-proposals@isocpp.org NNTP-Posting-Host: plane.gmane.org Mime-Version: 1.0 Content-Type: multipart/mixed; boundary="----=_Part_769_1774317915.1462225115022" X-Trace: ger.gmane.org 1462225120 17609 80.91.229.3 (2 May 2016 21:38:40 GMT) X-Complaints-To: usenet@ger.gmane.org NNTP-Posting-Date: Mon, 2 May 2016 21:38:40 +0000 (UTC) To: ISO C++ Standard - Future Proposals Original-X-From: std-proposals+bncBDFLZXWOVYIBBW4RT64QKGQEK6TCGVI@isocpp.org Mon May 02 23:38:40 2016 Return-path: Envelope-to: gclcip-std-proposals@m.gmane.org Original-Received: from mail-qk0-f199.google.com ([209.85.220.199]) by plane.gmane.org with esmtp (Exim 4.69) (envelope-from ) id 1axLY6-00052Q-17 for gclcip-std-proposals@m.gmane.org; Mon, 02 May 2016 23:38:38 +0200 Original-Received: by mail-qk0-f199.google.com with SMTP id t184sf2626725qkh.3 for ; Mon, 02 May 2016 14:38:37 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=isocpp-org.20150623.gappssmtp.com; s=20150623; h=date:from:to: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=ulPOPffmKfnmo1fBfQmjbMtDOhhPdU/BWM0yj3v74/E=; b=eguKRhk3IsMOmN9Y2j9NxqH8yGeW0Zdd5bKF9UGt9YVjGJdokAoDUrG02ogAyU7Q24 hRrYtuKhPmh09I25/htxm/gio7IX1QPmOTNLiNe+CZdbscJgpMaB7uksdjb/vHMiKOq1 jZ/T9RY7OJ6PI4+8twp/G4FEFavfFJpEfDiB6kzRq9yo+F21OSBLYv59eeFKzJjTQHhb PNrAYCeWtPf/J5f0/VqcuPTCR6tP7fMSoCLYf8qjZ8zQ3PGVXD2h1PJXYRWyCbMSQyIw mr6hNT1nciYax+QN8P/wEnzT1o+MXmS2qQxb/rTju4DQxq54ZVJ/nNQWZxXZWC18PgkK mkfQ== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20130820; h=x-gm-message-state:date:from:to: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=ulPOPffmKfnmo1fBfQmjbMtDOhhPdU/BWM0yj3v74/E=; b=N/Pmv57/dd23rJG4etXxq0l4mcDOYZHKwA661N5QAXaq/kWKM/M0/0TLvLQkSjCkWl ps/K1Cqn34KIzTTycljOju9m3OcAw69XLjlGy8ouVmP9WsWMKUaUthoLvQPZOlO3bzxY dfNREunWa4yg+U+tYuxsO0grmKGZSgznAh13Gmfdpk0FCKKKGInSdlIhgTjrYBQq/Fi/ CJg2d5akNGt19Sz6QNa6lVsmsBIN+6OG3iA/arYlNqTuRktRlgNLbt4mIeMbDBDZCTRJ DrkTLYAL7eOzSulLxBeNq8dHQGPSVQowc3qKgzIf0hLko3LDSOl/o0+inZzLt+jeJSIw BHlQ== X-Gm-Message-State: AOPr4FUdEWQxhtCGzPFCWyX4Fxpn90T6DngTKN/XVGA5fpLom6JnKz9iupKPAn8hQb2GZg== X-Received: by 10.141.41.70 with SMTP id s67mr26085199qhe.11.1462225117121; Mon, 02 May 2016 14:38:37 -0700 (PDT) X-BeenThere: std-proposals@isocpp.org Original-Received: by 10.107.170.210 with SMTP id g79ls1818412ioj.61.gmail; Mon, 02 May 2016 14:38:35 -0700 (PDT) X-Received: by 10.50.90.5 with SMTP id bs5mr360388igb.5.1462225115721; Mon, 02 May 2016 14:38:35 -0700 (PDT) In-Reply-To: X-Original-Sender: wkaras@yahoo.com X-Original-From: Walt Karas 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:25750 Archived-At: ------=_Part_769_1774317915.1462225115022 Content-Type: multipart/alternative; boundary="----=_Part_770_819221775.1462225115022" ------=_Part_770_819221775.1462225115022 Content-Type: text/plain; charset=UTF-8 I took a stab at trying to implement something along these lines in the current language: https://github.com/wkaras/C-plus-plus-library-default-operators On Saturday, February 27, 2016 at 4:18:34 PM UTC-5, Walt Karas wrote: > > A "primitive" template named "Components" with this interface: > > template < > class Target, > template > class Op, > typename Param> > struct Components > { > bool operator () (Param & param); > }; > > The Op parameter is presumed to fulfill the constraint: > > Param & param; > Op op; > bool b = op(param); > > for all data members and base classes of the class Target. The function > operator of Components would be generated implicitly. It would instantiate > Op for each component of Target, create an instance of the instantiated > class, and call its function operator as shown in the constraint above, > forwarding to it the parameter 'param'. If the call returned false, the > function operator of Components would return false immediately. If all the > calls for components returned true, the Components operator would also > return true. > > For example, for this class: > > struct A { int i, j; double x; }; > > the compiler would implicitly generate the equivalent of the following > partial specialization of Components: > > template < > template class Op, > typename Param> > struct Components > { > bool operator () (Param & param) > { > return( > Op()(param) && > Op()(param) && > Op()(param)); > } > }; > > This would make it possible to define a default operator == as a library > template: > > template > class Equality_op; > > template > class Equality_param > { > friend class Equality_op; > > const Target &op1, &op2; > > public: > Equality_param(const Target &op1_, const Target &op2_) > : op1(op1_), op2(op2_) { } > }; > > template > struct Equality_op > { > template > struct Op > { > bool operator () (Equality_param & param) > { > return(param.op1.*mptr == param.op2.*mptr); > } > }; > }; > > template > bool operator == (const Target &op1, const Target &op2) > { > Equality_param param(op1, op2); > > Components< > Target, Equality_op::template Op, Equality_param > c; > > return(c(param)); > } > > Likewise for default operators > and <. > > This implies that member pointers can point to base classes, which I'm > guessing is an idea that's been rejected before. So doing this would > create "pressure" to accept the syntax: > > Base Derived::*p = &Derived::Base; > > There would also be a need for another "primitive" template > Components_no_virt_base that would skip over virtual base classes I think. > > A nice thing about this approach is that the default == operator could > sometimes be "fixed" rather than discarded. For example, suppose the > default == operator worked for a class "Big" except for one data member > "argh", which was a pointer to a heap object. One could handle this with > an overriding instantiation of "Equality_op": > > template <> > struct Equality_op > { > template <> > struct Op > { > bool operator () (Equality_param & param) > { > // ... > } > }; > }; > -- 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/77c1b655-f29c-4836-aab9-334d6e77af73%40isocpp.org. ------=_Part_770_819221775.1462225115022 Content-Type: text/html; charset=UTF-8 Content-Transfer-Encoding: quoted-printable
I took a stab at trying to implement something along these= lines in the current language:

https://github.com/wkaras/C-plus-plu= s-library-default-operators

On Saturday, February 27, 2016 at 4:18:3= 4 PM UTC-5, Walt Karas wrote:
<= div dir=3D"ltr">A "primitive" template named "Components&quo= t; with this interface:

template <
=C2=A0 class Target,
=C2= =A0 template <typename Member_type, Member_type Target::*mptr>
=C2= =A0=C2=A0=C2=A0 class Op,
=C2=A0 typename Param>
struct Components=
=C2=A0 {
=C2=A0=C2=A0=C2=A0 bool operator () (Param & param);=C2=A0 };

The Op parameter is presumed to fulfill the constraint:
Param & param;
Op<Member_type, &Target::member> op; =
bool b =3D op(param);

for all data members and base classes of t= he class Target.=C2=A0 The function operator of Components would be generat= ed implicitly.=C2=A0 It would instantiate Op for each component of Target, = create an instance of the instantiated class, and call its function operato= r as shown in the constraint above, forwarding to it the parameter 'par= am'.=C2=A0 If the call returned false, the function operator of Compone= nts would return false immediately.=C2=A0 If all the calls for components r= eturned true, the Components operator would also return true.

For ex= ample, for this class:

struct A { int i, j; double x; };

the = compiler would implicitly generate the equivalent of the following partial = specialization of Components:

template <
=C2=A0 template <t= ypename Member_type, Member_type A::*mptr> class Op,
=C2=A0 typename = Param>
struct Components<A, Op, Param>
=C2=A0 {
=C2=A0=C2= =A0=C2=A0 bool operator () (Param & param)
=C2=A0=C2=A0=C2=A0=C2=A0= =C2=A0 {
=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 return(
=C2=A0=C2= =A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 Op<int, &A::i>()(pa= ram) &&
=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 O= p<int, &A::j>()(param) &&
=C2=A0=C2=A0=C2=A0=C2=A0=C2= =A0=C2=A0=C2=A0=C2=A0=C2=A0 Op<double, &A::x>()(param));
=C2= =A0=C2=A0=C2=A0=C2=A0=C2=A0 }
=C2=A0 };

This would make it possib= le to define a default operator =3D=3D as a library template:

templa= te <class Target>
class Equality_op;

template <class Tar= get>
class Equality_param
=C2=A0 {
=C2=A0=C2=A0=C2=A0 friend cl= ass Equality_op<Target>;

=C2=A0=C2=A0=C2=A0 const Target &= op1, &op2;

=C2=A0 public:
=C2=A0=C2=A0=C2=A0 Equality_param(c= onst Target &op1_, const Target &op2_)
=C2=A0=C2=A0=C2=A0=C2=A0= =C2=A0 : op1(op1_), op2(op2_) { }
=C2=A0 };

template <class Ta= rget>
struct Equality_op
=C2=A0 {
=C2=A0=C2=A0=C2=A0 template &= lt;typename Member_type, Member_type Target::*mptr>
=C2=A0=C2=A0=C2= =A0 struct Op
=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 {
=C2=A0=C2=A0=C2=A0=C2= =A0=C2=A0=C2=A0=C2=A0 bool operator () (Equality_param<Target> & = param)
=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 {
=C2= =A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 return(para= m.op1.*mptr =3D=3D param.op2.*mptr);
=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2= =A0=C2=A0=C2=A0=C2=A0 }
=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 };
=C2=A0 };
template <class Target>
bool operator =3D=3D (const Target &= amp;op1, const Target &op2)
=C2=A0 {
=C2=A0=C2=A0=C2=A0 Equality_= param<Target> param(op1, op2);

=C2=A0=C2=A0=C2=A0 Components&l= t;
=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 Target, Equality_op<Target>::tem= plate Op, Equality_param<Target> > c;

=C2=A0=C2=A0=C2=A0 re= turn(c(param));
=C2=A0 }

Likewise for default operators > and = <.

This implies that member pointers can point to base classes, w= hich I'm guessing is an idea that's been rejected before.=C2=A0 So = doing this would create "pressure" to accept the syntax:

B= ase Derived::*p =3D &Derived::Base;

There would also be a need f= or another "primitive" template Components_no_virt_base that woul= d skip over virtual base classes I think.

A nice thing about this ap= proach is that the default =3D=3D operator could sometimes be "fixed&q= uot; rather than discarded.=C2=A0 For example, suppose the default =3D=3D o= perator worked for a class "Big" except for one data member "= ;argh", which was a pointer to a heap object.=C2=A0 One could handle t= his with an overriding instantiation of "Equality_op":

tem= plate <>
struct Equality_op<Big>
=C2=A0 {
=C2=A0=C2=A0= =C2=A0 template <>
=C2=A0=C2=A0=C2=A0 struct Op<Argh, &Big:= :argh>
=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 {
=C2=A0=C2=A0=C2=A0=C2=A0= =C2=A0=C2=A0=C2=A0 bool operator () (Equality_param<Target> & par= am)
=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 {
=C2=A0= =C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 // ...
=C2= =A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 }
=C2=A0=C2=A0=C2=A0= =C2=A0=C2=A0 };
=C2=A0 };

--
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/77c1b655-f29c-4836-aab9-334d6e77af73= %40isocpp.org.
------=_Part_770_819221775.1462225115022-- ------=_Part_769_1774317915.1462225115022-- .