220 24927 <7db63114-8e12-4d38-94e4-39f0061362de@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: Fri, 4 Mar 2016 11:48:07 -0800 (PST) Lines: 157 Approved: news@gmane.org Message-ID: <7db63114-8e12-4d38-94e4-39f0061362de@isocpp.org> References: <92a6bfcf-0100-40b2-8946-a9c634783ec9@isocpp.org> <3c8bb230-c83b-4c18-95b6-1cc32caae76d@isocpp.org> Reply-To: std-proposals@isocpp.org NNTP-Posting-Host: plane.gmane.org Mime-Version: 1.0 Content-Type: multipart/mixed; boundary="----=_Part_1228_1211116281.1457120887341" X-Trace: ger.gmane.org 1457120899 30749 80.91.229.3 (4 Mar 2016 19:48:19 GMT) X-Complaints-To: usenet@ger.gmane.org NNTP-Posting-Date: Fri, 4 Mar 2016 19:48:19 +0000 (UTC) To: ISO C++ Standard - Future Proposals Original-X-From: std-proposals+bncBDFLZXWOVYIBB6GM463AKGQEOJGHLYA@isocpp.org Fri Mar 04 20:48:11 2016 Return-path: Envelope-to: gclcip-std-proposals@m.gmane.org Original-Received: from mail-yk0-f198.google.com ([209.85.160.198]) by plane.gmane.org with esmtp (Exim 4.69) (envelope-from ) id 1abvhq-0001Nx-LN for gclcip-std-proposals@m.gmane.org; Fri, 04 Mar 2016 20:48:10 +0100 Original-Received: by mail-yk0-f198.google.com with SMTP id m4sf69358649yke.2 for ; Fri, 04 Mar 2016 11:48:10 -0800 (PST) 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=MEX0rSUeG4PimnSdqV6OENhHC7yROUMXD7mht/YmHwk=; b=Mg4uqt/Ocu8FnfDEqs/0Yr3V/X9NuP0fDBFg2oYm5amom2yoC0Ogcl2S3zkNK9ZBU+ G3datlFTIM3t8WTrQdPqa1EydFQdO5NH3DyLv9xcBdydN0woDzGPn6RmBvcP/n100NXH +ZY0/MMpXb0ipYu6woF4LU6UBgrUwPvjff3uodHUQKA5gfCJBiAdSWuUg1Pk5fFSODpY jBcqkcJDN5tivof1M7ILplCD60+ds4ldFYm0S65DtyQZpJVaw9l0C+ZjJ52duSVUzfgE zZN3hQAVW74WTKR17njY4yoSM2uHiUUNFyGlxXAlVCL4yQ/H/221Q8HzAo2cUZrBTfJq 6iOQ== 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=MEX0rSUeG4PimnSdqV6OENhHC7yROUMXD7mht/YmHwk=; b=HgLHF5eXcvs2M/FYnFJ0gXcr5YZKrd9rYptNQPFwv75xdWBfJCi/rV3wE0798srVis OyOlTsBXDMtSpJOUwXXW6r5Jn+EN2ZQi0j3/7kHCdampaorD8xuntl2vUuZ/HaNGADRw L9aYU++OukeDe1OxHOXps0abIKyRmsxySKiz1YGe3kbdZ3K9T7tLmZLNRpkRwxNLfP7o gvOKBj1z11bKhWb7e1SIQkNDhtMs/gXFDCYJLXeVF+7OLmijmIG1qfSQYDGAjhaCxteV a5f+OQM1BV53mLLHvQhvHaTPiKqbaiC3PlhUAUSdBi4QIExu+cm6h+f40D7Jydeb+yI4 HjpQ== X-Gm-Message-State: AD7BkJIQVWS09GYPZiGQ9nanepWI+6FAKJr8ua9SGWlpSfOYadv4eN/rwLd3Xydt+k00dA== X-Received: by 10.140.28.71 with SMTP id 65mr7186602qgy.1.1457120889665; Fri, 04 Mar 2016 11:48:09 -0800 (PST) X-BeenThere: std-proposals@isocpp.org Original-Received: by 10.107.14.138 with SMTP id 132ls297840ioo.27.gmail; Fri, 04 Mar 2016 11:48:08 -0800 (PST) X-Received: by 10.50.109.230 with SMTP id hv6mr24458igb.2.1457120888279; Fri, 04 Mar 2016 11:48:08 -0800 (PST) In-Reply-To: <3c8bb230-c83b-4c18-95b6-1cc32caae76d@isocpp.org> X-Original-Sender: wkaras@yahoo.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: , X-Original-From: Walt Karas Xref: news.gmane.org gmane.comp.lang.c++.isocpp.proposals:24927 Archived-At: ------=_Part_1228_1211116281.1457120887341 Content-Type: multipart/alternative; boundary="----=_Part_1229_67359125.1457120887341" ------=_Part_1229_67359125.1457120887341 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: quoted-printable I'm a complete beginner with variadic templates. I think the equivalent=20 with your proposal is maybe: template struct A_component { Target_ Target; Member_type_ Member_type; static const Target::Member_type *mptr =3D mptr_ }; #define COMPONENTS(C) A_component, typedef >... But still there is the issue of needing member pointers to parent/base=20 classes. On Friday, March 4, 2016 at 1:02:44 PM UTC-5, Cleiton Santoia wrote: > > > Reflection ? p0255r0=20 > > > X...> solves your case.... > > > > > Em sexta-feira, 4 de mar=C3=A7o de 2016 02:01:15 UTC-3, Walt Karas escrev= eu: >> >> Another possible approach generic mapping of operations to components=20 >> would be to accept the idiom ++C , C a class name, as a template=20 >> parameter. The compiler would implicitly expand ++C to multiple=20 >> parameters, namely C itself, followed by a pair of parameters for each= =20 >> component of C. Each pair would consist of the type T of the component= ,=20 >> followed by the member pointer of type T C::* to the component. (The=20 >> reason for using ++ is to suggest iteration of over the components.) Fo= r=20 >> example, for this class: >> >> struct A { int i, j; double x; }; >> >> the template instantiation: >> >> X<++A> >> >> would be equivalent to: >> >> X >> >> This hopefully would allow variadic templates to create reasonable=20 >> default comparison, input and display, persistence, endiance swap, hashi= ng,=20 >> etc. operations for many classes. A difficult issue would be wether the= =20 >> context where X<++A> is instantiated would be required to have access to= =20 >> all the implicitly reverenced component types and components themselves. >> >> Perhaps the compiler could also except the idiomatic template parameter= =20 >> --C, which would lay out the components in reverse order (of destruction= )=20 >> rather than forward order (of construction). >> > --=20 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 e= mail 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/7db63114-8e12-4d38-94e4-39f0061362de%40isocpp.or= g. ------=_Part_1229_67359125.1457120887341 Content-Type: text/html; charset=UTF-8 Content-Transfer-Encoding: quoted-printable
I'm a complete beginner with variadic templates.=C2=A0= I think the equivalent with your proposal is maybe:

template <cl= ass Target_, typename Member_type_, Member_type_ Target_::* mptr_>
st= ruct A_component
=C2=A0 {
=C2=A0=C2=A0=C2=A0 Target_ Target;
=C2= =A0=C2=A0=C2=A0 Member_type_ Member_type;
=C2=A0=C2=A0=C2=A0 static cons= t Target::Member_type *mptr =3D mptr_
=C2=A0 };

#define COMPONENT= S(C) A_component<C, typename<C, is_member_object_pointer || is_parent= >, typedef<C, is_member_object_pointer || is_parent> >...
But still there is the issue of needing member pointers to parent/base cl= asses.

On Friday, March 4, 2016 at 1:02:44 PM UTC-5, Cleiton Santoia= wrote:
<= br>
=C2=A0 =C2=A0Reflection ?=C2=A0p0255r0

X<typedef<A>...> solves you= r case....




Em sexta-feira, = 4 de mar=C3=A7o de 2016 02:01:15 UTC-3, Walt Karas escreveu:
Another possible approach generi= c mapping of operations to components would be to accept the idiom ++C , C = a class name, as a template parameter.=C2=A0 The compiler would implicitly = expand ++C to multiple parameters, namely C itself, followed by a pair of p= arameters for each component of C.=C2=A0 Each pair would consist of=C2=A0 t= he type T of the component, followed by the member pointer of type T C::* t= o the component.=C2=A0 (The reason for using ++ is to suggest iteration of = over the components.)=C2=A0 For example, for this class:

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

the template instantiation:

X<++A&g= t;

would be equivalent to:

X<A, int, &C::i, int, &= C::j, double, &C::x>

This hopefully would allow variadic temp= lates to create reasonable default comparison, input and display, persisten= ce, endiance swap, hashing, etc. operations for many classes.=C2=A0 A diffi= cult issue would be wether the context where X<++A> is instantiated w= ould be required to have access to all the implicitly reverenced component = types and components themselves.

Perhaps the compiler could also exc= ept the idiomatic template parameter --C, which would lay out the component= s in reverse order (of destruction) rather than forward order (of construct= ion).

--
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/7db63114-8e12-4d38-94e4-39f0061362de= %40isocpp.org.
------=_Part_1229_67359125.1457120887341-- ------=_Part_1228_1211116281.1457120887341-- .