220 28303 <4630771e-c4d2-41f2-b133-d4272c37bb81@isocpp.org> article Path: news.gmane.org!.POSTED!not-for-mail From: Nicol Bolas Newsgroups: gmane.comp.lang.c++.isocpp.proposals Subject: Re: keyword to make a class function unique per object Date: Tue, 20 Sep 2016 21:02:35 -0700 (PDT) Lines: 131 Approved: news@gmane.org Message-ID: <4630771e-c4d2-41f2-b133-d4272c37bb81@isocpp.org> References: <746c9278-4fcc-4022-b377-8a77e1afd043@isocpp.org> Reply-To: std-proposals@isocpp.org NNTP-Posting-Host: blaine.gmane.org Mime-Version: 1.0 Content-Type: multipart/mixed; boundary="----=_Part_1464_488597594.1474430555616" X-Trace: blaine.gmane.org 1474430571 13751 195.159.176.226 (21 Sep 2016 04:02:51 GMT) X-Complaints-To: usenet@blaine.gmane.org NNTP-Posting-Date: Wed, 21 Sep 2016 04:02:51 +0000 (UTC) To: ISO C++ Standard - Future Proposals Original-X-From: std-proposals+bncBCEKFTV6ZUMBBXEMRC7QKGQEWT5YVZQ@isocpp.org Wed Sep 21 06:02:47 2016 Return-path: Envelope-to: gclcip-std-proposals@m.gmane.org Original-Received: from mail-qt0-f200.google.com ([209.85.216.200]) by blaine.gmane.org with esmtp (Exim 4.84_2) (envelope-from ) id 1bmYk3-00023b-1p for gclcip-std-proposals@m.gmane.org; Wed, 21 Sep 2016 06:02:39 +0200 Original-Received: by mail-qt0-f200.google.com with SMTP id p53sf75968774qtp.0 for ; Tue, 20 Sep 2016 21:02:41 -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=wLAWtpBDzJyaP9qST4mGkfjZcWdQhOJJXa2xpxK60Nw=; b=gp20jvqTqS9mvqFzLQDvbMT2SscqTE30Zst7WONxo+OjFFc9ifaDq95ZdoppEOumvW jh3pudi/GXgTMvJ6hMFRESv4ldv8P8QpNXvFF+Jlm5hDUIQxRodcLNcdI1HGaBe+FSsz ySW4Rh5a3KymYiJK2MThedllf570ND42b0wNdGJZC+6U248I5Pzla5OJgmCxx5E4Et6N qdRoQbm9z44IiYKWFIfqLTYYPv9ZF4N94z+y3RDsfIy3fFV3S2YuA+SP/Qg0o+CbjMqn BA/wcChTVL1StzFTEjCByI3EC5sFcIr5r2MgbVnMxd4zePyX0i2sBWzMFosxYpPRcDx3 8Bhg== DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; 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=wLAWtpBDzJyaP9qST4mGkfjZcWdQhOJJXa2xpxK60Nw=; b=VL7gk/g2YIGkySwqL1H0w8lxqn9xtMeqJj8m1TO3591a0KUutTG1n1Pfuy5y2K2QGS HaTulgy9bDMp3VbAuxbpRskicuB9HmHUp3EjFJxaKcdqcYrY5fDszkKNTiG4sIERWWZe 4TnA5yjWezr6BQ2DQsA4t04DtZT/Or+neEtqXcQSf5EJeQqIeoBjoH3te64hS6pJusEg NwowOjEkQoFwQYDRFRQUp4yFKDJX3zghuJDE9EC9R5FfzUvhn1YIpYQGQlw8rHuOm9SA jt8gfl4MFbKd1pkiBxI+QNZcfMwob+fySS7xquLhwYoU8hph7cl6A2dAznn151/AR59I pjNg== 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=wLAWtpBDzJyaP9qST4mGkfjZcWdQhOJJXa2xpxK60Nw=; b=nDGee6PN8szbhlf1x9mCY2dsrt/MBbPxubblj+3/65mFxEjKmHQrXK0fviiu9zEfpR XzT1vaXBMWrIgJHZma/6agMvYQLogyn2Rm0LXLyoBogIN+IcMqLMLXFwR/rBUv5TcDhN dSArLVwBPa/leLNHVr8I5Cb69EqPajwxXxsqdz1N4nRZG18OIC3O2z0Uq3+spFzqq1OR fRALp94NCUeNCU8T1NXrTfZGp4poFBrnuYiCg+//KmTxU7kgtxa5w0kYMxCNcaC946p7 yOGwMDuJOtEdDBZFWdbvEozBAUh8vT2kzCu1gNU/aSoJCl7gOa8Vfqvc2e5/uhXwiInn kE/Q== X-Gm-Message-State: AE9vXwPq96tv/8bixpwmkAkKW8OFbyqCJ9B3d4IPF7AssZs7wLEFX0RfU8l5rykjb8Qe4w== X-Received: by 10.237.34.194 with SMTP id q2mr17193430qtc.3.1474430560485; Tue, 20 Sep 2016 21:02:40 -0700 (PDT) X-BeenThere: std-proposals@isocpp.org Original-Received: by 10.107.153.9 with SMTP id b9ls4297186ioe.23.gmail; Tue, 20 Sep 2016 21:02:36 -0700 (PDT) X-Received: by 10.36.111.147 with SMTP id x141mr75369itb.6.1474430556562; Tue, 20 Sep 2016 21:02:36 -0700 (PDT) In-Reply-To: <746c9278-4fcc-4022-b377-8a77e1afd043@isocpp.org> X-Original-Sender: jmckesson@gmail.com Precedence: list Mailing-list: list std-proposals@isocpp.org; contact std-proposals+owners@isocpp.org List-ID: 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:28303 Archived-At: ------=_Part_1464_488597594.1474430555616 Content-Type: multipart/alternative; boundary="----=_Part_1465_1368959221.1474430555617" ------=_Part_1465_1368959221.1474430555617 Content-Type: text/plain; charset=UTF-8 On Tuesday, September 20, 2016 at 6:16:23 PM UTC-4, Przemyslaw Szuplak wrote: > > Hello, > > I would like to have a keyword which would make a method unique (points to > different memory area) for an instance of class. > > What is the point? Please let me explain using below example. > > > > > > > > > > > > > > *class CA{ template int Get() { static int val = > foo(no); return val; }};//...obj1->Get<4>();obj2->Get<4>();* > > Currently *Get<4>()* would return the same value for 2 different objects > (*obj1 *and *obj2*), because *Get()* points to the same memory area. > There is a possibility that "foo()" may return different value depends on > the objects members. > > Basically *foo()* makes a lot of calculations depends on *no* and data > loaded in objects constructor. That data will not change, so i may expect > that *Get<4>() *will return the same value each time without repeating > calculations. > So I want to get this straight. You have a constexpr value. And you're going to do some computation on it which is based on the runtime state of an object. And you want to cache this value. So... why not actually *do that*? There are ways to cache values based on an object. Furthermore, such a cache would (presumably) be global. And therefore, if you happen to create two separate `CA` instances which happen to have the same data, you'll still get the cached value. The "keyword" idea you're talking about would recalculate the value in that case, since they are two separate instances. Plus, by using an explicit cache, you don't have to restrict yourself to immutable data. That is, your way *only* works if the dependent state never changes. With a real cache, it would work in cases other than yours. Why make this a language feature? -- 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/4630771e-c4d2-41f2-b133-d4272c37bb81%40isocpp.org. ------=_Part_1465_1368959221.1474430555617 Content-Type: text/html; charset=UTF-8 Content-Transfer-Encoding: quoted-printable
On Tuesday, September 20, 2016 at 6:16:23 PM UTC-4, Przemy= slaw Szuplak wrote:
Hello,

I would like to have a keyword which would make a metho= d unique (points to different memory area) for an instance of class.
What is the point? Please let me explain using below example.

<= span style=3D"color:rgb(11,83,148)">class CA
{
=C2=A0=C2=A0 template<int no>
=C2=A0= =C2=A0 int Get()
=C2=A0=C2=A0 {=C2=A0=C2=A0=C2=A0
=C2=A0=C2=A0=C2=A0= =C2=A0=C2=A0 static int val =3D foo(no);
=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 = return val;=C2=A0=C2=A0=C2=A0 =C2=A0=C2=A0=C2=A0
=C2=A0=C2=A0 }
};
//...
obj1->Get<4>();
obj2->Get<4>();<= /span>

Currently Get<= 4>()=C2=A0 would return the same value for 2 different object= s (obj1 and obj2), because Get() points to the same memory area.
Ther= e is a possibility that "foo()" may return different value depend= s on the objects members.

Basically foo() mak= es a lot of calculations depends on no and data loaded in objects constructor. That data wil= l not change, so i may expect that = Get<4>() will return the same value each time without repeating c= alculations.

So I want to get this= straight.

You have a constexpr value. And you're going to do so= me computation on it which is based on the runtime state of an object. And = you want to cache this value. So... why not actually do that? There = are ways to cache values based on an object.

Furthermore, such a cac= he would (presumably) be global. And therefore, if you happen to create two= separate `CA` instances which happen to have the same data, you'll sti= ll get the cached value. The "keyword" idea you're talking ab= out would recalculate the value in that case, since they are two separate i= nstances.

Plus, by using an explicit cache, you don't have to re= strict yourself to immutable data. That is, your way only works if t= he dependent state never changes. With a real cache, it would work in cases= other than yours.

Why make this a language feature?

--
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/4630771e-c4d2-41f2-b133-d4272c37bb81= %40isocpp.org.
------=_Part_1465_1368959221.1474430555617-- ------=_Part_1464_488597594.1474430555616-- .