220 15179 <76e7d5fe-9ae9-42c4-ac0f-dbfb38673da0@isocpp.org> article Path: news.gmane.org!not-for-mail From: sasho648 Newsgroups: gmane.comp.lang.c++.isocpp.proposals Subject: Re: Inline variables - superior of templates and constexpr. Date: Sat, 20 Dec 2014 06:43:56 -0800 (PST) Lines: 283 Approved: news@gmane.org Message-ID: <76e7d5fe-9ae9-42c4-ac0f-dbfb38673da0@isocpp.org> References: <84da1bdc-3356-4c40-a29a-bdd4c0b058b4@isocpp.org> Reply-To: std-proposals@isocpp.org NNTP-Posting-Host: plane.gmane.org Mime-Version: 1.0 Content-Type: multipart/mixed; boundary="----=_Part_56_361848433.1419086636219" X-Trace: ger.gmane.org 1419086645 28198 80.91.229.3 (20 Dec 2014 14:44:05 GMT) X-Complaints-To: usenet@ger.gmane.org NNTP-Posting-Date: Sat, 20 Dec 2014 14:44:05 +0000 (UTC) To: std-proposals@isocpp.org Original-X-From: std-proposals+bncBDBNZXHN2QPRBLEW22SAKGQEOAHNI4Y@isocpp.org Sat Dec 20 15:43:59 2014 Return-path: Envelope-to: gclcip-std-proposals@m.gmane.org Original-Received: from mail-vc0-f200.google.com ([209.85.220.200]) by plane.gmane.org with esmtp (Exim 4.69) (envelope-from ) id 1Y2LGA-0005EO-4e for gclcip-std-proposals@m.gmane.org; Sat, 20 Dec 2014 15:43:58 +0100 Original-Received: by mail-vc0-f200.google.com with SMTP id im6sf6784290vcb.11 for ; Sat, 20 Dec 2014 06:43:57 -0800 (PST) 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:content-type:x-original-sender:reply-to :precedence:mailing-list:list-id:list-post:list-help:list-archive :list-subscribe:list-unsubscribe; bh=7aXaOauXlyu3NVR7FDfEIXacMLtyrBxRqFMZQEnfLdY=; b=mWKeEUpd8zk+cZeVf/B1Hde+FMlko2+Q9pDHHErmk9ZfopdVEm+8fmj71LeMayV5Ac Ui3MeRykEzRBKbCV3+wTUDN9r+KgcrIQwtWN39symbkTNrPNFMynswHNFMhFEzjNZu+t Hmg1VgR5bX7QdI8xiztY4lsi5eF5NQ9jux/sO+8cLCauQ3FEB+wNpDjA2DRbGqAu8TV7 /+WmxAUGXx6b0BsUa3K1FQl0ybAMmCEM9/C+Wo6h7rfywy1pBk1z8QV+9Fas/NIphlcI w+Q/YoPHmWrKZUO1Nmvd3g1lO12MNcGrXMgY+aPsIJr3tVJOolbbfb2uXS1p/vAFez8G PD9A== X-Gm-Message-State: ALoCoQlu8N8cPT294eOkLwb/GFpHGneeGXhSOcRILM4tKO6QAtPx0LFgXMTdDRr75hNwNYsbHpOB X-Received: by 10.236.207.136 with SMTP id n8mr11398624yho.5.1419086637145; Sat, 20 Dec 2014 06:43:57 -0800 (PST) X-BeenThere: std-proposals@isocpp.org Original-Received: by 10.140.104.213 with SMTP id a79ls4933449qgf.61.gmail; Sat, 20 Dec 2014 06:43:56 -0800 (PST) X-Received: by 10.140.33.229 with SMTP id j92mr26399qgj.7.1419086636667; Sat, 20 Dec 2014 06:43:56 -0800 (PST) In-Reply-To: <84da1bdc-3356-4c40-a29a-bdd4c0b058b4@isocpp.org> X-Original-Sender: sasho648@mail.bg 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:15179 Archived-At: ------=_Part_56_361848433.1419086636219 Content-Type: multipart/alternative; boundary="----=_Part_57_2060408774.1419086636219" ------=_Part_57_2060408774.1419086636219 Content-Type: text/plain; charset=UTF-8 It wouldn't work in the case because as I mentioned above, inline references will be bound either to another inline reference or to actual object name (not a reference to one of those). So such copy-constructor will be illegal (unless it's parameter is inline reference too): struct S { S(const S &arg) : NumberOfSomething (arg.NumberOfSomething) //error 'arg.NumberOfSomething' doesn't name object (but instead is a reference to such) { } B arr[5]; int & inline NumberOfSomething = arr[3].FieldName; }; However I don't see a reason why would you want to copy such a reference as it's purpose is just to be another name of some data member. You could write the copy-constructor only for the data: struct S { S(const S &arg) { memcpy((void*)&arr, (void*)&arg.arr, sizeof(arr)); } B arr[5]; int & inline NumberOfSomething = arr[3].FieldName; }; Also class members inlines can be only 'const' (inline references doesn't count as they're implicit constant). Some examples of 'inline' references: int *pa; int a; int & inline var_0 = (new S())->NumberOfSomething; //invalid '(new S())->NumberOfSomething' doesn't name an object int & inline var_1 = *pa; //same as above int & inline var_2 = a; //ok 'a' names an object var_2 = 9; //same as 'a = 9' I wanted to add that constructors which initialize 'inline' members, if their value can't be determined at the time of parsing the class which they belong to, and methods which use them are implicit inline and should be executed at compiler time. In such cases the class type is incomplete. In the above example this is not true because the value of 'NumberOfSomething' can be determined at class parsing time. -- --- 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. Visit this group at http://groups.google.com/a/isocpp.org/group/std-proposals/. ------=_Part_57_2060408774.1419086636219 Content-Type: text/html; charset=UTF-8 Content-Transfer-Encoding: quoted-printable
It wouldn't work in the case because as I mentioned above,= inline references will be bound either to another inline reference or to a= ctual object name (not a reference to one of those). So such copy-construct= or will be illegal (unless it's parameter is inline reference too):

=
struct S {
    S(const S &am= p;arg) : NumberOfSomething (arg.N= umberOfSomething) //error 'arg= ..NumberOfSomething' doesn't name object (but instead is a reference to such= )
 = ;   {= }

    B arr<= /span>[5];  
     
=     int &= inline NumberOfSomething =3D arr[3].F= ieldName;<= /span>
= };

However I= don't see a reason why would you want to copy such a reference as it's pur= pose is just to be another name of some data member. You could write t= he copy-constructor only for the data:

st= ruct S {
    S(const S &arg)
    { memcpy((void*)&= arr, (void*)&arg.arr,= sizeof(arr)); }

    B arr[5];  
     
    int & inline NumberOfSomething =3D arr[3].FieldName;
};

Also cl= ass members inlines can be only 'const' (inline references doesn't count as= they're implicit constant).

Some ex= amples of 'inline' references:

int *p= a;<= span style=3D"color: #000;" class=3D"styled-by-prettify">

int a;

int & inline va= r_0 =3D (new S())->NumberOfSomething; //invalid '(new S())->NumberOfSomething' doesn't name an= object
int & inline var_1 =3D *pa= ; <= span style=3D"color: #800;" class=3D"styled-by-prettify">//same as above

int & inline var_2 =3D a;= //ok 'a' = names an object

var_2 =3D 9; //same as 'a =3D 9'

I wanted to add that construc= tors which initialize 'inline' members, if their value can't be determined = at the time of parsing the class which they belong to, and methods whi= ch use them are implicit inline and should be executed at compiler time. In= such cases the class type is incomplete. In the above example this is not = true because the value of 'NumberOfSomething' can be determined at class pa= rsing time.

--

---
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.
Visit this group at http://groups.google.com/a/isocpp.org/group/std-proposals/<= /a>.
------=_Part_57_2060408774.1419086636219-- ------=_Part_56_361848433.1419086636219-- .