220 15708 article Path: news.gmane.org!not-for-mail From: Nevin Liber Newsgroups: gmane.comp.lang.c++.isocpp.proposals Subject: Fwd: vector vs vector> problems Date: Fri, 16 Jan 2015 13:10:49 -0600 Lines: 200 Approved: news@gmane.org Message-ID: References: Reply-To: std-proposals@isocpp.org NNTP-Posting-Host: plane.gmane.org Mime-Version: 1.0 Content-Type: multipart/alternative; boundary=089e0158b50c70b6f5050cc9bde0 X-Trace: ger.gmane.org 1421435494 23211 80.91.229.3 (16 Jan 2015 19:11:34 GMT) X-Complaints-To: usenet@ger.gmane.org NNTP-Posting-Date: Fri, 16 Jan 2015 19:11:34 +0000 (UTC) To: "std-proposals@isocpp.org" Original-X-From: std-proposals+bncBCE35H5S6IDBBYWE4WSQKGQEUCTL7VY@isocpp.org Fri Jan 16 20:11:34 2015 Return-path: Envelope-to: gclcip-std-proposals@m.gmane.org Original-Received: from mail-ob0-f198.google.com ([209.85.214.198]) by plane.gmane.org with esmtp (Exim 4.69) (envelope-from ) id 1YCCIu-0003Zy-1f for gclcip-std-proposals@m.gmane.org; Fri, 16 Jan 2015 20:11:32 +0100 Original-Received: by mail-ob0-f198.google.com with SMTP id vb8sf119544892obc.1 for ; Fri, 16 Jan 2015 11:11:31 -0800 (PST) X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20130820; h=x-gm-message-state:mime-version:sender:in-reply-to:references:from :date:message-id:subject:to:content-type:x-original-sender :x-original-authentication-results:reply-to:precedence:mailing-list :list-id:list-post:list-help:list-archive:list-subscribe :list-unsubscribe; bh=yWrJs9oDsVdu/7u1DxioKcD5q2c5XKCc1giPEq+XlzQ=; b=AZvWIQV6evJpdfx6UDWUMaQ1C8aRZyQg1Aew7JUxvSNHI0L0dTsbw5kwzNPIiLYkRS xbEhpXGaa6osQdM/n9iFHN+BLodoUTGiG1Sz/lF9OJFSy62JEqPkW5QgexrbrEaXVCyg +mn/Zd1nL/32iBnwjSCBdb2na7J9XJVJLxn1t4v1KsOjcQeld7GTtNle1OxtypYnFzWM QhxXInUCOvSsLBwyhg4vsauBLcnaoX3zu7ZGg2n54NmH8gDXsLs3K9VbuRsMQtrNFecr UL09+u3gEFG3HNjHBpq5nJDmYv7/Fd39rz/YM5twVeAlq9RrFT33SZeqBOPj/rxMHd7k frsw== X-Gm-Message-State: ALoCoQnhZAf4rAfrXVEU7uVhDqzf4N11mgse0FC4Z95pDDh11DIRnX0AxFi4Z1audjy2ljjlwaRK X-Received: by 10.42.22.145 with SMTP id o17mr13990690icb.17.1421435491094; Fri, 16 Jan 2015 11:11:31 -0800 (PST) X-BeenThere: std-proposals@isocpp.org Original-Received: by 10.140.106.201 with SMTP id e67ls1512272qgf.63.gmail; Fri, 16 Jan 2015 11:11:30 -0800 (PST) X-Received: by 10.229.70.133 with SMTP id d5mr27937757qcj.2.1421435490141; Fri, 16 Jan 2015 11:11:30 -0800 (PST) Original-Received: from mail-yk0-x22b.google.com (mail-yk0-x22b.google.com. [2607:f8b0:4002:c07::22b]) by mx.google.com with ESMTPS id a11si7071208qge.86.2015.01.16.11.11.29 for (version=TLSv1 cipher=ECDHE-RSA-RC4-SHA bits=128/128); Fri, 16 Jan 2015 11:11:29 -0800 (PST) Received-SPF: pass (google.com: domain of nliber@gmail.com designates 2607:f8b0:4002:c07::22b as permitted sender) client-ip=2607:f8b0:4002:c07::22b; Original-Received: by mail-yk0-f171.google.com with SMTP id 142so10278995ykq.2 for ; Fri, 16 Jan 2015 11:11:29 -0800 (PST) X-Received: by 10.236.29.17 with SMTP id h17mr10635147yha.188.1421435489791; Fri, 16 Jan 2015 11:11:29 -0800 (PST) Original-Sender: nliber@gmail.com Original-Received: by 10.170.85.8 with HTTP; Fri, 16 Jan 2015 11:10:49 -0800 (PST) In-Reply-To: X-Original-Sender: nevin@eviloverlord.com X-Original-Authentication-Results: mx.google.com; spf=pass (google.com: domain of nliber@gmail.com designates 2607:f8b0:4002:c07::22b as permitted sender) smtp.mail=nliber@gmail.com; dkim=pass header.i=@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:15708 Archived-At: --089e0158b50c70b6f5050cc9bde0 Content-Type: text/plain; charset=UTF-8 On 14 January 2015 at 17:55, Matthew Fioravante wrote: > We have unique_ptr and vector> but there is one big > use case where both are inadequate. > > > struct FooSlow { > std::vector getSubset(int i) const { return { _v.begin() + i, _v.end > () }; } > std::vector> _v; > }; > > struct FooFast { > carray_view getSubset(int i) const { return { _v.begin() + 1, _v.end > () }; } > std::vector _v; > > ~FooFast() { while(!_v.empty()) { delete v.back(); v.pop_back(); } } > }; > > > Here we have a linear array of pointers in our Foo object and want to > return a view to a subsection of the array. With vector>, we > are forced to create a copy because of the type system. > > With vector, we can directly return a view referencing a subset but > now we have to take care of copy(), copy=, move(), move=, and the > destructor. > Given that you said this is a big use case already, how are you dealing with copying FooSlow objects? If you really want to do something like this, why can't you just use vector>? You at least get the pointer syntax with any range you return. > Another alternative would be to have some guarantee in the standard that > reinterpret_cast(std::unique_ptr()) is valid. > Seems extremely unlikely C++ will move in this direction. I don't see why this case is different than any other case where you have a vector and want to return a range. I don't see that the unique_ptr -> T* case is all that special (or even common). -- Nevin ":-)" Liber (847) 691-1404 -- --- 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/. --089e0158b50c70b6f5050cc9bde0 Content-Type: text/html; charset=UTF-8 Content-Transfer-Encoding: quoted-printable On 14 January 2015 at 17:55, Matthew Fioravante <fm= atthew5876@gmail.com> wrote:
=C2=A0We have unique_ptr<T[]> and vector<unique_ptr<T>>= ; but there is one big use case where both are inadequate.
=C2=A0=
=C2=A0
struct FooS= low { =C2=A0 std= ::vector<T*> getSubset(int i)= const { return { _v.begin() + i, _v.end() }; } =C2=A0 std::vector<std::unique_ptr<= T>> _v; = }; struct<= span style=3D"color:rgb(0,0,0)"> FooFast { =C2=A0 carr= ay_view<T*> getSubset(int i)<= /span> const { return { _v.begin() + 1, _v.end() }; } =C2=A0 std:= :vector<T*>= _v; =C2=A0 ~FooFast() { while(!_v.empty()) { delete v.back(); v.pop_back(); }= } };

H= ere we have a linear array of pointers in our Foo object and want to return= a view to a subsection of the array. With vector<unique_ptr<T>>= ;, we are forced to create a copy because of the type system.
=C2= =A0
With vector<T*>, we can directly return a view referenc= ing a subset but now we have to=C2=A0take=C2=A0care of=C2=A0copy(), copy=3D= , move(), move=3D, and the destructor.

Given that you said this is a big use case already, how are you dea= ling with copying FooSlow objects?

If you really want to = do something like this, why can't you just use vector<shared_ptr<= T>>?=C2=A0 You at least get the pointer syntax with any range you ret= urn.
=C2=A0
=C2=A0 Another alternative would be to have some guarantee in the = standard that reinterpret_cast<T*>(std::unique_ptr<T>()) is val= id.

Seems extremely unlikely C+= + will move in this direction.

I don't see why this c= ase is different than any other case where you have a vector<A> and w= ant to return a range<B>.=C2=A0 I don't see that the unique_ptr&l= t;T> -> T* case is all that special (or even common).
= --
=C2=A0Nevin ":-)" Liber=C2= =A0 <mailto:= nevin@eviloverlord.com>=C2=A0 (847) 691-1404

--

---
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>.
--089e0158b50c70b6f5050cc9bde0-- .