220 33546 article Path: news.gmane.org!.POSTED!not-for-mail From: Edward Catmur Newsgroups: gmane.comp.lang.c++.isocpp.proposals Subject: Re: Re: Catch block for multiple exceptions Date: Wed, 2 Aug 2017 13:05:33 -0700 (PDT) Lines: 202 Approved: news@gmane.org Message-ID: References: <4126837.jZlGhgVnua@tjmaciei-mobl1> Reply-To: std-proposals@isocpp.org NNTP-Posting-Host: blaine.gmane.org Mime-Version: 1.0 Content-Type: multipart/mixed; boundary="----=_Part_6499_559883321.1501704333569" X-Trace: blaine.gmane.org 1501704336 5906 195.159.176.226 (2 Aug 2017 20:05:36 GMT) X-Complaints-To: usenet@blaine.gmane.org NNTP-Posting-Date: Wed, 2 Aug 2017 20:05:36 +0000 (UTC) To: ISO C++ Standard - Future Proposals Original-X-From: std-proposals+bncBDZLZTXF7UJBBDXBRDGAKGQEY7GGB2A@isocpp.org Wed Aug 02 22:05:31 2017 Return-path: Envelope-to: gclcip-std-proposals@m.gmane.org Original-Received: from mail-qt0-f199.google.com ([209.85.216.199]) by blaine.gmane.org with esmtp (Exim 4.84_2) (envelope-from ) id 1dczta-0001DL-8k for gclcip-std-proposals@m.gmane.org; Wed, 02 Aug 2017 22:05:30 +0200 Original-Received: by mail-qt0-f199.google.com with SMTP id u19sf25689548qtc.14 for ; Wed, 02 Aug 2017 13:05:36 -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 :list-post:list-help:list-archive:list-subscribe:list-unsubscribe; bh=p2RytbtSlnY2AEx8365QQUg2pyG8SoyUT+iK1R/kzX8=; b=m1+KIsEe804GDVGNh9CERvg1gRnWB3WGmGhoXs0DBG+a4w97+mFVLxUp6+guNYAJBb 1DpbGVP1/OhqkSL4fdNui9DyWHmR24rhtF70XVp4TAjrVQsScmVip/W4fCVWSexLBMHe diPdY1515uJQ4bhGIC6jlPNqfr/H0F4747PTKAXrb15QXKaG+ykikYrdVWpqnurb8zIi RN9mJ6mdY84ekRQGKetbFXfukdaBzKZSKXZNo+id8sXWBdWTwaKYNYEx7T6hW/IDMiXx E0k7P15Ul/Z64+/nHi55UTklKu1Qt7tE3QoH8SPly4+NSHuhwBRDfevlqLgwTGFQOoem FJxg== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20161025; 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=p2RytbtSlnY2AEx8365QQUg2pyG8SoyUT+iK1R/kzX8=; b=SCtSX3s6Y80GkeUZuCvjgzd2zrGWM+9GfP+hbzB7oKMkkAs+thqN2nOnuqZaqieQz5 IfP/ah+UjN/UvTmPwyvngUmoSsQSRt0v8SqmXaxu/nNNnKivoYyyy3DdLECQDbLYpVuG 7woKFCEVZKebepxML3zyJovH7Q8JNixARWbaMUGfzafPszRE7IlN/qSw9QNidsPUNe2c 5hyEjDP6k+dxXDCXui4KBusjebCJZBEFr/HhEE9HiE16RNNc+e3NkwKkzjMh7P+XKInw QhH98TmyyIFEaJ8v4Cf8RWFC/EILpjekAL04Udmh9iM8QQhYf38GRleMT6pYqGaC2WrJ tmkA== X-Gm-Message-State: AIVw111WGdr//gb5Hari3Ah9XRYzYDKSMjFrFmsMUXN0aZ0U+fIucgBg g9KHGgnIfbluQwTm X-Received: by 10.237.62.245 with SMTP id o50mr16966649qtf.35.1501704335971; Wed, 02 Aug 2017 13:05:35 -0700 (PDT) X-BeenThere: std-proposals@isocpp.org Original-Received: by 10.36.111.9 with SMTP id x9ls131660itb.16.gmail; Wed, 02 Aug 2017 13:05:34 -0700 (PDT) X-Received: by 10.31.151.77 with SMTP id z74mr101408vkd.3.1501704334183; Wed, 02 Aug 2017 13:05:34 -0700 (PDT) In-Reply-To: X-Original-Sender: ed@catmur.co.uk 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:33546 Archived-At: ------=_Part_6499_559883321.1501704333569 Content-Type: multipart/alternative; boundary="----=_Part_6500_271610059.1501704333570" ------=_Part_6500_271610059.1501704333570 Content-Type: text/plain; charset="UTF-8" On Wednesday, 2 August 2017 20:07:03 UTC+1, Ville Voutilainen wrote: > > On 2 August 2017 at 21:47, Nicol Bolas > > wrote: > > However, I think a small library feature could help make it less > painful. > > Right now, we have `current_exception`, which gives us an > `exception_ptr`, > > which is completely type-erased. We could give `exception_ptr` a way to > > un-erase the type. After all, the implementation has to have that type > > somewhere, since we can throw and re-catch it. > > > > So perhaps an `any_cast`-like mechanism, that returns either `nullptr` > or a > > pointer to the exception object. Only it should work as if it were a > > `dynamic_cast`. Then, we could implement an `exception_visitor` that > takes a > > sequence of types and tries them, in order. > > The ability to do an any_cast on an exception_ptr would indeed be > nice. I'll see what I can > do. There's a chance that adding such a beast isn't particularly hard > implementation-wise. > For libstdc++ it's trivial: template target_type* exception_cast(exception_ptr& ep, bool exact = false) { void* p = ep._M_get(); std::typeinfo target_type = typeid(target), exc_type = ep. __cxa_exception_type(); return (exact ? target_type == exc_type : target_type.__do_catch( exc_type, &p, 1)) ? static_cast(p) : nullptr; } I suppose MSVC might require a bit more work though. I remember when I last looked at Windows exception handling (trying to peek at the typeid of an unknown exception, no __cxa_exception_type there) it got hairy pretty fast. -- 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/f85a2e67-4d76-4dac-9e45-a560e3e8872f%40isocpp.org. ------=_Part_6500_271610059.1501704333570 Content-Type: text/html; charset="UTF-8" Content-Transfer-Encoding: quoted-printable
On Wednesday, 2 August 2017 20:07:03 UTC+1, Ville Voutilai= nen wrote:
On 2 August 2017 at= 21:47, Nicol Bolas <jmck...@gmail.com> wrote:
> However, I think a small library feature could help make it less p= ainful.
> Right now, we have `current_exception`, which gives us an `excepti= on_ptr`,
> which is completely type-erased. We could give `exception_ptr` a w= ay to
> un-erase the type. After all, the implementation has to have that = type
> somewhere, since we can throw and re-catch it.
>
> So perhaps an `any_cast`-like mechanism, that returns either `null= ptr` or a
> pointer to the exception object. Only it should work as if it were= a
> `dynamic_cast`. Then, we could implement an `exception_visitor` th= at takes a
> sequence of types and tries them, in order.

The ability to do an any_cast on an exception_ptr would indeed be
nice. I'll see what I can
do. There's a chance that adding such a beast isn't particularl= y hard
implementation-wise.

For libstdc++ it's trivial:
<= div>
template<class target>
target_type* exception_cast(exception_ptr& ep, b= ool exact = =3D= false) {
=C2=A0 =C2=A0 void* p =3D ep<= /span>._M_get();
=C2=A0 =C2=A0 std::typeinfo target_type = =3D typeid(target), exc_type =3D ep.__cxa_= exception_type();=C2=A0 =C2=A0 return (= exact ? target_type =3D=3D exc_type : target_type.__do_catch(exc_type, &p,= 1)) ?
=C2=A0 =C2=A0 =C2=A0 =C2=A0 static_cast<target*>(p) :
=C2=A0 = =C2=A0 =C2=A0 =C2=A0 nullptr;<= br>}

I suppose MSVC might r= equire a bit more work though. I remember when I last looked at Windows exc= eption handling (trying to peek at the typeid of an unknown exception, no _= _cxa_exception_type there) it got hairy pretty fast.

--
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/f85a2e67-4d76-4dac-9e45-a560e3e8872f= %40isocpp.org.
------=_Part_6500_271610059.1501704333570-- ------=_Part_6499_559883321.1501704333569-- .