From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 32161 invoked by alias); 28 Feb 2004 14:52:22 -0000 Mailing-List: contact gcc-bugs-help@gcc.gnu.org; run by ezmlm Precedence: bulk List-Archive: List-Post: List-Help: Sender: gcc-bugs-owner@gcc.gnu.org Received: (qmail 32150 invoked by uid 48); 28 Feb 2004 14:52:21 -0000 Date: Sat, 28 Feb 2004 14:52:00 -0000 Message-ID: <20040228145221.32149.qmail@sources.redhat.com> From: "giovannibajo at libero dot it" To: gcc-bugs@gcc.gnu.org In-Reply-To: <20040213140825.14140.christian.engstrom@glindra.org> References: <20040213140825.14140.christian.engstrom@glindra.org> Reply-To: gcc-bugzilla@gcc.gnu.org Subject: [Bug c++/14140] Unwanted call to derived constructor after implicit conversion X-Bugzilla-Reason: CC X-SW-Source: 2004-02/txt/msg02711.txt.bz2 List-Id: ------- Additional Comments From giovannibajo at libero dot it 2004-02-28 14:52 ------- I don't see any reason why the derived copy constructor should be called. I also double-checked with MSVC71 and Comeau 4.3.3 (which uses the latest EDG), and they both agree that only the base constructor should be called. It looks like the conversion operator is responsible for this behaviour, and calls the copy constructor withour realizing the operator* returns the same type. Can anybody check if this is a regression? I don't have anything older than 3.2 (and yes, I should really rebuild those). -- What |Removed |Added ---------------------------------------------------------------------------- Known to fail| |3.3.3 3.2.3 3.4.0 3.5.0 http://gcc.gnu.org/bugzilla/show_bug.cgi?id=14140