Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!swrinde!mips!daver!tscs!tct!chip
From: chip@tct.uucp (Chip Salzenberg)
Newsgroups: comp.lang.c++
Subject: Re: "foo operator+()" vs "foo& operator+()"
Message-ID: <2753DDF4.7359@tct.uucp>
Date: 28 Nov 90 15:55:32 GMT
References: <11782@hubcap.clemson.edu> <1990Nov22.001208.26092@athena.mit.edu> <rae.659300778@earth>
Organization: Teltronics/TCT, Sarasota, FL
Lines: 21

According to Reid Ellis <rae@gpu.utcs.toronto.edu>:
>	foo & foo::operator+=(const foo &f) { ... }
>
>	foo & foo::operator+(const foo &f)
>	{
>		static foo sf;
>
>		sf = *this;
>		return sf += f;
>	}
>
>The only way to get "caught" using this is if you hold
>the return value in a reference, as in:
>
>	foo &rf = foo1 + foo2;

I believe that:

    foo f = (foo1 + (foo2 + foo3));

is likely to break using the reference-to-static trick.