Path: utzoo!utgpu!news-server.csri.toronto.edu!rpi!zaphod.mps.ohio-state.edu!wuarchive!emory!wa4mei!nanovx!msa3b!kevin
From: kevin@msa3b.UUCP (Kevin P. Kleinfelter)
Newsgroups: comp.lang.c++
Subject: Re: Public vs Private header files in C++
Message-ID: <1667@msa3b.UUCP>
Date: 9 Jun 91 18:57:20 GMT
References: <5243@servax0.essex.ac.uk> <25231@well.sf.ca.us> <WARSAW.91Jun5121952@warsaw.nlm.nih.gov>
Organization: Dun and Bradstreet Software, Inc., Atlanta, GA
Lines: 38

warsaw@nlm.nih.gov (Barry A. Warsaw) writes:


>>>>>> "Alfred" == Alfred Fontes <al@well.sf.ca.us> writes:

>	Alfred> ... If you have inlines, they can immediately follow
>	Alfred> the end of the class definition.

>This is a Really Good Suggestion.  I'll second that, as this is the
>C++ coding style we've adopted and have been using for several months.
>...

>// foo.h
>class foo
>{
> ...
>inline int foo::number( void )
>{
>	return( priv_number );
>};

Oddly enough, this can produce multiple definitions of 
	int foo::number (void);
with BC++, if you have "out of line inline functions" enabled.
It will literally create a copy of the function in each module compiled.
The solution I found was to turn off this option, or to use
	static inline int foo::number (void) ...

Is the need to specify static a compiler anomaly or a language ambiguity?

Of course, if you want it to be a static function (referencing no class
members), you get something like:
	static inline static int foo::number(void);

-- 
Kevin Kleinfelter @ DBS, Inc (404) 239-2347   ...gatech!nanoVX!msa3b!kevin

Dun&Bradstreet Software, 3445 Peachtree Rd, NE, Atlanta GA 30326-1276