Subject: Re: duplicate members fix on Solaris Netatalk
From: Bill Studenmund (skippy@macro.stanford.edu)
Date: Sat Jun 21 1997 - 19:25:46 EDT
On Sat, 21 Jun 1997, Dejan Ilic wrote:
> On Fri, 20 Jun 1997, Bill Studenmund wrote:
>
> > Shouldn't it actually be "BYTE_ORDER == BIG_ENDIAN" becomes " == " which
> > is a syntax error???
>
> Well, GCC 2.7.2.2 passes that line as no error has happend. Maybe another
> compiler would have triggered a error there but not GCC. I dont have
> Sun-CC to check what happends on another compiler.
Weird.
> > Also, I thought that all the include files were arranged so that endian.h
> > got included anyway. I.E. if a file includes ddp.h, it should be including
> > endian.h anyway.
>
> As you can see it doesn't, at least no the "dpp.h" in solaris directory. I
> complained about the existance of several files with the same name in
> different directorys. It is maybe correct in another directory.
No. ddp.h isn't supposed to include endian.h. Any program file which
includes ddp.h is already supposed to be including endian.h.
I did a search on the files in libatalk, and of the five files which
include atalk/ddp.h, only one (libatalk/atp/atp_packet.c) doesn't also
include atalk/endian.h.
In bin, bin/aecho/aecho.c is fine (includes both). In /etc, only files in
atalkd include ddp.h. Of them, main.c, rtmp.c, and zip.c are fine, but
aep.c and nbp.c don't also include endian.h.
So rather than change ddp.h, I think the better fix is for all files which
include ddp.h to also be including endian.h. That means adding an include
line to three files. :-)
Take care,
Bill
This archive was generated by hypermail 2b28 : Sat Dec 18 1999 - 16:25:07 EST