[tlbuild] dvibook failure on Cygwin

[Ken Brown]

On 2/25/2012 5:03 PM, Karl Berry wrote:
>      I don't understand why we need unistd.h (and why only on Cygwin)
>
> Of course I can't guess what Cygwin is doing in the system headers, but
> I can't say I'm surprised.  The "dual nature" of kpathsea/getopt.h,
> being both kpathsea-specific but also a system library routine, is
> clearly a source of potential confusion, and in the case of Cygwin, the
> potential has been realized.

Hi Karl,

It turned out not to be so hard to sort out.  It was a matter of making 
sure that getopt gets properly declared (with its prototype).  The 
attached patch fixes the problem.  With this patch, the declaration is 
either in the system getopt.h (if unistd.h is included) or in 
kpathsea/getopt.h.

Ken

P.S. Apparently this has in fact always been broken, but it never got 
noticed because most of the texlive programs that use getopt include 
unistd.h.
-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: getopt.patch
URL: <http://tug.org/pipermail/tlbuild/attachments/20120226/5ce52482/attachment.ksh>