Re: [PATCH] -D, --define.

[Joel E. Denny]

On Tue, 11 Nov 2008, Akim Demaille wrote:

> I was not referring to locations displayed by Bison, but of locations left as
> synclines.

Oh right, you said that in your first report, but I forgot.

> Maybe that's a case where we should not issue the #lines at all.

That makes sense to me.  In the eyes of the user, Bison-specified or 
skeleton-specified default values fall in the generated portion of 
parser.tab.c just like any other code from Bison or from a skeleton.

Perhaps the syncline macro could detect this case by looking for a 
negative line number.