Re: glr.c cleanup

[Joel E. Denny]

On Wed, 7 Dec 2005, Paul Eggert wrote:

> "Joel E. Denny" <address@hidden> writes:
> 
> > I've been generating patches with cvs diff -p -u.  Am I missing a
> > flag? Is this the wrong procedure?
> 
> No, that's right.  But there's a mixup somewhere between the time you
> generate that output, and what appears in the emails that Paul
> Hilfinger and I receive.  The lines of the patch that start with a
> space are munged so that they start with two spaces instead.  You can
> see some of the effect here:

You're right.  It's pine.  I've reconfigured.  If it happens again by some 
chance, just let me know and I'll be happy to resubmit.

> > I just need something consistent in order for the patch to work.
> >
> > Let me know what the consensus is
> 
> For glr.c Paul likes the current naming convention (append "p" to
> names of pointers to objects, to disambiguate them from the actual
> objects), so that seems to be the way to go.

Would it be ok to submit all of the following as a single patch?  Each is 
a separate cosmetic change like the rename.  I'll save the real meat of 
the original for the next patch.

        * data/glr.c: Reorder typedef declarations for structs to match order
        of struct declarations.
        (yyglrShift): Change parameter YYSTYPE yysval to YYSTYPE* yyvalp for
        consistency. [Pending Paul's approval.]
        Rename yystack everywhere to yystackp except in yyparse where it's not
        a pointer.

Joel