On Tue, 3 Jan 2006, Akim Demaille wrote:

> >>> "Joel" == Joel E Denny <[EMAIL PROTECTED]> writes:
> 
>  > On Mon, 2 Jan 2006, Akim Demaille wrote:
>  >> Installed.
> 
>  > Thanks.
> 
>  >> For some reason, glr.c sometimes uses yydefact[], and at others
>  >> yydefaultAction.
> 
>  > I don't know why.  I'd prefer it to be consistent.  Even though I'd 
>  > normally vote for the function, yydefact is more consistent with yacc.c.
> 
> Right, but the glr skeleton is much more readable, and I won't prefer
> the converse :)

I can't argue with that.

> The glr skeleton tries to keep some abstraction over
> how the automaton is encoded.  Its yygetLRActions for instance is
> something that would clarify yacc.c.

So, since we're looking at it, maybe yacc.c should have a yydefaultAction?  
One small step for abstraction.

Joel


Reply via email to