Hi Pablo, On Mon, Nov 18, 2019 at 03:07:18PM +0100, Pablo Neira Ayuso wrote: > From: Phil Sutter <phil@xxxxxx> > > With set references being recognized by symbol_expr and that being part > of primary_expr as well as primary_rhs_expr, they could basically occur > anywhere while in fact they are allowed only in quite a few spots. > > Untangle things a bit by introducing set_ref_expr and adding that only > in places where it is needed to pass testsuites. > > Make sure we still allow to define variables as set references, eg. > > define xyz = @setref > > And allow to use them from set expressions and statements. > > Signed-off-by: Phil Sutter <phil@xxxxxx> > Signed-off-by: Pablo Neira Ayuso <pablo@xxxxxxxxxxxxx> > --- > v2: Extend Phil's original patch to support for set reference from > variable definitions. We have no test to cover this, I will follow > up with a patch for this. That's great! I tried to allow for that but struggled with reduce/reduce conflicts and ultimately gave up. Glad you found a solution. Thanks, Phil
