On Sat, 29 Aug 2020, Markus Elfring wrote:
> >> @replacement@
> >> identifier array, instance, size;
> >> type T;
> >> @@
> >> (
> >> -sizeof(T)
> >> |
> >> -sizeof(*instance->array)
> >> )
> >> *
> >> +flex_array_size(instance, array,
> >> instance->size
> >> + )
> >
> > This semantic patch will fail if the sizeof(T) option is matched, becuse
> > then it won't be able to create the + code, since it won't know what array
> > should be.
>
> Will this information trigger further consequences for the clarification
> of the topic “[RFC PATCH] coccinelle: api: add flex_array_size.cocci script”?
> https://lore.kernel.org/cocci/20200828163134.496386-1-efre...@linux.com/
> https://systeme.lip6.fr/pipermail/cocci/2020-August/008169.html
As far as I can see, in that semantic patch, array is always inherited
from a previous rule.
julia
_______________________________________________
Cocci mailing list
Cocci@systeme.lip6.fr
https://systeme.lip6.fr/mailman/listinfo/cocci
