On 12/06/2019 11:25, Juergen Schoenwaelder wrote: > That said, they do seem to declare something like > oc-ext:regexp-posix; but it would have been much smarter to use for > example oc-posix:regex instead of changing the semantics of the > pattern statement.
Yes, this would be preferable, but it has a downside: now you need to carry the patterns twice and must make sure they are in sync. If they are not, all sorts of implementation-dependent breakage would ensue. The differences are far from minor where specification is concerned, but they certainly minor in the vast majoritiy of the use cases and most users do not even understand REs enough to know they are dealing with a maze of twisty flavors, all alike. Do we have an easy way of understanding the corpus of pattern arguments being 'out there' in real use? Can we engineer a workable solution for the general case without getting everyone know the differences in RE engines? Something along the lines of this: https://github.com/openconfig/public/issues/44#issuecomment-501629497 perhaps? Thanks, Robert
signature.asc
Description: OpenPGP digital signature
_______________________________________________ netmod mailing list [email protected] https://www.ietf.org/mailman/listinfo/netmod
