Re: [w3ctag/design-reviews] WG New Spec: RDF 1.2 N-Triples (Issue #1161)

ylafon left a comment (w3ctag/design-reviews#1161)

Hi @pchampin 
(note that it is a personal comment)
The [PR](https://github.com/w3c/rdf-n-triples/pull/107) is an improvement, but does not really make processing really predictable.

How about designing multiple error types that define if the processing stopped or just ignored what's not known and recovered?
The application would know what really happened at parsing level and act accordingly. There can even be a parameter for the parser so that the application can enforce a specific behaviour if it relies on that. I understand to be as flexible as possible in terms of behaviour, but reliable behaviour in case you change parser or upgrade to RDF 1.3, which was the origin of the discussion about the reuse of the RDF media type, is important for long term interop.

-- 
Reply to this email directly or view it on GitHub:
https://github.com/w3ctag/design-reviews/issues/1161#issuecomment-5477863547
You are receiving this because you are subscribed to this thread.

Message ID: <w3ctag/design-reviews/issues/1161/5477863547@github.com>