- From: <bugzilla@jessica.w3.org>
- Date: Wed, 05 Jun 2013 12:25:30 +0000
- To: public-qt-comments@w3.org
https://www.w3.org/Bugs/Public/show_bug.cgi?id=19597 --- Comment #25 from Jonathan Robie <jonathan.robie@gmail.com> --- (In reply to comment #23) > (In reply to comment #22) > > However, I also think that require/prohibit options should have been moved > > to the very top of the query prolog. It feels irritating to me that a > > statement can be placed in a query, and will later be prohibited. > > I completely agree with this and pointed it out some time ago, however it's > not easy to fix because of the handling of namespace declarations in the > prolog. Yes, if this were easy to do with our grammar, we would have done so. But I think this is only irritating for implementers, users will not be irritated by getting the prohibit-feature error. In any implementation that does static analysis first, the static error for prohibit-feature will be raised before the dynamic-error ever occurs. In an implementation that at least parses the query first, it's easy enough to detect the prohibited features before evaluation. -- You are receiving this mail because: You are the QA Contact for the bug.
Received on Wednesday, 5 June 2013 12:25:33 UTC