On Tue, May 8, 2012 at 12:15 PM, Brian Manthos <brianman@microsoft.com> wrote: > On multiple occasions, I’ve run into spec concerns related to not indicating > directly in the grammar that negative values are not allowed. > > Shouldn’t there be tokens such as <positive-integer> that can be used > directly rather than using <integer> and then relying on prose to say zero > and negative are invalid? Since this isn't a problem with what's already in the draft, and adding it would either require some careful (and verbose) manipulation of calc() and attr(), or more careful definition so that we can avoid the verbosity, we'd like to defer this to level 4. We definitely agree that it's a good idea, though, for making the intent of property grammars clearer. ~TJ and fantasaiReceived on Tuesday, 22 May 2012 22:31:12 UTC
This archive was generated by hypermail 2.4.0 : Monday, 23 January 2023 02:14:15 UTC