On Tue, Jul 28, 2009 at 12:04 PM, Marcin Hanclik<[email protected]> wrote: > Hi Marcos, > >>>Unless it's broken (?), I would prefer to leave it as is. > I think that P&C should either change the grammar (as suggested in my email) > or specify that the zip-rel-path operates on characters. > The first would be well adjusted to Zip filename encoding in UTF-8 as a kind > of natural match, whereas the latter could be clearer wrt the characters that > can be used in the path. > > Anyway, I think some clarification - e.g. on of the above - is required. > > Thanks.
Ok, I see what you mean now. I made the following change in the spec: 1. deleted the P&C's ABNF definition of utf8-char. 2. added "The utf8-char rule is defined in section 4 of [UTF-8]." Kind regards, Marcos -- Marcos Caceres http://datadriven.com.au
