On Fri, May 31, 2013 at 3:56 AM, Markus Lanthaler <markus.lanthaler@gmx.net>wrote: > Gavin, > > This is not entirely correct, I think. > > On Friday, May 31, 2013 3:48 AM, Gavin Carothers wrote: > > @base <//example.org/> . # Creates a new base with either http or > > https depending on the current document base > > @base <http://example.org/pointless> . # Creates the base > > URI http://example.org/ > > No, this would create the base http://example.org/pointless, i.e., > <#frag> later in that document would be expanded to > http://example.org/pointless#frag > Err... indeed. Not sure what I was thinking yesterday, I think I had spent too long dealing with path switching in history.js yesterday. > > > > @base <> . # Does nothing. > > @base <b/> . # Adds b/ to the current base > > More or less. http://example.com/a would be changed to > http://example.com/b/ > Yep, should have just done all the examples showing a retaliative reference being resolved with a given base. :\ > > > > @base <#> . # Still does nothing > > @base <http://example.org/a/b#> . # Creates the base URI > > http://example.org/a/ note the lack of a 'b' > > Same as above. The new base would be http://example.org/a/b > ... sigh... points for consistency in screwing up writing examples? <c> would resolve to <http://example.org/a/c> is where I was going with that, but <#c> would resolve to <http://example.org/a/b#c> Cheers, Gavin > > > Cheers, > Markus > > > -- > Markus Lanthaler > @markuslanthaler > >
Received on Friday, 31 May 2013 15:13:16 UTC