David Carlisle scripsit:
> And you can even do that if you specify an ibm encoding and the
> parser does the "obvious" thing with NEL and maps it to #10
> (whilst it is arguable that the mapping for NEL should be 85 which then
> wouldn't work without xml 1.1 , this is dubious, it is clearly less
> useful and it seems unnatural, one assumes that other text transfers
> (eg ftp) between ibm and the ascii world map line to line ends, so it
> see,s unreasonable that the mapping for xml should map ibm new lines to
> #85 which isn't a line end in any ascii/unicode system and is just
> presumably there to allow some kind of round tripping.
And what is to be done with ASCII-encoded IBM plain text, which actually
uses 85 to represent #x85?
--
John Cowan http://www.ccil.org/~cowan
Please leave your values | Check your assumptions. In fact,
at the front desk. | check your assumptions at the door.
--sign in Paris hotel | --Miles Vorkosigan
