Re: para termination

On Fri, Nov 23, 2012 at 10:53 AM, Dave Pawson <dave.pawson@gmail.com> wrote:

> On 23 November 2012 15:06, Dave Pawson <dave.pawson@gmail.com> wrote:
> > Given
> > text\n\s\s\s\n
> > more stuff
> >
> >
> > where \s is a space character, \n a newline.
> >
> > Where does the para containing 'text' terminate? Is 'more stuff' a
> > part of that para?
> >
> > Tested at http://johnmacfarlane.net/babelmark2/?text=para%0A++++%0Apara
>
> Proposal, for the core profile.
>
> \s+\n|\EOF be ignored  that is any sequence of white space followed by
> newline or EOF.
>
> I.e. A blank line on its own is normalized to nothing.
>

I still don't think I agree. First, this should cover all whitespace
characters, not just spaces. Second, I think it should be normalized to
just a newline.

In my opinion, the para definition should be:

0 to 3 space characters, arbitrary text, then EOF or (a newline character
followed by a blank line).

and "blank line" should be defined as

Any number of whitespace characters followed by a newline character or EOF.

This preserves Gruber's intent ("A blank line is any line that looks like a
blank line"), which I think is important to the readability/usability of
the markdown source. The definition of "blank line" will also affect the
definition of lists and embedded HTML.

Ryan