Re: "special" spaces in Texinfo parsing and output

[Patrice Dumas]

On Tue, Mar 26, 2013 at 09:19:03PM +0000, Karl Berry wrote:
> (Switching to texinfo-devel.)
> 
> The form feed itself should not, it seems to me, necessarily start a
> paragraph.  It would be strange to output
> <p>^L</p>
> but I suppose if that is what is natural to implement, then ok.
> (I'm not sure form feeds make sense in HTML anyway, but that's a
> different question.)

Ok.  This is in line with the current handling of form feed in the
parser.

> Then the blank line before BB definitely ends the form-feed paragraph,
> if one had been started.  Thus, we end up with (HTML for example):
> <p>AA</p>
> ^L
> <p>BB</p>

This should be what happens.

> A more problematic case is if there are no blank lines in the input:
> 
> AA
> ^L
> BB
> 
> But I think the output, with ^L specifically, should be the same.  In
> other words, ^L acts like a blank input line, ending the current
> paragraph.  Thus, to switch to Info output, AA and BB would receive
> normal paragraph indentation.  As in:
> 
>   AA
> ^L
>   BB

Ok, this should be the current output (except that ^L is removed and
replaced by a simple end of line, but I will correct that).

> Finally, if instead of a ^L character, one of the other Unicode space
> characters is used, I don't think the behavior should be the same.
> Instead, the Unicode spaces should either be treated like SPC, or like
> regular non-space characters.  Per the other msg and my other reply,
> treating them as non-space is fine with me.

Would it be ok to treat them as spaces?


Another question.  What happens if there is a form feed within a
paragraph, such as

  Blah^Lblam.

Or
  Blah ^L blam.

-- 
Pat