Re: dealing with new-lines

Thomas A. Fine <fine@cis.ohio-state.edu>
Date: Fri, 8 Jan 93 15:52:19 -0500
From: Thomas A. Fine <fine@cis.ohio-state.edu>
Message-id: <9301082052.AA14305@soccer.cis.ohio-state.edu>
To: connolly@pixel.convex.com, timbl@nxoc01.cern.ch
Subject: Re: dealing with new-lines 
Cc: www-talk@nxoc01.cern.ch
X-Mailer: Perl Mail System v1.1

>> Beacuse
>>each XMP section is like a black box, any white spce inside it would not be  
>>seen by the white space management logic which overlaps the white spaces  
>>required around successive paragraphs, and extra white spce would result.
>
>Unfortunately, this conflicts with the SGML standard. An SGML parser
>will report
><XMP>
>foo
></XMP>
>
>as
>an XMP start tag
>the string "foo" [NOT "\nfoo\n"]
>an XMP end tag
>
>And so it is, strictly speaking, illegal to treat the above markup
>differently from
><XMP>foo</XMP>
>
>I'm not sure how we should resolve this.

This was almost the exact example I was playing with when I decided to start
this messy discussion.  Also some paragraph examples.  Both of the following
should mean the same:

some text<P>more text

and

some text
<P>
more text

The latter being commonly used.  I realized that unless you could remove
surround new-lines in cases like this, the second paragraph would start
with a space.  Similarly, you might find headings, or other things with
a space at the beginning.

>>	PRE
>>
>>By the way, I think we agreed (I gave in) that the <PRE> sections would have
>>siugnificant newlines.  Your manuals, Tom, have <p> as well as newlines, which
>>gives double spacing on my browsers.  So I tread newlines as newlines in all
>>the <PRE> element just as XMP.
>
>This is a situation we need to resolve, one way or another. I'm inclined
>to decide that <p> elements are allowed in PRE elements, but they have
>no significance. They were put there to cause the old linemode browser
>to break the line, so they have always been redundant. So Tim's treatment
>of newlines in PRE is correct, but his treatment of <P> there is not
>[by this new convention.]

Note that the method I proposed has the advantage of automatically dealing
with this (sort of).  If the document was converted simply by adding a <P>
to the end of each line, then for such lines, I remove the newline but keep
the <P>.  Then the code that deals with PRE sections can happily honor
both new-lines and <P> tags, thus making it compatible with either style.

	tom