> It may be that the TEI documentation gives that impression, but
> it is not so. This is perfectly legal TEI syntax.
Ahhh... no. With all due respect, this is invalid TEI because of the
naked <p>. Proof appended. Christian is correct, in TEI a <div> can
have a series of <p> elements (or similar component-level stuff) as
children followed by a series of <div> (or <divGen>) elements as
children (either, but not both, series can have 0 members), but once
there has been a <div>, everything following must be <div>s (or
global stuff). I like to say that, except for initial <p>s, <div>s
must tessellate a <div>.
I will try to post a more detailed explanation of the content of
<div> within the next few days.
But that's not what Christian asked, he asked "why", a question to
which I do not have an answer right now; I might after 8 hours sleep,
but then again I might not.
--------- begin test file instance ---------
<?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE TEI.2 PUBLIC "-//TEI//DTD TEI Lite P4 XML 2002-05//EN" "/opt/local/sgml/TEI/P4/teixlite.dtd">
<p>Some stuff that does not belong to the previous diff and not to
the following div</p>
<div><p>some more stuff</p></div>
--------- end test file instance ---------
--------- begin nsgmls output [whitespace added for readability] ---------
nsgmls:t.xml:21:3:E: document type does not allow element "p" here;
missing one of "epigraph", "div" start-tag
--------- end nsgmls output ---------
--------- begin xmlparse output [warnings deleted, whitespace added] ---------
error (1152): element violates enclosing tag's content model, /tmp/t.xml, line 22:
p (expecting: divGen, div, pb, milestone, lb,
interpGrp, interp, index, gap, anchor, trailer,
signed, salute, epigraph, dateline, closer, byline)
--------- end xmlparse output ---------