Dear All:
Just this afternoon, chapters which were displaying perfectly this morning in Live preview on VS now show empty <note>s--Here's the curious part.--when I view only that chapter in Division preview. If I instead view main.ptx in Division preview, then the <note> text populates correctly, and all looks perfect. The logs show no errors. The web and PDF output behave perfectly.
I have tracked it down to the following curiosity. (Offending code shown below.). In the first paragraph of the note is an xref pointing at the figure below with matching xml:id. If I change the xref to point to a non-existent figure, then all displays correctly in Live Preview.
<note xml:id="bridges-of-konigsberg-problem">
<title>
Bridges of Königsberg Problem
</title>
<p>
In the city of Königsberg are seven bridges which cross the Pregolya River. At one point, the Pregolya splits into two branches which flow around two islands before rejoining to form one river.<fn>The Susquehanna River splits into two branches just to the east of Lock Haven. The two branches form Great Island<mdash />a large island with houses, farms, and roads.</fn> The river partitions the city into four pieces of land; the seven bridges connect the four pieces of land as shown in the sixteenth-century drawing in <xref ref="konigsberg-drawing" />. Is it possible to walk a path through Königsberg which allows a person to cross each of these bridges exactly once?
</p>
<p>
A more challenging version of the problem adds the requirement that the walk must begin and end on the same piece of land.
</p>
</note>
<figure xml:id="konigsberg-drawing">
<caption>
domain, via Wikimedia Commons</url>)
</caption>
<image source="BridgesKonigsberg-color.jpg" width="85%" xml:id="konigsberg-drawing-a">
<shortdescription>Drawing of the seven bridges of Königsberg</shortdescription>
</image>
</figure>
Clearly, I'm doing something wrong, but this wasn't happening this morning.
I did upgrade node.js, but it's still happening.
I would be grateful to anyone who can spot what I'm doing incorrectly. (I don't wish to become an annoyance to this group as Fermat was to Wallis. You may, though, refer to me as 'that damned Pennsylvanian', if you wish.)
With much gratitude,
Charles