@Hitch
I feared that there's no such table
May be I would be a good practice of all professionel ePub producers to demonstrate their disrespect against ignorant eReader producers, when they deliver all their work in two files:
one clean, crisp, beautiful coded epub.
and another ugly one for all the bad devices.
To blame them. Not the users of course. But the eReader producer.
When I understand you correctly, than the situation is like the web in the stone age.
"Best viewed with".
I still disagree with you about the sources of joy in professionell work.
And be sure, I have I great respect for all (you of course included) who struggle with ugly technology. Who try their best.
Joy, in my humble opinion, comes out of great layouts, great typografy. And it starts, after a professionell has overcome all the boring defects.
To the "software which interpretes the tree of an XML file":
May be you're right and it will never come to the market. Bad luck for me and for many others.
I'm not at all an expert in estimate the software market situation under that special aspect.
But even a guro of RegEx like Jeffrey Friedel writes, that RegEx is not a good tool for editing html.
I wonder that languages like XLST exist but still there's no tool to perform things in a simple gui like:
Delete all childs of the elemtent foo which contain the class "zot".
Strange.
I say to Santa.