Interpreting t-h0- mechanism

[Matthew Stuckwisch via CLDR-Users]

The T extension for BCP-47 are contained in CLDR and described at length in TR35, so I hope this is the correct forum for the question.  I'm implementing a module for handling language tags and have a few questions relating to the h0 tag:

1. What is the difference between 'en-t-es-h0-hybrid' and 'en-t-h0-es'?  Both styles are given as example encodings for what would be Spanglish (Spanish-English hybrid), but surely there ought to be a difference
2. What would be the interpretation of 'en-t-pt-h0-es'?  English, transformed from Portuñol (Spanish-Portuguese hybrid), or Spanglish, transformed from Portuguese?
3. What would the correct method for indicating, for instance, Portuñol transformed from Franglish (French-English hybrid)?
4. What is the format for a triple language hybrid?  Would Asturian-Portuguese-Spanish be 'ast-t-h0-pt-h0-es'?

TR35 is fairly anemic in the h0 mechanism documentation, and particularly with regards to (4), it seems like much was intentionally left as an open question to be handled at a later date, hence my confusion.

Matéu