Cool, I should be able to do that. I don't have an android phone but others in my household do, and that might give really interesting information.

In the meantime, I have a suspicion about what might have caused it. Looking at my own transaction logs from the session, I see that there was a moment during the first query for track data (the one which succeeded) where my code failed to respond to one of the queries in the group of metadata queries. The conversation looked like this:

Code:
AVRCP 0 GET_ELEMENT_ATTRIBUTES 02 05 04?

AVRCP 0 GET_ELEMENT_ATTRIBUTES 01 01?
AVRCP RSP 1 1 "Sultans of Swing"
AVRCP 0 GET_ELEMENT_ATTRIBUTES 01 03?
AVRCP RSP 1 3 "(Album N/A)"
AVRCP 0 GET_ELEMENT_ATTRIBUTES 01 02?
AVRCP RSP 1 2 "Dire Straits"


The first of those queries is doing a multi-query, it's asking for two items at once:
02 = Number of attributes being queried for
05 = Attribute code for the total number of tracks in the album.
04 = Attribute code for the current track number.

My code doesn't yet know how to respond to a multi-query. It only knows how to respond to single queries where the first number in the query is "01". The Honda stereo only ever sent single queries, never a multi-query.

I think what happened (this is just a hunch that I haven't tested yet) is that since I didn't respond to that first query in a timely fashion, its parent routine for making track metadata queries is hung waiting for a response and it won't ask again. It managed to get that one group of queries but now it won't do it again.

So I guess I have to write the full parser for the "GET_ELEMENT_ATTRIBUTES" responses that I was putting off writing.
_________________________
Tony Fabris