Way to go.

[Ralph Little]

Hi,
Further to my question regarding typesetting grobs after timestep stop,
which works fine...

The issue was that I didn't know the text until the end of the measure. 
However, so that I can integrate multiple voices, I am planning to
arrange things so that there is a BrailleStaff and multiple BrailleVoice
contexts.
Each BrailleVoice context will handle separate note sequences as in
Staff and Voice. 
However the final typeset version is combined in textual form in a
different sequence from the way that it was processed.

For example,

\notes {
        {c d e f \\ f e d c}
}

.. actually comes out as the Braille equivalent of "c d e f" <join
sequence> "f e d c".
That's OK, but it is processed in Lilypond in the separate contexts
ordered by timestep as "c f", "d e", "e d" and "f c".

To get around this, I was going to collect the notes for each voice in
the Braille_note_engraver separately in BrailleVoice, then at the end of
the measure, copy these strings to the end of a property in the
BrailleStaff parent context, and get an engraver in BrailleStaff
(Braille_measure_engraver) to actually put the text together for the
measure.

This method sounds a bit mad, but I cannot think of a better way to do
this.

Any ideas for doing this better? 
I guess I'm really talking to Han-Wen or Jan here, but I'm open to any
suggestions.

Regards,
Ralph

---------
Tribal Data Solutions has moved, please visit our website for more details 
http://www.tribaldata.co.uk. 
This e-mail and any attachments are confidential and are sent on the basis of 
our copyright, e-mail and security policy which can be inspected by visiting 
http://www.tribaldata.co.uk/policies.asp.
If you are not the intended recipient, please notify the sender and delete this 
message. Thank you.
-----------