[Kde-accessibility] kttsd using speech-dispatcher

[Jeremy Whiting]

Gary,

I hope all is well with you.  As you may or may not know I've been working 
toward getting kttsd to use speech-dispatcher exclusively for it's speech 
synthesis and have a couple of questions for you.  I'm not sure if you are 
allowed to give advice in this regard, if not please let me know and I'll 
continue to get help from other KDE devs.

I've been porting ktts using the speech-dispatcher C API and it has been 
working alright so far.  I removed all the plugins completely, and combined 
speechdata and speaker classes into one class, made it a singleton, and it 
uses the C API to talk to speech-dispatcher.  I've commented out the filtering 
for now, as I'm not sure how that will work.  speech-dispatcher gives us a job 
number when we tell it to say something, so I'm wondering about possibly 
having our joblists in kttsd not tied directly to those in speech-dispatcher, 
so one SpeechJob can send each sentence as a separate say request.  This would 
mean we'd need to handle priority of jobs ourselves also though, which speech-
dispatcher already handles internally for all requests it receives.

So, I think I've got 2 options.
1. continue to make SpeechJob's directly correlate to speech-dispatcher jobs 
and do the filtering synchronously as they come in to kttsd via the dbus 
connection.

2. make SpeechJob's keep a speech-dispatcher jobnum internally, and keep a 
mapping of speech-dispatcher job numbers to our speechjob numbers to use when 
notifications come from speech-dispatcher about it's job's state changes.

Any advice about either option?

thanks,
Jeremy