Re: Bug in mac speech server

[Tyler Spivey <tspivey@xxxxxxxxxxx>]

That worked. Next one.
As far as I understand, s should silence the speech and clear the queue,
and d should speak any queued items.

I have a case where the server is skipping text. Run the attached
script, and press enter to get it speaking. While the first string is
speaking, press enter again.
At this point, I would expect to hear the first string over again, since
I silenced and added it to the queue. Instead, I only hear the second
one. If I let both of them finish, I get both of them again.

On 7/24/2015 10:49 AM, Haden Pike wrote:
> In short, the braces around the argument were being sent to the command, which caused it to fail when trying to cast the argument to an int.
> 
> Long explanation: the server wasn’t checking for an _ when trying to match the command to a form with braces. In effect, this meant the server was assuming the braces were included in the argument. I’ve attached a patch to fix this if Bart or David or Raman if they want it.
> Haden
> 
> 
> 
> 
>> On Jul 24, 2015, at 12:43 AM, Tyler Spivey <tspivey@xxxxxxxxxxx> wrote:
>>
>> As I found out while trying to make yasr work, the following will crash
>> the mac speech server:
>> tts_set_speech_rate {300}
>>
>> -----------------------------------------------------------------------------
>> To unsubscribe from the emacspeak list or change your address on the
>> emacspeak list send mail to "emacspeak-request@xxxxxxxxxxx" with a
>> subject of "unsubscribe" or "help".
>>
> 

import subprocess
p = subprocess.Popen(["python", "mac"], stdin=subprocess.PIPE)
def speak(t):
	p.stdin.write('q {%s}\n' % t)
	p.stdin.write('d\n')

while 1:
	raw_input()
	p.stdin.write("s\n")
	speak("This is line 1.")
	speak('Here is another string.')