Re: Is there an off-by-1 bug in notification callbacks?

[Nolan Darilek]

Thanks. \r\n.\r\n. is probably for multiline commands. The text I read was:

wfr is either SPD_WAIT_REPLY (integer value of 1) or SPD_NO_REPLY (0). This specifies if you expect to get a reply on the sent data according to SSIP. For example, if you are sending ordinary text inside a SPEAK command, you don’t expect to get a reply, but you expect a reply after sending the final sequence \r\n.\r\n.

"final sequence" is what confused me. Maybe "final sequence of a multiline command" might be a bit clearer? As it stands, I didn't know if I needed to grep for batch commands, multiline commands, etc. when looking through the SSIP docs.

Thanks, I'll see what I can do with this.

On 8/19/20 3:02 PM, Samuel Thibault wrote:

> Nolan Darilek, le mer. 19 août 2020 12:45:14 -0500, a ecrit:
> > On 8/19/20 11:34 AM, Samuel Thibault wrote:
> >
> > > I can't find any function in the API doing this, but it'd be simple to
> > > add: just send HISTORY GET CLIENT_ID to the server.
> > >         
> > OK, thanks. What do I send, specifically?
> >
> > "HISTORY GET CLIENT_ID\r\n"
> >       
> That should just be enough. Try by hand:
>
> $ socat UNIX-CONNECT:/run/user/$UID/speech-dispatcher/speechd.sock,crlf STDIO
> HISTORY GET CLIENT_ID
> 245-4
> 245 OK CLIENT ID SENT
>
> > "HISTORY GET CLIENT_ID\r\n.\r\n."
> >
> > Or something else? I'm reading the spd_send_data docs, and they seem to imply that "\r\n.\r\n." is a thing separate from "\r\n".
> >       
> Where does it talk about \r\n.\r\n?
> Are you thinking about the line to be used to end a multi-line text such
> as for SPEAK? There is no reason for a HISTORY GET to need multi-line.
>
> Samuel
>