[asterisk-dev] say.conf options (m/f)

Tzafrir Cohen tzafrir.cohen at xorcom.com
Thu May 31 01:59:56 MST 2007


On Sun, May 27, 2007 at 03:11:30PM +0300, Tzafrir Cohen wrote:
> Hi
> 
> I'm trying to write a say.conf syntax for Hebrew and run into an issue 
> that will probably affect other languages.
> 
> The function ast_say_number_full in say.c has an "options" field. It is 
> generally used for the "gender" field of SayNumber.
> 
> >From what I understand say.conf parsing is only implemented in 
> app_playback.c . From looking at the implementation of ast_say_number 
> there it seems that the "options" variable is just past along
> to recusrsive calls, but not really used.
> 
> The say.conf syntax does not seem to relate to the gender in any
> way: I cannot see a way to generate a different number for the male 
> form and for the female form.

>From the rest of the thread you can see that the problem is there.

Initial suggested syntax. I have not yet looked at implementing it:
append the optional "options" field to the prefix field. If I need
to say  the number 328/f  (explicit Female form) I will try first:

  numf:238

and only then:

  num:328

But then I'm not exactly sure what should happen then to the options 
field on recursive calls to do_say(). If the default form was used, the
options field should be preserved because maybe the num is the same but
digits are different.

But if 'numf:238' has some explicit handling of the number, maybe saving 
the options field on recursive calls can create some wierd behaviours.
Though I don't have anything specific yet.

-- 
               Tzafrir Cohen       
icq#16849755                    jabber:tzafrir at jabber.org
+972-50-7952406           mailto:tzafrir.cohen at xorcom.com       
http://www.xorcom.com  iax:guest at local.xorcom.com/tzafrir


More information about the asterisk-dev mailing list