Hi,
Now I am using lua's session:getDigits function, and it seems that "read_result" and "
read_terminator_used" are not set at all.
Besides, lua's implementation of this function accept only three parameters( lenght, terminator and timeout),
however, some other modules allow more parameters  (
http://wiki.freeswitch.org/wiki/Session_getDigits)
Why is that?

Cheers,
Peter



Anthony Minessale wrote:
We have such a variable called "read_result"
possible vals: success, timeout, failure

"read_terminator_used" contains the terminator


On Tue, Nov 3, 2009 at 2:20 PM, Peter Volchek <[email protected]> wrote:
Hi,
I am having some troubles adopting [session:playAndGetDigits] function
to my application needs.

There are three possible outcomes that the function should report on:
1. Success. The required number of of digits was collected and they
match the mask.
2. Fail 1.The entered digits do not match the mask
3. Fail2. Timeout failure

Right now, I cannot distinguish between 2 & 3. In both cases, an empty
string is returned, so I have no idea what was the real failure.

4. Success upon termination
That is a tricky one, and I believe, there is a bug in this function.
This function accept the terminators, which should stop digits collection.
Consider the following function call:

digits = session:playAndGetDigits(
   0, 5, 1, 3000, "#",
   "enter_mailbox_number_or_hit_pound_to_login.wav",
   "error.wav",
   "\\d*"
   );


Here, I want to collect up to 5 digits, that represent a mailbox number.
Note that I set the minimum value to be 0, as I want "#" itself to be
the sign, that I want to login to my system.
However, when I execute this code and hit "#" immediately, the system
play me an error message. Oops! Well, I changed my digits mask to be
"#", and it did not help either.
BTW, about the last case. If I my digits mask contains the symbol used
in the terminators, then it has to be collected and returned. The
current implementation just uses terminates the collection and does not
include the terminator to the returned string.

Possible solutions to determine the function's outcome:
1. Return multiple values (state, digits). That may break tons of the
existing code
2. Set a channel (or session) variable. Similar to C's errno and define
some variables defining the outcome. Below is a summary table that
covers all the cases:

------------------------------------------------------
ERRNO       | Description
------------------------------------------------------
RET_OK      | Maximum number of digits was collected or
           | terminated when in range [min_digits..max_digits-1]
           | or timeout triggered after collecting enough digits
[min_digits..max_digits-1].
           | The collected digits match digits_regex
           | Returns the collected digits

RET_TERM    | Terminating before collecting enough digits or when
           | the collected digits do not match digits_regex
           | Return whatever was collected (for logging)


RET_TIMEOUT | Did not collect enough digits for the time specified or when
           | the collected digits do not match digits_regex
           | Return whatever was collected (for logging)


Does it all make sense?
Cheers,
PV



_______________________________________________
FreeSWITCH-dev mailing list
[email protected]
http://lists.freeswitch.org/mailman/listinfo/freeswitch-dev
UNSUBSCRIBE:http://lists.freeswitch.org/mailman/options/freeswitch-dev
http://www.freeswitch.org



--
Anthony Minessale II

FreeSWITCH http://www.freeswitch.org/
ClueCon http://www.cluecon.com/
Twitter: http://twitter.com/FreeSWITCH_wire

AIM: anthm
MSN:[email protected]
GTALK/JABBER/PAYPAL:[email protected]
IRC: irc.freenode.net #freeswitch

FreeSWITCH Developer Conference
sip:[email protected]
iax:[email protected]/888
googletalk:[email protected]
pstn:213-799-1400

_______________________________________________ FreeSWITCH-dev mailing list [email protected] http://lists.freeswitch.org/mailman/listinfo/freeswitch-dev UNSUBSCRIBE:http://lists.freeswitch.org/mailman/options/freeswitch-dev http://www.freeswitch.org



_______________________________________________
FreeSWITCH-dev mailing list
[email protected]
http://lists.freeswitch.org/mailman/listinfo/freeswitch-dev
UNSUBSCRIBE:http://lists.freeswitch.org/mailman/options/freeswitch-dev
http://www.freeswitch.org

Reply via email to