[asterisk-dev] DTMF detection and generation code

[Kevin P. Fleming]

Dan Evans wrote:

> Of course, the situation is different when Asterisk needs to originate
> the event; still, if possible, I think it should try to replicate the
> duration.  But in the relay situation, it should definitely send what it
> receives.

We already decided during last week's developer conference call that
this will be fixed for Asterisk 1.4. Since all DTMF passing through
Asterisk is an attempt to model a 'real world' event, it makes sense to
try to preserve as much information as is practical.

While I agree with Steve Underwood that in general
DTMF-duration-sensitive applications are not ideal (<G>), the current
situation with Asterisk causes problems even with applications that
don't care about the duration. This is especially true when Asterisk
itself is creating the DTMF events; it makes no attempt to try get them
to conform to any 'real world' model at all.

Over the next week or two we will get this taken care of; we will have a
_lot_ of development effort put into Asterisk in the next two weeks, as
I will be traveling to Europe for two dedicated development events.
Start watching the commit list and be ready to test!