[asterisk-dev] [svn-commits] file: branch 11 r371619 - /branches/11/channels/chan_motif.c

[Matthew Jordan]

----- Original Message -----
> From: "Paul Belanger" <paul.belanger at polybeacon.com>
> To: "Asterisk Developers Mailing List" <asterisk-dev at lists.digium.com>
> Sent: Wednesday, August 22, 2012 2:03:56 PM
> Subject: Re: [asterisk-dev] [svn-commits] file: branch 11 r371619 -	/branches/11/channels/chan_motif.c
> 
> On 12-08-22 02:49 PM, Joshua Colp wrote:
> > ----- Original Message -----
> >
> > The fundamental call-id logging feature was obviously in before
> > that date, as was chan_motif itself. Both things were developed at
> > the same time and implementing the minor changes required in
> > chan_motif for call-id logging didn't quite make it. Personally I
> > made the call to put it into 11 because it's on the line between a
> > new feature and something that should have been there. It's also
> > very non-invasive, limited to a single new module. If this
> > discussion leads to keeping this out, so be it.
> >
> It's not my call to keep it in or out, I'm just bringing up the fact
> we
> explicitly say feature freeze is the 3rd Wednesday in July. If we, as
> a
> project, cannot make that date we should send out notifications of
> the
> slip.  This way other people have the opportunity to get their (late)
> patches in.

You are misconstruing what occurred.

Both the chan_motif and call ID logging features were put up for code review by
the feature freeze deadline.  It was a minor oversight during the code review
that chan_motif did not use the call ID logging feature.  Some might consider
that a bug in chan_motif; some might consider adding the usage of call ID
logging in chan_motif an improvement.  Both new features however made it in
during the time frame specified; both were reviewed; both were tested.

I fail to see how two new features getting linked together - after
both went through the process correctly and everyone missed the fact that
the channel driver should naturally use the new non-intrusive feature -
is somehow a violation of a feature freeze.

As I view it, bug or improvement, it is *wrong* that chan_motif failed to use
call ID logging.  We are choosing to correct that problem - and improve the
end user experience - during the beta period, before release.

> Personally, I would rather slip the date, get all the new features
> in,
> then branch and release. I just don't like branching and releasing,
> then
> trying to add more features afterwards.  From a testing perspective,
> this is just wrong.

It isn't released yet.  It is in beta.  Things are going to get cleaned up and
fixed.  That is why its a beta.  If you can't make two new features work
together by fixing it in a -beta2, than why even bother making a beta release?

--
Matthew Jordan
Digium, Inc. | Engineering Manager
445 Jan Davis Drive NW - Huntsville, AL 35806 - USA
Check us out at: http://digium.com & http://asterisk.org