<br>Comments inline...<br><div class="gmail_quote">On Mon, Jun 4, 2012 at 7:53 AM, Dmitry Sytchev <span dir="ltr"><<a href="mailto:kbdfck@gmail.com" target="_blank">kbdfck@gmail.com</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
<div>I asked this question many times on mailing list, and now I'm sure this can't be really done with loopback.</div>The only alternative for loopback is to re-inject call into FS via some separate Sofia profile, and specify that profile in string for att_xfer.<br>
</blockquote><div>Maybe there's a more elegant way to do this. <br><br></div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">
This brings up large amount of troubles including DTMF transcoding, sequential att_xfer attempt recognition and overall voice/dtmf delay introduced by chained channels. Maybe some channels can be moved out of scene by using 'simplify' api on correct channels, but this needs tests.</blockquote>
<div>This makes me wonder if att_xfer is the right tool. <br><br></div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex"><div>
<br><div>Anyway, loopback channel in FS is completely unusable, so we do need to have some best practices on how to do things without it in FS wiki... Maybe I have time and will describe our experience soon.<br></div></div>
</blockquote><div>Please do. I get the feeling that we (the community attempting to help) have just enough information to be frustrated. I suspect that if we knew the big picture we could probably offer some more useful suggestions.<br>
<br>-MC<br></div></div><br>