Hi,
I didn't look at asterisk-chan-misdn in the past (as I was using asterisk-chan-capi) but the recent versions of asterisk shipping with chan_misdn seem fine - I'm not sure what the issue is with synchronizing asterisk and mISDN - I've used several mISDN and several Asterisk versions together.
The interfaces are a huge mess and not properly versioned. There are reasons why I asked for the package to be pulled from etch.
Using the chan_misdn in the asterisk tree will make it impossible to keep mISDN out of the release until it can be said that it is stable enough for us to be able to support it for three years.
Whenever the kernel version changes, you need at least two approvals from ftpmaster before the package can enter unstable (the first upload only allows the new version for that particular architecture it was uploaded for, the other architectures need to go through NEW again when the autobuilders have picked up the package as they use different binary package names). Usually a new kernel version is out before that has happened.
The userspace code is heavily patched to allow for it to work on other architectures than i386, and building shared libraries is disabled to allow for the library ABI to change. However, the kernel ABI also changed since the last upload, and it will be necessary to add proper runtime detection code, which I want to do only when there is a chance that the code ends up in a release (i.e. it is stable enough for me to commit to it for a few years); doing it now would add more compatibility code to be pushed along for the next two releases.
This package needs to be maintained in a very conservative way. I doubt this can be done in a team setting.
Simon -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
