On Wed, May 18, 2011 at 05:33:01PM +0200, Alan DeKok wrote:
I see how that's useful, but that kind of thing worries me a lot. It can be used/abused in many ways.
I could see the code going in for 3.0, but only in a way where it takes work to enable it. Otherwise, most people using it will get it wrong.
OK. Thanks for the feedback Alan. If there were a loop scanning for Cleartext-Password attributes, instead of using pairfine(), there could still be a flag which makes it stop after the first one. I'll see if the need arises, and then follow your guidance if I end up implementing it. Regarding 3.0: will this be forked from v2.1.x? The 'master' branch has had hardly any commits for a while, and the last time I tried to build it it failed. Regards, Brian.