On Wednesday 28 November 2001 09.04, Robert Collins wrote:
> I finally got some time today. The digest code hasn't been fixed yet,
> but that should not hold releasing pre-stable versions. Those changes
> will be very isolated.
Fine. Not even an reason to delay STABLE then as the "digest" scheme is a new
addition, not likely to gain a wide coverage and can be documented as
"experimental".
> I've included Kinkies new winbindd helper in HEAD on the following
> logic:
> 1) There are no in-squid changes for the helper.
> 2) It needs wide exposure.
> 3) Even brand new it's likely better than the NTLMSSP helper, which
> we've known all along to be a stop-gap.
Have no problems with including new helpers at any time, or patches to
existing code. New helpers, error pages etc are free to go in even between
STABLE releases.
What I want is that when we make a 2.5.STABLE1 release, there is no known
bugs in any pieces of the code that people are very likely to use.
* digest scheme is not in this category
* diskd most certainly is
* basic scheme most certainly is
* ntlm scheme is also, but is a new feature so it is acceptable with some
unknown bugs or documented shortcomings (i.e. the "fail-open" thingy)
Regards
Henrik
Received on Wed Nov 28 2001 - 03:08:28 MST
This archive was generated by hypermail pre-2.1.9 : Tue Dec 09 2003 - 16:14:39 MST