On Sun, Oct 15, 2006 at 01:02:08AM -0200, Henrique de Moraes Holschuh wrote: > On Sat, 14 Oct 2006, Roberto C. Sanchez wrote: > > A patch [0] was proposed a while back on the cyrus-sasl list to allow > > crypt in libsasl2. The original message [1] is also available. Anyhow, > > We should not accept that patch *ever* in any other format than a "optional, > *disabled by default* thing you should enable only if you know what you are > doing". And I would not include it even in that form. > > It breaks auxprop plugins, which is a fundamental way of how Cyrus SASL > works. It requires disabling globally some auth methods [that require the > cleartext password to generate challenges] when the feature is enabled too, > if the patch doesn't do this, please reject it without futher consideration. > > > I'd like to hear people's opinions on this. My vote (for the Debian > > people) is that if upstream says that they don't want the patch, that we > > close the bug and be done with it. > > Upstream didn't want the patch. They may change their mind (they have done > so with the 8-bit stuff in cyrus imapd headers, which goes against *all* > RFCs), though. > OK. I thought it would cause at least some breakage, though I had no idea of the scope. Given Henrique's comments, I have decided to close the bug. Upstream can disregard it from consideration. Unless someone has an exceptionally compelling case for why this should be different, I think this is how it should stay. Regards, -Roberto -- Roberto C. Sanchez http://people.connexer.com/~roberto http://www.connexer.com
Attachment:
signature.asc
Description: Digital signature
