Re: [PATCH] doc: try to clarify owner- alias handling

Previous Topic Next Topic
classic Classic list List threaded Threaded
1 message Options
Reply | Threaded
Open this post in threaded view

Re: [PATCH] doc: try to clarify owner- alias handling

Wietse Venema
Simon Ruderich:

Checking application/pgp-signature: FAILURE
-- Start of PGP signed section.

> On Sun, Nov 26, 2017 at 10:40:11AM -0500, Wietse Venema wrote:
> >> This patch (based on the 3.2.4 sources) tries to improve the
> >> documentation of the owner- alias handling. I hope I patched the
> >> correct files as I'm not entirely sure which files are generated.
> >>
> >> The first part adds a "link" to reset_owner_alias to help the
> >> reader to understand the effects of owner- handling in more
> >> detail (at first I found only the short comment in man 5 aliases
> >> and was confused what actually happens).
> >>
> >> The second part explains that the last owner- alias "wins" when
> >> choosing the sender of the generated mails which I found by
> >> trial-and-error when looking at my owner- aliases configuration
> >> (at first I thought the first owner- alias is chosen).
> Hello,
> Sorry for the late reply.
> > I have updated some text, however it already says that owner-name
> Thank you. Is this change publicly available (before the next
> release)?
> > will direct delivery errors, and I don't see why that would change
> > for nested aliases. That is:
> >
> > aliases:
> > list-a: aa, list-b, ac, ...
> > owner-list-a: [hidden email]
> >
> > list-b: ba, bb bc, ...
> > owner-list-b: [hidden email]
> >
> > Sending to list-b will always direct delivery errors to owner-list-b.
> I think it's obvious if you already know how this feature works,
> but I thought it can't hurt to clearly mention the fact. The
> whole owner- feature felt arcane and underdocumented to me so I
> tried to make it clearer to casual readers.

The updated text will be in the next 'unstable' release, when I
have time.

#     In addition, when an alias exists for \fBowner-\fIname\fR, this
#     will replace the envelope sender address, so that delivery
#     diagnostics are directed to that address, instead of the originator
#     of the message (for details, see \fBexpand_owner_alias\fR and
#     \fBreset_owner_alias\fR).

But I agree that the documentation is incomplete. I have added
that to the TODO list.