digest buffer behavior

digest buffer behavior

[Laura Conrad]

I just ran a digest produced by majordomo through the "Enter Digest
Buffer" (A D) command, and checked out what r and f do to those
messages.

This is a digest that does include a Reply-to field in the headers of
the digest, which points to the list.  

Undigestified digests did not include a reply-to field, so both r and
f resulted in mail to the individual, not to the list.

If there's a global reply-to field for the digest, wouldn't it make
sense for the digest buffer to be able to use that?

-- 
Laura (mailto:lconrad@world.std.com , http://www.world.std.com/~lconrad/ )
(617) 661-8097	fax: (801) 365-6574 
233 Broadway, Cambridge, MA 02139

Re: digest buffer behavior

[Shenghuo ZHU]

>>>>> "LC" == Laura Conrad <lconrad@gamesville.com> writes:

LC> I just ran a digest produced by majordomo through the "Enter
LC> Digest Buffer" (A D) command, and checked out what r and f do to
LC> those messages.

LC> This is a digest that does include a Reply-to field in the headers
LC> of the digest, which points to the list.

LC> Undigestified digests did not include a reply-to field, so both r
LC> and f resulted in mail to the individual, not to the list.

LC> If there's a global reply-to field for the digest, wouldn't it
LC> make sense for the digest buffer to be able to use that?

[...]

The problem is ephemeral groups can not save parameters.  In early
verison of pgnus, I solved the default charset for ephemeral groups by
inheriting parent group's varibles.  Now it is the time to inherit all
parameters from ancestor groups (an ephemeral group may be generated
from another ephemeral group). I think adding a variable
`gnus-newsgroup-ancestor-list' to keep the ancestor groups tractable
would works in most case.

-- 
Shenghuo ZHU

Re: digest buffer behavior

[Lars Magne Ingebrigtsen]

Laura Conrad <lconrad@gamesville.com> writes:

> I just ran a digest produced by majordomo through the "Enter Digest
> Buffer" (A D) command, and checked out what r and f do to those
> messages.
> 
> This is a digest that does include a Reply-to field in the headers of
> the digest, which points to the list.  
> 
> Undigestified digests did not include a reply-to field, so both r and
> f resulted in mail to the individual, not to the list.
> 
> If there's a global reply-to field for the digest, wouldn't it make
> sense for the digest buffer to be able to use that?

Hm.  Hmm.  Hmmm.

Perhaps `A D' should make the Reply-To or From address of the digest
the `to-address' group params of the digest group?  That would make
`a' and `f' work as intended, I think.

Yup.  Fix in Gnus v5.8.0.

-- 
(domestic pets only, the antidote for overdose, milk.)
   larsi@gnus.org * Lars Magne Ingebrigtsen

Re: digest buffer behavior

[Lars Magne Ingebrigtsen]

Shenghuo ZHU <zsh@cs.rochester.edu> writes:

> The problem is ephemeral groups can not save parameters.  In early
> verison of pgnus, I solved the default charset for ephemeral groups by
> inheriting parent group's varibles.  Now it is the time to inherit all
> parameters from ancestor groups (an ephemeral group may be generated
> from another ephemeral group).

In this case, I don't think that's necessary.  (It's fixed in cvs in a 
few jiffies.)

> I think adding a variable `gnus-newsgroup-ancestor-list' to keep the
> ancestor groups tractable would works in most case.

When creating ephemeral groups based on other groups (that would only
be in virtual groups, right?), then that would probably be necessary.
I think.  Hm.

-- 
(domestic pets only, the antidote for overdose, milk.)
   larsi@gnus.org * Lars Magne Ingebrigtsen