Re: gmail and gnus-summary-move-article

[Eric Abrahamsen <eric@ericabrahamsen.net>]

George McNinch <gmcninch@gmail.com> writes:

>>> Say I'm in the gmail group INBOX, put point on an article, enter
>>> the key sequence B m (to perform gnus-summary-move-article ) and
>>> specify an nnml group as the target, everything works fine *except*
>>> that the article is not really deleted on the gmail account (all
>>> "gmail tags" are removed, but the article is still present on the
>>> server -- it can be found in "All Mail"). 
>
>> Hm.  That sounds like a bug.  Or is it at Gmail quirk? 
>
> I'm basically certain this is just a quirk of Gmail.

Right, Gmail treats groups as tags (one message can be in many) instead
of groups (the message is the message, and it's in one of your groups).
Deletion from a group means "remove this group/tag from this message",
not "delete this message". You have to move a message to "Trash" to
delete it.

>>> Mostly I manage this without too many errors, but surely (!?) there
>>> is some automagical way to apply gnus-summary-copy-article followed
>>> by gnus-summary-mark-as-expirable to all process-marked articles in
>>> a summary buffer? 
>
>> Well, you could write a function that calls those two functions. :-) 
>
> Right... After writing this, I think I had decided to do that. But I
> realized that I'm not sure I understand how to recognize/interact with
> the process marks, though I confess I really only had a quick look.
> Then I got distracted.
>
> Looking again, I recall noting that gnus-summary-mark-as-expirable
> doesn't interact with the process marks, as-is. (Well, outside of my
> perhaps odd-ball proposed usage, there is no reason it should...)

Expiration won't actually delete a Gmail message, unless the expiry
process moves the message to Trash.

E