tag:deleted messages immediately deleted ?

Mark Walters markwalters1009 at gmail.com
Fri Nov 21 07:08:59 PST 2014


On Fri, 21 Nov 2014, Olivier Berger <olivier.berger at telecom-sudparis.eu> wrote:
> David Bremner <david at tethera.net> writes:
>
>> Olivier Berger <olivier.berger at telecom-sudparis.eu> writes:
>>
>>>
>>> So, I've tried and removed the spam tag from the exclude_tags, and
>>> suddenly, the search in emacs responds with the 981... which means that
>>> most of the deleted ones had the spam tag too.
>>>
>>>
>>> So it means that if one explicitely requests an excluded tag, other
>>> exclude tags still apply. Not sure this is the desirable option : maybe
>>> if one exclusion is waved, then others should too ?
>>>
>>> What do you think ?
>>
>> I'm not sure. What you suggest sounds sensible enough. On the other hand
>> the way it behaves now is precisely as documented; I'm not sure whether
>> this is because of a design choice or ease of implementation. Maybe Mark
>> can comment further on that.  I guess there are even people who
>> like/rely on the current functionality, since there always are ;).

This was definitely a design choice (I think probably by Austin/jrollins) and I
think it makes sense: why would you want to include one include spam
messages when you are searching for deleted messages?

A change would break my setup - not in itself a problem as I setup this
way to make sure I exercised the exclude code. I tag all my notmuch
mailing list mail tag:notmuch and have that as an excluded tag. Then
mailing list results do not clutter up results when I am doing personal
searches.

>>
>
> In any case, there has been a change in the way this worked.

I don't think anyone has touched this code for over two years: git blame
seems to suggest March 2012.

> For the moment, I'm using the following saved search :
>     (tag:deleted or tag:spam) and tag:deleted
> which will display the deleted mails.

I do think it would be nice to have a clear way of turning excludes off
in the emacs frontend. Without a query parser it's not clear what the
best way to do it is: I suggested a hack which allowed --exclude=false
to be passed as part of the search. We could add a toggle to rerun a
search with exclude=false but that doesn't help much for saved searches
or manually entered searches.

Best wishes

Mark



More information about the notmuch mailing list