Re: completion for gnus-group-jump-to-group

Re: completion for gnus-group-jump-to-group

[Julien Cubizolles]

Eric Abrahamsen <eric-5ibeKGLO59aIK5dEoMBc7KxOck334EZe@public.gmane.org>
writes:

> Julien Cubizolles <j.cubizolles-GANU6spQydw@public.gmane.org> writes:
>
>> I don't really know if this is a helm of gnus question, but here goes:
>> where does helm get its completion list when running
>> gnus-group-jump-to-group ? I end up with a lot of very old and obsolete
>> group names dangerously close to some real ones.
>
> This is Gnus question (helm doesn't make it's own list of groups), and
> the answer -- in `gnus-group-completing-read' is it uses the variable
> `gnus-active-hashtb', which is an obarray holding the active groups. All
> those groups ought to be subscribed and in play: if you hit "L" in the
> Group buffer, you see them all listed, right?

Not all of them. There are some groups in gnus-group-completing-read
that aren't displayed in the list of groups displayed by
gnus-group-list-all-groups, that I can't find in any "active" file and
that can't be found when browsing a giver server in the *Server*
buffer. They are for example groups associated to some old nnimap
methods that I don't use anymore. Is there any way to prune this hash
table ?

Julien.




-- 
You received this message because you are subscribed to the Google Groups "emacs-helm" group.
To unsubscribe from this group and stop receiving emails from it, send an email to emacs-helm+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
For more options, visit https://groups.google.com/d/optout.

Re: completion for gnus-group-jump-to-group

[Eric Abrahamsen]

Julien Cubizolles <j.cubizolles-GANU6spQydw@public.gmane.org> writes:

> Eric Abrahamsen <eric-5ibeKGLO59aIK5dEoMBc7KxOck334EZe@public.gmane.org>
> writes:
>
>> Julien Cubizolles <j.cubizolles-GANU6spQydw@public.gmane.org> writes:
>>
>>> I don't really know if this is a helm of gnus question, but here goes:
>>> where does helm get its completion list when running
>>> gnus-group-jump-to-group ? I end up with a lot of very old and obsolete
>>> group names dangerously close to some real ones.
>>
>> This is Gnus question (helm doesn't make it's own list of groups), and
>> the answer -- in `gnus-group-completing-read' is it uses the variable
>> `gnus-active-hashtb', which is an obarray holding the active groups. All
>> those groups ought to be subscribed and in play: if you hit "L" in the
>> Group buffer, you see them all listed, right?
>
> Not all of them. There are some groups in gnus-group-completing-read
> that aren't displayed in the list of groups displayed by
> gnus-group-list-all-groups, that I can't find in any "active" file and
> that can't be found when browsing a giver server in the *Server*
> buffer. They are for example groups associated to some old nnimap
> methods that I don't use anymore. Is there any way to prune this hash
> table ?

The hash table is constructed at load time -- the entries are probably
in your newsrc.eld file, in the `gnus-newsrc-alist' variable. Take a
look in there and see if that's the case, maybe? You can always shut
down Gnus and then (very carefully) manually remove those entries.

E

-- 
You received this message because you are subscribed to the Google Groups "emacs-helm" group.
To unsubscribe from this group and stop receiving emails from it, send an email to emacs-helm+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
For more options, visit https://groups.google.com/d/optout.

Re: completion for gnus-group-jump-to-group

[Julien Cubizolles]

Eric Abrahamsen <eric-5ibeKGLO59aIK5dEoMBc7KxOck334EZe@public.gmane.org>
writes:


> The hash table is constructed at load time -- the entries are probably
> in your newsrc.eld file, in the `gnus-newsrc-alist' variable. Take a
> look in there and see if that's the case, maybe? You can always shut
> down Gnus and then (very carefully) manually remove those entries.

Thanks, that was easy. I had to do it manually since I was getting stack
overflow errors with the regexp matching but I could trim it down quite
a lot.

Julien.

-- 
You received this message because you are subscribed to the Google Groups "emacs-helm" group.
To unsubscribe from this group and stop receiving emails from it, send an email to emacs-helm+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
For more options, visit https://groups.google.com/d/optout.