--- Begin Message ---
Subject: |
29.0.50; Gnus' nnvirtual-retrieve-headers shouldn't force usage of the cache |
Date: |
Thu, 05 May 2022 08:28:39 -0700 |
User-agent: |
Gnus/5.13 (Gnus v5.13) Emacs/29.0.50 (gnu/linux) |
In ./lisp/gnus/nnvirtual.el, the `nnvirtual-retrieve-headers' deffoo
let-binds `gnus-use-cache' to t when it uses `gnus-retrieve-headers' on
its various constituent groups. For users not using the cache, this
will result in an error.
I can put an additional check in gnus-cache.el when we call
`gnus-cache-retrieve-headers', but it would be nice to know why this
necessary in the first place. The code comment on
`nnvirtual-retrieve-headers' says:
;; FIX FIX FIX we want to check the cache!
;; This is probably evil if people have set
;; gnus-use-cache to nil themselves, but I
;; have no way of finding the true value of it.
All the above is true, except I don't understand why we want to check
the cache. This code is ancient and I'm only medium-hopeful Lars
remembers the reasoning here, but on the face of it I don't know why we
need to force a check of the cache. It seems like we could simply remove
that, and the cache will still be checked iff the user has set
`gnus-use-cache' to t.
--- End Message ---
--- Begin Message ---
Subject: |
Re: bug#55275: 29.0.50; Gnus' nnvirtual-retrieve-headers shouldn't force usage of the cache |
Date: |
Fri, 06 May 2022 10:17:54 -0700 |
User-agent: |
Gnus/5.13 (Gnus v5.13) Emacs/29.0.50 (gnu/linux) |
On 05/06/22 14:17 PM, Lars Ingebrigtsen wrote:
> Eric Abrahamsen <eric@ericabrahamsen.net> writes:
>
>> All the above is true, except I don't understand why we want to check
>> the cache. This code is ancient and I'm only medium-hopeful Lars
>> remembers the reasoning here, but on the face of it I don't know why we
>> need to force a check of the cache. It seems like we could simply remove
>> that, and the cache will still be checked iff the user has set
>> `gnus-use-cache' to t.
>
> I have no recollection of that code. :-/
It was a while ago :)
> But, yes, I think removing that binding would be the right thing to do.
Done.
--- End Message ---