Gnus configuration

Gnus configuration

[Edouard Debry]

Hello,

I configured gnus to work with an imap server.

The network is rather slow so, I turned on cache and put all messages in
the gnus cache.

Trouble is that the cached mark seems to take precedence over other
secondary marks, like "replied" or "forwarded". Is it possible to
configure gnus to revert the precedence order ?

In my case, I think it is more useful to see any other secondary mark
than the cached one, before the cached mark and, see this later only if
no other secondary marks apply to the message.

Secondary marks show up in the gnus-summary-line-format with %R.

Another trouble is that, as soon as a message enters the cache, the
number of characters (%k) is no more displayed and is replaced by "n/a".
Is it possible to configure gnus so as to keep the number of characters
event if the article is cached ?

Regards

Re: Gnus configuration

[Michael Heerdegen]

Hello Edouard,

> In my case, I think it is more useful to see any other secondary mark
> than the cached one, before the cached mark and, see this later only if
> no other secondary marks apply to the message.

I never have used caching.  It doesn't seem possible out of the box.

The definitions of the format specifiers are in
`gnus-summary-line-format-alist'.  The format line is calculated in
`gnus-summary-prepare-threads'.  After reading that code, it doesn't
seem to be the case that what you want can be gotten by configuration.

So I think you want to define an appropriate "gnus-user-format-function"
that offers what you want.  AFAIU it can refer to any dynamical bindinds
in effect when the

  (eval gnus-summary-line-format-spec t)

call in `gnus-summary-prepare-threads' is processed.


> Another trouble is that, as soon as a message enters the cache, the
> number of characters (%k) is no more displayed and is replaced by "n/a".
> Is it possible to configure gnus so as to keep the number of characters
> event if the article is cached ?

I don't have a clue about this one.  Seems to come from
`gnus-summary-line-message-size'.  I guess the answer is probably "no".


Regards,

Michael.

Re: Gnus configuration

[Edouard Debry]

Michael Heerdegen <michael_heerdegen@web.de> writes:

> Hello Edouard,

Hello Michael,


>
>> In my case, I think it is more useful to see any other secondary mark
>> than the cached one, before the cached mark and, see this later only if
>> no other secondary marks apply to the message.
>
> I never have used caching.  It doesn't seem possible out of the box.
>
> The definitions of the format specifiers are in
> `gnus-summary-line-format-alist'.  The format line is calculated in
> `gnus-summary-prepare-threads'.  After reading that code, it doesn't
> seem to be the case that what you want can be gotten by configuration.
>
> So I think you want to define an appropriate "gnus-user-format-function"
> that offers what you want.  AFAIU it can refer to any dynamical bindinds
> in effect when the
>
>   (eval gnus-summary-line-format-spec t)
>
> call in `gnus-summary-prepare-threads' is processed.

Thanks for giving me some hints, I'll look into it

>
>> Another trouble is that, as soon as a message enters the cache, the
>> number of characters (%k) is no more displayed and is replaced by "n/a".
>> Is it possible to configure gnus so as to keep the number of characters
>> event if the article is cached ?
>
> I don't have a clue about this one.  Seems to come from
> `gnus-summary-line-message-size'.  I guess the answer is probably "no".

I am sure it can be done, once an article is cached, it definitely can
be reached by gnus and its number of characters counted ? No ? One could
do it by hand. It looks like a bug.

>
> Regards,
>
> Michael.

Regards

Edouard

Re: Gnus configuration

[Emanuel Berg via Users list for the GNU Emacs text editor]

In `gnus-summary-line-format', it is the %R.

  "A" if this article has been replied to, " "
  otherwise (character)

No mention of cache but it seems it uses that as well, yes...

-- 
underground experts united
https://dataswamp.org/~incal

Re: Gnus configuration

[Emanuel Berg via Users list for the GNU Emacs text editor]

Edouard Debry wrote:

> I configured gnus to work with an imap server.
>
> The network is rather slow so, I turned on cache and put all
> messages in the gnus cache.

Did you read

  If you have an _extremely_ slow NNTP connection, you may
  consider turning article caching on. [...]
  (info "(gnus) Article Caching")

?

Also

  Closely related to article caching, we have “persistent
  articles”. In fact, it’s just a different way of looking at
  caching, and much more useful in my opinion. [...]
  (info "(gnus) Persistent Articles")

-- 
underground experts united
https://dataswamp.org/~incal

Re: Gnus configuration

[Edouard Debry]

Hello,

Thanks for your answeer.

Actually, I understand that persistent and cached articles
refer to the same process. I talked about caching articles
but in fact, in light of what you say, I am really doing
persistent articles : I do not rely on ticked or dormant
mark but, on * and M-* to move articles to or from the cache.

Trouble is that it does not change the mark, whether you cache an
article or make it persistent, the secondary mark is "cached".


Emanuel Berg via Users list for the GNU Emacs text editor
<help-gnu-emacs@gnu.org> writes:

> Edouard Debry wrote:
>
>> I configured gnus to work with an imap server.
>>
>> The network is rather slow so, I turned on cache and put all
>> messages in the gnus cache.
>
> Did you read
>
>   If you have an _extremely_ slow NNTP connection, you may
>   consider turning article caching on. [...]
>   (info "(gnus) Article Caching")
>
> ?
>
> Also
>
>   Closely related to article caching, we have “persistent
>   articles”. In fact, it’s just a different way of looking at
>   caching, and much more useful in my opinion. [...]
>   (info "(gnus) Persistent Articles")

Re: Gnus configuration

[Emanuel Berg via Users list for the GNU Emacs text editor]

Edouard Debry wrote:

> Trouble is that it does not change the mark, whether you
> cache an article or make it persistent, the secondary mark
> is "cached".

Did you try this?

  To avoid having all ticked articles (and stuff) entered into
  the cache, you should set ‘gnus-use-cache’ to ‘passive’ if
  you’re just interested in persistent articles:

    (setq gnus-use-cache 'passive)

  (info "(gnus) Persistent Articles")

?

-- 
underground experts united
https://dataswamp.org/~incal

Re: Gnus configuration

[Emanuel Berg via Users list for the GNU Emacs text editor]

Edouard Debry wrote:

> Trouble is that the cached mark seems to take precedence
> over other secondary marks, like "replied" or "forwarded".
> Is it possible to configure gnus to revert the precedence
> order ?

Aha, now I see it, yes, it is the same here!

But, remove the cache mark and the old A for example for
Answered returns! So that data is still there somewhere.

-- 
underground experts united
https://dataswamp.org/~incal

Re: Gnus configuration

[Emanuel Berg via Users list for the GNU Emacs text editor]

See

  gnus-summary-update-secondary-mark is a compiled Lisp
  function in ‘gnus-sum.el’.

  (gnus-summary-update-secondary-mark ARTICLE)

  Update the secondary (read, process, cache) mark.

and

  Articles stored in the article cache will be marked with an
  ‘*’ in the second column (‘gnus-cached-mark’).
  (info "(gnus) Other Marks")
  
-- 
underground experts united
https://dataswamp.org/~incal

Re: Gnus configuration

[Eric S Fraga]

Instead of caching, you could use the gnus agent which, although
intended primarily to enable off-line use, it has the side effect of
caching all articles and would achieve what you asking for.

-- 
Eric S Fraga via Emacs 28.0.50 & org 9.4.6 on Debian bullseye/sid

Re: Gnus configuration

[Edouard Debry]

It seems interesting if the agent caches all articles but without
marking them as cached. Is it the case ?


Eric S Fraga <e.fraga@ucl.ac.uk> writes:

> Instead of caching, you could use the gnus agent which, although
> intended primarily to enable off-line use, it has the side effect of
> caching all articles and would achieve what you asking for.

Re: Gnus configuration

[Eric S Fraga]

On Tuesday,  6 Jul 2021 at 15:37, Edouard Debry wrote:
> It seems interesting if the agent caches all articles but without
> marking them as cached. Is it the case ?

Yes.  The agent provides a separate mark (i.e. different column in the
summary line, not interfering with other marks like A and R).

-- 
Eric S Fraga via Emacs 28.0.50 & org 9.4.6 on Debian bullseye/sid

Re: Gnus configuration

[Emanuel Berg via Users list for the GNU Emacs text editor]

Eric S Fraga wrote:

>> It seems interesting if the agent caches all articles but
>> without marking them as cached. Is it the case ?
>
> Yes. The agent provides a separate mark (i.e.
> different column in the summary line, not interfering with
> other marks like A and R).

Can't one have one place holder/possible indicator for every
state (at least those who are independent of each other) - if
there are two many of them that won't matter actually because
the user can still configure `gnus-summary-line-format' to not
show them if it gets too much, and the default value won't
have to change, either.

Because this problem feels ... unworthy of software of this
caliber, somehow.

Like in the docstring to gnus-summary-line-format

  %C          "*" if the article is cached, "" otherwise

?

-- 
underground experts united
https://dataswamp.org/~incal

Re: Gnus configuration

[Edouard Debry]

Emanuel Berg via Users list for the GNU Emacs text editor
<help-gnu-emacs@gnu.org> writes:

> Eric S Fraga wrote:
>
>>> It seems interesting if the agent caches all articles but
>>> without marking them as cached. Is it the case ?
>>
>> Yes. The agent provides a separate mark (i.e.
>> different column in the summary line, not interfering with
>> other marks like A and R).
>
> Can't one have one place holder/possible indicator for every
> state (at least those who are independent of each other) - if
> there are two many of them that won't matter actually because
> the user can still configure `gnus-summary-line-format' to not
> show them if it gets too much, and the default value won't
> have to change, either.

I do not really understand what you are saying here. My issue is that
the "reply" or "forward" marks are hidden by the cache mark when the
message is indeed cached. Removing the secondary mark would remove all
of them. What I want is to distinguish the cache mark from others.


>
> Because this problem feels ... unworthy of software of this
> caliber, somehow.

I do not understand

>
> Like in the docstring to gnus-summary-line-format
>
>   %C          "*" if the article is cached, "" otherwise
>

C-h v RET gnus-summary-line-format RET

I did not see this. But it would not be useful to solve my problem : %R
gathers many marks, including the cache mark which takes precedence when
article is cached.


> ?

Regards