Gnus development mailing list
 help / color / mirror / Atom feed
From: Kevin Greiner <kevin.greiner@compsol.cc>
Subject: Re: Problems with Gnus Agent-Mode (5.10 and 5.11)
Date: Sun, 17 Sep 2006 19:12:17 -0500	[thread overview]
Message-ID: <450DE461.30300@compsol.cc> (raw)
In-Reply-To: <uac4yifee.fsf@rjap.de>

Robert Rakowicz wrote:
> Kevin Greiner <kevin.greiner@compsol.cc> writes:
>
>   
>> Robert Rakowicz wrote:
>>     
>>> Hi all,
>>>
>>> hmm I can't believe it, really nobody has any tips/ hints for my problem
>>> with agent-mode?
>>>
>>>   
>>>       
>> It could be because your problem sounds more like a scoring issue than
>> an agent issue.
>>
>>     
>
> Hmm, and why do you think that? My Gnus setup is really easy.
>  
>   
>> What is the value of your gnus-summary-line-format?  To be more
>> specific, does it contain the '%z' format specifier?  If you do not have
>> this specifier, you would not know which articles have been scored down
>> vs. up.
>>
>>     
>
> I do not have changed this variable. He has default values
>
> ,----[ C-h v gnus-summary-line-format RET ]
> | gnus-summary-line-format is a variable defined in `gnus.el'.
> | Its value is 
> | "%U%R%z%I%(%[%4L: %-20,20n%]%) %s\n"
> | 
> | Documentation:
> | *The format specification of the lines in the summary buffer.
> | 
> | ....
> |
> | You can customize this variable.
> `----
>
>  
>   
>> When you have a group with "missing articles", reopen it then try the '/
>> o' command.  This will display old articles that hopefully includes your
>> missing articles.  Move the cursor over any of the missing articles then
>> enter 'V S' to display that article's score.  If the number is negative,
>> gnus most-likely excluding it per your scoring rules.
>>     
>
> I have not changed my scoring rules between 5.9 and 5.10 or 5.11
>
> ,----[ ~/.gnus.el(Score) ]
> | (setq gnus-parameters
> |         '(
> |           ("de\\.comm\\.software\\.gnus"
> |            (agent-predicate . high)
> |            (agent-score "~/News/agentscore/de.comm.software.gnus"))
> |           ))
> `----
>
> ,----[ ~/News/agentscore/de.comm.software.gnus ]
> | ;; -*- mode: Gnus-Score -*-
> | (
> |  ("subject"
> |   ("[SPAM]" -1 nil s)
> |   ("[SPAM?]" -1 nil s))
> |  ("message-id"
> |   ("@rjap.de\\|@localhost.localdomain" 1000 nil s))
> |  ("references"
> |   ("@rjap.de\\|@localhost.localdomain" 1000 nil s)))
> `----
>
> I have only one "default" category
>
> ,----
> | ;;; Editing the predicate for category default
> | ;; Type `C-c C-c' after you've finished editing.
> | 
> | false
> `----
>   
Interesting. That may clear scoring but it still doesn't indicate any 
problem with the agent. Did you try the gnus-summary-insert-old-articles 
('/ o') command at all?

Your original email read, in part:

    Now, I start Gnus an go *online* with 'J-j', then I refresh
    *Group-Buffer* with 'g' and receive new Headers with 'J-s'. Now as
    example I see 

    ,----
    | 4: pl.comp.os.linux 
    `----

    this group has 4 new Messages.

      

Up to this point, you apparently feel that everything is working. That's 
good as neither J-J nor J-s are capable of marking articles.

    I join the group and mark all messages as
    donloadable with 'J-#'

    ,----
    | %. [  17: rz                     ] Re: Czy ktoś mógłby...
    | %.     [  22: Konrad Kosmowski       ] 
    | %.         [  26: rz                     ] 
    | %. [  11: incuś                  ] Re: Regiony DVD - jak sprawdzić?
    `----

      

The 'J-#' command (gnus-agent-mark-article) simply applies a download 
mark (a character) to an article using gnus-summary-update-mark. That 
function happens to be the primary mechanism for setting any sort of 
mark. A problem here should result in wide spread errors for many people.

    Then I leve the group with 'q' ans I see

    ,----
    | 1: pl.comp.os.linux 
    `----

      

OK, so leaving a group results in some articles disappearing. That's 
when scoring is applies which is why I pointed you in that direction. 
You really must try the gnus-summary-insert-old-articles command. If you 
can get these articles back, you'll have someway to reproduce the 
problem at will and then you can experiment with changing your 
configuration to fix it.

To fully clear, or indict, the agent, try setting your gnus-parameters 
to nil. You can set up the same configuration using group customization 
(G c) in the group buffer to set these agent parameters.

Kevin




  reply	other threads:[~2006-09-18  0:12 UTC|newest]

Thread overview: 6+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2006-09-08 10:21 Robert Rakowicz
2006-09-13 20:55 ` Robert Rakowicz
2006-09-17  7:26   ` Kevin Greiner
2006-09-17 17:14     ` Robert Rakowicz
2006-09-18  0:12       ` Kevin Greiner [this message]
2006-09-18 16:32         ` Robert Rakowicz

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=450DE461.30300@compsol.cc \
    --to=kevin.greiner@compsol.cc \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).