Gnus development mailing list
 help / color / mirror / Atom feed
* agent again
@ 1997-09-21  3:22 Matt Simmons
  1997-09-21  5:16 ` Lars Magne Ingebrigtsen
  1997-09-21  5:28 ` Matt Simmons
  0 siblings, 2 replies; 6+ messages in thread
From: Matt Simmons @ 1997-09-21  3:22 UTC (permalink / raw)


Am I the only one who is getting errors on "J a" in the Group buffer?
Whenever I do it, I get the below:

Signaling: (void-variable function)
  gnus-group-iterate(nil #<compiled-function (from "gnus-agent.elc") (group) "...(40)" [gnus-group-category group c delete groups] 4>)
  #<compiled-function (from "gnus-agent.elc") (category arg) "...(42)" [category gnus-category-alist nil groups c cat gnus-group-iterate arg #<compiled-function (from "gnus-agent.elc") ... "...(40)" ... 4> gnus-category-write] 3 ("/home/simmonmt/gnus/lisp/gnus-agent.elc" . 8889) (list (intern ...) current-prefix-arg)>(default nil)
  call-interactively(gnus-agent-add-group)

I tried byte-compiling the 0.8 source with 20.2 and 20.3b22 and tried
running both sets with both 20.2 and 20.3b22.  I kept getting the
above.

There also seems to be some disagreement with the 0.8 patch as posted
and the 0.7 tarball on ftp.gnus.org.  Nothing failed outright, but I
got several patches that were applied offset.

Matt

-- 
     Matt Simmons  -  simmonmt@acm.org  -  http://www.netcom.com/~simmonmt
      They told me it would disrupt my life less if I got killed sooner. 
                                                         --Joseph Heller


^ permalink raw reply	[flat|nested] 6+ messages in thread

* Re: agent again
  1997-09-21  3:22 agent again Matt Simmons
@ 1997-09-21  5:16 ` Lars Magne Ingebrigtsen
  1997-09-21  5:28 ` Matt Simmons
  1 sibling, 0 replies; 6+ messages in thread
From: Lars Magne Ingebrigtsen @ 1997-09-21  5:16 UTC (permalink / raw)


Matt Simmons <simmonmt@acm.org> writes:

> Am I the only one who is getting errors on "J a" in the Group buffer?
> Whenever I do it, I get the below:
> 
> Signaling: (void-variable function)
>   gnus-group-iterate(nil #<compiled-function (from "gnus-agent.elc") (group) "...(40)" [gnus-group-category group c delete groups] 4>)

Yup.  Fix in Quassia Gnus v0.9.

-- 
(domestic pets only, the antidote for overdose, milk.)
  larsi@gnus.org * Lars Magne Ingebrigtsen


^ permalink raw reply	[flat|nested] 6+ messages in thread

* agent again
  1997-09-21  3:22 agent again Matt Simmons
  1997-09-21  5:16 ` Lars Magne Ingebrigtsen
@ 1997-09-21  5:28 ` Matt Simmons
  1997-09-21 16:29   ` Lars Magne Ingebrigtsen
  1 sibling, 1 reply; 6+ messages in thread
From: Matt Simmons @ 1997-09-21  5:28 UTC (permalink / raw)


[-- Attachment #1: Type: text/plain, Size: 1684 bytes --]

Matt Simmons <simmonmt@acm.org> writes:

> Am I the only one who is getting errors on "J a" in the Group buffer?
> Whenever I do it, I get the below:
> [ ... backtrace removed ... ]

OK, I found out what was happening in the above, but it still breaks.
The cause of the above was:

(defun gnus-group-iterate (arg gnus-group-iterate-function)
[....]                     ______________/
 	(save-excursion   /
 	  (funcall function group))))))

(Hopefully you're looking a this in a fixed font =) )

gnus-group-iterate was declaring g-g-i-f but not using it.  I replaced 
function with g-g-i-f (see attachment) and gnus-group-iterate function 
began to work again.  It looks like the change crept in in the
qgnus-0.0-0.1 patch.

Once that's fixed, however, another problem pops up.  The function
lambda'd up to be passed to gnus-group-iterate has (push group groups) 
at the end.  This would be OK, except for the fact that
gnus-group-iterate depends on groups emptying out at some point so it
can exit.  If you try to gnus-agent-add-group a group that has not
previously been assigned[1], group gets passes down to the lambda'd
function, processed, and then pushed right back on groups.  This
ensures that gnus-group-iterate will never exit.  I tried to figure
out what the lambda'd function is trying to do, but I gave up.

Matt

Footnotes: 
[1]  I don't know about other cases - I've just checked g-a-a-g on a
     group that has not previously been assigned.  I'd check more, but 
     I'm trying to get this in before Lars goes to sleep. =)
-- 
     Matt Simmons  -  simmonmt@acm.org  -  http://www.netcom.com/~simmonmt
	    If a parsley farmer is sued, can they garnish his wages?

[-- Attachment #2: group-patch --]
[-- Type: application/octet-stream, Size: 582 bytes --]

*** /tmp/T0a0000o	Sun Sep 21 00:11:41 1997
--- gnus-group.el	Sat Sep 20 23:31:58 1997
***************
*** 1457,1465 ****
        (gnus-group-remove-mark group)
        (save-selected-window
   	(save-excursion
!  	  (funcall function group))))))
    
! (put 'gnus-group-iterate 'lisp-indent-function 1)
  
  ;; Selecting groups.
  
--- 1457,1465 ----
        (gnus-group-remove-mark group)
        (save-selected-window
   	(save-excursion
!  	  (funcall gnus-group-iterate-function group))))))
    
! ;(put 'gnus-group-iterate 'lisp-indent-function 1)
  
  ;; Selecting groups.
  

^ permalink raw reply	[flat|nested] 6+ messages in thread

* Re: agent again
  1997-09-21  5:28 ` Matt Simmons
@ 1997-09-21 16:29   ` Lars Magne Ingebrigtsen
  0 siblings, 0 replies; 6+ messages in thread
From: Lars Magne Ingebrigtsen @ 1997-09-21 16:29 UTC (permalink / raw)


Matt Simmons <simmonmt@acm.org> writes:

> Once that's fixed, however, another problem pops up.  The function
> lambda'd up to be passed to gnus-group-iterate has (push group groups) 
> at the end.  This would be OK, except for the fact that
> gnus-group-iterate depends on groups emptying out at some point so it
> can exit.

I've now redefined that function to use gensymmed variables.

-- 
(domestic pets only, the antidote for overdose, milk.)
  larsi@gnus.org * Lars Magne Ingebrigtsen


^ permalink raw reply	[flat|nested] 6+ messages in thread

* Re: Agent again
  2010-12-20 16:21 Agent again Richard Riley
@ 2010-12-20 17:06 ` Robert Pluim
  0 siblings, 0 replies; 6+ messages in thread
From: Robert Pluim @ 2010-12-20 17:06 UTC (permalink / raw)
  To: ding

Richard Riley <rileyrg@googlemail.com> writes:

> J s and "g" when plugged dont do the same thing for me. "J s" does
> something, but "plugged" groups dont get a new article count. "g" does
> get the new article count. What does, in the docs,
>
> "Fetch all articles and headers that are eligible for fetching."
>
> mean? I assumed all articles which pass the category predicate.
>
> I have a category "fun" with a predicate of "true". I am assuming this
> means fetch all articles. The groups in question are added to this
> category. Is that correct?
>

Yes

> Or another way of asking, what are the conditions that mean J s and g
> DONT do the same thing when plugged?

As I understand it:

g => query servers for info about any new articles that have
arrived. For 'mail-like' backends this may end up retrieving mail, for
nntp/nnimap it just updates article counts.

J s => download articles based on currently available headers according
to configured category predicates.

Informally: g means 'do I have anything new to read', and J s is 'download
stuff'.

Robert




^ permalink raw reply	[flat|nested] 6+ messages in thread

* Agent again
@ 2010-12-20 16:21 Richard Riley
  2010-12-20 17:06 ` Robert Pluim
  0 siblings, 1 reply; 6+ messages in thread
From: Richard Riley @ 2010-12-20 16:21 UTC (permalink / raw)
  To: nognus


J s and "g" when plugged dont do the same thing for me. "J s" does
something, but "plugged" groups dont get a new article count. "g" does
get the new article count. What does, in the docs,

"Fetch all articles and headers that are eligible for fetching."

mean? I assumed all articles which pass the category predicate.

I have a category "fun" with a predicate of "true". I am assuming this
means fetch all articles. The groups in question are added to this
category. Is that correct?

Or another way of asking, what are the conditions that mean J s and g
DONT do the same thing when plugged?




^ permalink raw reply	[flat|nested] 6+ messages in thread

end of thread, other threads:[~2010-12-20 17:06 UTC | newest]

Thread overview: 6+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
1997-09-21  3:22 agent again Matt Simmons
1997-09-21  5:16 ` Lars Magne Ingebrigtsen
1997-09-21  5:28 ` Matt Simmons
1997-09-21 16:29   ` Lars Magne Ingebrigtsen
2010-12-20 16:21 Agent again Richard Riley
2010-12-20 17:06 ` Robert Pluim

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).