Re: split-fancy-with-parent

2011-04-03 Thread Lars Magne Ingebrigtsen
Vincent Beffara vbeff...@ens-lyon.fr writes:

 LMI Could you repost the backtrace after doing the following?
 LMI (setq eval-expression-print-level 20)
 LMI (setq eval-expression-print-length 80)

 There you are - it didn't seem to change anything:

 ,
 | Debugger entered--Lisp error: (wrong-type-argument stringp nil)
 |   encode-coding-string(nil utf-7-imap)
 |   utf7-encode(nil t)
 |   nnimap-request-accept-article(nil local t)
 |   gnus-request-accept-article(nil (nnimap local
 | (nnimap-address localhost) (nnimap-inbox INBOX)
 | (nnimap-split-methods (quote nnimap-split-fancy))
 | (nnimap-split-fancy (| ... INBOX))) t t)

Hm.  Does anybody know what the magic incantation to get complete
backtraces is in Emacs 23?

-- 
(domestic pets only, the antidote for overdose, milk.)
  bloggy blog http://lars.ingebrigtsen.no/
___
info-gnus-english mailing list
info-gnus-english@gnu.org
http://lists.gnu.org/mailman/listinfo/info-gnus-english


Re: split-fancy-with-parent

2011-03-30 Thread Vincent Beffara

LMI Could you repost the backtrace after doing the following?
LMI (setq eval-expression-print-level 20)
LMI (setq eval-expression-print-length 80)

There you are - it didn't seem to change anything:

,
| Debugger entered--Lisp error: (wrong-type-argument stringp nil)
|   encode-coding-string(nil utf-7-imap)
|   utf7-encode(nil t)
|   nnimap-request-accept-article(nil local t)
|   gnus-request-accept-article(nil (nnimap local
| (nnimap-address localhost) (nnimap-inbox INBOX)
| (nnimap-split-methods (quote nnimap-split-fancy))
| (nnimap-split-fancy (| ... INBOX))) t t)
|   eval((gnus-request-accept-article nil
| (quote (nnimap local ... ... ... ...)) t t))
|   nnimap-request-move-article(375 vt local
| (gnus-request-accept-article nil (quote
| (nnimap local ... ... ... ...)) t t) t nil)
|   gnus-request-move-article(375 vt local
| (gnus-request-accept-article nil (quote
| (nnimap local ... ... ... ...)) t t) t nil)
|   gnus-summary-move-article(nil nil (nnimap local
| (nnimap-address localhost) (nnimap-inbox INBOX)
| (nnimap-split-methods (quote nnimap-split-fancy))
| (nnimap-split-fancy (| ... INBOX
|   gnus-summary-respool-article(nil (nnimap local
| (nnimap-address localhost) (nnimap-inbox INBOX)
| (nnimap-split-methods (quote nnimap-split-fancy))
| (nnimap-split-fancy (| ... INBOX
|   call-interactively(gnus-summary-respool-article nil nil)
`

/v

-- 
| |   UMPA - ENS Lyon   | Mél: vbeff...@ens-lyon.fr |
| Vincent Beffara |  46 allée d'Italie  | Tél: (+33) 4 72 72 85 25  |
| | 69364 Lyon Cedex 07 | Fax: (+33) 4 72 72 84 80  |


___
info-gnus-english mailing list
info-gnus-english@gnu.org
http://lists.gnu.org/mailman/listinfo/info-gnus-english


Re: split-fancy-with-parent

2011-03-29 Thread Lars Magne Ingebrigtsen
Vincent Beffara vbeff...@ens-lyon.fr writes:

 Here you go (rewrapped to make gnus happy):

 ,
 | Debugger entered--Lisp error: (wrong-type-argument stringp nil)
 |   encode-coding-string(nil utf-7-imap)

Could you repost the backtrace after doing the following?

(setq eval-expression-print-level 20)
(setq eval-expression-print-length 80)

-- 
(domestic pets only, the antidote for overdose, milk.)
  la...@gnus.org * Lars Magne Ingebrigtsen
___
info-gnus-english mailing list
info-gnus-english@gnu.org
http://lists.gnu.org/mailman/listinfo/info-gnus-english


Re: split-fancy-with-parent

2011-03-24 Thread Vincent Beffara

VB PS: BTW, now B t to test where a message would end up consistenty
VB answers mail.misc even on a message that was correctly split. And
VB respooling with B r fails saying
VB wrong type argument: stringp, nil
VB But I can live with that.

TZ I haven't looked at that code in a lng time.  Could you do `M-x
TZ toggle-debug-on-error', generate the wrong type error, and post the
TZ backtrace, to be sure it's not something local to you?

Here you go (rewrapped to make gnus happy):

,
| Debugger entered--Lisp error: (wrong-type-argument stringp nil)
|   encode-coding-string(nil utf-7-imap)
|   utf7-encode(nil t)
|   nnimap-request-accept-article(nil local t)
|   gnus-request-accept-article(nil (nnimap local
| (nnimap-address localhost) (nnimap-stream network)
| (nnimap-inbox INBOX)
| (nnimap-split-methods (quote nnimap-split-fancy))
| (nnimap-split-fancy (| ... INBOX))) t t)
|   eval((gnus-request-accept-article nil
| (quote (nnimap local ... ... ... ... ...)) t t))
|   nnimap-request-move-article(363 vt local
| (gnus-request-accept-article nil (quote
| (nnimap local ... ... ... ... ...)) t t) t nil)
|   gnus-request-move-article(363 vt local
| (gnus-request-accept-article nil (quote
| (nnimap local ... ... ... ... ...)) t t) t nil)
|   gnus-summary-move-article(nil nil (nnimap local
| (nnimap-address localhost) (nnimap-stream network)
| (nnimap-inbox INBOX) (nnimap-split-methods
| (quote nnimap-split-fancy))
| (nnimap-split-fancy (| ... INBOX
|   gnus-summary-respool-article(nil (nnimap local
| (nnimap-address localhost) (nnimap-stream network)
| (nnimap-inbox INBOX) (nnimap-split-methods
| (quote nnimap-split-fancy)) (nnimap-split-fancy (| ... INBOX
|   call-interactively(gnus-summary-respool-article nil nil)
`

/v

-- 
| |   UMPA - ENS Lyon   | Mél: vbeff...@ens-lyon.fr |
| Vincent Beffara |  46 allée d'Italie  | Tél: (+33) 4 72 72 85 25  |
| | 69364 Lyon Cedex 07 | Fax: (+33) 4 72 72 84 80  |


___
info-gnus-english mailing list
info-gnus-english@gnu.org
http://lists.gnu.org/mailman/listinfo/info-gnus-english


Re: split-fancy-with-parent

2011-03-23 Thread Vincent Beffara

 Invariably when I run B t gnus tells me this message would be
 split to INBOX (but the (: nnmail-split-fancy-with-parent) does
 appear in the window). What am I missing here ?

EA I don't know if what you're doing is supposed to work or not, but
EA I'm also using the registry, and I use (:
EA gnus-registry-split-fancy-with-parent). That works nicely. I'm using
EA No Gnus, though, don't know if this is a newly-introduced function…

I tried that as well, and the behavior is exactly the same, that line
seems to be completely ignored ... BTW, using No Gnus as well (from the
git repository).

How can I investigate further ?

/v

-- 
| |   UMPA - ENS Lyon   | Mél: vbeff...@ens-lyon.fr |
| Vincent Beffara |  46 allée d'Italie  | Tél: (+33) 4 72 72 85 25  |
| | 69364 Lyon Cedex 07 | Fax: (+33) 4 72 72 84 80  |


___
info-gnus-english mailing list
info-gnus-english@gnu.org
http://lists.gnu.org/mailman/listinfo/info-gnus-english


Re: split-fancy-with-parent

2011-03-23 Thread Eric Abrahamsen
Vincent Beffara vbeff...@ens-lyon.fr writes:

 Invariably when I run B t gnus tells me this message would be
 split to INBOX (but the (: nnmail-split-fancy-with-parent) does
 appear in the window). What am I missing here ?

 EA I don't know if what you're doing is supposed to work or not, but
 EA I'm also using the registry, and I use (:
 EA gnus-registry-split-fancy-with-parent). That works nicely. I'm using
 EA No Gnus, though, don't know if this is a newly-introduced function…

 I tried that as well, and the behavior is exactly the same, that line
 seems to be completely ignored ... BTW, using No Gnus as well (from the
 git repository).

I'm very much the last person who should be tackling any else's gnus
problems, but I'm thinking that since you're using imap, you should be
setting =nnimap-split-fancy=, not =nnml-split-fancy=. That variable is
mentioned in the docs for =gnus-registry-split-fancy-with-parent=, but
nowhere else, so I'm going to bow out here before I give any
particularly dumb advice :)

E


___
info-gnus-english mailing list
info-gnus-english@gnu.org
http://lists.gnu.org/mailman/listinfo/info-gnus-english


Re: split-fancy-with-parent

2011-03-23 Thread Ted Zlatanov
On Wed, 23 Mar 2011 19:41:38 +0800 Eric Abrahamsen e...@ericabrahamsen.net 
wrote: 

EA Vincent Beffara vbeff...@ens-lyon.fr writes:
 Invariably when I run B t gnus tells me this message would be
 split to INBOX (but the (: nnmail-split-fancy-with-parent) does
 appear in the window). What am I missing here ?
 
EA I don't know if what you're doing is supposed to work or not, but
EA I'm also using the registry, and I use (:
EA gnus-registry-split-fancy-with-parent). That works nicely. I'm using
EA No Gnus, though, don't know if this is a newly-introduced function…
 
 I tried that as well, and the behavior is exactly the same, that line
 seems to be completely ignored ... BTW, using No Gnus as well (from the
 git repository).

EA I'm very much the last person who should be tackling any else's gnus
EA problems, but I'm thinking that since you're using imap, you should be
EA setting =nnimap-split-fancy=, not =nnml-split-fancy=. That variable is
EA mentioned in the docs for =gnus-registry-split-fancy-with-parent=, but
EA nowhere else, so I'm going to bow out here before I give any
EA particularly dumb advice :)

The split rules changed in recent versions of No Gnus.  I think they are
only a server parameter now, so you have to modify your server
definition.

The `gnus-registry-split-fancy-with-parent' documentation is wrong
(outdated, really).  Sorry about that.  I'm pretty sure you don't want
`nnmail-split-fancy-with-parent' though.

Ted
___
info-gnus-english mailing list
info-gnus-english@gnu.org
http://lists.gnu.org/mailman/listinfo/info-gnus-english


Re: split-fancy-with-parent

2011-03-23 Thread Vincent Beffara

TZ The split rules changed in recent versions of No Gnus.  I think they
TZ are only a server parameter now, so you have to modify your server
TZ definition.

TZ The `gnus-registry-split-fancy-with-parent' documentation is wrong
TZ (outdated, really).  Sorry about that.  I'm pretty sure you don't
TZ want `nnmail-split-fancy-with-parent' though.

Aha. So I did this:

,
| (setq gnus-select-method '(nnimap local
|   (nnimap-address localhost)
|   (nnimap-stream network)
|   (nnimap-inbox INBOX)
|   (nnimap-split-methods 'nnimap-split-fancy)
|   (nnimap-split-fancy
|(| (: gnus-registry-split-fancy-with-parent)
|   INBOX))
|   ))
`

and everything works great. That is fantastic, thanks a lot !

If I understand correctly, this will only split messages arriving in
INBOX ? Is there a way I can split all incoming messages in all
mailboxes, but setting the default rule to wherever the message was
delivered rather than always INBOX ?

Thanks again !

   /v

PS: BTW, now B t to test where a message would end up consistenty
answers mail.misc even on a message that was correctly split. And
respooling with B r fails saying

wrong type argument: stringp, nil

But I can live with that.

--
| |   UMPA - ENS Lyon   | Mél: vbeff...@ens-lyon.fr |
| Vincent Beffara |  46 allée d'Italie  | Tél: (+33) 4 72 72 85 25  |
| | 69364 Lyon Cedex 07 | Fax: (+33) 4 72 72 84 80  |


___
info-gnus-english mailing list
info-gnus-english@gnu.org
http://lists.gnu.org/mailman/listinfo/info-gnus-english


Re: split-fancy-with-parent

2011-03-23 Thread Ted Zlatanov
On Wed, 23 Mar 2011 18:36:46 +0100 Vincent Beffara vbeff...@ens-lyon.fr 
wrote: 

VB Aha. So I did this:

VB ,
VB | (setq gnus-select-method '(nnimap local
VB |   (nnimap-address localhost)
VB |   (nnimap-stream network)
VB |   (nnimap-inbox INBOX)
VB |   (nnimap-split-methods 'nnimap-split-fancy)
VB |   (nnimap-split-fancy
VB |(| (: 
gnus-registry-split-fancy-with-parent)
VB |   INBOX))
VB |   ))
VB `

VB and everything works great. That is fantastic, thanks a lot !

I'm glad.

VB If I understand correctly, this will only split messages arriving in
VB INBOX ? Is there a way I can split all incoming messages in all
VB mailboxes, but setting the default rule to wherever the message was
VB delivered rather than always INBOX ?

I don't think so, currently.  But you could write a group-entry function
(there's a hook for those) that looks at the unseen messages, runs
`gnus-registry-split-fancy-with-parent' on each one, and moves them to
the return value if it's a string (group name).  I would gladly put that
in gnus-registry.el so everyone can benefit from it.

VB PS: BTW, now B t to test where a message would end up consistenty
VB answers mail.misc even on a message that was correctly split. And
VB respooling with B r fails saying

VB wrong type argument: stringp, nil

VB But I can live with that.

I haven't looked at that code in a lng time.  Could you do
`M-x toggle-debug-on-error', generate the wrong type error, and post the
backtrace, to be sure it's not something local to you?

Thanks
Ted
___
info-gnus-english mailing list
info-gnus-english@gnu.org
http://lists.gnu.org/mailman/listinfo/info-gnus-english