Limit nnmail-split-fancy-with-parent to certain groups?

Limit nnmail-split-fancy-with-parent to certain groups?

[Raymond Scholz]

Hi!

Is there a simple way of limiting the application of
nnmail-split-fancy-with-parent to certain groups?  Can this be done by
not recording message-IDs in nnmail-message-id-cache-file for these
groups?  And if so, how could this be done?

Background: I overlooked some personal replies in high volume mailing
lists.

Furthermore I suspect this feature not to work as described in the Info
documentation:

,----
| When `nnmail-cache-accepted-message-ids' is non-`nil', Gnus
| also records the message ids of moved articles, so that the followup
| messages goes into the new group.
`----

I've moved a personal reply from folder A into another folder B, went
to B, composed my reply (which ends up in B too because of gcc-self)
but the answer to my reply ended up in A.  Can anyone confirm this?

Cheers, Ray
-- 
"I heard if you play the NT-4.0-CD  backwards, you get a satanic message."
  "Thats nothing, if you play it forward, it installs NT-4.0"

Re: Limit nnmail-split-fancy-with-parent to certain groups?

[Kai Großjohann]

On Tue, 24 Jul 2001, Raymond Scholz wrote:

> Is there a simple way of limiting the application of
> nnmail-split-fancy-with-parent to certain groups?  Can this be done
> by not recording message-IDs in nnmail-message-id-cache-file for
> these groups?  And if so, how could this be done?

You would have to frob the code.  But it wouldn't be difficult.

> Background: I overlooked some personal replies in high volume
> mailing lists.

Hm.  Yeah.  Bad.  Hm.  Your feature sounds useful.

Lessee.

Ah.  In the function nnmail-split-fancy-with-parent there is a section
where it looks whether the group is "drafts".  If so, nil is returned
rather than the group name.

So the extension ought to be simple: define a variable which is a
regex.  And augment the above mentioned statement to check against
that regex rather than just the string "drafts".

Can you make that change and submit a patch?

> Furthermore I suspect this feature not to work as described in the
> Info documentation:
> 
> ,----
> | When `nnmail-cache-accepted-message-ids' is non-`nil', Gnus also
> | records the message ids of moved articles, so that the followup
> | messages goes into the new group.
> `----
> 
> I've moved a personal reply from folder A into another folder B,
> went to B, composed my reply (which ends up in B too because of
> gcc-self) but the answer to my reply ended up in A.  Can anyone
> confirm this?

Sadly, I can't reproduce this bug.  So I can't fix it.  Anyone?

kai
-- 
~/.signature: No such file or directory

Re: Limit nnmail-split-fancy-with-parent to certain groups?

[Raymond Scholz]

[-- Attachment #1: Type: text/plain, Size: 234 bytes --]

Kai.Grossjohann@CS.Uni-Dortmund.DE (Kai Großjohann) writes:

> Can you make that change and submit a patch?

Most probably I screwed something up, so please have a look at the
code.  It's stolen from nnmail-list-identifiers's usage.


[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #2: nnmail.el.patch --]
[-- Type: text/x-patch, Size: 1425 bytes --]

--- nnmail.el.old	Wed Jul 25 22:04:04 2001
+++ nnmail.el	Wed Jul 25 22:20:40 2001
@@ -115,6 +115,14 @@
   :group 'nnmail-split
   :type 'boolean)
 
+(defcustom nnmail-split-fancy-with-parent-ignore-groups nil
+  "Regexp that matches group names to be ignored when applying
+`nnmail-split-fancy-with-parent'.  This can also be a list of regexps."
+  :group 'nnmail-split
+  :type '(choice (const :tag "none" nil)
+		 (regexp :value ".*")
+		 (repeat :value (".*") regexp)))
+
 ;; Added by gord@enci.ucalgary.ca (Gordon Matzigkeit).
 (defcustom nnmail-keep-last-article nil
   "If non-nil, nnmail will never delete/move a group's last article.
@@ -1472,14 +1480,18 @@
   (let* ((refstr (or (message-fetch-field "references")
 		     (message-fetch-field "in-reply-to")))
 	 (references nil)
-	 (res nil))
+	 (res nil)
+	 (regexp (if (consp nnmail-split-fancy-with-parent-ignore-groups)
+		     (mapconcat 'nnmail-split-fancy-with-parent-ignore-groups " *\\|")
+		   nnmail-split-fancy-with-parent-ignore-groups)))
     (when refstr
       (setq references (nreverse (gnus-split-references refstr)))
       (unless (gnus-buffer-live-p nnmail-cache-buffer)
 	(nnmail-cache-open))
       (mapcar (lambda (x)
 		(setq res (or (nnmail-cache-fetch-group x) res))
-		(when (string= "drafts" res)
+		(when (or (string= "drafts" res)
+			  (and regexp (string-match regexp res)))
 		  (setq res nil)))
 	      references)
       res)))

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #3: gnus.texi.patch --]
[-- Type: text/x-patch, Size: 1209 bytes --]

--- gnus.texi.old	Wed Jul 25 22:17:19 2001
+++ gnus.texi	Wed Jul 25 22:07:25 2001
@@ -12199,11 +12199,12 @@
 @code{nnmail-split-fancy-with-parent} then looks at the References (and
 In-Reply-To) header of each message to split and searches the file
 specified by @code{nnmail-message-id-cache-file} for the message ids.
-When it has found a parent, it returns the corresponding group name.  It
-is recommended that you set @code{nnmail-message-id-cache-length} to a
-somewhat higher number than the default so that the message ids are
-still in the cache.  (A value of 5000 appears to create a file some 300
-kBytes in size.)
+When it has found a parent, it returns the corresponding group name
+unless the group name matches the regexp
+@code{nnmail-split-fancy-with-parent-ignore-groups}.  It is recommended
+that you set @code{nnmail-message-id-cache-length} to a somewhat higher
+number than the default so that the message ids are still in the cache.
+(A value of 5000 appears to create a file some 300 kBytes in size.)
 @vindex nnmail-cache-accepted-message-ids
 When @code{nnmail-cache-accepted-message-ids} is non-@code{nil}, Gnus
 also records the message ids of moved articles, so that the followup

[-- Attachment #4: Type: text/plain, Size: 473 bytes --]


and

2001-07-25 22:22:22  Raymond Scholz  <rscholz@zonix.de>

	* nnmail.el (nnmail-split-fancy-with-parent-ignore-groups): New variable.
	(nnmail-split-fancy-with-parent): Ignore certain groups.


and

2001-07-25 22:22:22  Raymond Scholz  <rscholz@zonix.de>

	* gnus.texi (Fancy Mail Splitting): New variable
        nnmail-split-fancy-with-parent-ignore-groups

Cheers, Ray
-- 
Q: What have american beer and sex in a canoe in common?
A: Both are fucking close to water.

Re: Limit nnmail-split-fancy-with-parent to certain groups?

[ShengHuo ZHU]

Raymond Scholz <ray-2001@zonix.de> writes:

> Kai.Grossjohann@CS.Uni-Dortmund.DE (Kai Großjohann) writes:
> 
>> Can you make that change and submit a patch?
> 
> Most probably I screwed something up, so please have a look at the
> code.  It's stolen from nnmail-list-identifiers's usage.

> and
> 
> 2001-07-25 22:22:22  Raymond Scholz  <rscholz@zonix.de>
> 
> 	* nnmail.el (nnmail-split-fancy-with-parent-ignore-groups): New variable.
> 	(nnmail-split-fancy-with-parent): Ignore certain groups.
> 
> 
> and
> 
> 2001-07-25 22:22:22  Raymond Scholz  <rscholz@zonix.de>
> 
> 	* gnus.texi (Fancy Mail Splitting): New variable
>         nnmail-split-fancy-with-parent-ignore-groups

Installed.

ShengHuo

Re: Limit nnmail-split-fancy-with-parent to certain groups?

[Raymond Scholz]

Raymond Scholz <ray-2001@zonix.de> writes:

> Most probably I screwed something up, so please have a look at the
> code.

Ouch, I certainly did.  And I can't find the error.  Symptom: messages
with references (in References: or In-Reply-To:) but no matching
message-id in the nnmail cache end up in bogus.
`nnmail-split-fancy-with-parent' obviously returns some return value
it should not return.

,----[ *Messages* ]
| Error in `nnmail-split-methods'; using `bogus' mail group
| This message would go to bogus
`----

The error must be in this part of `nnmail-split-fancy-with-parent':

  		(when (or (string= "drafts" res)
  			  (and regexp (string-match regexp res)))
  		    (setq res nil)))

I really feel bad for providing a buggy feature and not knowing how to
fix it :(
           
Cheers, Ray
-- 
WYSIWYG is a step backwards. Human labor is used to do that which the
computer can do better.                         (Andrew S. Tanenbaum)

Re: Limit nnmail-split-fancy-with-parent to certain groups?

[Kai Großjohann]

On Sat, 04 Aug 2001, Raymond Scholz wrote:

> The error must be in this part of `nnmail-split-fancy-with-parent':
> 
>   		(when (or (string= "drafts" res)
>   			  (and regexp (string-match regexp res)))
>   		    (setq res nil)))
> 
> I really feel bad for providing a buggy feature and not knowing how
> to fix it :(

I think the error was in the let statement which computed the value of
regexp.  The mapconcat syntax was bogus.

I've tried to fix it, but I now use nnimap, so I can't test that part
of the code.  Please try it:

  - with a list of regexes in
    nnmail-split-fancy-with-parent-ignore-groups
  - with a single regex in the variable

New version in CVS.

Thanks,
kai
-- 
~/.signature: No such file or directory

Re: Limit nnmail-split-fancy-with-parent to certain groups?

[Raymond Scholz]

Kai.Grossjohann@CS.Uni-Dortmund.DE (Kai Großjohann) writes:

> I think the error was in the let statement which computed the value of
> regexp.  The mapconcat syntax was bogus.

Ah, does that mean `nnmail-remove-list-identifiers' is buggy too?
 
> I've tried to fix it, but I now use nnimap, so I can't test that part
> of the code.  Please try it:
> 
>   - with a list of regexes in
>     nnmail-split-fancy-with-parent-ignore-groups
>   - with a single regex in the variable

Sorry, I still get the same error message as before.

Cheers, Ray
-- 
Hfr fgebat rapelcgvba!

Re: Limit nnmail-split-fancy-with-parent to certain groups?

[Kai Großjohann]

Raymond Scholz <ray-2001@zonix.de> writes:

> Kai.Grossjohann@CS.Uni-Dortmund.DE (Kai Großjohann) writes:
> 
>> I think the error was in the let statement which computed the value of
>> regexp.  The mapconcat syntax was bogus.
> 
> Ah, does that mean `nnmail-remove-list-identifiers' is buggy too?

No, you made a typo.  You changed from

    (mapconcat 'identity x "bla")

to

    (mapconcat 'x "bla")

>> I've tried to fix it, but I now use nnimap, so I can't test that part
>> of the code.  Please try it:
>> 
>>   - with a list of regexes in
>>     nnmail-split-fancy-with-parent-ignore-groups
>>   - with a single regex in the variable
> 
> Sorry, I still get the same error message as before.

Argh.

kai
-- 
~/.signature: No such file or directory

Re: Limit nnmail-split-fancy-with-parent to certain groups?

[Kai Großjohann]

Raymond Scholz <ray-2001@zonix.de> writes:

> Sorry, I still get the same error message as before.

Just a sanity check: did you remember to byte-compile?

kai
-- 
~/.signature: No such file or directory

Re: Limit nnmail-split-fancy-with-parent to certain groups?

[Kai Großjohann]

Raymond Scholz <ray-2001@zonix.de> writes:

> Ouch, I certainly did.  And I can't find the error.  Symptom: messages
> with references (in References: or In-Reply-To:) but no matching
> message-id in the nnmail cache end up in bogus.
> `nnmail-split-fancy-with-parent' obviously returns some return value
> it should not return.

Please check whether this is indeed the case.  Go to a message which
was put in bogus.  Do C-u g on it.  Go to the article buffer.  Invoke
nnmail-split-fancy-with-parent:

    M-: (nnmail-split-fancy-with-parent) RET

Does it return nil, as it should?

kai
-- 
~/.signature: No such file or directory

Re: Limit nnmail-split-fancy-with-parent to certain groups?

[Raymond Scholz]

Kai.Grossjohann@CS.Uni-Dortmund.DE (Kai Großjohann) writes:

> Please check whether this is indeed the case.  Go to a message which
> was put in bogus.  Do C-u g on it.  Go to the article buffer.  Invoke
> nnmail-split-fancy-with-parent:
> 
>     M-: (nnmail-split-fancy-with-parent) RET
> 
> Does it return nil, as it should?

Gah, I think I got it.  Thanks for showing this way of debugging the
function.

The string-match will fail if res is nil.

                 (when (or (string= "drafts" res)
                           (and regexp (string-match regexp res)))
                     (setq res nil)))

So it should look like this:

                 (when (or (string= "drafts" res)
                           (and res regexp (string-match regexp res)))
                     (setq res nil)))
                     

Right?

Cheers, Ray
-- 
Look Ma, this man can twist his fingers as if they were made of rubber,
isn't that amazing? -- Not really, he's been using Emacs for years...!

Re: Limit nnmail-split-fancy-with-parent to certain groups?

[Kai Großjohann]

Raymond Scholz <ray-2001@zonix.de> writes:

> The string-match will fail if res is nil.

Good shot!  New version in CVS.  Please test.

kai
-- 
~/.signature: No such file or directory

Re: Limit nnmail-split-fancy-with-parent to certain groups?

[Raymond Scholz]

Kai.Grossjohann@CS.Uni-Dortmund.DE (Kai Großjohann) writes:

> Good shot!  New version in CVS.  Please test.

Works.

Re: Limit nnmail-split-fancy-with-parent to certain groups?

[ShengHuo ZHU]

Raymond Scholz <ray-2001@zonix.de> writes:

[...]

> Furthermore I suspect this feature not to work as described in the Info
> documentation:
> 
> ,----
>| When `nnmail-cache-accepted-message-ids' is non-`nil', Gnus
>| also records the message ids of moved articles, so that the followup
>| messages goes into the new group.
> `----
> 
> I've moved a personal reply from folder A into another folder B, went
> to B, composed my reply (which ends up in B too because of gcc-self)
> but the answer to my reply ended up in A.  Can anyone confirm this?

I think I've fixed this before, but maybe only for some backends.
What kind of backends are you using?

ShengHuo

Re: Limit nnmail-split-fancy-with-parent to certain groups?

[Raymond Scholz]

ShengHuo ZHU <zsh@cs.rochester.edu> writes:

> I think I've fixed this before, but maybe only for some backends.
> What kind of backends are you using?

Just nnfolder.

Cheers, Ray
-- 
C-x i ~/.signature

Re: Limit nnmail-split-fancy-with-parent to certain groups?

[ShengHuo ZHU]

Raymond Scholz <ray-2001@zonix.de> writes:

> Hi!
> 
> Is there a simple way of limiting the application of
> nnmail-split-fancy-with-parent to certain groups?  Can this be done by
> not recording message-IDs in nnmail-message-id-cache-file for these
> groups?  And if so, how could this be done?
> 
> Background: I overlooked some personal replies in high volume mailing
> lists.
> 
> Furthermore I suspect this feature not to work as described in the Info
> documentation:
> 
> ,----
>| When `nnmail-cache-accepted-message-ids' is non-`nil', Gnus
>| also records the message ids of moved articles, so that the followup
>| messages goes into the new group.
> `----
> 
> I've moved a personal reply from folder A into another folder B, went
> to B, composed my reply (which ends up in B too because of gcc-self)
> but the answer to my reply ended up in A.  Can anyone confirm this?

I just committed a patch of nnfolder.el. Maybe it is related. Please
test.

ShengHuo

Re: Limit nnmail-split-fancy-with-parent to certain groups?

[Raymond Scholz]

ShengHuo ZHU <zsh@cs.rochester.edu> writes:

>> I've moved a personal reply from folder A into another folder B, went
>> to B, composed my reply (which ends up in B too because of gcc-self)
>> but the answer to my reply ended up in A.  Can anyone confirm this?
> 
> I just committed a patch of nnfolder.el. Maybe it is related. Please
> test.

Looks like it's fixed now.  Thanks!

Cheers, Ray
-- 
"The obvious mathematical breakthrough would be development of an easy way
to factor large prime numbers."     (Bill Gates from The Road Ahead, p265)