Quoting the active region when doing `S W' ?

Quoting the active region when doing `S W' ?

[Bastien]

Hi Gnusers,

when wide-replying-with-original to an email, the active region is not
taken as the default quoted text but the entire email is quoted instead.
I would expect it to behave like `R' or `F' do.  Did I miss something
obvious here?

Thanks,

-- 
Bastien

Re: Quoting the active region when doing `S W' ?

[Bastien]

Bastien <bzg@altern.org> writes:

> when wide-replying-with-original to an email, the active region is not
> taken as the default quoted text but the entire email is quoted instead.
> I would expect it to behave like `R' or `F' do.  Did I miss something
> obvious here?

So I discovered the solution is to use `C-u R' instead of `S W'.  

Actually, there is both:

  `gnus-summary-reply-with-original'
  `gnus-article-reply-with-original'

and both take care of the active region.

But there is only 

  `gnus-summary-wide-reply-with-original' 

and *no*: 

  `gnus-article-wide-reply-with-original'

Maybe this is a bit confusing?  Or is there any strong argument behind
this restriction?

Regards,

-- 
Bastien

Re: Quoting the active region when doing `S W' ?

[Leo]

On 2007-11-25 21:34 +0000, Bastien wrote:
> Bastien <bzg@altern.org> writes:
>
>> when wide-replying-with-original to an email, the active region is not
>> taken as the default quoted text but the entire email is quoted instead.
>> I would expect it to behave like `R' or `F' do.  Did I miss something
>> obvious here?
>
> So I discovered the solution is to use `C-u R' instead of `S W'.  
>
> Actually, there is both:
>
>   `gnus-summary-reply-with-original'
>   `gnus-article-reply-with-original'
>
> and both take care of the active region.
>
> But there is only 
>
>   `gnus-summary-wide-reply-with-original' 
>
> and *no*: 
>
>   `gnus-article-wide-reply-with-original'
>
> Maybe this is a bit confusing?  Or is there any strong argument behind
> this restriction?
>
> Regards,

What a coincidence! I have noted the same problem in:

 gmane.emacs.gnus.user:9031

HTH,
-- 
.:  Leo  :.  [ sdl.web AT gmail.com ]  .:  [ GPG Key: 9283AA3F ]  :.

          Use the best OS -- http://www.fedoraproject.org/

Re: Quoting the active region when doing `S W' ?

[Bastien]

Leo <sdl.web@gmail.com> writes:

>> Maybe this is a bit confusing?  Or is there any strong argument behind
>> this restriction?
>>
>> Regards,
>
> What a coincidence! I have noted the same problem in:
>
>  gmane.emacs.gnus.user:9031

:)

I didn't know about gnmane.emacs.gnus.user -- is it reachable from a
mailing list (since ding@gnus.org is not sync'ed with it) ?   

BTW, does anyone here use a local news server with Gnus?  I've given a
shot to inn2, but any relevant pointer to get some help on how quickly
setup inn2/suck and Gnus (and maybe nntprss..) work together would be
nice.  

Thanks!

-- 
Bastien

Re: Quoting the active region when doing `S W' ?

[Leo]

On 2007-11-26 00:36 +0000, Bastien wrote:
> :)
>
> I didn't know about gnmane.emacs.gnus.user -- is it reachable from a
> mailing list (since ding@gnus.org is not sync'ed with it) ?   

Yes.

info-gnus-english@gnu.org

HTH,
-- 
.:  Leo  :.  [ sdl.web AT gmail.com ]  .:  [ GPG Key: 9283AA3F ]  :.

          Use the best OS -- http://www.fedoraproject.org/

Re: Quoting the active region when doing `S W' ?

[Bastien]

Does this have been taken care of?

Thanks,

Leo <sdl.web@gmail.com> writes:

> On 2007-11-25 21:34 +0000, Bastien wrote:
>> Bastien <bzg@altern.org> writes:
>>
>>> when wide-replying-with-original to an email, the active region is not
>>> taken as the default quoted text but the entire email is quoted instead.
>>> I would expect it to behave like `R' or `F' do.  Did I miss something
>>> obvious here?
>>
>> So I discovered the solution is to use `C-u R' instead of `S W'.  
>>
>> Actually, there is both:
>>
>>   `gnus-summary-reply-with-original'
>>   `gnus-article-reply-with-original'
>>
>> and both take care of the active region.
>>
>> But there is only 
>>
>>   `gnus-summary-wide-reply-with-original' 
>>
>> and *no*: 
>>
>>   `gnus-article-wide-reply-with-original'
>>
>> Maybe this is a bit confusing?  Or is there any strong argument behind
>> this restriction?
>>
>> Regards,
>
> What a coincidence! I have noted the same problem in:
>
>  gmane.emacs.gnus.user:9031
>
> HTH,

-- 
Bastien

Re: Quoting the active region when doing `S W' ?

[Leo]

Hi There,

On 2007-12-10 00:50 +0000, Bastien wrote:
> Does this have been taken care of?
>
> Thanks,
[...]

This feature is very useful; hope someone get around to do it.

Thanks,
-- 
.:  Leo  :.  [ sdl.web AT gmail.com ]  .:  [ GPG Key: 9283AA3F ]  :.

          Use the best OS -- http://www.fedoraproject.org/

Re: Quoting the active region when doing `S W' ?

[Katsumi Yamaoka]

[-- Attachment #1: Type: text/plain, Size: 804 bytes --]

>>>>> Leo wrote:
> On 2007-12-10 00:50 +0000, Bastien wrote:
>> Does this have been taken care of?

> This feature is very useful; hope someone get around to do it.

`C-u R' in the article buffer runs wide reply (you can also use
a numeric prefix like `1' instead of `C-u'):

(info "(gnus)Article Keymap")
,----
| `R'
|      Send a reply to the current article and yank the current article
|      (`gnus-article-reply-with-original').  If given a prefix, make a
|      wide reply.  If the region is active, only yank the text in the
|      region.
`----

However, the prefix argument for the `R' command in the summary
buffer has a different meaning.  I'd like to suggest adding

`SW' (gnus-article-wide-reply-with-original)

and removing the prefix argument from `g-a-reply-with-original'.
Objections?


[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #2: Patch for the Gnus trunk --]
[-- Type: text/x-patch, Size: 6547 bytes --]

--- lisp/gnus-art.el~	2007-12-19 08:40:24 +0000
+++ lisp/gnus-art.el	2008-01-08 11:51:49 +0000
@@ -4225,6 +4225,13 @@
 (substitute-key-definition
  'undefined 'gnus-article-read-summary-keys gnus-article-mode-map)
 
+(gnus-define-keys (gnus-article-send-map "S" gnus-article-mode-map)
+  "W" gnus-article-wide-reply-with-original)
+(if (featurep 'xemacs)
+    (set-keymap-default-binding gnus-article-send-map
+				'gnus-article-read-summary-send-keys)
+  (define-key gnus-article-send-map [t] 'gnus-article-read-summary-send-keys))
+
 (defun gnus-article-make-menu-bar ()
   (unless (boundp 'gnus-article-commands-menu)
     (gnus-summary-make-menu-bar))
@@ -6243,8 +6250,9 @@
     (save-excursion
       (set-buffer gnus-article-current-summary)
       (let (gnus-pick-mode)
-	(push (or key last-command-event) unread-command-events)
-	(setq keys (if (featurep 'xemacs)
+	(setq unread-command-events (nconc unread-command-events
+					   (list (or key last-command-event)))
+	      keys (if (featurep 'xemacs)
 		       (events-to-keys (read-key-sequence nil))
 		     (read-key-sequence nil)))))
 
@@ -6339,24 +6347,30 @@
 	      (signal (car err) (cdr err))
 	    (ding))))))))
 
+(defun gnus-article-read-summary-send-keys ()
+  (interactive)
+  (let ((unread-command-events (list (if (featurep 'xemacs)
+					 (character-to-event ?S)
+				       ?S))))
+    (gnus-article-read-summary-keys)))
+
 (defun gnus-article-describe-key (key)
   "Display documentation of the function invoked by KEY.  KEY is a string."
   (interactive "kDescribe key: ")
   (gnus-article-check-buffer)
-  (if (eq (key-binding key) 'gnus-article-read-summary-keys)
+  (if (memq (key-binding key t) '(gnus-article-read-summary-keys
+				  gnus-article-read-summary-send-keys))
       (save-excursion
 	(set-buffer gnus-article-current-summary)
 	(let (gnus-pick-mode)
 	  (if (featurep 'xemacs)
-	      (progn
-		(push (elt key 0) unread-command-events)
-		(setq key (events-to-keys
-			   (read-key-sequence "Describe key: "))))
+	      (setq unread-command-events (append key nil)
+		    key (events-to-keys (read-key-sequence nil)))
 	    (setq unread-command-events
 		  (mapcar
 		   (lambda (x) (if (>= x 128) (list 'meta (- x 128)) x))
 		   (string-to-list key)))
-	    (setq key (read-key-sequence "Describe key: "))))
+	    (setq key (read-key-sequence nil))))
 	(describe-key key))
     (describe-key key)))
 
@@ -6364,20 +6378,19 @@
   "Display documentation of the function invoked by KEY.  KEY is a string."
   (interactive "kDescribe key: \nP")
   (gnus-article-check-buffer)
-  (if (eq (key-binding key) 'gnus-article-read-summary-keys)
+  (if (memq (key-binding key t) '(gnus-article-read-summary-keys
+				  gnus-article-read-summary-send-keys))
       (save-excursion
 	(set-buffer gnus-article-current-summary)
 	(let (gnus-pick-mode)
 	  (if (featurep 'xemacs)
-	      (progn
-		(push (elt key 0) unread-command-events)
-		(setq key (events-to-keys
-			   (read-key-sequence "Describe key: "))))
+	      (setq unread-command-events (append key nil)
+		    key (events-to-keys (read-key-sequence nil)))
 	    (setq unread-command-events
 		  (mapcar
 		   (lambda (x) (if (>= x 128) (list 'meta (- x 128)) x))
 		   (string-to-list key)))
-	    (setq key (read-key-sequence "Describe key: "))))
+	    (setq key (read-key-sequence nil))))
 	(describe-key-briefly key insert))
     (describe-key-briefly key insert)))
 
@@ -6385,7 +6398,7 @@
   "Start composing a reply mail to the current message.
 The text in the region will be yanked.  If the region isn't active,
 the entire article will be yanked."
-  (interactive "P")
+  (interactive)
   (let ((article (cdr gnus-article-current))
 	contents)
     (if (not (gnus-region-active-p))
@@ -6400,6 +6413,13 @@
 	(gnus-summary-reply
 	 (list (list article contents)) wide)))))
 
+(defun gnus-article-wide-reply-with-original ()
+  "Start composing a wide reply mail to the current message.
+The text in the region will be yanked.  If the region isn't active,
+the entire article will be yanked."
+  (interactive)
+  (gnus-article-reply-with-original t))
+
 (defun gnus-article-followup-with-original ()
   "Compose a followup to the current article.
 The text in the region will be yanked.  If the region isn't active,
--- lisp/lpath.el~	2007-12-10 01:18:01 +0000
+++ lisp/lpath.el	2008-01-08 11:51:49 +0000
@@ -51,13 +51,15 @@
 		    show-trailing-whitespace temporary-file-directory
 		    timer-list tool-bar-mode transient-mark-mode)))
   (maybe-fbind '(bbdb-complete-name
-		 delete-annotation delete-extent device-connection dfw-device
-		 events-to-keys find-face font-lock-set-defaults frame-device
-		 get-char-table glyph-height glyph-width ldap-search-entries
+		 character-to-event delete-annotation delete-extent
+		 device-connection dfw-device events-to-keys find-face
+		 font-lock-set-defaults frame-device get-char-table
+		 glyph-height glyph-width ldap-search-entries
 		 mail-aliases-setup make-annotation make-event make-glyph
 		 map-extents message-xmas-redefine put-char-table
 		 run-mode-hooks set-extent-property set-itimer-function
-		 temp-directory valid-image-instantiator-format-p
+		 set-keymap-default-binding temp-directory
+		 valid-image-instantiator-format-p
 		 w3-coding-system-for-mime-charset w3-do-setup
 		 w3-prepare-buffer w3-region window-pixel-height
 		 window-pixel-width))
--- texi/gnus.texi~	2008-01-06 22:51:28 +0000
+++ texi/gnus.texi	2008-01-08 11:51:49 +0000
@@ -11900,17 +11900,22 @@
 @kindex R (Article)
 @findex gnus-article-reply-with-original
 Send a reply to the current article and yank the current article
-(@code{gnus-article-reply-with-original}).  If given a prefix, make a
-wide reply.  If the region is active, only yank the text in the
-region.
+(@code{gnus-article-reply-with-original}).  If the region is active,
+only yank the text in the region.
+
+@item S W
+@kindex S W (Article)
+@findex gnus-article-wide-reply-with-original
+Send a wide reply to the current article and yank the current article
+(@code{gnus-article-reply-with-original}).  If the region is active,
+only yank the text in the region.
 
 @item F
 @kindex F (Article)
 @findex gnus-article-followup-with-original
 Send a followup to the current article and yank the current article
-(@code{gnus-article-followup-with-original}).  If given a prefix, make
-a wide reply.  If the region is active, only yank the text in the
-region.
+(@code{gnus-article-followup-with-original}).  If the region is active,
+only yank the text in the region.
 
 
 @end table

Re: Quoting the active region when doing `S W' ?

[Leo]

On 2008-01-08 11:55 +0000, Katsumi Yamaoka wrote:
> `C-u R' in the article buffer runs wide reply (you can also use
> a numeric prefix like `1' instead of `C-u'):
>
> (info "(gnus)Article Keymap")
> ,----
> | `R'
> |      Send a reply to the current article and yank the current article
> |      (`gnus-article-reply-with-original').  If given a prefix, make a
> |      wide reply.  If the region is active, only yank the text in the
> |      region.
> `----

'R' does not reply to all address while 'S W' or 'S V' does.

To see the difference:

,----[ R ]
| To: Katsumi Yamaoka <yamaoka@jpl.org>
| Subject: Re: Quoting the active region when doing `S W' ?
`----

,----[ S W ]
| To: Katsumi Yamaoka <yamaoka@jpl.org>
| Cc: ding@gnus.org
| Subject: Re: Quoting the active region when doing `S W' ?
`----

> However, the prefix argument for the `R' command in the summary
> buffer has a different meaning.  I'd like to suggest adding
>
> `SW' (gnus-article-wide-reply-with-original)
>
> and removing the prefix argument from `g-a-reply-with-original'.
> Objections?

-- 
.:  Leo  :.  [ sdl.web AT gmail.com ]  .:  [ GPG Key: 9283AA3F ]  :.

          Use the best OS -- http://www.fedoraproject.org/

Re: Quoting the active region when doing `S W' ?

[Katsumi Yamaoka]

>>>>> Leo wrote:
> 'R' does not reply to all address while 'S W' or 'S V' does.

Did you try `C-u R' (or `1 R'), not `R', in the article buffer?

> On 2008-01-08 11:55 +0000, Katsumi Yamaoka wrote:
>> `C-u R' in the article buffer runs wide reply (you can also use
>> a numeric prefix like `1' instead of `C-u'):
>>
>> (info "(gnus)Article Keymap")
>> ,----
>>| `R'
>>|      Send a reply to the current article and yank the current article
>>|      (`gnus-article-reply-with-original').  If given a prefix, make a
                                                *************************
>>|      wide reply.  If the region is active, only yank the text in the
         **********
>>|      region.
>> `----

[...]

>> However, the prefix argument for the `R' command in the summary
>> buffer has a different meaning.  I'd like to suggest adding
>>
>> `SW' (gnus-article-wide-reply-with-original)
>>
>> and removing the prefix argument from `g-a-reply-with-original'.
>> Objections?

A patch that realizes it was in my last article
(<b4mhchobkfx.fsf@jpl.org> or
 <URL:http://article.gmane.org/gmane.emacs.gnus.general/66083>).
I'll install it with a slight modification if no one comments.

Re: Quoting the active region when doing `S W' ?

[Leo]

On 2008-01-08 22:51 +0000, Katsumi Yamaoka wrote:
> A patch that realizes it was in my last article
> (<b4mhchobkfx.fsf@jpl.org> or
>  <URL:http://article.gmane.org/gmane.emacs.gnus.general/66083>).
> I'll install it with a slight modification if no one comments.

I am looking forward to it being applied.

-- 
.:  Leo  :.  [ sdl.web AT gmail.com ]  .:  [ GPG Key: 9283AA3F ]  :.

          Use the best OS -- http://www.fedoraproject.org/

Re: Quoting the active region when doing `S W' ?

[Katsumi Yamaoka]

>>>>> Leo wrote:
> On 2008-01-08 22:51 +0000, Katsumi Yamaoka wrote:
>> I'll install it with a slight modification if no one comments.

> I am looking forward to it being applied.

Installed.  Please test `S W' in the article buffer.

There is one problem; XEmacs doesn't show all the S-prefixed keys
when hitting `S C-h', which invokes `describe-bindings', in the
article buffer.

Re: Quoting the active region when doing `S W' ?

[Reiner Steib]

On Tue, Jan 08 2008, Katsumi Yamaoka wrote:

> However, the prefix argument for the `R' command in the summary
> buffer has a different meaning.  I'd like to suggest adding
> `SW' (gnus-article-wide-reply-with-original)
> and removing the prefix argument from `g-a-reply-with-original'.
> Objections?

I'm not sure if I understand the differences in the old and new
behavior of `C-u R' and `S W'.

The behavior of `C-u R' was:

  Summary: Reply to author of the current article,
           citing the next 4 articles.
           Ignore active region.

  Article: Reply to all (wide reply),
           citing the current article.
           Ignore active region.

The behavior of `S W' was:

  Summary: Reply to all (wide reply),
           citing the current article.
           Ignore active region.

  Article: Ditto.

Correct?  Could you summarize the new behavior, too, and point out the
differences, please?

I think the ChangeLog entry might more clear with the following
changes.  May I change it?

| -	* gnus-art.el (gnus-article-send-map): New keymap for S prefix keys;
| -	bind SW key to gnus-article-wide-reply-with-original; set default
| +	* gnus-art.el (gnus-article-send-map): New keymap for `S' prefix keys;
| +	bind `S W' to gnus-article-wide-reply-with-original; set default
|  	binding to gnus-article-read-summary-send-keys.
|  	(gnus-article-read-summary-keys): Fix the order of keys; display
|  	continuation keys correctly in the echo area; describe bindings
| -	correctly when keys end with C-h.
| +	correctly when keys end with `C-h'.

Bye, Reiner.
-- 
       ,,,
      (o o)
---ooO-(_)-Ooo---  |  PGP key available  |  http://rsteib.home.pages.de/

Re: Quoting the active region when doing `S W' ?

[Katsumi Yamaoka]

>>>>> Reiner Steib wrote:
> On Tue, Jan 08 2008, Katsumi Yamaoka wrote:

>> However, the prefix argument for the `R' command in the summary
>> buffer has a different meaning.  I'd like to suggest adding
>> `SW' (gnus-article-wide-reply-with-original)
>> and removing the prefix argument from `g-a-reply-with-original'.
>> Objections?

> I'm not sure if I understand the differences in the old and new
> behavior of `C-u R' and `S W'.

> The behavior of `C-u R' was:

>   Summary: Reply to author of the current article,
>            citing the next 4 articles.
>            Ignore active region.

>   Article: Reply to all (wide reply),
>            citing the current article.
>            Ignore active region.

In the version in the Gnus trunk now the `R' command in the article
buffer ignores the prefix argument.  It is useless to make it behave
like the one in the summary buffer, isn't it?  Or should it behave
with the prefix argument as before (i.e., wide reply)?

> The behavior of `S W' was:

>   Summary: Reply to all (wide reply),
>            citing the current article.
>            Ignore active region.

>   Article: Ditto.

> Correct?

Yes.  But in the article buffer, only a text in the active region,
if it is set, is yanked.

> Could you summarize the new behavior, too, and point out the
> differences, please?

> I think the ChangeLog entry might more clear with the following
> changes.  May I change it?

>| -	* gnus-art.el (gnus-article-send-map): New keymap for S prefix keys;
>| -	bind SW key to gnus-article-wide-reply-with-original; set default
>| +	* gnus-art.el (gnus-article-send-map): New keymap for `S' prefix keys;
>| +	bind `S W' to gnus-article-wide-reply-with-original; set default
>|  	binding to gnus-article-read-summary-send-keys.
>|  	(gnus-article-read-summary-keys): Fix the order of keys; display
>|  	continuation keys correctly in the echo area; describe bindings
>| -	correctly when keys end with C-h.
>| +	correctly when keys end with `C-h'.

Please do it.

Re: Quoting the active region when doing `S W' ?

[Katsumi Yamaoka]

>>>>> Katsumi Yamaoka wrote:
>>>>>> Reiner Steib wrote:
>> I think the ChangeLog entry might more clear with the following
>> changes.  May I change it?

>>| -	* gnus-art.el (gnus-article-send-map): New keymap for S prefix keys;
>>| -	bind SW key to gnus-article-wide-reply-with-original; set default
>>| +	* gnus-art.el (gnus-article-send-map): New keymap for `S' prefix keys;
>>| +	bind `S W' to gnus-article-wide-reply-with-original; set default
>>|  	binding to gnus-article-read-summary-send-keys.
>>|  	(gnus-article-read-summary-keys): Fix the order of keys; display
>>|  	continuation keys correctly in the echo area; describe bindings
>>| -	correctly when keys end with C-h.
>>| +	correctly when keys end with `C-h'.

> Please do it.

I've already done it together with the other changes, that fixed
the problem I mentioned as follows:

> There is one problem; XEmacs doesn't show all the S-prefixed keys
> when hitting `S C-h', which invokes `describe-bindings', in the
> article buffer.

Regards,

Re: Quoting the active region when doing `S W' ?

[Reiner Steib]

On Thu, Jan 10 2008, Katsumi Yamaoka wrote:

>>>>>> Reiner Steib wrote:
>> The behavior of `C-u R' was:
>
>>   Summary: Reply to author of the current article,
>>            citing the next 4 articles.
>>            Ignore active region.
>
>>   Article: Reply to all (wide reply),
>>            citing the current article.
>>            Ignore active region.
>
> In the version in the Gnus trunk now the `R' command in the article
> buffer ignores the prefix argument.  It is useless to make it behave
> like the one in the summary buffer, isn't it?  Or should it behave
> with the prefix argument as before (i.e., wide reply)?
[...]
>> The behavior of `S W' was:
>
>>   Summary: Reply to all (wide reply),
>>            citing the current article.
>>            Ignore active region.
>
>>   Article: Ditto.
>
>> Correct?
>
> Yes.  But in the article buffer, only a text in the active region,
> if it is set, is yanked.

I think your change is fine.  (But I never used `C-u R'.)  Could you
please add a brief entry about this in `texi/gnus-news.texi'?

Bye, Reiner.
-- 
       ,,,
      (o o)
---ooO-(_)-Ooo---  |  PGP key available  |  http://rsteib.home.pages.de/

Re: Quoting the active region when doing `S W' ?

[Katsumi Yamaoka]

>>>>> Reiner Steib wrote:

> I think your change is fine.  (But I never used `C-u R'.)  Could you
> please add a brief entry about this in `texi/gnus-news.texi'?

I've added it as follows:

,----
| ** Now the new command `S W' (`gnus-article-wide-reply-with-original') for
| a wide reply in the article buffer yanks a text that is in the active
| region, if it is set, as well as the `R'
| (`gnus-article-reply-with-original') command.  Note that the `R' command
| in the article buffer no longer accepts a prefix argument, which was
| used to make it do a wide reply.  *Note Article Keymap::.
`----

In addition, I've added the new article command which the `C-h b'
key is bound to (`gnus-article-describe-bindings').  It shows
the real summary commands instead of `gnus-article-read-summary-key'
that is meaningless for a user.

Regards,

Re: Quoting the active region when doing `S W' ?

[Reiner Steib]

On Fri, Jan 11 2008, Katsumi Yamaoka wrote:

> I've added it as follows:
> ,----
> | ** Now the new command `S W' (`gnus-article-wide-reply-with-original') for
[...]

Thanks.

> In addition, I've added the new article command which the `C-h b'
> key is bound to (`gnus-article-describe-bindings').  It shows the
> real summary commands instead of `gnus-article-read-summary-key'
> that is meaningless for a user.

Cool!  Please also add it to `gnus-news.texi'.  Thanks.

Bye, Reiner.
-- 
       ,,,
      (o o)
---ooO-(_)-Ooo---  |  PGP key available  |  http://rsteib.home.pages.de/

Re: Quoting the active region when doing `S W' ?

[Leo]

On 2008-01-09 22:30 +0000, Katsumi Yamaoka wrote:
>> I am looking forward to it being applied.
>
> Installed.  Please test `S W' in the article buffer.

I've just updated my Emacs to the latest CVS. And now this command
supports partial citation. This is pretty cool.

Can you also extend `S V' to behave similarly?

> There is one problem; XEmacs doesn't show all the S-prefixed keys
> when hitting `S C-h', which invokes `describe-bindings', in the
> article buffer.

Thanks,
-- 
.:  Leo  :.  [ sdl.web AT gmail.com ]  .:  [ GPG Key: 9283AA3F ]  :.

          Use the best OS -- http://www.fedoraproject.org/

Re: Quoting the active region when doing `S W' ?

[Katsumi Yamaoka]

>>>>> Leo wrote:

> Can you also extend `S V' to behave similarly?

How is `S V' used in the article buffer?  AFAICS, `S V' in the
summary buffer is for gathering recipient addresses in two or
more articles that are marked.  If it is run for only one
article, it works like `S W'.  Am I wrong?

Re: Quoting the active region when doing `S W' ?

[Katsumi Yamaoka]

>>>>> Reiner Steib wrote:

>> In addition, I've added the new article command which the `C-h b'
>> key is bound to (`gnus-article-describe-bindings').  It shows the
>> real summary commands instead of `gnus-article-read-summary-key'
>> that is meaningless for a user.

> Cool!  Please also add it to `gnus-news.texi'.  Thanks.

Thanks.  I'll add it but after improving the function (now it
doesn't show all the S-prefixed commands).

Re: Quoting the active region when doing `S W' ?

[Leo]

On 2008-01-15 00:12 +0000, Katsumi Yamaoka wrote:
>>>>>> Leo wrote:
>
>> Can you also extend `S V' to behave similarly?
>
> How is `S V' used in the article buffer?  AFAICS, `S V' in the
> summary buffer is for gathering recipient addresses in two or
> more articles that are marked.  If it is run for only one
> article, it works like `S W'.  Am I wrong?

It looks like so but it does not support partial citation as 'S W' now
does.

-- 
.:  Leo  :.  [ sdl.web AT gmail.com ]  .:  [ GPG Key: 9283AA3F ]  :.

          Use the best OS -- http://www.fedoraproject.org/

Re: Quoting the active region when doing `S W' ?

[Katsumi Yamaoka]

>>>>> Leo wrote:
> On 2008-01-15 00:12 +0000, Katsumi Yamaoka wrote:
>>>>>>> Leo wrote:
>>> Can you also extend `S V' to behave similarly?

>> How is `S V' used in the article buffer?  AFAICS, `S V' in the
>> summary buffer is for gathering recipient addresses in two or
>> more articles that are marked.  If it is run for only one
>> article, it works like `S W'.  Am I wrong?

> It looks like so but it does not support partial citation as 'S W' now
> does.

Is it okay to bind `S V' to the same function which `S W' is
bound to?  Though I'm not sure it is the right thing.  Though it
might be useful to a user who usually uses `S V' even for replying
to a single article, I think that he had better get used to use
`S W' and should learn that `S V' is special.  WDYT?

Re: Quoting the active region when doing `S W' ?

[Leo]

On 2008-01-15 01:15 +0000, Katsumi Yamaoka wrote:
>>>>>> Leo wrote:
>> On 2008-01-15 00:12 +0000, Katsumi Yamaoka wrote:
>>>>>>>> Leo wrote:
>>>> Can you also extend `S V' to behave similarly?
>
>>> How is `S V' used in the article buffer?  AFAICS, `S V' in the
>>> summary buffer is for gathering recipient addresses in two or
>>> more articles that are marked.  If it is run for only one
>>> article, it works like `S W'.  Am I wrong?
>
>> It looks like so but it does not support partial citation as 'S W' now
>> does.
>
> Is it okay to bind `S V' to the same function which `S W' is
> bound to?  Though I'm not sure it is the right thing.  Though it
> might be useful to a user who usually uses `S V' even for replying
> to a single article, I think that he had better get used to use
> `S W' and should learn that `S V' is special.  WDYT?

Could someone comment on this?

-- 
.:  Leo  :.  [ sdl.web AT gmail.com ]  .:  [ GPG Key: 9283AA3F ]  :.

          Use the best OS -- http://www.fedoraproject.org/

Re: Quoting the active region when doing `S W' ?

[Bastien]

[-- Attachment #1: Type: text/plain, Size: 450 bytes --]

Hi Katsumi,

Katsumi Yamaoka <yamaoka@jpl.org> writes:

>>>>>> Leo wrote:
>> On 2007-12-10 00:50 +0000, Bastien wrote:
>>> Does this have been taken care of?
>
>> This feature is very useful; hope someone get around to do it.
>
> `C-u R' in the article buffer runs wide reply (you can also use
> a numeric prefix like `1' instead of `C-u'):

Thanks for this. 

But for me (No Gnus v0.7) `C-u R' didn't make a wide reply.  I had to
apply this patch:


[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #2: gnus-msg.el.patch --]
[-- Type: text/x-diff, Size: 733 bytes --]

diff -u /home/guerry/install/cvs/gnus/lisp/ /home/guerry/elisp/testing/bzg/gnus-msg.el
--- /home/guerry/install/cvs/gnus/lisp/gnus-msg.el	2007-11-26 00:04:09.000000000 +0000
+++ /home/guerry/elisp/testing/bzg/gnus-msg.el	2008-01-16 18:43:13.000000000 +0000
@@ -1160,7 +1160,8 @@
   "Start composing a reply mail to the current message.
 The original article will be yanked."
   (interactive "P")
-  (gnus-summary-reply (gnus-summary-work-articles n) wide))
+  (gnus-summary-reply (gnus-summary-work-articles n) 
+		      (or current-prefix-arg wide)))
 
 (defun gnus-summary-reply-broken-reply-to (&optional yank wide very-wide)
   "Like `gnus-summary-reply' except removing reply-to field.

Diff finished.  Wed Jan 16 18:43:24 2008

[-- Attachment #3: Type: text/plain, Size: 402 bytes --]


> (info "(gnus)Article Keymap")
> ,----
> | `R'
> |      Send a reply to the current article and yank the current article
> |      (`gnus-article-reply-with-original').  If given a prefix, make a
> |      wide reply.  If the region is active, only yank the text in the
> |      region.
> `----

Maybe I didn't digg enough into this, but what is the difference between
`C-u R' and `S W' ?

-- 
Bastien

Re: Quoting the active region when doing `S W' ?

[Katsumi Yamaoka]

>>>>> Bastien wrote:
> Katsumi Yamaoka <yamaoka@jpl.org> writes:

>> `C-u R' in the article buffer runs wide reply (you can also use
>> a numeric prefix like `1' instead of `C-u'):

> But for me (No Gnus v0.7) `C-u R' didn't make a wide reply.

[...]

You meant `gnus-summary-reply-with-original' which is used in
the *SUMMARY* buffer.  In that case, the prefix argument is
interpreted as the number of articles to be replied.  I.e.,
since `C-u' is interpreted into 4, `C-u R' means replying to a
series of four articles.

cf. (info "(gnus)Summary Mail Commands")
    (info "(gnus)Process/Prefix")

OTOH,

>> (info "(gnus)Article Keymap")
>> ,----
>>| `R'
>>|      Send a reply to the current article and yank the current article
>>|      (`gnus-article-reply-with-original').  If given a prefix, make a
>>|      wide reply.  If the region is active, only yank the text in the
>>|      region.
>> `----

this is `gnus-article-reply-with-original' which is used in the
*ARTICLE* buffer.  In the Gnus CVS trunk, I've changed it so as
not to accept the prefix argument.

,---- [ GNUS-NEWS ]
| ** Now the new command `S W' (`gnus-article-wide-reply-with-original') for
| a wide reply in the article buffer yanks a text that is in the active
| region, if it is set, as well as the `R'
| (`gnus-article-reply-with-original') command.  Note that the `R' command
| in the article buffer no longer accepts a prefix argument, which was
| used to make it do a wide reply.  *Note Article Keymap::.
`----

Regards,

Re: Quoting the active region when doing `S W' ?

[Bastien]

Katsumi Yamaoka <yamaoka@jpl.org> writes:

>>>>>> Bastien wrote:
>> Katsumi Yamaoka <yamaoka@jpl.org> writes:
>
>>> `C-u R' in the article buffer runs wide reply (you can also use
>>> a numeric prefix like `1' instead of `C-u'):
>
>> But for me (No Gnus v0.7) `C-u R' didn't make a wide reply.
>
> [...]
>
> You meant `gnus-summary-reply-with-original' which is used in
> the *SUMMARY* buffer.  

[snip]

Thanks for the explanations.  I've checked the last CVS HEAD and it's
working as expected now.   

-- 
Bastien

Re: Quoting the active region when doing `S W' ?

[Leo]

On 2008-01-15 01:15 +0000, Katsumi Yamaoka wrote:
>> It looks like so but it does not support partial citation as 'S W' now
>> does.
>
> Is it okay to bind `S V' to the same function which `S W' is
> bound to?  Though I'm not sure it is the right thing.  Though it
> might be useful to a user who usually uses `S V' even for replying
> to a single article, I think that he had better get used to use
> `S W' and should learn that `S V' is special.  WDYT?

Sorry I can't comment on this as I don't really know the differences
between all the reply commands.

-- 
.:  Leo  :.  [ sdl.web AT gmail.com ]  .:  [ GPG Key: 9283AA3F ]  :.

          Use the best OS -- http://www.fedoraproject.org/

Re: Quoting the active region when doing `S W' ?

[Katsumi Yamaoka]

[-- Attachment #1: Type: text/plain, Size: 1004 bytes --]

>>>>> Katsumi Yamaoka wrote:

>>> In addition, I've added the new article command which the `C-h b'
>>> key is bound to (`gnus-article-describe-bindings').  It shows the
>>> real summary commands instead of `gnus-article-read-summary-keys'
>>> that is meaningless for a user.

In relation to this, I tried letting the *article* menu bar items
also display the keyboard equivalents for those items as follows:

Treatment
---------
Hide headers    W W h
Hide signature  W W s
Hide citation   W W c
[...]

(Note that your Gnus doesn't display "W W h", etc. when clicking
the `Treatment' menu bar in the *article* buffer.)

The Lisp code which achieves this is attached below.  I think it
is useful especially for users who haven't gotten used to using
such command keys.  However, I hesitate to install it to the Gnus
CVS trunk because it wastes about one second when beginning to
read an article only for the first time and it works only with
Emacs 22 and greater (and both seem to be hard to solve. :-<)


[-- Attachment #2: Type: application/emacs-lisp, Size: 1066 bytes --]