gnus-summary-edit-article vs. Unicode

gnus-summary-edit-article vs. Unicode

[jidanni]

e runs the command gnus-summary-edit-article
But when used on a Draft, Chinese characters present become \333.
It is much safer to use
D e which runs the command gnus-draft-edit-message

Re: gnus-summary-edit-article vs. Unicode

[Reiner Steib]

On Wed, Jul 22 2009, jidanni@jidanni.org wrote:

> e runs the command gnus-summary-edit-article

In a draft summary buffer, `e' runs `gnus-draft-edit-message'.

> But when used on a Draft, Chinese characters present become \333.
> It is much safer to use
> D e which runs the command gnus-draft-edit-message

... only in a draft summary buffer.  So I'm not sure if I understand
what you mean.

Anyhow, if M-x gnus-summary-edit-article RET results in broken
messages (even in a draft group), it is a bug.

Please give a complete reproducible test case.

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

Re: gnus-summary-edit-article vs. Unicode

[jidanni]

Let's say you make a draft,

 To: zzz
 Subject: 測試
 X-Draft-From: ("nndraft:drafts")
 --text follows this line--
 測試

and save it with message-dont-send (C-c C-d). OK, now later one day
reenter the *Drafts* group, and view that draft. Now put the cursor in
that draft's buffer, and hit
>> e runs the command gnus-summary-edit-article
RS> In a draft summary buffer, `e' runs `gnus-draft-edit-message'.
I know. I'm talking about when you have already put the cursor into the
Draft itself, and then you hit e.
>> Chinese characters present become \333.

Re: gnus-summary-edit-article vs. Unicode

[Katsumi Yamaoka]

[-- Attachment #1: Type: text/plain, Size: 784 bytes --]

>>>>> jidanni@jidanni.org wrote:
> Let's say you make a draft,

>  To: zzz
>  Subject: 測試
>  X-Draft-From: ("nndraft:drafts")
>  --text follows this line--
>  測試

> and save it with message-dont-send (C-c C-d). OK, now later one day
> reenter the *Drafts* group, and view that draft. Now put the cursor in
> that draft's buffer, and hit
>>> e runs the command gnus-summary-edit-article
RS> In a draft summary buffer, `e' runs `gnus-draft-edit-message'.
> I know. I'm talking about when you have already put the cursor into the
> Draft itself, and then you hit e.
>>> Chinese characters present become \333.

The cause is that the "e" key is bound to `gnus-summary-edit-article'
in the article buffer unconditionally.  Here is a patch.  Not tested
thoroughly, though.


[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #2: Type: text/x-patch, Size: 903 bytes --]

--- gnus-art.el~	2009-06-01 11:45:19 +0000
+++ gnus-art.el	2009-07-23 12:15:49 +0000
@@ -4270,7 +4270,7 @@
   "s" gnus-article-show-summary
   "\C-c\C-m" gnus-article-mail
   "?" gnus-article-describe-briefly
-  "e" gnus-summary-edit-article
+  "e" gnus-article-read-summary-keys
   "<" beginning-of-buffer
   ">" end-of-buffer
   "\C-c\C-i" gnus-info-find-node
@@ -6374,8 +6374,13 @@
 		 (select-window win))))
 	(setq in-buffer (current-buffer))
 	;; We disable the pick minor mode commands.
-	(if (and (setq func (let (gnus-pick-mode)
-			      (lookup-key (current-local-map) keys)))
+	(if (and (let ((maps (current-minor-mode-maps))
+		       gnus-pick-mode)
+		   (while maps
+		     (when (setq func (lookup-key (pop maps) keys))
+		       (setq maps nil)))
+		   (or func
+		       (setq func (lookup-key (current-local-map) keys))))
 		 (functionp func)
 		 (condition-case code
 		     (progn

Re: gnus-summary-edit-article vs. Unicode

[Katsumi Yamaoka]

>>>>> Katsumi Yamaoka wrote:
> The cause is that the "e" key is bound to `gnus-summary-edit-article'
> in the article buffer unconditionally.

The "e" key in the article buffer has to run
`gnus-article-read-summary-keys' and it has to find the proper
command for the `gnus-draft-mode' minor mode in the summary buffer.

> Here is a patch.  Not tested thoroughly, though.

I've installed a better fix in the Gnus trunk.  In addition,
I've deleted two old functions `gnus-article-summary-command'
and `gnus-article-summary-command-nosave', that used to be used
before `gnus-article-read-summary-keys' was introduced.