Changes in Gnus from Aug 17 to Sep 1st cause hang on article-decode-encoded-words.

Changes in Gnus from Aug 17 to Sep 1st cause hang on article-decode-encoded-words.

[Nelson Ferreira]

Hi all,

The subject kinda says it all. Let me know what should I look for in
configuration or test in order to troubleshoot.

I toggled-debug-on-quit and when it hanged it was running
gnus-article-decode-encoded-words from a run-hooks.

It is definitely not the article or configuration, because this
version from Aug 17 has no problem whatsoever.

Thanks,

        Nelson

-- 
Nelson Ferreira

Re: Changes in Gnus from Aug 17 to Sep 1st cause hang on article-decode-encoded-words.

[Nelson Ferreira]

"njsf" == Nelson Ferreira <nelson.ferreira@ieee.org> writes:

    njsf> Hi all,
    njsf> The subject kinda says it all. Let me know what should I look for in
    njsf> configuration or test in order to troubleshoot.

    njsf> I toggled-debug-on-quit and when it hanged it was running
    njsf> gnus-article-decode-encoded-words from a run-hooks.

    njsf> It is definitely not the article or configuration, because this
    njsf> version from Aug 17 has no problem whatsoever.


Complementing...

I believe this is probably the offending change...

Any clues why it has this bad behavior in SXEmacs (And I suspect
XEmacs 21.4 too) ?

Index: gnus-art.el
===================================================================
RCS file: /usr/local/cvsroot/gnus/lisp/gnus-art.el,v
retrieving revision 7.186
diff -u -r7.186 gnus-art.el
--- gnus-art.el 22 Jun 2006 08:54:45 -0000      7.186
+++ gnus-art.el 2 Sep 2006 04:45:08 -0000
@@ -2515,10 +2525,23 @@
                             (set-buffer gnus-summary-buffer)
                           (error))
                         gnus-newsgroup-ignored-charsets))
-       (inhibit-read-only t))
+       (inhibit-read-only t)
+       start)
     (save-restriction
       (article-narrow-to-head)
-      (funcall gnus-decode-header-function (point-min)
       (point-max)))))
+      (while (not (eobp))
+       (setq start (point))
+       (if (prog1
+               (looking-at "\
+\\(?:Resent-\\)?\\(?:From\\|Cc\\|To\\|Bcc\\|\\(?:In-\\)?Reply-To\\|Sender\
+\\|Mail-Followup-To\\|Mail-Copies-To\\|Approved\\):")
+             (while (progn
+                      (forward-line)
+                      (if (eobp)
+                          nil
+                        (memq (char-after) '(?\t ? ))))))
+           (funcall gnus-decode-address-function start (point))
+         (funcall gnus-decode-header-function start (point)))))))

-- 
Nelson Ferreira

Re: Changes in Gnus from Aug 17 to Sep 1st cause hang on

[Katsumi Yamaoka]

>>>>> In <m3hczqzxea.fsf@tuxie.homelinux.net> Nelson Ferreira wrote:

njsf> The subject kinda says it all. Let me know what should I look for in
njsf> configuration or test in order to troubleshoot.

njsf> I toggled-debug-on-quit and when it hanged it was running
njsf> gnus-article-decode-encoded-words from a run-hooks.

njsf> It is definitely not the article or configuration, because this
njsf> version from Aug 17 has no problem whatsoever.

> Complementing...

> I believe this is probably the offending change...

> Any clues why it has this bad behavior in SXEmacs (And I suspect
> XEmacs 21.4 too) ?

I'm sorry for that.  I've fixed it in both the v5-10 branch and
the trunk just now.  The cause of this trouble is that the form

(decode-coding-region START (point) CODING-SYSTEM)

moves the point to START in XEmacs but doesn't in Emacs.

Regards,

Re: Changes in Gnus from Aug 17 to Sep 1st cause hang on

[Nelson Ferreira]

"Katsumi" == Katsumi Yamaoka <yamaoka@jpl.org> writes:


[...]

    Katsumi> I'm sorry for that.  I've fixed it in both the v5-10 branch and
    Katsumi> the trunk just now.  The cause of this trouble is that the form

    Katsumi> (decode-coding-region START (point) CODING-SYSTEM)

    Katsumi> moves the point to START in XEmacs but doesn't in Emacs.

Thanks,

        Nelson

-- 
Nelson Ferreira

Re: Changes in Gnus from Aug 17 to Sep 1st cause hang on

[Katsumi Yamaoka]

> "Katsumi" == Katsumi Yamaoka <yamaoka@jpl.org> writes:

> I'm sorry for that.  I've fixed it in both the v5-10 branch and
> the trunk just now.

I've fixed it further.  Please use the latest one if you've
checked it out just after reading my last message.  Sorry.