No Gnus on Sunday

No Gnus on Sunday

[Lars Magne Ingebrigtsen]

So if y'all have any last-minute bug fixes for Gnus 5.10, please get
them in during the next 36 hours...

-- 
(domestic pets only, the antidote for overdose, milk.)
  larsi@gnus.org * Lars Magne Ingebrigtsen

Re: No Gnus on Sunday

[Steve Youngs]

[-- Attachment #1: Type: text/plain, Size: 1905 bytes --]

|--==> "LMI" == Lars Magne Ingebrigtsen <larsi@gnus.org> writes:

  LMI> So if y'all have any last-minute bug fixes for Gnus 5.10,
  LMI> please get them in during the next 36 hours...

Quick!  Somebody give me a way of implementing `with-timeout' in
XEmacs _without_ relying on timer.el.

In mm-url.el there is...

(defun mm-url-insert (url &optional follow-refresh)
  "Insert the contents from an URL in the current buffer.
If FOLLOW-REFRESH is non-nil, redirect refresh url in META."
  (let ((times mm-url-retries)
	(done nil)
	(first t)
	result)
    (while (and (not (zerop (decf times)))
		(not done))
      (with-timeout (mm-url-timeout)
	(unless first
	  (message "Trying again (%s)..." (- mm-url-retries times)))
	(setq first nil)
	(if follow-refresh
	    (save-restriction
	      (narrow-to-region (point) (point))
	      (mm-url-insert-file-contents url)
	      (goto-char (point-min))
	      (when (re-search-forward
		     "<meta[ \t\r\n]*http-equiv=\"Refresh\"[^>]*URL=\\([^\"]+\\)\"" nil t)
		(let ((url (match-string 1)))
		  (delete-region (point-min) (point-max))
		  (setq result (mm-url-insert url t)))))
	  (setq result (mm-url-insert-file-contents url)))
	(setq done t)))
    result))

It's the _only_ timing related thing in Gnus that XEmacs can't seem to
do with itimer.  Because of this single form the XEmacs Gnus package
must require the "fsf-compat" package to get `with-timeout' from
timer.el.

This obviously isn't a show-stopper, or even a bug for that matter,
but I'd be one happy camper if someone could help me with this.  I'd
_*REALLY REALLY*_ like the XEmacs Gnus package to _*NOT*_ need the
fsf-compat package.

Thanks!

-- 
|---<Steve Youngs>---------------<GnuPG KeyID: A94B3003>---|
|              Ashes to ashes, dust to dust.               |
|      The proof of the pudding, is under the crust.       |
|------------------------------<sryoungs@bigpond.net.au>---|

[-- Attachment #2: Type: application/pgp-signature, Size: 256 bytes --]

Re: No Gnus on Sunday

[Xavier Maillard]

[-- Attachment #1: Type: text/plain, Size: 219 bytes --]

On 2 Jan 2004, Lars Magne Ingebrigtsen<larsi@gnus.org> spake:
> So if y'all have any last-minute bug fixes for Gnus 5.10, please get
> them in during the next 36 hours...

at last ... :)

zeDek
-- 
 In Gruuik we trust


[-- Attachment #2: Type: application/pgp-signature, Size: 188 bytes --]

Re: No Gnus on Sunday

[Kai Grossjohann]

Steve Youngs <sryoungs@bigpond.net.au> writes:

> It's the _only_ timing related thing in Gnus that XEmacs can't seem to
> do with itimer.  Because of this single form the XEmacs Gnus package
> must require the "fsf-compat" package to get `with-timeout' from
> timer.el.

FWIW, that's also the reason why Tramp needs fsf-compat...

Kai

Re: No Gnus on Sunday

[Steve Youngs]

[-- Attachment #1: Type: text/plain, Size: 705 bytes --]

|--==> "KG" == Kai Grossjohann <kai@emptydomain.de> writes:

  KG> Steve Youngs <sryoungs@bigpond.net.au> writes:
  >>It's the _only_ timing related thing in Gnus that XEmacs can't seem to
  >>do with itimer.  Because of this single form the XEmacs Gnus package
  >>must require the "fsf-compat" package to get `with-timeout' from
  >>timer.el.

  KG> FWIW, that's also the reason why Tramp needs fsf-compat...

Maybe we can kill two birds with one stone.

-- 
|---<Steve Youngs>---------------<GnuPG KeyID: A94B3003>---|
|              Ashes to ashes, dust to dust.               |
|      The proof of the pudding, is under the crust.       |
|------------------------------<sryoungs@bigpond.net.au>---|

[-- Attachment #2: Type: application/pgp-signature, Size: 256 bytes --]

Re: No Gnus on Sunday

[Kai Grossjohann]

Steve Youngs <sryoungs@bigpond.net.au> writes:

> Maybe we can kill two birds with one stone.

The expansion of with-timeout looks rather complicated, I would be
happy if I didn't have to include it in Tramp.

Hm.  It seems to be the same in Emacs and in XEmacs.  So it appears
that XEmacs has no alternative implementation for doing the same job
-- else I'd guess that the compat macro would use it.

So one approach would be to declare with-timeout a new feature instead
of a compat thing.

But I might be misunderstanding, and the with-timeout implementation
in XEmacs is the same as the one in Emacs for a different reason.  I
don't know much about XEmacs specifics.

Kai

Re: No Gnus on Sunday

[Frank Schmitt]

Lars Magne Ingebrigtsen <larsi@gnus.org> writes:

> So if y'all have any last-minute bug fixes for Gnus 5.10, please get
> them in during the next 36 hours...

Well, there's at least one major bug left which I reported in
<m3y8sr27r1.fsf@fed1.frank-schmitt.net> in gnus.bug. I don't know about
the usage patterns of other Gnus users, but I use "Resend edit" quite
frequently and I really think this should be fixed.

MFG Frank

-- 
Did you ever realize how much text fits in eighty columns? If you now consider
that a signature usually consists of up to four lines, this gives you enough
space to spread a tremendous amount of information with your messages. So seize
this opportunity and don't waste your signature with bullshit nobody will read.

Re: No Gnus on Sunday

[Lars Magne Ingebrigtsen]

Frank Schmitt <ich@Frank-Schmitt.net> writes:

> Well, there's at least one major bug left which I reported in
> <m3y8sr27r1.fsf@fed1.frank-schmitt.net> in gnus.bug. I don't know about
> the usage patterns of other Gnus users, but I use "Resend edit" quite
> frequently and I really think this should be fixed.

Fixing it would (if I understand the interactions with the PGP code
and stuff, which I don't) be somewhat complicated, so it'll have to
remain broken for Gnus 5.10, I think.

-- 
(domestic pets only, the antidote for overdose, milk.)
  larsi@gnus.org * Lars Magne Ingebrigtsen

Re: No Gnus on Sunday

[Frank Schmitt]

Lars Magne Ingebrigtsen <larsi@gnus.org> writes:

> Frank Schmitt <ich@Frank-Schmitt.net> writes:
>
>> Well, there's at least one major bug left which I reported in
>> <m3y8sr27r1.fsf@fed1.frank-schmitt.net> in gnus.bug. I don't know about
>> the usage patterns of other Gnus users, but I use "Resend edit" quite
>> frequently and I really think this should be fixed.
>
> Fixing it would (if I understand the interactions with the PGP code
> and stuff, which I don't) be somewhat complicated, so it'll have to
> remain broken for Gnus 5.10, I think.

Well, it brake after the first release of Gnus 5.10 and as far as I
understood all following commits should have been bug fixes and no
changes in functionality.

-- 
Did you ever realize how much text fits in eighty columns? If you now consider
that a signature usually consists of up to four lines, this gives you enough
space to spread a tremendous amount of information with your messages. So seize
this opportunity and don't waste your signature with bullshit nobody will read.

Re: No Gnus on Sunday

[Lars Magne Ingebrigtsen]

Frank Schmitt <ich@Frank-Schmitt.net> writes:

> Well, it brake after the first release of Gnus 5.10 and as far as I
> understood all following commits should have been bug fixes and no
> changes in functionality.

Yes.  What it fixed was the ability to send PGP-encoded articles.

-- 
(domestic pets only, the antidote for overdose, milk.)
  larsi@gnus.org * Lars Magne Ingebrigtsen

Re: No Gnus on Sunday

[Frank Schmitt]

Lars Magne Ingebrigtsen <larsi@gnus.org> writes:

> Yes.  What it fixed was the ability to send PGP-encoded articles.

Ok, I agree that this is more important than the feature which broke.

MFG Frank

-- 
Did you ever realize how much text fits in eighty columns? If you now consider
that a signature usually consists of up to four lines, this gives you enough
space to spread a tremendous amount of information with your messages. So seize
this opportunity and don't waste your signature with bullshit nobody will read.

Re: No Gnus on Sunday

[Steve Youngs]

[-- Attachment #1: Type: text/plain, Size: 1031 bytes --]

* Kai Grossjohann <kai@emptydomain.de> writes:

[`with-timeout' in XEmacs]

  > Hm.  It seems to be the same in Emacs and in XEmacs.  So it appears
  > that XEmacs has no alternative implementation for doing the same job
  > -- else I'd guess that the compat macro would use it.

  > So one approach would be to declare with-timeout a new feature
  > instead of a compat thing.

Well XEmacs uses itimer, with-timeout uses timer.  That's why it's a
compat thing.  What's needed is an itimer equivalent of
`with-timeout'. 

  > But I might be misunderstanding, and the with-timeout implementation
  > in XEmacs is the same as the one in Emacs for a different reason.

Yeah, the person who added it to the XEmacs fsf-compat package was too
lazy to implement it using XEmacs API's. :-)

-- 
|---<Steve Youngs>---------------<GnuPG KeyID: A94B3003>---|
|              Ashes to ashes, dust to dust.               |
|      The proof of the pudding, is under the crust.       |
|------------------------------<sryoungs@bigpond.net.au>---|

[-- Attachment #2: Type: application/pgp-signature, Size: 256 bytes --]

Re: No Gnus on Sunday

[Kai Grossjohann]

Steve Youngs <sryoungs@bigpond.net.au> writes:

> Well XEmacs uses itimer, with-timeout uses timer.  That's why it's a
> compat thing.  What's needed is an itimer equivalent of
> `with-timeout'. 

Ah!  OK.  So you're looking for with-itimeout ;-)

Kai