* Re: Marking as already read when using gcc-self?
1999-01-04 23:42 ` Lloyd Zusman
@ 1999-01-05 1:56 ` Nelson Jose dos Santos Ferreira
1999-01-06 0:53 ` Lloyd Zusman
1999-01-05 9:37 ` Stefan Waldherr
` (2 subsequent siblings)
3 siblings, 1 reply; 13+ messages in thread
From: Nelson Jose dos Santos Ferreira @ 1999-01-05 1:56 UTC (permalink / raw)
Cc: ding
[-- Attachment #1: Type: text/plain, Size: 2468 bytes --]
On 04 Jan 1999 18:42:17 -0500, Lloyd Zusman (aka "Lloyd"),
regarding 'Re: Marking as already read when using gcc-self?', said:
Lloyd> Kai.Grossjohann@CS.Uni-Dortmund.DE writes:
>> Lloyd Zusman <ljz@asfast.com> writes:
>>
>> > If I set the `gcc-self' group parameter for a given group to `t', is
>> > there any way to cause any messages that are archived into that group
>> > via this method to automatically be marked as `read' as soon as they
>> > appear in the group?
>>
>> This is in the todo list. See the node `Newest Features' in the Gnus
>> info file.
Lloyd> Ah yes ... I found it. It only contained the string "Gcc" and not
Lloyd> "gcc-self", so I didn't find it in my previous search.
Lloyd> Well, I'll devote a little bit of my time to this task.
Maybe you want to start from where I left off.
The patch attach has the problem where it doesn't set the
mark as read if the summary buffer for the group is the
current one (defeating the gcc-self purpose, but making
other gcc as read ok). Maybe you can figure out what's wrong
with it. It applies cleanly to pgnus 0.69.
Lloyd> Perhaps Gcc could cause a special header to be inserted into its saved
Lloyd> messages, such as ...
Lloyd> X-Gnus-Gcc-Action: <what to do>
Lloyd> Where <what to do> might be `tick' or `mark read' or something
Lloyd> similar. This could be specified via some sort of option, such as ...
Lloyd> (defvar gnus-gcc-action "mark read"
Lloyd> "*Action to be taken when the message is Gcc'ed.
Lloyd> Could be `tick', `mark read', etc.")
Good! I think this can be easily accomodated to my code
provide someone more elisp/gnus knowledgeable will help
track the problem.
Lloyd> Then, when entering summary mode, some routine (which one? ... well,
Lloyd> I'll dig around and find it if no one can make a suggestion) could
Lloyd> then recognize this `X-Gnus-Gcc-Action' header and perform the
Lloyd> specified action.
Lloyd> Does this sound like a good approach? Before I start hacking, I'd
Lloyd> appreciate any feedback and opinions from those of you who are
Lloyd> familiar with this part of Gnus.
[...]
As you can see by the patch I hacked the gcc treatment in
the message sending part, which is a diferent approach from
yours.Note that yours will show the message you sent on the
group buffer until you enter the group.
Best regards,
Nelson
[-- Attachment #2: GCC Mark as read patch for pgnus --]
[-- Type: application/x-patch, Size: 2151 bytes --]
[-- Attachment #3: Type: text/plain, Size: 269 bytes --]
---
Nelson Jose dos Santos Ferreira | . . . | INESC/PT-Servicos
Systems Integrator | |. / / . | Lisbon,PORTUGAL
mailto:Nelson.Ferreira@inesc.pt |.||\| | / | Phone:+351 (1) 3100050
| .|<:/ \. | Fax :+351 (1) 3100008
^ permalink raw reply [flat|nested] 13+ messages in thread* Re: Marking as already read when using gcc-self?
1999-01-05 1:56 ` Nelson Jose dos Santos Ferreira
@ 1999-01-06 0:53 ` Lloyd Zusman
0 siblings, 0 replies; 13+ messages in thread
From: Lloyd Zusman @ 1999-01-06 0:53 UTC (permalink / raw)
Nelson Jose dos Santos Ferreira <Nelson.Ferreira@inesc.pt> writes:
> Maybe you want to start from where I left off.
> The patch attach has the problem where it doesn't set the
> mark as read if the summary buffer for the group is the
> current one (defeating the gcc-self purpose, but making
> other gcc as read ok). Maybe you can figure out what's wrong
> with it. It applies cleanly to pgnus 0.69.
After realizing that the `X-Gnus-Gcc-Action' header isn't such a good
idea after all (others pointed out that people could send email or
news with that header set, thereby screwing up my newsreader), I took
a different approach, which seems to be similar to yours ... and
unfortunately, I hit the same problem that you did. :(
But I'll keep plugging away ... and once I figure this out, I'll let
you know.
And thanks for the patch!
Sincerely,
- Lloyd
--
Lloyd Zusman
ljz@asfast.com
^ permalink raw reply [flat|nested] 13+ messages in thread
* Re: Marking as already read when using gcc-self?
1999-01-04 23:42 ` Lloyd Zusman
1999-01-05 1:56 ` Nelson Jose dos Santos Ferreira
@ 1999-01-05 9:37 ` Stefan Waldherr
1999-01-05 12:44 ` Hrvoje Niksic
1999-01-06 4:45 ` Matt Simmons
1999-01-05 16:00 ` Justin Sheehy
1999-01-05 22:14 ` Kai.Grossjohann
3 siblings, 2 replies; 13+ messages in thread
From: Stefan Waldherr @ 1999-01-05 9:37 UTC (permalink / raw)
>>>>> "Lloyd" == Lloyd Zusman <ljz@asfast.com> writes:
Lloyd> Perhaps Gcc could cause a special header to be inserted into its
Lloyd> saved messages, such as ...
Lloyd> X-Gnus-Gcc-Action: <what to do>
Excellent idea.
Lloyd> Then, when entering summary mode, some routine (which one? ... well,
Lloyd> I'll dig around and find it if no one can make a suggestion) could
Lloyd> then recognize this `X-Gnus-Gcc-Action' header and perform the
Lloyd> specified action.
I would rather see the action while _incorporating_ the mail and not when
entering the summary mode. Advantage: mail doesn't show up as new mail at all
in the group buffer and you don't have to go thru the groups to catch-up (even
if it is auto-catch-up in this case).
cu
Stefan.
--
Stefan Waldherr fax +49 431 8058 136
e-Mail stefan@waldherr.org
www http://www.waldherr.org/
^ permalink raw reply [flat|nested] 13+ messages in thread* Re: Marking as already read when using gcc-self?
1999-01-05 9:37 ` Stefan Waldherr
@ 1999-01-05 12:44 ` Hrvoje Niksic
1999-01-06 4:45 ` Matt Simmons
1 sibling, 0 replies; 13+ messages in thread
From: Hrvoje Niksic @ 1999-01-05 12:44 UTC (permalink / raw)
Stefan Waldherr <stefan@waldherr.org> writes:
> I would rather see the action while _incorporating_ the mail and not
> when entering the summary mode. Advantage: mail doesn't show up as
> new mail at all in the group buffer and you don't have to go thru
> the groups to catch-up (even if it is auto-catch-up in this case).
I agree. Adding an "action" header to be processed later looks like
the wrong approach to me.
^ permalink raw reply [flat|nested] 13+ messages in thread
* Re: Marking as already read when using gcc-self?
1999-01-05 9:37 ` Stefan Waldherr
1999-01-05 12:44 ` Hrvoje Niksic
@ 1999-01-06 4:45 ` Matt Simmons
1 sibling, 0 replies; 13+ messages in thread
From: Matt Simmons @ 1999-01-06 4:45 UTC (permalink / raw)
>>>>> "Stefan" == Stefan Waldherr <stefan@waldherr.org> writes:
Stefan> I would rather see the action while _incorporating_ the
Stefan> mail and not when entering the summary mode. Advantage:
Stefan> mail doesn't show up as new mail at all in the group
Stefan> buffer and you don't have to go thru the groups to
Stefan> catch-up (even if it is auto-catch-up in this case).
Could the behavior should be configurable? I can easily see myself
wanting to refer back to the message I just sent. Having to A T or
C-u g just to find it would be annoying. Much better to Z G, see it
once, and then have it vanish into the ether.
Matt
--
Matt Simmons - simmonmt@acm.org - http://www.netcom.com/~simmonmt
Why do they call it the Department of Interior when they are in charge of
everything outdoors?
^ permalink raw reply [flat|nested] 13+ messages in thread
* Re: Marking as already read when using gcc-self?
1999-01-04 23:42 ` Lloyd Zusman
1999-01-05 1:56 ` Nelson Jose dos Santos Ferreira
1999-01-05 9:37 ` Stefan Waldherr
@ 1999-01-05 16:00 ` Justin Sheehy
1999-01-05 16:53 ` Hrvoje Niksic
1999-01-06 7:41 ` Stefan Waldherr
1999-01-05 22:14 ` Kai.Grossjohann
3 siblings, 2 replies; 13+ messages in thread
From: Justin Sheehy @ 1999-01-05 16:00 UTC (permalink / raw)
Lloyd Zusman <ljz@asfast.com> writes:
> Then, when entering summary mode, some routine (which one? ... well,
> I'll dig around and find it if no one can make a suggestion) could
> then recognize this `X-Gnus-Gcc-Action' header and perform the
> specified action.
Hm.
So, one could send me mail with a X-Gnus-Gcc-Action: <something>
header and make Gnus do something for me when it processes that mail?
I don't think I'd like that very much.
--
Justin Sheehy
In a cloud bones of steel.
^ permalink raw reply [flat|nested] 13+ messages in thread
* Re: Marking as already read when using gcc-self?
1999-01-05 16:00 ` Justin Sheehy
@ 1999-01-05 16:53 ` Hrvoje Niksic
1999-01-05 23:45 ` Lloyd Zusman
1999-01-06 7:41 ` Stefan Waldherr
1 sibling, 1 reply; 13+ messages in thread
From: Hrvoje Niksic @ 1999-01-05 16:53 UTC (permalink / raw)
Justin Sheehy <justin@linus.mitre.org> writes:
> Lloyd Zusman <ljz@asfast.com> writes:
>
> > Then, when entering summary mode, some routine (which one? ... well,
> > I'll dig around and find it if no one can make a suggestion) could
> > then recognize this `X-Gnus-Gcc-Action' header and perform the
> > specified action.
>
> Hm.
>
> So, one could send me mail with a X-Gnus-Gcc-Action: <something>
> header and make Gnus do something for me when it processes that
> mail?
>
> I don't think I'd like that very much.
Why not? I think it's be totally hip -- see the headers. :-)
^ permalink raw reply [flat|nested] 13+ messages in thread
* Re: Marking as already read when using gcc-self?
1999-01-05 16:53 ` Hrvoje Niksic
@ 1999-01-05 23:45 ` Lloyd Zusman
0 siblings, 0 replies; 13+ messages in thread
From: Lloyd Zusman @ 1999-01-05 23:45 UTC (permalink / raw)
Hrvoje Niksic <hniksic@srce.hr> writes:
> Justin Sheehy <justin@linus.mitre.org> writes:
>
> > Lloyd Zusman <ljz@asfast.com> writes:
> >
> > > Then, when entering summary mode, some routine (which one? ... well,
> > > I'll dig around and find it if no one can make a suggestion) could
> > > then recognize this `X-Gnus-Gcc-Action' header and perform the
> > > specified action.
> >
> > Hm.
> >
> > So, one could send me mail with a X-Gnus-Gcc-Action: <something>
> > header and make Gnus do something for me when it processes that
> > mail?
> >
> > I don't think I'd like that very much.
>
> Why not? I think it's be totally hip -- see the headers. :-)
Sure ... why not! Might as well allow any and all sexp's that appear
data portion of that header to be evaluated, instead of just
interpreting a few specific actions. :)
But yes, I agree with all of you who pointed out that even with a few
specific action keywords, the header is indeed a bad idea, given the
fact that anyone could send a message with that header already
defined.
Hmmm ... OK. What about this idea:
Create a new group parameter, perhaps called `gcc-action'. It can
take only a few specified values such as 'tick, 'mark-read, etc.
See to it that `gnus-inews-do-gcc' in `gnus-msg.el' has access to the
value of `gcc-action' for the given group (I have to look more closely
at the code to determine the details as to how to get that value into
`gnus-inews-do-gcc', but I'm certain that this can be done). If this
`gcc-action' value is present, pass it as a new parameter to the
invocation of `gnus-request-accept-article' near the end of this
routine.
Then, add another optional argument called `action' to the function
`gnus-request-accept-article' in `gnus-int.el'. If this optional
argument is `nil', then this function will behave as it currently
does. However, if `action' is set to one of the possible values
of `gcc-action', then tick or mark the article accordingly after
the `funcall' at the end of `gnus-request-accept-article' returns
(the article number of the new article should be contained within
the return value of this funcall, and this plus the group name
should allow us to set the appropriate ticking or marking flags).
Any comments?
--
Lloyd Zusman
ljz@asfast.com
^ permalink raw reply [flat|nested] 13+ messages in thread
* Re: Marking as already read when using gcc-self?
1999-01-05 16:00 ` Justin Sheehy
1999-01-05 16:53 ` Hrvoje Niksic
@ 1999-01-06 7:41 ` Stefan Waldherr
1 sibling, 0 replies; 13+ messages in thread
From: Stefan Waldherr @ 1999-01-06 7:41 UTC (permalink / raw)
>>>>> "Justin" == Justin Sheehy <justin@linus.mitre.org> writes:
Justin> So, one could send me mail with a X-Gnus-Gcc-Action: <something>
Justin> header and make Gnus do something for me when it processes that
Justin> mail?
Well, one could choose the header to be a list of predefined actions (not rm
-rf) and (since it is gcc) check whether the sender of the mail is you. But
that would still leave a hole.
Here's a better idea: make the action a simple local variable. And leave the
gcc header like it is now. Whenever you discover the header, mark them as read
(or whatever that local variable sez).
cu
Stefan.
--
Stefan Waldherr fax +49 431 8058 136
e-Mail stefan@waldherr.org
www http://www.waldherr.org/
^ permalink raw reply [flat|nested] 13+ messages in thread
* Re: Marking as already read when using gcc-self?
1999-01-04 23:42 ` Lloyd Zusman
` (2 preceding siblings ...)
1999-01-05 16:00 ` Justin Sheehy
@ 1999-01-05 22:14 ` Kai.Grossjohann
3 siblings, 0 replies; 13+ messages in thread
From: Kai.Grossjohann @ 1999-01-05 22:14 UTC (permalink / raw)
Lloyd Zusman <ljz@asfast.com> writes:
> Then, when entering summary mode, some routine (which one? ... well,
> I'll dig around and find it if no one can make a suggestion) could
> then recognize this `X-Gnus-Gcc-Action' header and perform the
> specified action.
IMHO, this action should be performed when the Gcc itself is processed
(maybe directly after that).
kai
--
Just because I'm paranoid doesn't mean they're not out there to get me!
^ permalink raw reply [flat|nested] 13+ messages in thread