Completely removing an article from drafts?

Completely removing an article from drafts?

[Lloyd Zusman]

How to I completely remove an article from my drafts folder?

In the old days before I started using gnus-agent (i.e., up until two
days ago), I had a home-grown way of completely expunging articles from
any group.  I used one function I wrote to mark the articles in a
certain way (I forget exactly which marks I used, since I wrote this
code eons ago), and then had another function which would completely
expunge all articles marked in this way when I leave the group.

But now that I'm using gnus-agent, the marks are different and my
functions no longer work.  Rather than change my functions (which are
pretty much hacks), I'd rather remove articles from my drafts folder in
the "proper" way, whatever that is.

Thanks in advance.

-- 
 Lloyd Zusman
 ljz@asfast.com

Re: Completely removing an article from drafts?

[Jochen Küpper]

[-- Attachment #1: Type: text/plain, Size: 471 bytes --]

On Mon, 12 May 2003 05:32:49 -0400 Lloyd Zusman wrote:

Lloyd> How to I completely remove an article from my drafts folder?

I use
,----
| B DEL           gnus-summary-delete-article
`----

(Although I still have problems with it in nnimap groups...)

Greetings,
Jochen
-- 
Einigkeit und Recht und Freiheit                http://www.Jochen-Kuepper.de
    Liberté, Égalité, Fraternité                GnuPG key: CC1B0B4D
        Sex, drugs and rock-n-roll

[-- Attachment #2: Type: application/pgp-signature, Size: 185 bytes --]

Re: Completely removing an article from drafts?

[Lloyd Zusman]

"Jochen Küpper" <jochen@jochen-kuepper.de> writes:

> On Mon, 12 May 2003 05:32:49 -0400 Lloyd Zusman wrote:
>
> Lloyd> How to I completely remove an article from my drafts folder?
>
> I use
> ,----
> | B DEL           gnus-summary-delete-article
> `----
>
> (Although I still have problems with it in nnimap groups...)

Yep.  That worked.  I was looking for functions with the name 'expunge'
in them.  IMHO '-delete-' is misleading as part of this function's name.

Thanks for your help.


-- 
 Lloyd Zusman
 ljz@asfast.com

Re: Completely removing an article from drafts?

[Karl Kleinpaste]

Lloyd Zusman <ljz@asfast.com> writes:
> Yep.  That worked.  I was looking for functions with the name 'expunge'
> in them.  IMHO '-delete-' is misleading as part of this function's name.

Why?

How is "remove an article" (your request) less represented by "delete"
than by "expunge"?

Re: Completely removing an article from drafts?

[Lloyd Zusman]

Karl Kleinpaste <k@charcoal.com> writes:

> Lloyd Zusman <ljz@asfast.com> writes:
>> Yep.  That worked.  I was looking for functions with the name 'expunge'
>> in them.  IMHO '-delete-' is misleading as part of this function's name.
>
> Why?
>
> How is "remove an article" (your request) less represented by "delete"
> than by "expunge"?

First of all, I meant to type "expire" instead of "expunge" (I have been
in the IMAP world a lot lately).

As for why I don't like "delete" in the function name, most (all?) of
the other 'delete' functions simply put a mark on the article and don't
actually get rid of it.  The removal itself is handled later on via
expiry.  Since `gnus-summary-delete-article' does actual expiry, I think
that it would be more clear if "expire" appeared in the function name.

Of course, this proposed `gnus-summary-expire-article' function would
have to simply be an aliaas for `gnus-summary-delete-article', so as not
to break existing code.

-- 
 Lloyd Zusman
 ljz@asfast.com

Re: Completely removing an article from drafts?

[Karl Kleinpaste]

Lloyd Zusman <ljz@asfast.com> writes:
> Since `gnus-summary-delete-article' does actual expiry, I think
> that it would be more clear if "expire" appeared in the function name.

No, I don't agree.  "Expiry" in Gnus is generally a time-based
concept; an article expires after a timeout passes, causing deletion.

gnus-summary-delete-article simply deletes, right now.  It does not
expire anything; it annihilates something on the spot.  For "delete,"
dictionary.reference.com says "To remove by striking out or
canceling", and that's what the function does.

I think it is named exactly right.

Re: Completely removing an article from drafts?

[Lloyd Zusman]

Karl Kleinpaste <k@charcoal.com> writes:

> Lloyd Zusman <ljz@asfast.com> writes:
>> Since `gnus-summary-delete-article' does actual expiry, I think
>> that it would be more clear if "expire" appeared in the function name.
>
> No, I don't agree.  "Expiry" in Gnus is generally a time-based
> concept; an article expires after a timeout passes, causing deletion.
>
> gnus-summary-delete-article simply deletes, right now.  It does not
> expire anything; it annihilates something on the spot.  For "delete,"
> dictionary.reference.com says "To remove by striking out or
> canceling", and that's what the function does.
>
> I think it is named exactly right.

If we are to adhere to that dictionary definition, then the other
functions with "delete" in their names should then have a name change.
They should then contain the string "mark-for-deletion" or some such
thing, since those functions do not delete articles in the way that the
dictionary specifies.

-- 
 Lloyd Zusman
 ljz@asfast.com

Re: Completely removing an article from drafts?

[Karl Kleinpaste]

Lloyd Zusman <ljz@asfast.com> writes:
> They should then contain the string "mark-for-deletion" or some such
> thing, since those functions do not delete articles in the way that the
> dictionary specifies.

I'm having a hard time finding even one example of a Gnus function
named with "delete" which _doesn't_ delete something on the spot.

b  gnus-article-delete-invisible-text - Delete all invisible text in the current buffer.
b  gnus-article-delete-text-of-type - Delete text of TYPE in the current buffer.
b  gnus-async-delete-prefetched-entry - Delete ENTRY from buffer and alist.
b  gnus-cite-delete-overlays     
   gnus-delete-alist              - Delete by side effect any elements of ALIST whose car is `eq' to KEY.
b  gnus-delete-file               - Delete FILE if it exists.
b  gnus-delete-first              - Delete by side effect the first occurrence of ELT as a member of LIST.
b  gnus-delete-images             - Delete all images in CATEGORY.
m  gnus-delete-line              
   gnus-delete-overlay            - Remove EXTENT from its buffer and destroy it.
b  gnus-delete-wash-type          - Add a washing of TYPE to the current status.
b  gnus-delete-windows-in-gnusey-frames - Do a `delete-other-windows' in all frames that have Gnus windows.
b  gnus-dribble-delete-file      
b* gnus-group-delete-group        - Delete the current group.  Only meaningful with editable groups.
b* gnus-group-delete-groups       - Delete the current group.  Only meaningful with editable groups.
b* gnus-mime-delete-part          - Delete the MIME part under point.
b  gnus-request-delete-group     
b* gnus-summary-delete-article    - Delete the N next (mail) articles.
 * gnus-summary-delete-marked-as-read - Limit the summary buffer to articles that are not marked as read.
 * gnus-summary-delete-marked-with - Exclude articles that are marked with MARKS (e.g. "DK").
b* gnus-topic-delete              - Delete a topic.
a  gnus-uu-delete-work-dir       

Re: Completely removing an article from drafts?

[Lloyd Zusman]

You're right, except for `gnus-summary-delete-article',
`gnus-summary-delete-marked-as-read', and
`gnus-summary-delete-marked-wuth'.

The first one does what `nnmail-expiry-function' tells it to do.  As I
mentioned in my other message, that means that these functions do NOT do
a dictionary "delete" unless `nnmail-expiry-function' is set to 'delete.

The second two simply limit what's shown in the summary buffer to
those articles that are unread, or accordingly marked, without actually
removing anything.



Karl Kleinpaste <k@charcoal.com> writes:

> Lloyd Zusman <ljz@asfast.com> writes:
>> They should then contain the string "mark-for-deletion" or some such
>> thing, since those functions do not delete articles in the way that the
>> dictionary specifies.
>
> I'm having a hard time finding even one example of a Gnus function
> named with "delete" which _doesn't_ delete something on the spot.
>
> b  gnus-article-delete-invisible-text - Delete all invisible text in the current buffer.
> b  gnus-article-delete-text-of-type - Delete text of TYPE in the current buffer.
> b  gnus-async-delete-prefetched-entry - Delete ENTRY from buffer and alist.
> b  gnus-cite-delete-overlays     
>    gnus-delete-alist              - Delete by side effect any elements of ALIST whose car is `eq' to KEY.
> b  gnus-delete-file               - Delete FILE if it exists.
> b  gnus-delete-first              - Delete by side effect the first occurrence of ELT as a member of LIST.
> b  gnus-delete-images             - Delete all images in CATEGORY.
> m  gnus-delete-line              
>    gnus-delete-overlay            - Remove EXTENT from its buffer and destroy it.
> b  gnus-delete-wash-type          - Add a washing of TYPE to the current status.
> b  gnus-delete-windows-in-gnusey-frames - Do a `delete-other-windows' in all frames that have Gnus windows.
> b  gnus-dribble-delete-file      
> b* gnus-group-delete-group        - Delete the current group.  Only meaningful with editable groups.
> b* gnus-group-delete-groups       - Delete the current group.  Only meaningful with editable groups.
> b* gnus-mime-delete-part          - Delete the MIME part under point.
> b  gnus-request-delete-group     
> b* gnus-summary-delete-article    - Delete the N next (mail) articles.
>  * gnus-summary-delete-marked-as-read - Limit the summary buffer to articles that are not marked as read.
>  * gnus-summary-delete-marked-with - Exclude articles that are marked with MARKS (e.g. "DK").
> b* gnus-topic-delete              - Delete a topic.
> a  gnus-uu-delete-work-dir       

-- 
 Lloyd Zusman
 ljz@asfast.com

Re: Completely removing an article from drafts?

[Karl Kleinpaste]

Lloyd Zusman <ljz@asfast.com> writes:
> You're right, except for `gnus-summary-delete-article',
> `gnus-summary-delete-marked-as-read', and
> `gnus-summary-delete-marked-with'.

> The first one does what `nnmail-expiry-function' tells it to do.

No, not merely so.  The code near the top of -delete-article has
forced nnmail-expiry-target to 'delete since we had this argument some
months back.  (See archives, subject "B DEL being treated as expiry?")
See today's gnus-sum.el, line 9235.  The doc string is reflective of
this.  And that was merely to re-assert the previous always-deleting
behavior of -delete-article anyway, following another change which
broke this longstanding behavior.

> The second two simply limit what's shown in the summary buffer to
> those articles that are unread, or accordingly marked, without actually
> removing anything.

In the 1st place, if you will check the documentation, you will find
that these are aliases:

-as-read: Obsolete; use `gnus-summary-limit-to-unread' instead.
-with: Obsolete; use `gnus-summary-limit-exclude-marks' instead.

And in the 2nd place, what those functions do, native or aliased, is
to DELETE things from the summary buffer, as advertised.  They were
merely renamed (and left as aliases) because they fit better
conceptually into the general "limit" scheme.

> Also, I want to add that `gnus-summary-delete-article' does not delete
> an article using the dictionary definition of "delete" that you mention
> above.  Check the documentation and the source code.  What that function
> actually does is to expire the article, in the exact manner that Gnus
> defines "expire".

Lloyd, all this started out as, for me, was curiosity as to why you
think "delete" doesn't mean "delete."  That's all.  I wasn't intending
to get anyone's shorts in a knot over it.

I did check the source; you should do the same; look again at
nnmail-expiry-target.  And the doc string on -delete-article says:

    Delete the N next (mail) articles.
    This command actually deletes articles. This is not a marking
    command.  The article will disappear forever from your life, never to
    return.

Enough already.  Delete means delete.

Re: Completely removing an article from drafts?

[Lloyd Zusman]

> Lloyd Zusman <ljz@asfast.com> writes:
>> You're right, except for `gnus-summary-delete-article',
>> `gnus-summary-delete-marked-as-read', and
>> `gnus-summary-delete-marked-with'.
>
>> The first one does what `nnmail-expiry-function' tells it to do.
>
> No, not merely so.  The code near the top of -delete-article has
> forced nnmail-expiry-target to 'delete since we had this argument some
> months back.  (See archives, subject "B DEL being treated as expiry?")
> See today's gnus-sum.el, line 9235.  The doc string is reflective of
> this.  And that was merely to re-assert the previous always-deleting
> behavior of -delete-article anyway, following another change which
> broke this longstanding behavior.

Ah ... I missed that 'delete argument.


>> The second two simply limit what's shown in the summary buffer to
>> those articles that are unread, or accordingly marked, without actually
>> removing anything.
>
> In the 1st place, if you will check the documentation, you will find
> that these are aliases:
>
> -as-read: Obsolete; use `gnus-summary-limit-to-unread' instead.
> -with: Obsolete; use `gnus-summary-limit-exclude-marks' instead.
>
> And in the 2nd place, what those functions do, native or aliased, is
> to DELETE things from the summary buffer, as advertised.  They were
> merely renamed (and left as aliases) because they fit better
> conceptually into the general "limit" scheme.

I had misunderstood `make-obsolete', so yes, I am in agreement.


> Lloyd, all this started out as, for me, was curiosity as to why you
> think "delete" doesn't mean "delete."  That's all.  I wasn't intending
> to get anyone's shorts in a knot over it.
>
> I did check the source; you should do the same; look again at
> nnmail-expiry-target.  And the doc string on -delete-article says:
>
>     Delete the N next (mail) articles.
>     This command actually deletes articles. This is not a marking
>     command.  The article will disappear forever from your life, never to
>     return.
>
> Enough already.  Delete means delete.

The doc string didn't seem to correspond to what I thought the function
actually did, because I overlooked the 'delete argument that you
mentioned above.

So yes, they all do a real "delete".

Shorts are duly unknotted.

-- 
 Lloyd Zusman
 ljz@asfast.com

Re: Completely removing an article from drafts?

[Lloyd Zusman]

Lloyd Zusman <ljz@asfast.com> writes:

> Karl Kleinpaste <k@charcoal.com> writes:
>
>> Lloyd Zusman <ljz@asfast.com> writes:
>>> Since `gnus-summary-delete-article' does actual expiry, I think
>>> that it would be more clear if "expire" appeared in the function name.
>>
>> No, I don't agree.  "Expiry" in Gnus is generally a time-based
>> concept; an article expires after a timeout passes, causing deletion.
>>
>> gnus-summary-delete-article simply deletes, right now.  It does not
>> expire anything; it annihilates something on the spot.  For "delete,"
>> dictionary.reference.com says "To remove by striking out or
>> canceling", and that's what the function does.
>>
>> I think it is named exactly right.
>
> If we are to adhere to that dictionary definition, then the other
> functions with "delete" in their names should then have a name change.
> They should then contain the string "mark-for-deletion" or some such
> thing, since those functions do not delete articles in the way that the
> dictionary specifies.

Also, I want to add that `gnus-summary-delete-article' does not delete
an article using the dictionary definition of "delete" that you mention
above.  Check the documentation and the source code.  What that function
actually does is to expire the article, in the exact manner that Gnus
defines "expire".

What `gnus-summary-delete-article' actually does is to apply
`nnimap-expiry-function' to the article.  If `nnimap-expiry-function' is
set to 'delete, then `gnus-summary-delete-article' will indeed do a
deletion as the expiry process that it performs.  However, if that
variable contains a function symbol, that function is invoked to perform
the expiry.

Therefore, `gnus-summary-delete-article' is indeed a misnomer for this
function, and `gnus-summary-expire-article' is the perfectly precise
name for what this function actually does.


-- 
 Lloyd Zusman
 ljz@asfast.com

Re: Completely removing an article from drafts?

[Kai Großjohann]

Lloyd Zusman <ljz@asfast.com> writes:

> Also, I want to add that `gnus-summary-delete-article' does not delete
> an article using the dictionary definition of "delete" that you mention
> above.  Check the documentation and the source code.  What that function
> actually does is to expire the article, in the exact manner that Gnus
> defines "expire".

Ayee.  This has been discussed before.  I think you're right.  `B
DEL' should always delete the article, even if nnmail-expiry-target
is something other than delete.

But I think that the Gnus Towers had a different opinion, or
something.  Who remembers more?
-- 
This line is not blank.

Re: Completely removing an article from drafts?

[Lars Magne Ingebrigtsen]

kai.grossjohann@gmx.net (Kai Großjohann) writes:

> Ayee.  This has been discussed before.  I think you're right.  `B
> DEL' should always delete the article, even if nnmail-expiry-target
> is something other than delete.
>
> But I think that the Gnus Towers had a different opinion, or
> something.  Who remembers more?

I think `B DEL' should always actually delete articles.  It doesn't
if you've set expiry-target to something funky?  

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

Re: Completely removing an article from drafts?

[Kai Großjohann]

Lloyd Zusman <ljz@asfast.com> writes:

> If we are to adhere to that dictionary definition, then the other
> functions with "delete" in their names should then have a name change.

Which other functions with "delete" in their names?  (I looked and
couldn't find any relevant ones.  Probably I'm having a blonde day.)
-- 
This line is not blank.

Re: Completely removing an article from drafts?

[Lloyd Zusman]

Karl Kleinpaste <k@charcoal.com> writes:

> Lloyd Zusman <ljz@asfast.com> writes:
>> Since `gnus-summary-delete-article' does actual expiry, I think
>> that it would be more clear if "expire" appeared in the function name.
>
> No, I don't agree.  "Expiry" in Gnus is generally a time-based
> concept; an article expires after a timeout passes, causing deletion.
>
> gnus-summary-delete-article simply deletes, right now.  It does not
> expire anything; it annihilates something on the spot.  For "delete,"
> dictionary.reference.com says "To remove by striking out or
> canceling", and that's what the function does.
>
> I think it is named exactly right.

In addition to what I wrote in my previous message, I want to add that
`gnus-summary-delete-article' does _not_ delete an article using the
dictionary definition of "delete" that you mention above.  Check the
documentation and the source code.  What that function actually does is
to _expire_ the article, in the exact manner that Gnus defines _expire_.

What `gnus-summary-delete-article' actually does is to apply
`nnimap-expiry-function' to the article.  If `nnimap-expiry-function' is
set to 'delete, then `gnus-summary-delete-article' will indeed do a
deletion as the expiry process that it performs.  However, if that
variable contains a function symbol, that function is invoked to perform
the expiry.

Therefore, `gnus-summary-delete-article' is indeed a misnomer for this
function, and `gnus-summary-expire-article' is the perfectly precise
name for what this function actually does.

-- 
 Lloyd Zusman
 ljz@asfast.com

Re: Completely removing an article from drafts?

[Lloyd Zusman]

This message was delayed by the mailing-list server and posted now.

Therefore, the items mentioned here have already been discussed, and are
no longer relevant.


Lloyd Zusman <ljz@asfast.com> writes:

> Karl Kleinpaste <k@charcoal.com> writes:
>
>> Lloyd Zusman <ljz@asfast.com> writes:
>>> Since `gnus-summary-delete-article' does actual expiry, I think
>>> that it would be more clear if "expire" appeared in the function name.
>>
>> No, I don't agree.  "Expiry" in Gnus is generally a time-based
>> concept; an article expires after a timeout passes, causing deletion.
>>
>> gnus-summary-delete-article simply deletes, right now.  It does not
>> expire anything; it annihilates something on the spot.  For "delete,"
>> dictionary.reference.com says "To remove by striking out or
>> canceling", and that's what the function does.
>>
>> I think it is named exactly right.
>
> In addition to what I wrote in my previous message, I want to add that
> `gnus-summary-delete-article' does _not_ delete an article using the
> dictionary definition of "delete" that you mention above.  Check the
> documentation and the source code.  What that function actually does is
> to _expire_ the article, in the exact manner that Gnus defines _expire_.
>
> What `gnus-summary-delete-article' actually does is to apply
> `nnimap-expiry-function' to the article.  If `nnimap-expiry-function' is
> set to 'delete, then `gnus-summary-delete-article' will indeed do a
> deletion as the expiry process that it performs.  However, if that
> variable contains a function symbol, that function is invoked to perform
> the expiry.
>
> Therefore, `gnus-summary-delete-article' is indeed a misnomer for this
> function, and `gnus-summary-expire-article' is the perfectly precise
> name for what this function actually does.
>
> -- 
>  Lloyd Zusman
>  ljz@asfast.com

-- 
 Lloyd Zusman
 ljz@asfast.com

Re: Completely removing an article from drafts?

[Simon Josefsson]

Jochen Küpper <jochen@jochen-kuepper.de> writes:

[B-DEL]
> (Although I still have problems with it in nnimap groups...)

What was those?  Perhaps they are related to what Niklas Morberg was
seeing.  Do you still see them in current CVS?

Re: Completely removing an article from drafts?

[Lloyd Zusman]

Simon Josefsson <jas@extundo.com> writes:

> Jochen Küpper <jochen@jochen-kuepper.de> writes:
>
> [B-DEL]
>> (Although I still have problems with it in nnimap groups...)
>
> What was those?  Perhaps they are related to what Niklas Morberg was
> seeing.  Do you still see them in current CVS?

FWIW, I ran that function inside of one of my agentized nnimap groups,
and it worked just fine: the article was properly expired.

Maybe this is a problem when the nnimap group is not agentized?

-- 
 Lloyd Zusman
 ljz@asfast.com

Re: Completely removing an article from drafts?

[Jochen Küpper]

[-- Attachment #1: Type: text/plain, Size: 1061 bytes --]

On Mon, 12 May 2003 13:57:37 +0200 Simon Josefsson wrote:

Simon> Jochen Küpper <jochen=W2i0oK7yYu0+t3nig+EZI7NAH6kLmebB@public.gmane.org> writes:

Simon> [B-DEL]
>> (Although I still have problems with it in nnimap groups...)

Simon> What was those? Perhaps they are related to what Niklas Morberg
Simon> was seeing. Do you still see them in current CVS?

I was referring to 
,----
| Message-ID: <86k7eq402b.fsf=bhz6AtcEt9J/F1eftIH1yA@public.gmane.org>
`----
and references therein.

It seems to work just fine with latest cvs. Sorry for the line noise.


I do have a few old messages which forget their expiry marks whenever
I try to get rid of them and which I also cannot delete (B DEL), but I
(strongly) assume this is a server problem. (Related to the admin
moving messages form one server to another after an upgrade...)

Greetings,
Jochen
-- 
Einigkeit und Recht und Freiheit                http://www.Jochen-Kuepper.de
    Liberté, Égalité, Fraternité                GnuPG key: CC1B0B4D
        Sex, drugs and rock-n-roll

[-- Attachment #2: Type: application/pgp-signature, Size: 185 bytes --]

Re: Completely removing an article from drafts?

[David S Goldberg]

>>>>> On Mon, 12 May 2003 15:45:41 +0200, Jochen Küpper
>>>>> <jochen@jochen-kuepper.de> said:

> I do have a few old messages which forget their expiry marks whenever
> I try to get rid of them and which I also cannot delete (B DEL), but I
> (strongly) assume this is a server problem. (Related to the admin
> moving messages form one server to another after an upgrade...)

Could they be in the agent, but not on the server?  Although expiry
seems to do the right thing for me agentwise now (articles in the
agent are deleted from the agent when expiry deletes them from the
server) I recently found a whole slew of really ancient (well several
months anyway :-) articles in the agent which B DEL would not get rid
of because there was no such article on the server.  Running
gns-agent-expire from the group buffer got rid of them for me.

-- 
Dave Goldberg
david.goldberg6@verizon.net