[PROPOSAL]: Date header and Queued messages

[PROPOSAL]: Date header and Queued messages

[Xavier Maillard]

[-- Attachment #1.1: Type: text/plain, Size: 1410 bytes --]

Hi,

Why do we put the 'Date' header even if the message is pushed into the
queued group ?

For my case it result to something quite annoying since I write my
mails when I am in the train and can't send them before the evening so
it generally result in something with 14/15h late !

Can't we just set this header when the mail is really sent (ie:
dequeued and sent) or at least remove the header when sending it.

I have personally thought to this a long time ago and I am currently
thinking it would be better to do this way. 

My current proposal is to remove the header in `gnus-draft-send`
function if we can ensure we can regenerate the Date header after
otherwise just pass through ;)

So I did something quite simple and it works perfectly for me (I have
been using it since a week now). So in case it is of interest by one of
you, I have enclosed a very small patch that just does the trick.

Any opinion ?

I know I can remove the "date" header just by editing the queued
message again but it is obvious for me to edit all of the queued
messages just to send them with a correct date header.


Regards,

zeDek

P.S: since the patch is really small and given the fact I still do not
have sent my FSF copyrights assignment papers, I hope you will able to
add this upstream otherwise be sure I will send them (the papers) on
Saturday. 

-----%<---------%<---------%<---------%<---------%<---------%<----

[-- Attachment #1.2: draft_date_is_good.diff --]
[-- Type: text/plain, Size: 632 bytes --]

--- gnus-draft.el	2004-01-05 23:37:34.000000000 +0100
+++ /tmp/gnus-draft.el	2004-01-15 09:10:17.000000000 +0100
@@ -154,6 +154,13 @@
 	(skip-syntax-forward "-")
 	(setq move-to (buffer-substring (point) (point-at-eol)))
 	(message-remove-header gnus-agent-target-move-group-header))
+
+      ;; Force Gnus to delete Date to have a newer one
+      ;; First ensure Date is to be regenerated :)
+      (when (memq 'Date message-required-mail-headers)
+        (message-remove-header "Date" t))
+
       (goto-char (point-min))
       (when (re-search-forward
 	     (concat "^" (regexp-quote gnus-agent-meta-information-header) ":")

[-- Attachment #1.3: Type: text/plain, Size: 204 bytes --]

-----%<---------%<---------%<---------%<---------%<---------%<----
-- 
Xavier Maillard
7 rue Jeanne Jugan, 51100 Reims, France
phone: +33 3 26 77 02 21, mobile: +33 6 68 04 64 37
email: zedek@gnu-rox.org

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

Re: [PROPOSAL]: Date header and Queued messages

[Simon Josefsson]

Xavier Maillard <zedek@gnu-rox.org> writes:

> For my case it result to something quite annoying since I write my
> mails when I am in the train and can't send them before the evening so
> it generally result in something with 14/15h late !

2822 has this to say on the semantics of Date:

,----
| 3.6.1. The origination date field
| 
|    The origination date field consists of the field name "Date" followed
|    by a date-time specification.
| 
| orig-date       =       "Date:" date-time CRLF
| 
|    The origination date specifies the date and time at which the creator
|    of the message indicated that the message was complete and ready to
|    enter the mail delivery system.  For instance, this might be the time
|    that a user pushes the "send" or "submit" button in an application
|    program.  In any case, it is specifically not intended to convey the
|    time that the message is actually transported, but rather the time at
|    which the human or other creator of the message has put the message
|    into its final form, ready for transport.  (For example, a portable
|    computer user who is not connected to a network might queue a message
|    for delivery.  The origination date is intended to contain the date
|    and time that the user queued the message, not the time when the user
|    connected to the network to send the message.)
`----

It says it is "not intended to convey the time that the message is
actually transported", although I can see how some of the other
wording can be read as meaning something else -- by using the argument
that the queue might not be the "final" form of the message, since you
can edit messages in the queue.

Further, having the Date: header say when Gnus sent a message through
SMTP or /usr/bin/sendmail make it impossible for the receiver to find
out when the message was created.  The time the message is sent is
visible in other headers already.

My vote is for keeping things as is, so that Date: indicate when you
finished creating a message that is eventually sent.

Perhaps if you edit a message in the queue, it should replace the
Date: header.  Maybe it already does.

But it should be simple to achieve the behaviour you want, via some
hook.  I couldn't find any appropriate hook.  Perhaps replacing the
code in your patch with some code that invoke a
gnus-agent-send-queue-article-hook or something.

Re: [PROPOSAL]: Date header and Queued messages

[Steve Youngs]

[-- Attachment #1: Type: text/plain, Size: 685 bytes --]

* Simon Josefsson <jas@extundo.com> writes:

  > Xavier Maillard <zedek@gnu-rox.org> writes:
  >> For my case it result to something quite annoying since I write my
  >> mails when I am in the train and can't send them before the evening so
  >> it generally result in something with 14/15h late !

How can that possibly matter to the readers of your email messages?

  > My vote is for keeping things as is

That would be my vote too.

-- 
|---<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: [PROPOSAL]: Date header and Queued messages

[Katsumi Yamaoka]

>>>>> In <microsoft-free.871xq073fl.fsf@eicq.dnsalias.org>
>>>>>	Steve Youngs <sryoungs@bigpond.net.au> wrote:

> * Simon Josefsson <jas@extundo.com> writes:

>> Xavier Maillard <zedek@gnu-rox.org> writes:
>>> For my case it result to something quite annoying since I write my
>>> mails when I am in the train and can't send them before the evening so
>>> it generally result in something with 14/15h late !

> How can that possibly matter to the readers of your email messages?

>> My vote is for keeping things as is

> That would be my vote too.

I vote to Xavier Maillard.  Gnus is fully customizable, isn't it?
I regard the proposal of RFC2822 as nonsense.  Even if he
violates it, nobody is troubled and nobody notices.

Of course, he can do what he wants to do using some hooks, e.g.,
message-header-hook, maybe.
-- 
Katsumi Yamaoka <yamaoka@jpl.org>

Re: : Date header and Queued messages

[Adam Sjøgren]

On Fri, 16 Jan 2004 05:17:41 +0100, Xavier wrote:

> Why do we put the 'Date' header even if the message is pushed into
> the queued group ?

Perhaps because the message is regarded as "done" when it's queued? In
view of what rfc 2822 says...

Especially when you see that messsages saved in drafts, do not get a
Date (they aren't "done")... a logical difference between messages in
'drafts' and in 'queue'. Does it make sense?

(Maybe that could be another solution in your particular case: Save
the messages in drafts and then send them all when connected (D S)?

I've used that when I've written something late at night and didn't
feel like revealing that O:-))


  Best regards,

-- 
 "My baby is my sun                                           Adam Sjøgren
  Not the burning one"                                   asjo@koldfront.dk

Re: [PROPOSAL]: Date header and Queued messages

[Reiner Steib]

On Fri, Jan 16 2004, Katsumi Yamaoka wrote:

>>>>>> In <microsoft-free.871xq073fl.fsf@eicq.dnsalias.org>
>>>>>>	Steve Youngs <sryoungs@bigpond.net.au> wrote:
>> * Simon Josefsson <jas@extundo.com> writes:
[...]
>>> My vote is for keeping things as is
>> That would be my vote too.

Mine too.

> I vote to Xavier Maillard.  Gnus is fully customizable, isn't it?
> I regard the proposal of RFC2822 as nonsense.  Even if he
> violates it, nobody is troubled and nobody notices.

If the patch gets installed, removing of "Date" should be customizable
(defaulting to the current behavior).

> Of course, he can do what he wants to do using some hooks, e.g.,
> message-header-hook, maybe.

And if there's no suitable hook yet, we could add one.

BTW: Xavier, why do you set the IS-REGEXP argument of
`message-remove-header' to `t'?  It should be nil, IMHO.

+      (when (memq 'Date message-required-mail-headers)
+        (message-remove-header "Date" t))

Additionally, your code also applies to messages from "nndraft:drafts"
when sent with `D s' (gnus-draft-send-message).  If there already is a
date header, I don't think it's a good idea to remove it in this case.

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

Re: [PROPOSAL]: Date header and Queued messages

[Xavier Maillard]

[-- Attachment #1: Type: text/plain, Size: 1596 bytes --]

On 16 Jan 2004, Reiner Steib said:

> On Fri, Jan 16 2004, Katsumi Yamaoka wrote:
> 
> > > > > > > In <microsoft-free.871xq073fl.fsf@eicq.dnsalias.org>
> > > > > > > 	Steve Youngs <sryoungs@bigpond.net.au> wrote:
> > > * Simon Josefsson <jas@extundo.com> writes:
> [...]
> > > > My vote is for keeping things as is
> > > That would be my vote too.
> 
> Mine too.
> 
> > I vote to Xavier Maillard.  Gnus is fully customizable, isn't it?
> > I regard the proposal of RFC2822 as nonsense.  Even if he
> > violates it, nobody is troubled and nobody notices.
> 
> If the patch gets installed, removing of "Date" should be customizable
> (defaulting to the current behavior).

I agree with that.
 
> > Of course, he can do what he wants to do using some hooks, e.g.,
> > message-header-hook, maybe.
> 
> And if there's no suitable hook yet, we could add one.
> 
> BTW: Xavier, why do you set the IS-REGEXP argument of
> `message-remove-header' to `t'?  It should be nil, IMHO.

Typo :)
 
> +      (when (memq 'Date message-required-mail-headers)
> +        (message-remove-header "Date" t))
> 
> Additionally, your code also applies to messages from "nndraft:drafts"
> when sent with `D s' (gnus-draft-send-message).  If there already is a
> date header, I don't think it's a good idea to remove it in this case.

My guess is that 'gnus-draft-send-message' actually calls
'gnus-draft-send' function. So normally it should not matter.

zeDek
-- 
No e-patents, pas de brevets logiciels
Pétition contre les brevets logiciels : http://petition.eurolinux.org


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

Re: [PROPOSAL]: Date header and Queued messages

[Xavier Maillard]

[-- Attachment #1: Type: text/plain, Size: 697 bytes --]

On 16 Jan 2004, Steve Youngs spake thusly:

> * Simon Josefsson <jas@extundo.com> writes:
> 
> > Xavier Maillard <zedek@gnu-rox.org> writes:
> > > For my case it result to something quite annoying since I write my
> > > mails when I am in the train and can't send them before the
> > > evening so it generally result in something with 14/15h late !
> 
> How can that possibly matter to the readers of your email messages?

Not the readers but me :) In fact it doesn't matter that much but I
find it more convenient and I almost fear people won't see the messages
I post when the date header is quite old (ie 2/3 days old).

zeDek
-- 
Xavier Maillard
http://www.gnu-rox.org/~zedek/cgi-bin/wiki.pl


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

Re: : Date header and Queued messages

[Xavier Maillard]

[-- Attachment #1: Type: text/plain, Size: 1418 bytes --]

On 16 Jan 2004, Adam Sjøgren said:

> On Fri, 16 Jan 2004 05:17:41 +0100, Xavier wrote:
> 
> > Why do we put the 'Date' header even if the message is pushed into
> > the queued group ?
> 
> Perhaps because the message is regarded as "done" when it's queued? In
> view of what rfc 2822 says...

Yeah i was not aware of that point. I do not think my mail exchange is
done when I just say 'Send' I considerate it is done when I press
'Send' and gnus is effectively plugged.
 
> Especially when you see that messsages saved in drafts, do not get a
> Date (they aren't "done")... a logical difference between messages in
> 'drafts' and in 'queue'. Does it make sense?

I was talking about 'queued' messages that *do* have a Date header (for
sure).
 
> (Maybe that could be another solution in your particular case: Save
> the messages in drafts and then send them all when connected (D S)?

Yeah maybe but my problem is I keep 'templates' of mails in my draft
group so it may be a problem ;)
 
> I've used that when I've written something late at night and didn't
> feel like revealing that O:-))

So do I :) It is one of the point I wanted to modify because my
employer maybe perplex if he would see how weird is my life :)
 
Regards,

zeDek
-- 
Xavier Maillard
7 rue Jeanne Jugan, 51100 Reims, France
phone: +33 3 26 77 02 21, mobile: +33 6 68 04 64 37
email: zedek@gnu-rox.org


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

Re: [PROPOSAL]: Date header and Queued messages

[Reiner Steib]

On Fri, Jan 16 2004, Xavier Maillard wrote:

> On 16 Jan 2004, Reiner Steib said:
[...]
>> Additionally, your code also applies to messages from "nndraft:drafts"
>> when sent with `D s' (gnus-draft-send-message).  If there already is a
>> date header, I don't think it's a good idea to remove it in this case.
>
> My guess is that 'gnus-draft-send-message' actually calls
> 'gnus-draft-send' function. 

Yes.

> So normally it should not matter.

?  When a user has set a Date header in a postpone message, it should
_not_ be removed, when the message is finally sent using `D s'.

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

Re: [PROPOSAL]: Date header and Queued messages

[Steve Youngs]

[-- Attachment #1: Type: text/plain, Size: 461 bytes --]

* Xavier Maillard <zedek@gnu-rox.org> writes:

  > In fact it doesn't matter that much but I find it more convenient

If it is just a convenience to you then a local hack via some hook
would be more appropriate.

-- 
|---<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: [PROPOSAL]: Date header and Queued messages

[Jochen Küpper]

On Fri, 16 Jan 2004 06:43:36 +0100 Simon Josefsson wrote:

Simon> Xavier Maillard <zedek-icAv0V0KtPVAfugRpC6u6w@public.gmane.org> writes:

>> For my case it result to something quite annoying since I write my
>> mails when I am in the train and can't send them before the evening so
>> it generally result in something with 14/15h late !

Simon> My vote is for keeping things as is, so that Date: indicate
Simon> when you finished creating a message that is eventually sent.

I totally agree here.

I often write messages at night and only send them the next morning,
bypassing additional emails (often from other timezones). Specifying
the time the message was actually /written/ allows the receiver to see
that I could not possibly have known any information sent later.

Simon> Perhaps if you edit a message in the queue, it should replace
Simon> the Date: header.

IMHO it should:)

Greetings,
Jochen
-- 
Einigkeit und Recht und Freiheit                http://www.Jochen-Kuepper.de
    Liberté, Égalité, Fraternité                GnuPG key: CC1B0B4D
        (Part 3 you find in my messages before fall 2003.)

Re: [PROPOSAL]: Date header and Queued messages

[Karl Pflästerer]

On 16 Jan 2004, Steve Youngs <- sryoungs@bigpond.net.au wrote:

> * Xavier Maillard <zedek@gnu-rox.org> writes:

>   > In fact it doesn't matter that much but I find it more convenient

> If it is just a convenience to you then a local hack via some hook
> would be more appropriate.

That's also my opinion.  Hooks at neraly every stage of e-mail
processing exist.

   KP

-- 
And as in uffish thought he stood,
The Jabberwock, with eyes of flame,
Came whiffling through the tulgey wood,
And burbled as it came!                "Lewis Carroll" "Jabberwocky"

Re: [PROPOSAL]: Date header and Queued messages

[Xavier Maillard]

[-- Attachment #1: Type: text/plain, Size: 883 bytes --]

On 17 Jan 2004, Jochen Küpper told this:

> Simon> My vote is for keeping things as is, so that Date: indicate
> Simon> when you finished creating a message that is eventually sent.
> 
> I totally agree here.
> 
> I often write messages at night and only send them the next morning,
> bypassing additional emails (often from other timezones). Specifying
> the time the message was actually /written/ allows the receiver to see
> that I could not possibly have known any information sent later.

Hmm yes, I see your point. It is valid (at least it seems to me) and I
must admit I didn't think about that :)

Ok now I think I clearly don't want Date header to be overloaded when
_really_ sending mails ;)

Thank your for this explanation
 
zeDek
-- 
No e-patents, pas de brevets logiciels
Pétition contre les brevets logiciels : http://petition.eurolinux.org


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

Re: [PROPOSAL]: Date header and Queued messages

[Lars Magne Ingebrigtsen]

Xavier Maillard <zedek@gnu-rox.org> writes:

>> Simon> My vote is for keeping things as is, so that Date: indicate
>> Simon> when you finished creating a message that is eventually sent.
>> 
>> I totally agree here.
>> 
>> I often write messages at night and only send them the next morning,
>> bypassing additional emails (often from other timezones). Specifying
>> the time the message was actually /written/ allows the receiver to see
>> that I could not possibly have known any information sent later.
>
> Hmm yes, I see your point. It is valid (at least it seems to me) and I
> must admit I didn't think about that :)
>
> Ok now I think I clearly don't want Date header to be overloaded when
> _really_ sending mails ;)

So, did anybody fix this?  :-)

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