It often happens to me that I send a message (Gnus) and need to keep a reference of the mail just sent as link in an org file. In Gnus copies of sent messages can be store in one or more groups using the `Gcc' header. So I usually select the group where the copy went right after sending and do an `org-store-link' on the newly created copy. It would of course be much nicer if the org link could be stored automagically upon sending the message (or more precisely upon creating the copy of the message). Such functionality could certainly be introduced in `gnus-inews-do-gcc' where group and article number of the copy are determined and the actual copy is written. But that would mean to actually modify Gnus code which is probably not a good and clean approach. As I'm not very familiar with the Gnus code I'll probably have to ask at the Gnus towers if it's possible to get hold of the group and article number of the last Gcc message created. But maybe someone around here has a partly or completely different idea how to achieve the described behaviour? Ulf
Ulf Stegemann <ulf-news@zeitform.de> writes: > It often happens to me that I send a message (Gnus) and need to keep a > reference of the mail just sent as link in an org file. [...] > It would of course be much nicer if the org link could be stored > automagically upon sending the message (or more precisely upon creating > the copy of the message). In Gnus, just like in other Emacs applications, there is a lot of hooks and I am quite sure there is one we could use. I am going to investigate the problem soon as I also need the feature you descibed. > As I'm not very familiar with the Gnus code I'll probably have to ask at > the Gnus towers if it's possible to get hold of the group and article > number AFAIR, message-id is used for org links rather than the articla number. > of the last Gcc message created. But maybe someone around here > has a partly or completely different idea how to achieve the described > behaviour? Which simplifies the problem quite a lot because we need to be able to see the buffer just before sending or copying it to Gcc folder. All the information is there. -- Miłego dnia, Łukasz Stelmach
Łukasz Stelmach <lukasz.stelmach@iem.pw.edu.pl> wrote: > Ulf Stegemann <ulf-news@zeitform.de> writes: [snipped 15 lines] > AFAIR, message-id is used for org links rather than the articla > number. ah, right, that's at least what raw gnus-article links look like. I remember vaguely that this was different in the past and I obviously hadn't checked again. >> of the last Gcc message created. But maybe someone around here >> has a partly or completely different idea how to achieve the described >> behaviour? > > Which simplifies the problem quite a lot because we need to be able to > see the buffer just before sending or copying it to Gcc folder. All the > information is there. Yes, this makes things a lot easier :) IIUC we just need to parse the Gcc header for the group(s), grab the message-id and put things together. We should however keep in mind that we need to parse the Gcc header _before_ `gnus-inews-do-gcc' copies the message as the Gcc header is removed in the process. But the org link should probably be built _after_ `gnus-inews-do-gcc' to avoid bogus org links in case the Gcc fails. Additionally, it's probably not necessary to create a link for every group in Gcc but only for one (first one?). Furthermore, storing an org link should only happen when somehow requested by the user and not on every send. Although prefix args of `message-send' and friends already serve a different purpose that nevertheless seems to be trivial. Ulf
I came back to the matter of storing an org link to a copy of a message upon sending that message. The function below does just that and proves to be quite useful together with a sensible key binding (if you use Gnus and Gcc that is). --8<------------------cut here----------------start---------------->8--- (defun ulf-message-send-and-org-gnus-store-link (&optional arg) "Send message with `message-send-and-exit' and store org link to message copy. If multiple groups appear in the Gcc header, the link refers to the copy in the last group." (interactive "P") (save-excursion (save-restriction (message-narrow-to-headers) (let ((gcc (car (last (message-unquote-tokens (message-tokenize-header (mail-fetch-field "gcc" nil t) " ,"))))) (buf (current-buffer)) id to from subject desc link newsgroup xarchive) (message-send-and-exit arg) (or ;; gcc group found ... (and gcc (save-current-buffer (progn (set-buffer buf) (setq id (org-remove-angle-brackets (mail-fetch-field "Message-ID"))) (setq to (mail-fetch-field "To")) (setq from (mail-fetch-field "From")) (setq subject (mail-fetch-field "Subject")))) (org-store-link-props :type "gnus" :from from :subject subject :message-id id :group gcc :to to) (setq desc (org-email-link-description)) (setq link (org-gnus-article-link gcc newsgroup id xarchive)) (setq org-stored-links (cons (list link desc) org-stored-links))) ;; no gcc group found ... (message "Can not create Org link: No Gcc header found.")))))) (define-key message-mode-map [(control c) (control meta c)] 'ulf-message-send-and-org-gnus-store-link) --8<------------------cut here-----------------end----------------->8--- Ulf
Ulf Stegemann <ulf-news@zeitform.de> writes:
> I came back to the matter of storing an org link to a copy of a message
> upon sending that message. The function below does just that and proves
> to be quite useful together with a sensible key binding (if you use Gnus
> and Gcc that is).
Very nice! Works very well. Thanks.
Instead of binding this to a command sequence, is there any hook we can
attach the function to so that it's invoked for every message I send
(I'm somewhat of a gnus noob unfortunately although I'm learning...)? I
ask because I alternative between =C-c C-c= and =C-c C-j= for sending
emails and I would need to provide two alternatives to incorporate your
function.
Thanks again,
eric
--
: Eric S Fraga (GnuPG: 0xC89193D8FFFCF67D) in Emacs 23.2.1
: using Org-mode version 7.3 (release_7.3.42.g0fd5ec)
Eric S. Fraga <ucecesf@ucl.ac.uk> wrote:
> Ulf Stegemann <ulf-news@zeitform.de> writes:
>
>> I came back to the matter of storing an org link to a copy of a message
>> upon sending that message. The function below does just that and proves
>> to be quite useful together with a sensible key binding (if you use Gnus
>> and Gcc that is).
>
> Very nice! Works very well. Thanks.
>
> Instead of binding this to a command sequence, is there any hook we can
> attach the function to so that it's invoked for every message I send
> (I'm somewhat of a gnus noob unfortunately although I'm learning...)? I
> ask because I alternative between =C-c C-c= and =C-c C-j= for sending
> emails and I would need to provide two alternatives to incorporate your
> function.
hmmm, never thought of this, probably because I never use
`gnus-delay-article'. You could try to put the function into
`message-send-hook' (but you should remove the call to
`message-send-and-exit' first). I don't know if this would work as I
don't know if the Gcc magic (incl. removal of Gcc header) is done before
or after that hook. If the Gcc header has been removed when
`message-send-hook' is called than you'll have to look for a different
hook to use. If the Gcc header is still present, it should probably
work as expected but note that in that case the org link will be created
before the actual copying of the message took place. If something goes
wrong with the latter this will leave you with a bogus org link (which,
however, is probably not much of a problem).
Ulf
Ulf Stegemann <ulf-news@zeitform.de> writes:
> Eric S. Fraga <ucecesf@ucl.ac.uk> wrote:
>
>> Ulf Stegemann <ulf-news@zeitform.de> writes:
>>
>>> I came back to the matter of storing an org link to a copy of a message
>>> upon sending that message. The function below does just that and proves
>>> to be quite useful together with a sensible key binding (if you use Gnus
>>> and Gcc that is).
>>
>> Very nice! Works very well. Thanks.
>>
>> Instead of binding this to a command sequence, is there any hook we can
>> attach the function to so that it's invoked for every message I send
>> (I'm somewhat of a gnus noob unfortunately although I'm learning...)? I
>> ask because I alternative between =C-c C-c= and =C-c C-j= for sending
>> emails and I would need to provide two alternatives to incorporate your
>> function.
>
> hmmm, never thought of this, probably because I never use
> `gnus-delay-article'. You could try to put the function into
> `message-send-hook' (but you should remove the call to
> `message-send-and-exit' first). I don't know if this would work as I
> don't know if the Gcc magic (incl. removal of Gcc header) is done before
> or after that hook. If the Gcc header has been removed when
> `message-send-hook' is called than you'll have to look for a different
> hook to use. If the Gcc header is still present, it should probably
> work as expected but note that in that case the org link will be created
> before the actual copying of the message took place. If something goes
> wrong with the latter this will leave you with a bogus org link (which,
> however, is probably not much of a problem).
Thanks. I realised after sending my earlier message that this is going
to be quite complicated. There's no point in storing a link when the
message is actually sent (after all, that will be sometime later). I
would want to store the link when I send the message to the delay queue
(so that I can do something with the link, of course) so there really
need to be two versions of your function.
Should be doable. I'll think about it some more.
In any case, most of my emails are sent immediately so your function is
very useful already.
Thanks again,
eric
--
: Eric S Fraga (GnuPG: 0xC89193D8FFFCF67D) in Emacs 23.2.1
: using Org-mode version 7.3 (release_7.3.42.g0fd5ec)
Ulf Stegemann <ulf-news@zeitform.de> wrote:
> Eric S. Fraga <ucecesf@ucl.ac.uk> wrote:
>
>> Ulf Stegemann <ulf-news@zeitform.de> writes:
>>
>>> I came back to the matter of storing an org link to a copy of a message
>>> upon sending that message. The function below does just that and proves
>>> to be quite useful together with a sensible key binding (if you use Gnus
>>> and Gcc that is).
>>
>> Very nice! Works very well. Thanks.
>>
>> Instead of binding this to a command sequence, is there any hook we can
>> attach the function to so that it's invoked for every message I send
>> (I'm somewhat of a gnus noob unfortunately although I'm learning...)? I
>> ask because I alternative between =C-c C-c= and =C-c C-j= for sending
>> emails and I would need to provide two alternatives to incorporate your
>> function.
>
> hmmm, never thought of this, probably because I never use
> `gnus-delay-article'. You could try to put the function into
> `message-send-hook' (but you should remove the call to
> `message-send-and-exit' first). I don't know if this would work as I
> don't know if the Gcc magic (incl. removal of Gcc header) is done before
> or after that hook. If the Gcc header has been removed when
> `message-send-hook' is called than you'll have to look for a different
> hook to use. If the Gcc header is still present, it should probably
> work as expected but note that in that case the org link will be created
> before the actual copying of the message took place. If something goes
> wrong with the latter this will leave you with a bogus org link (which,
> however, is probably not much of a problem).
If the hook approach does not work you could also advise
`gnus-inews-do-gcc', the function that actually creates the message
copies ...
... or you could replace `gnus-inews-do-gcc', e.g. replace the
call to `message-send-and-exit' in
`ulf-message-send-and-org-gnus-store-link' with something like
`gnus-inews-do-gcc-orig' and save it as e.g.
`my-gnus-gcc-and-org-store-link'. Then you could do something like
(fset 'gnus-inews-do-gcc-orig (symbol-function 'gnus-inews-do-gcc))
(fset 'gnus-inews-do-gcc 'my-gnus-gcc-and-org-store-link)
... but that's all untested ;)
Ulf
Ulf Stegemann <ulf-news@zeitform.de> writes:
[...]
>
> If the hook approach does not work you could also advise
> `gnus-inews-do-gcc', the function that actually creates the message
> copies ...
>
> ... or you could replace `gnus-inews-do-gcc', e.g. replace the
> call to `message-send-and-exit' in
> `ulf-message-send-and-org-gnus-store-link' with something like
> `gnus-inews-do-gcc-orig' and save it as e.g.
> `my-gnus-gcc-and-org-store-link'. Then you could do something like
>
> (fset 'gnus-inews-do-gcc-orig (symbol-function 'gnus-inews-do-gcc))
> (fset 'gnus-inews-do-gcc 'my-gnus-gcc-and-org-store-link)
>
> ... but that's all untested ;)
two very good suggestions. Thanks! I'll try to play with these over
the weekend if I get a chance.
--
: Eric S Fraga (GnuPG: 0xC89193D8FFFCF67D) in Emacs 23.2.1
: using Org-mode version 7.3 (release_7.3.42.g0fd5ec)
Eric S. Fraga <ucecesf@ucl.ac.uk> wrote:
> Thanks. I realised after sending my earlier message that this is going
> to be quite complicated. There's no point in storing a link when the
> message is actually sent (after all, that will be sometime later). I
> would want to store the link when I send the message to the delay queue
> (so that I can do something with the link, of course) so there really
> need to be two versions of your function.
>
> Should be doable. I'll think about it some more.
ah, now I seem to understand. With your scenario I'd probably remove the
call to `message-send-and-exit' from
`ulf-message-send-and-org-gnus-store-link' and advise both
`message-send-and-exit' and `gnus-delay-article' to run the modified
function before execution. This would leave you with a stored link
right after a `C-c C-c' and `C-c C-j'. But in case of the latter, the
org link will be bogus until the message has actually been sent. Could
that be a way to go for you?
Ulf
Ulf Stegemann <ulf-news@zeitform.de> writes:
> Eric S. Fraga <ucecesf@ucl.ac.uk> wrote:
>
>> Thanks. I realised after sending my earlier message that this is going
>> to be quite complicated. There's no point in storing a link when the
>> message is actually sent (after all, that will be sometime later). I
>> would want to store the link when I send the message to the delay queue
>> (so that I can do something with the link, of course) so there really
>> need to be two versions of your function.
>>
>> Should be doable. I'll think about it some more.
>
> ah, now I seem to understand. With your scenario I'd probably remove the
> call to `message-send-and-exit' from
> `ulf-message-send-and-org-gnus-store-link' and advise both
> `message-send-and-exit' and `gnus-delay-article' to run the modified
> function before execution. This would leave you with a stored link
> right after a `C-c C-c' and `C-c C-j'. But in case of the latter, the
> org link will be bogus until the message has actually been sent. Could
> that be a way to go for you?
Very much so. Brilliant. It doesn't matter (much) that the link be
bogus until the message is sent; in my usage scenario, it's about audit
trails so I can't imagine needing to follow an org link right away.
I'll try this out.
Thanks,
eric
--
: Eric S Fraga (GnuPG: 0xC89193D8FFFCF67D) in Emacs 23.2.1
: using Org-mode version 7.3 (release_7.3.42.g0fd5ec)
Hi Ulf,
Ulf Stegemann <ulf-news@zeitform.de> writes:
> (defun ulf-message-send-and-org-gnus-store-link (&optional arg)
This is something I've been trying to achieve for very long! Great
you found a solution. A minor suggestion: the function should make
sure the buffer is not killed after the message is sent.
I added (message-kill-buffer-on-exit nil) like this:
--8<---------------cut here---------------start------------->8---
(defun ulf-message-send-and-org-gnus-store-link (&optional arg)
"Send message with `message-send-and-exit' and store org link to message copy.
If multiple groups appear in the Gcc header, the link refers to
the copy in the last group."
(interactive "P")
(save-excursion
(save-restriction
(message-narrow-to-headers)
(let ((gcc (car (last
(message-unquote-tokens
(message-tokenize-header
(mail-fetch-field "gcc" nil t) " ,")))))
(buf (current-buffer))
(message-kill-buffer-on-exit nil)
id to from subject desc link newsgroup xarchive)
(message-send-and-exit arg)
(or
;; gcc group found ...
(and gcc
(save-current-buffer
(progn (set-buffer buf)
(setq id (org-remove-angle-brackets
(mail-fetch-field "Message-ID")))
(setq to (mail-fetch-field "To"))
(setq from (mail-fetch-field "From"))
(setq subject (mail-fetch-field "Subject"))))
(org-store-link-props :type "gnus" :from from :subject subject
:message-id id :group gcc :to to)
(setq desc (org-email-link-description))
(setq link (org-gnus-article-link
gcc newsgroup id xarchive))
(setq org-stored-links
(cons (list link desc) org-stored-links)))
;; no gcc group found ...
(message "Can not create Org link: No Gcc header found."))))))
--8<---------------cut here---------------end--------------->8---
I also added an entry in org-hacks.org.
Thanks!
--
Bastien
Hi Bastien, Bastien <bastien.guerry@wikimedia.fr> wrote: > Ulf Stegemann <ulf-news@zeitform.de> writes: > >> (defun ulf-message-send-and-org-gnus-store-link (&optional arg) > > This is something I've been trying to achieve for very long! Great you > found a solution. A minor suggestion: the function should make sure > the buffer is not killed after the message is sent. > > I added (message-kill-buffer-on-exit nil) like this: good point, thanks for adding it. > I also added an entry in org-hacks.org. Okay, thanks! Ulf