Re: [O] [PATCH RFC] subtree archive hook?

[Eric Abrahamsen]

Aaron Ecay <address@hidden> writes:

> Hi Eric,
>
> Looks like a sensible feature.  One comment:
>
> 2014ko urriak 12an, Eric Abrahamsen-ek idatzi zuen:
>> 
>> I think it would be useful to have a hook that runs before archiving a
>> subtree. I'm attaching two patches: one that includes a hook in the
>> archive process, and another (by way of an example) that adds a function
>> to that hook for the org-attach library. You can set the option
>> `org-attach-archive-delete' to a non-nil value to have org-attach delete
>> a subtree's attachments when you archive it.
>> 
>> Let me know what you think!
>> 
>> Eric
>> 
>> From 1bfc84570f29dd884c2759dfe19116f09228ed4e Mon Sep 17 00:00:00 2001
>> From: Eric Abrahamsen <address@hidden>
>> Date: Sun, 12 Oct 2014 22:01:29 +0800
>> Subject: [PATCH 2/3] Provide a hook during the archive process
>> 
>> * lisp/org-archive.el (org-archive-hook): New hook.
>>   (org-archive-subtree): Run hook.
>> ---
>>  lisp/org-archive.el | 13 +++++++++++--
>>  1 file changed, 11 insertions(+), 2 deletions(-)
>> 
>> diff --git a/lisp/org-archive.el b/lisp/org-archive.el
>> index 700e59b..c7f02b9 100644
>> --- a/lisp/org-archive.el
>> +++ b/lisp/org-archive.el
>> @@ -119,6 +119,13 @@ information."
>>            (const :tag "Outline path" olpath)
>>            (const :tag "Local tags" ltags)))
>>  
>> +(defvar org-archive-hook nil
>> +  "Hook run after successfully archiving a subtree.
>> +
>> +Hook functions are called with point on the subtree in the
>> +original file. At this stage, the subtree has been added to the
>> +archive location, but not yet deleted from the original file.")
>> +
>>  (defun org-get-local-archive-location ()
>>    "Get the archive location applicable at point."
>>    (let ((re "^[ \t]*#\\+ARCHIVE:[ \t]+\\(\\S-.*\\S-\\)[ \t]*$")
>> @@ -366,8 +373,10 @@ this heading."
>>          ;; Save and kill the buffer, if it is not the same buffer.
>>          (when (not (eq this-buffer buffer))
>>            (save-buffer))))
>> -    ;; Here we are back in the original buffer.  Everything seems to have
>> -    ;; worked.  So now cut the tree and finish up.
>> +    ;; Here we are back in the original buffer.  Everything seems
>> +    ;; to have worked.  So now run hooks, cut the tree and finish
>> +    ;; up.
>> +    (run-hooks 'org-archive-hook)
>>      (let (this-command) (org-cut-subtree))
>>      (when (featurep 'org-inlinetask)
>>        (org-inlinetask-remove-END-maybe))
>
> Can the above inlinetask thing also be moved into the hook?  That
> seems cleaner, and gives another demonstration of the usefulness of
> the feature.

Here's a patch that does it, though I'm a little more cautious about
this since I only did a minimal test.

Two things that worry me: 1) why is it called "remove-END-maybe" when it
appears to remove the whole inlinetask, and 2) it its original habitat
in org-attach, it came after the call to org-cut-subtree, meaning that
it couldn't have operated on the subtree to be archived at all! Or am I
misunderstanding something? I tried it on a test subtree, and the
org-cut-subtree took out the included inlinetask as well.

Anyway, it's a little mysterious, and I'm less confident about this bit.

0004-Move-deletion-of-inlinetasks-to-archive-hook.patch
Description: Text Data