Hi Rasmus,
thanks for your hint, indeed this particular commit changed the
check-box related algorithm; "git log" gives:
commit a4cc9d82d8069741be64c55f35f9d3ad7e2663d5
Author: Nicolas Goaziou <m...@nicolasgoaziou.fr>
Date: Tue Jun 9 17:06:17 2015 +0200
org-list: Fix checkbox update with inlinetasks
* lisp/org-list.el (org-update-checkbox-count): Change algorithm. Use
Element parser.
* testing/lisp/test-org-list.el (test-org-list/update-checkbox-count):
New test.
I then looked at this particular file's git log:
git log lisp/org-list.el
and checked out Nicolas' last revision BEFORE his commit on Tue Jun 9:
git checkout a0c1e490b30cc4c0dd9e9ecdd70c32a52b08a57e lisp/org-list.el
And indeed: everything works again as expected.
Nicolas, can you see what caused this regression and can it be fixed?
Thanks
Gerald.
On Mo, Aug 10 2015, Rasmus <ras...@gmx.us> wrote:
> Hi,
>
> I can reproduce.
>
> Gerald Wildgruber <gerald.wildgru...@unibas.ch> writes:
>
>> =====================================
>>
>> * TODO Heading [0/3]
>>
>> - [ ] entry 1
>> - [ ] entry 2 [0/0]
>> - [ ] sub-entry 1
>> - [ ] sub-entry 2
>> - [ ] sub-entry 3
>> - [ ] entry 3
>>
>> =====================================
>
> I don't have time to attempt to fix it now, but it's introduced in
> a4cc9d82d8069741be64c55f35f9d3ad7e2663d5 if my git bisect foo is correct.
>
> My guess would be that container goes too far back:
>
> + (container
> + (org-element-lineage
> + context
> + '(drawer center-block dynamic-block inlinetask
> plain-list
> + quote-block special-block verse-block))
>
> plain-list also includes entry 1, but it should only consider the sublist
> starting with entry 2.
>
> Rasmus
