Re: [O] An issue with org-agenda-todo-list-sublevels

[Marcin Borkowski]

On 2015-12-16, at 02:54, Nick Dokos  wrote:

> Marcin Borkowski  writes:
>
>> On 2015-12-14, at 21:48, Nick Dokos  wrote:
>>
>>> Marcin Borkowski  writes:
>>>
> On 2015-12-12, at 09:53, Nicolas Goaziou  wrote:
>
>> You can use a dedicated function in `org-agenda-skip-function' for that
>> (e.g., ignore task if one of its parents is a done task).

 OK, so it doesn't work (probably because I'm doing something wrong...)

 I did this:

 (defun mbork/org-agenda-skip-if-parent-done ()
   "Return t if any of the parents of the current entry is a DONE
   item."
   (save-excursion
 (catch 'done
   (while (org-up-heading-safe)
(if (org-entry-is-done-p)
(throw 'done t))

 (setq org-agenda-custom-commands
   '(("n"
 "Agenda and TODOs"
 ((agenda "")
  (alltodo "" ((org-agenda-skip-function 
 #'mbork/org-agenda-skip-if-parent-done)))

 and I see this:

 and: Wrong type argument: integer-or-marker-p, t

 What may be the problem?  How do I even debug this?

>>>
>>> Are you testing it with emacs -q -l /path/to/min/org-init.el?
>>> If not, you probably should: I don't get any errors with the
>>> stuff above in my minimal org file, which leads me to suspect
>>> it's something in (the rest of) your configuration.
>>
>> Strange.  I have the same error even with emacs -q, with minimal Org
>> config - the above function and one simple agenda file, containing this:
>>
>> * DONE done
>> ** TODO todo
>>
>> Org-mode version 8.2.10 (release_8.2.10 @ 
>> /usr/local/share/emacs/25.0.50/lisp/org/)
>>
>> Any hints?
>>
>> Here's the debugger output:
>>
>> Debugger entered--Lisp error: (wrong-type-argument integer-or-marker-p t)
>>   goto-char(t)
>>   (and (setq to (or (org-agenda-skip-eval org-agenda-skip-function-global) 
>> (org-agenda-skip-eval org-agenda-skip-function))) (goto-char to))
>
> Not so strange: I probably did not hit the (throw 'done t) in your
> function. The spec says that the function has to return a position
> in the buffer:
>
> ,
> | org-agenda-skip-function is a variable defined in ‘org-agenda.el’.
> | Its value is nil
> | 
> |   This variable may be risky if used as a file-local variable.
> | 
> | Documentation:
> | Function to be called at each match during agenda construction.
> | If this function returns nil, the current match should not be skipped.
> | Otherwise, the function must return a position from where the search
> | should be continued.
> `

Stupid me.  Thanks.

-- 
Marcin Borkowski
http://octd.wmi.amu.edu.pl/en/Marcin_Borkowski
Faculty of Mathematics and Computer Science
Adam Mickiewicz University

Re: [O] An issue with org-agenda-todo-list-sublevels

[Marcin Borkowski]

On 2015-12-14, at 21:48, Nick Dokos  wrote:

> Marcin Borkowski  writes:
>
>>> On 2015-12-12, at 09:53, Nicolas Goaziou  wrote:
>>>
 You can use a dedicated function in `org-agenda-skip-function' for that
 (e.g., ignore task if one of its parents is a done task).
>>
>> OK, so it doesn't work (probably because I'm doing something wrong...)
>>
>> I did this:
>>
>> (defun mbork/org-agenda-skip-if-parent-done ()
>>   "Return t if any of the parents of the current entry is a DONE
>>   item."
>>   (save-excursion
>> (catch 'done
>>   (while (org-up-heading-safe)
>>  (if (org-entry-is-done-p)
>>  (throw 'done t))
>>
>> (setq org-agenda-custom-commands
>>   '(("n"
>>   "Agenda and TODOs"
>>   ((agenda "")
>>(alltodo "" ((org-agenda-skip-function 
>> #'mbork/org-agenda-skip-if-parent-done)))
>>
>> and I see this:
>>
>> and: Wrong type argument: integer-or-marker-p, t
>>
>> What may be the problem?  How do I even debug this?
>>
>
> Are you testing it with emacs -q -l /path/to/min/org-init.el?
> If not, you probably should: I don't get any errors with the
> stuff above in my minimal org file, which leads me to suspect
> it's something in (the rest of) your configuration.

Strange.  I have the same error even with emacs -q, with minimal Org
config - the above function and one simple agenda file, containing this:

* DONE done
** TODO todo

Org-mode version 8.2.10 (release_8.2.10 @ 
/usr/local/share/emacs/25.0.50/lisp/org/)

Any hints?

Here's the debugger output:

--8<---cut here---start->8---
Debugger entered--Lisp error: (wrong-type-argument integer-or-marker-p t)
  goto-char(t)
  (and (setq to (or (org-agenda-skip-eval org-agenda-skip-function-global) 
(org-agenda-skip-eval org-agenda-skip-function))) (goto-char to))
  (or (save-excursion (goto-char p) (looking-at comment-start-skip)) (and 
org-agenda-skip-archived-trees (not org-agenda-archives-mode) 
(get-text-property p :org-archived) (org-end-of-subtree t)) (and 
org-agenda-skip-comment-trees (get-text-property p :org-comment) 
(org-end-of-subtree t)) (and (setq to (or (org-agenda-skip-eval 
org-agenda-skip-function-global) (org-agenda-skip-eval 
org-agenda-skip-function))) (goto-char to)) (org-in-src-block-p t))
  (if (or (save-excursion (goto-char p) (looking-at comment-start-skip)) (and 
org-agenda-skip-archived-trees (not org-agenda-archives-mode) 
(get-text-property p :org-archived) (org-end-of-subtree t)) (and 
org-agenda-skip-comment-trees (get-text-property p :org-comment) 
(org-end-of-subtree t)) (and (setq to (or (org-agenda-skip-eval 
org-agenda-skip-function-global) (org-agenda-skip-eval 
org-agenda-skip-function))) (goto-char to)) (org-in-src-block-p t)) (progn 
(throw :skip t)))
  (let ((p (point-at-bol)) to) (if (or (save-excursion (goto-char p) 
(looking-at comment-start-skip)) (and org-agenda-skip-archived-trees (not 
org-agenda-archives-mode) (get-text-property p :org-archived) 
(org-end-of-subtree t)) (and org-agenda-skip-comment-trees (get-text-property p 
:org-comment) (org-end-of-subtree t)) (and (setq to (or (org-agenda-skip-eval 
org-agenda-skip-function-global) (org-agenda-skip-eval 
org-agenda-skip-function))) (goto-char to)) (org-in-src-block-p t)) (progn 
(throw :skip t
  org-agenda-skip()
  org-agenda-get-todos()
  org-agenda-get-day-entries("/home/mbork/org/emacs.org" (12 16 2015) :todo)
  org-todo-list(nil)
  funcall-interactively(org-todo-list nil)
  call-interactively(org-todo-list)
  (let ((org-agenda-skip-function (function 
mbork/org-agenda-skip-if-parent-done))) (call-interactively (quote 
org-todo-list)))
  (let nil (let ((org-agenda-skip-function (function 
mbork/org-agenda-skip-if-parent-done))) (call-interactively (quote 
org-todo-list
  eval((let nil (let ((org-agenda-skip-function (function 
mbork/org-agenda-skip-if-parent-done))) (call-interactively (quote 
org-todo-list)
  org-let2(nil ((org-agenda-skip-function (function 
mbork/org-agenda-skip-if-parent-done))) (call-interactively (quote 
org-todo-list)))
  org-agenda-run-series("Agenda and TODOs" (((agenda "") (alltodo "" 
((org-agenda-skip-function (function mbork/org-agenda-skip-if-parent-done)))
  org-agenda(nil)
  funcall-interactively(org-agenda nil)
  call-interactively(org-agenda nil nil)
  command-execute(org-agenda)
--8<---cut here---end--->8---

-- 
Marcin Borkowski
http://octd.wmi.amu.edu.pl/en/Marcin_Borkowski
Faculty of Mathematics and Computer Science
Adam Mickiewicz University

Re: [O] An issue with org-agenda-todo-list-sublevels

[Nick Dokos]

Marcin Borkowski  writes:

> On 2015-12-14, at 21:48, Nick Dokos  wrote:
>
>> Marcin Borkowski  writes:
>>
 On 2015-12-12, at 09:53, Nicolas Goaziou  wrote:

> You can use a dedicated function in `org-agenda-skip-function' for that
> (e.g., ignore task if one of its parents is a done task).
>>>
>>> OK, so it doesn't work (probably because I'm doing something wrong...)
>>>
>>> I did this:
>>>
>>> (defun mbork/org-agenda-skip-if-parent-done ()
>>>   "Return t if any of the parents of the current entry is a DONE
>>>   item."
>>>   (save-excursion
>>> (catch 'done
>>>   (while (org-up-heading-safe)
>>> (if (org-entry-is-done-p)
>>> (throw 'done t))
>>>
>>> (setq org-agenda-custom-commands
>>>   '(("n"
>>>  "Agenda and TODOs"
>>>  ((agenda "")
>>>   (alltodo "" ((org-agenda-skip-function 
>>> #'mbork/org-agenda-skip-if-parent-done)))
>>>
>>> and I see this:
>>>
>>> and: Wrong type argument: integer-or-marker-p, t
>>>
>>> What may be the problem?  How do I even debug this?
>>>
>>
>> Are you testing it with emacs -q -l /path/to/min/org-init.el?
>> If not, you probably should: I don't get any errors with the
>> stuff above in my minimal org file, which leads me to suspect
>> it's something in (the rest of) your configuration.
>
> Strange.  I have the same error even with emacs -q, with minimal Org
> config - the above function and one simple agenda file, containing this:
>
> * DONE done
> ** TODO todo
>
> Org-mode version 8.2.10 (release_8.2.10 @ 
> /usr/local/share/emacs/25.0.50/lisp/org/)
>
> Any hints?
>
> Here's the debugger output:
>
> Debugger entered--Lisp error: (wrong-type-argument integer-or-marker-p t)
>   goto-char(t)
>   (and (setq to (or (org-agenda-skip-eval org-agenda-skip-function-global) 
> (org-agenda-skip-eval org-agenda-skip-function))) (goto-char to))

Not so strange: I probably did not hit the (throw 'done t) in your
function. The spec says that the function has to return a position
in the buffer:

,
| org-agenda-skip-function is a variable defined in ‘org-agenda.el’.
| Its value is nil
| 
|   This variable may be risky if used as a file-local variable.
| 
| Documentation:
| Function to be called at each match during agenda construction.
| If this function returns nil, the current match should not be skipped.
| Otherwise, the function must return a position from where the search
| should be continued.
`

-- 
Nick

Re: [O] An issue with org-agenda-todo-list-sublevels

[Marcin Borkowski]

> On 2015-12-12, at 09:53, Nicolas Goaziou  wrote:
>
>> You can use a dedicated function in `org-agenda-skip-function' for that
>> (e.g., ignore task if one of its parents is a done task).

OK, so it doesn't work (probably because I'm doing something wrong...)

I did this:

--8<---cut here---start->8---
(defun mbork/org-agenda-skip-if-parent-done ()
  "Return t if any of the parents of the current entry is a DONE
  item."
  (save-excursion
(catch 'done
  (while (org-up-heading-safe)
(if (org-entry-is-done-p)
(throw 'done t))

(setq org-agenda-custom-commands
  '(("n"
 "Agenda and TODOs"
 ((agenda "")
  (alltodo "" ((org-agenda-skip-function 
#'mbork/org-agenda-skip-if-parent-done)))
--8<---cut here---end--->8---

and I see this:

--8<---cut here---start->8---
and: Wrong type argument: integer-or-marker-p, t
--8<---cut here---end--->8---

What may be the problem?  How do I even debug this?

Best,

-- 
Marcin Borkowski
http://octd.wmi.amu.edu.pl/en/Marcin_Borkowski
Faculty of Mathematics and Computer Science
Adam Mickiewicz University

Re: [O] An issue with org-agenda-todo-list-sublevels

[Nick Dokos]

Marcin Borkowski  writes:

>> On 2015-12-12, at 09:53, Nicolas Goaziou  wrote:
>>
>>> You can use a dedicated function in `org-agenda-skip-function' for that
>>> (e.g., ignore task if one of its parents is a done task).
>
> OK, so it doesn't work (probably because I'm doing something wrong...)
>
> I did this:
>
> (defun mbork/org-agenda-skip-if-parent-done ()
>   "Return t if any of the parents of the current entry is a DONE
>   item."
>   (save-excursion
> (catch 'done
>   (while (org-up-heading-safe)
>   (if (org-entry-is-done-p)
>   (throw 'done t))
>
> (setq org-agenda-custom-commands
>   '(("n"
>"Agenda and TODOs"
>((agenda "")
> (alltodo "" ((org-agenda-skip-function 
> #'mbork/org-agenda-skip-if-parent-done)))
>
> and I see this:
>
> and: Wrong type argument: integer-or-marker-p, t
>
> What may be the problem?  How do I even debug this?
>

Are you testing it with emacs -q -l /path/to/min/org-init.el?
If not, you probably should: I don't get any errors with the
stuff above in my minimal org file, which leads me to suspect
it's something in (the rest of) your configuration.

--
Nick

Re: [O] An issue with org-agenda-todo-list-sublevels

[Nicolas Goaziou]

Hello,

Marcin Borkowski  writes:

> I have org-agenda-todo-list-sublevels set to nil.  However, given this
> tree:
>
> * DONE task
> ** TODO subtask
>
> it turns out that the todo list in the agenda shows the "subtask" anyway.
>
> Should it be so?  I would guess not.

It's not clear to to me it shouldn't be the case. AFAIU, the point of
`org-agenda-todo-list-sublevels' is not to skip TODO entries but to
limit the tasks you can process at a time.

> The reason I'd like to exclude this "subtask" from displaying is
> something like this: assume that I have a project, halfway done, which
> must be postponed for some reason.  I'd like to be able to do this:
>
> * SOMEDAY cool project
> ** DONE preparation
> ** TODO hard work
>
> where SOMEDAY is a done-type keyword.
>
> And of course, I don't want to be bothered by "hard work" in this case
> in my global todo list.
>
> Is there a way to achieve this?

You can use a dedicated function in `org-agenda-skip-function' for that
(e.g., ignore task if one of its parents is a done task).


Regards,

-- 
Nicolas Goaziou

Re: [O] An issue with org-agenda-todo-list-sublevels

[Marcin Borkowski]

On 2015-12-12, at 09:53, Nicolas Goaziou  wrote:

> Hello,
>
> Marcin Borkowski  writes:
>
>> I have org-agenda-todo-list-sublevels set to nil.  However, given this
>> tree:
>>
>> * DONE task
>> ** TODO subtask
>>
>> it turns out that the todo list in the agenda shows the "subtask" anyway.
>>
>> Should it be so?  I would guess not.
>
> It's not clear to to me it shouldn't be the case. AFAIU, the point of
> `org-agenda-todo-list-sublevels' is not to skip TODO entries but to
> limit the tasks you can process at a time.

I see.

>> The reason I'd like to exclude this "subtask" from displaying is
>> something like this: assume that I have a project, halfway done, which
>> must be postponed for some reason.  I'd like to be able to do this:
>>
>> * SOMEDAY cool project
>> ** DONE preparation
>> ** TODO hard work
>>
>> where SOMEDAY is a done-type keyword.
>>
>> And of course, I don't want to be bothered by "hard work" in this case
>> in my global todo list.
>>
>> Is there a way to achieve this?
>
> You can use a dedicated function in `org-agenda-skip-function' for that
> (e.g., ignore task if one of its parents is a done task).

Wow, thanks, I didn't know about it.  Now that I look into this, should
I use `org-agenda-skip-function' or `org-agenda-skip-function-global'?

Also, how do I check whether one of the parents is done?  I know how to
climb through the hierarchy of headlines, how do I get the TODO keyword?
Is `org-get-todo-state' the right function to do that?

Now that I think of it, maybe I should use a tag for that (like
:project: or something).  I'll have to think about it.

> Regards,

Best,

-- 
Marcin Borkowski
http://octd.wmi.amu.edu.pl/en/Marcin_Borkowski
Faculty of Mathematics and Computer Science
Adam Mickiewicz University

[O] An issue with org-agenda-todo-list-sublevels

[Marcin Borkowski]

Hi list,

I have org-agenda-todo-list-sublevels set to nil.  However, given this
tree:

* DONE task
** TODO subtask

it turns out that the todo list in the agenda shows the "subtask" anyway.

Should it be so?  I would guess not.

The reason I'd like to exclude this "subtask" from displaying is
something like this: assume that I have a project, halfway done, which
must be postponed for some reason.  I'd like to be able to do this:

* SOMEDAY cool project
** DONE preparation
** TODO hard work

where SOMEDAY is a done-type keyword.

And of course, I don't want to be bothered by "hard work" in this case
in my global todo list.

Is there a way to achieve this?
(Org-mode version 8.3beta (release_8.3beta-1136-g0e7062).)

If not, can it be considered a feature request?

A cursory examination of org-agenda.el shows that the culprit is the
function org-agenda-get-todos, and that (probably) this is responsible:

--8<---cut here---start->8---
 (regexp (format org-heading-keyword-regexp-format
 (cond
  ((and org-select-this-todo-keyword
(equal org-select-this-todo-keyword "*"))
   org-todo-regexp)
  (org-select-this-todo-keyword
   (concat "\\("
   (mapconcat 'identity
  (org-split-string
   org-select-this-todo-keyword
   "|")
  "\\|") "\\)"))
  (t org-not-done-regexp
--8<---cut here---end--->8---

(notice the last line, with `org-not-done-regexp'!)

A possible course of action might be to change that line to use
`org-todo-regexp'; OTOH, the exclusion of tasks marked as done should be
then done differently.

(If there is any interest, I might try to look into it further.)

Best,

-- 
Marcin Borkowski
http://octd.wmi.amu.edu.pl/en/Marcin_Borkowski
Faculty of Mathematics and Computer Science
Adam Mickiewicz University