I have tried in the cleanest environment I could think of, but still
get duplicated clock line. Here are the steps.

- CTRL+ALT+F1
  login to get a console
- env -i TERM=linux HOME=/tmp bash --login --norc --noprofile
  almost empty enviroment, except TERM and HOME, which emacs need
- git clone
  git HEAD, ccf52269aa670ef6f46012bafdec90aee953282b,
  Mon Aug 25 18:16:01 2014 +0200.
- cd org-mode; make autoloads; rm lisp/*.elc
  without make autoloads, org-clock-in will not be available.
- emacs -nw -q --no-site-file -l minimal-org.el
  minimal-org.el as said in section 1.4 Feedback.
  what added to load-path is org-mode/lisp. org-mode is the freshly
cloned work copy.
- create a new .org file
- add one header
- clock in
- add another header
- clock out
- c-u c-u c-u c-c c-x c-i

Are there any other place where an error in my Emacs or Org mode setup
might hide?

I do not know if these are relevant. But if I replace the recursively
call with setq, there will be no duplicated line.

There is a similar situation with superfluous clock lines. After
idling for more than org-clock-idle-time, there will be multiple
"[jkKgGSscCiq]?" question waiting for me in mini buffer. After
resolving 10, 10.1, 11 minutes of idle time, I get, again, multiple
clock lines, some of which contains start time bigger than end time.

I suppose the 2 lines bellow are meant to prevent re-entering
org-resolve-clocks-if-idle.

    876:  (let ((org-clock-resolving-clocks t))
   1056:    (let ((org-clock-resolving-clocks t))

But they do not suffice in my case. I have to replace (run-with-timer
60 60 with (run-with-time 60 nil, and add another timer at the end of
org-resolve-clocks-if-idle to be asked only once. Shadowing the value
with local one is not enough.

Does my emacs use different thread / variable binding / foo bar model
than yours? My Emacs Version is GNU Emacs 24.3.1 (x86_64-pc-linux-gnu,
GTK+ Version 3.12.2) of 2014-06-07 on barber, modified by Debian

On Fri, Aug 29, 2014 at 6:01 PM, Nicolas Goaziou <m...@nicolasgoaziou.fr> wrote:
> REN Lifeng <renlif...@gmail.com> writes:
>
>> Sorry. I gave the wrong steps. The last step should be C-u C-u C-u C-c
>> C-x C-i. Could you try again?
>
> This is what I did. I still cannot reproduce your problem. I also tried
> different values for `org-clock-into-drawer', namely nil, t and "CLOCK",
> to no avail.
>
>
> Regards,
>
> --
> Nicolas Goaziou

Reply via email to