I'm running planner 3.41, remember 1.9 and muse-latest (as of two days
ago), under GNU Emacs 21.3 and 21.4.

When I started using planner-timeclock, I discovered the following: If
the ~/.timelog file is non-existant (as it probably is the first time
one uses *timeclock), the first clock in
(e.g. planner-task-in-progress) goes okay, but the following clock out
(planner-task-done or timeclock-out) fails with "Wrong type argument:
number-or-marker-p, nil"

After the initial clock in, ~/.timelog contains one line with

i 2006/08/10 15:04:46 [[TaskPool]]: first task

The file is untouched by the (failed) clock out, but the .timelog
buffer contains the two lines

i 2006/08/10 15:04:46 [[TaskPool]]: first task
O 2006/08/10 15:05:11

I'm not a lisp expert, but a bit of debug-hunting showed that the
error is thrown by

(+ timeclock-discrepancy
   timeclock-last-period)

in timeclock-log(), defined in timeclock.el(*), the reason beging that
timeclock-discrepancy is nil.

Two work-arounds:  Either create an empty ~/.timelog prior to (require
planner-timeclock), or do M-x timeclock-reread-log after the initial
clock in (but prior to the clock out :-).

I don't know whether this is the intended behaviour.  If it is, it
would be nice to provide a hint in the planner documentation(**).  If it
isn't, perhaps something should be changed in timeclock.el and/or
planner-timeclock.el.

(*) The version that came with Emacs.  The file is marked as Version
2.6 (for both versions of Emacs).

(**) And if it is already documented, I'm sorry for wasting
everybody's time. :-)

-- 
Bjørn-Helge Mevik

_______________________________________________
Planner-el-discuss mailing list
[email protected]
https://mail.gna.org/listinfo/planner-el-discuss

Reply via email to