Bernt Hansen writes:
> Every so often I run into a situation where bulk refiling
> doesn't work anymore.
>
> I currently have 15 items in my refile.org file that I want
> to refile to other locations. I marked a few of them and
> bulk refiled them just fine. Then I marked a few more and B
> r fails with "Cannot find entry for marker #<marker at
> 297156 in norang.org>"
>
> I think this happens when I mark multiple tasks in the same
> subtree (i.e. the parent and a sibling) and refile both to
> the same location. After that it gets confused.
>
> If I have a task like this in refile.org
>
> #+FILETAGS: REFILE
> * Test
> ** Test 2
>
> and run a tags match on REFILE I see both tasks. Mark both
> with m in the agenda and B r to some other location. It
> refiles the first (and this moves the sibling too) and then
> it's broken after that.
>
> I get the following backtrace
>
> Debugger entered--Lisp error: (error "Cannot find entry for
> marker #<marker at 297156 in norang.org>")
Indeed the happens because, when a parent gets refiled or
achieved, any entries corresponding to its children are
removed from the agenda.
We address this issue by
- sorting the markers, to make sure parents will be handled
before children
- No longer throwing an error when a bulk action entry no
longer is present in the agenda - most likely it was taken
care of together with its parent.
Gregory Grubbs writes:
> When exporting a table with ido-mode active, an error is
> raised in org-ido-completing-read. I think
> ido-completing-read is being called with incorrect
> arguments, but the fix is beyond me.
>
> Steps to reproduce the error:
> Org-mode version: 6.28trans
> Emacs version: GNU Emacs 23.0.91.1 (i486-pc-linux-gnu, GTK+ Version
> 2.16.0) of 2009-04-05 on palmer, modified by Debian
>
>
> Turn on ido-mode: M-x ido-mode RET
> visit a file using C-x C-f /tmp/test.org RET
> Create a simple table:
> |column a|column b|
> |-
> |one|two|
> |three|four|
>
> Org-magic-tabelize it by hitting TAB somewhere in a column
>
> M-x org-table-export RET /tmp/test.csv
>
> Here's the backtrace I get:
>
> Debugger entered--Lisp error: (wrong-type-argument listp "orgtbl-to-tsv")
This error is due to the fact that org-ido-completing-read does
convert alists to flat lists for completion. Now we check if the list
really is an alist before converting it.
The first cycle command on a line did not work correctly, to always
switched to FOLD. This was due to an error in new, faster code to
find the next visible line.
* contrib/lisp/org-checklist.el
- Fix some problems with the print/export feature in org-checklist.
- Add custom options for this module
- Fix some interaction with a2ps
- Only reset checkboxes if TODO state is done.
* lisp/org.el
- Add org-checklist to org-modules
Gregory Grubbs writes:
I like to make org-mode tables in arbitrary buffers, then
save them as CVS files. The export fails when done from a
buffer with no associated file.
Here's a little patch that allows exporting a table from
any buffer (org-mode version 6.28trans):
Changing the priority of a task when the point is after the heading
(anywhere inside the task) worked but aligning the tags failed with
a "not on a heading" error due to the save-excursion not including
the tag alignment. This change moves back to the heading and
includes that during tag alignment to remove the error text.
Patch by Bernt Hansen.
Hsiu-Khuern writes:
> Hi all,
>
> The footnote at the bottom of section 13.1.4 ("Publishing
> action") of the Org manual says that publishing org files to
> the same directory using org-publish-org-to-org results in
> files named like file-source.org. It actually results in
> file.org-source, which is not as nice. I believe the
> problem is in the org-export-as-org function in org-exp.el.
Write \n instead of \xa in the regexp, this is clearer.
And make the \n optional so that also lines at the end of
the buffers will still be matched as headlines.
A relative row reference like @-1 in a table may now reach across a
horizontal separator line. I hope this will not break any important
tables out there, but I think it is the right thing to do.
The original reason for not-crossing was to implement running
averages of one column in the next. This can now be done using field
formulas near the beginning and end of the column, and a column
formula for the central part.
See the variable `org-table-relative-ref-may-cross-hline' for more
details.
Emacs recently (when?) changed the abbreviated format for days.
The clock was confused when trying to mach this format. This patch
by Nicolas Goaziou fixes it.
Brenton Kenkel writes:
> I'm using org-mode 6.28e on Aquamacs (based on GNU Emacs
> 22.3.1) on Mac OS X. I'm having an issue with horizontal
> rules and lists in HTML export. When I export the following
> to HTML, the horizontal rule tag is generated within the
> unordered list, rather than after it ends:
>
> ,-----
> | * test
> |
> | - this is a list
> |
> |
> | -----
> |
> | * next section
> |
> | No list.
> |
> |
> | -----
> |
> `-----
>
> I want the first horizontal rule to be even with the second;
> namely, extending all the way to the left. Any way to do
> this?
This commit defines three new functions in org-timer.el:
- org-timer-set-timer, bound to `C-c C-x ;' in Org buffers
and to `;' in Org agenda buffers. This function sets a
timer for the headline the cursor is currently it. Up to
three timers can be set.
- org-timer-show-remaining-time: this shows the remaining
time for the last set timer.
- org-timer-reset-timers: this reset all timers.
This functionality was requested by Samuel Wales and emulates
that of tea-time.el -- see the emacswiki doc about tea-time.el:
http://www.emacswiki.org/emacs/tea-time
When org-provide-todo-statistics is set to 'all-headlines, the
statistics are performed against all headlines, including those
with no TODO keyword.
This was requested by David A. Gershman.
Using :timetamp t as an option in a clock report now allows
insertion of the timestamp for the clocked entry. Timestamps
are searched for in this order: SCHEDULING, TIMESTAMP, DEADLINE
and TIMESTAMP_IA.
Allow whitespace in code references. Allow the -r switch to remove the
references in the source code even when the lines are not numbered: the
labels can be explicit enough. Note that "-r -k" is the same than no
switch - as expected.
Thanks to Ulf Stegemann for bring this up.
Non-nil means modify the regexp in org-set-font-lock-defaults so
that headlines are fontified till the end of the line. This is
useful when setting a background color for the org-level-* faces.
This defaults to nil so that the old behavior is preserved.
This option lets the user customize the notification mechanism.
For example, she might want to use todochiku.el.
This option defaults to nil, hence doesn't change the previous
behavior: if the program notify-send is installed on the system,
use it, and falls back on using (message [notification]) if not.
For this we took another look at when drawers actually have to be
hidden again and found that CONTENTS view does not need it, and that
CHILDREN view only needs it before the first child.
The second speed-up comes from advising outline-end-of-subtree to use
the Org version when in org-mode.
The third speed-up comes from using a better way to find the next
visible line, using `next-single-char-property-change'.
Finally, `org-forward-same-level' and `org-backward-same-level' are
faster versions of their outline equivalents and are now bound to
`C-c C-f' and `C-c C-b'.
`org-read-date' was loosing the focus when the calendar was displayed on
a separate frame. This patch by Robert Goldman solves it by introducing
a new macro `org-save-frame-excursion' which preserve the frame focus.
See this thread: http://article.gmane.org/gmane.emacs.orgmode/15528
(This fixes the issue I was trying to fix in a previous commit.)
This was a proposal by Samuel Wales. After the user selected a
refiles target, we make sure that the fully qualified target is in the
history, so that next time, UP will bring back exactly this target.
Before this patch the completion mechanism was this: TAB let's you
complete through link prefixes (gnus: file: bbdb: ...) then RET allows
completion if a completion mechanism is available for the chosen prefix.
Navigating through the history of stored links was a separate process,
available through the up/down M-n/M-p keys.
Now TAB not only completes through link prefixes but also through stored
links. This behavior matches other Emacs completion mechanisms a bit
more closely.
If you have a TODO keyword like "DELEGATED" and an entry like
"* Delegated this stuff to X", then you want this entry to be
in the list of possible targets. For that we need to distinguish
between DELEGATED as a keyword and "Delegated" as a simple word,
so (case-fold-search nil) seems relevant.
This default to t, so the default behavior of org-mode doesn't
change. But the user might want to keep at least one blank line
at the end of the remembered subtree, this option lets her do it.
This hook is called in org-clock-in earlier than the existing
org-clock-in-hook. This is useful for functions that need to
modify the Effort property, for example.
Carsten Dominik
James TD Smith
Bastien Guerry