On Mar 4, 2009, at 10:45 AM, Daniel Clemente wrote:
I'm starting to use column mode (C-c C-x C-c) and I'm surprised I
can't do some tasks from within, like for instance:
- move headings up/down with M-up M-down
- clock in/out tasks (C-c C-x C-i and … C-o)
- schedule tasks, etc. I always get: if: Text is read-only: "Type
`e' to edit property"
Technically, column view is a bitch to implement. It works by
putting overlays over characters in the headline. Column 1
is an overlay over the first character, column 2 is an overlay
over the second character etc. One could make the entire
line a single overlay, but this would no longer allow navigation
inside the table.
If the headline has less characters than your number of columns,
space characters will be appended to the line to make more
space. These characters have to be removed when you exit
column view.
Anything that would change the headline, even adding a newline
after it (to create space for a SCHEDULED time string) has the
potential to disrupt things. To protect against this, the entire
headline gets a read-only property, which then triggers an error
for any operation that somehow touches the headline.
With some thought, we probably could enable more commands,
but that has to be done one by one, explicitly.
If you and others compile a list of commands that would
be nice and important to have in column view, I can have
a look.
If SCHEDULED is one of the columns, you can go there and
press "e" to schedule the entry. The reason why this works
is that Org does call this as a special command, disables
the read-only, runs it, and then builds the column view
for this headline again from scratch.
To me, column view is like a *table* of *headings* (like a normal
org-table, I mean). I am used to the usual org's flexibility in
tables and in headings, and therefore it seems strange having to
quit and reenter column mode to do these changes.
Spoiled you are, I can tell ;-)
- Carsten
It would be an interesting experiment to have real Org tables which
get/set their rows from the headings in the subtree and their
columns from the properties of these headings. You would then have
all the usual features of tables: formulas, edit fields (properties
in this case) in a new buffer, …
Mmm… maybe that table/heading/columns mix would be too complex.
Anyway, do you also need to enter/quit column view many times? How
do you use it comfortably?
Daniel
_______________________________________________
Emacs-orgmode mailing list
Remember: use `Reply All' to send replies to the list.
Emacs-orgmode@gnu.org
http://lists.gnu.org/mailman/listinfo/emacs-orgmode
_______________________________________________
Emacs-orgmode mailing list
Remember: use `Reply All' to send replies to the list.
Emacs-orgmode@gnu.org
http://lists.gnu.org/mailman/listinfo/emacs-orgmode