I finally released TKDO 0.7, if anyone is interested. The main
change is a completely new text UI based on curses. It's faster
than the GUI, uses less memory, allows colorized tasks, and works
well over ssh. Also, this version has a variety of small new
features and bugfixes.
+ Added a curses-based text interface.
* Updated config library to allow saving only non-default values.
+ Added ability to snooze for units such as 'ddd' (3 days) or 'ww' (2 weeks).
+ Added ability to filter in 'and' or 'or' modes
(to display a union of contexts instead of an intersection).
* Made snoozed / completed tasks display as 'Zzz' / '[X]' instead of '0'.
* Handle missing task files by ignoring them instead of exiting.
* Added task title to log message for launching the editor.
+ Added an automation facility based on the TUI.
Basically, any extra parameters will be executed as if the user had
typed them into the TUI.
* Moved most of the GUI logic into a shared base class,
so the TUI and GUI can share behavior without writing everything twice.
+ Added a 'zoom' feature to auto-filter into the current branch.
(bound to the 'enter' key by default)
+ All new curses interface! Works well over ssh.
* Changed 'editor' config item to 'editor_gui'.
The GUI and TUI can use different editors.
On Thursday 12 November 2009 23:53:50 Scott Scriven wrote:
> I finally released TKDO 0.7, if anyone is interested. The main
> change is a completely new text UI based on curses.
You need better documentation. I made a todo list, and then I ran ./todo.py.
Then I tried ./todo.py todo.otl and it told me I had a bad command. There was
no ./tkdo.py at all. Here is the list of files created by unpacking your
I tried the make command, but it said target "all" had nothing to do. I finally
did make install and now I have a tkdo command, but I have no idea what to do
with it. I performed make todo and it listed what looked like a multilevel
todolist to the screen, but I had no idea what to do with it.
U need a tutorial. You can use me as the giny pig. What's my next step? How do
I do a proof of concept?
Thanks for looking at it, and for reminding me! :)
There are two files with documentation, README and example.otl.
But neither one really gives a tutorial or quickstart, and both
are probably getting a little old. So, there are a few things to
> I have no idea what to do with it. ...
> U need a tutorial. You can use me as the giny pig. What's my
> next step? How do I do a proof of concept?
Let's see... the quickstart should probably cover the following:
I've found that there are some things I do frequently which
should probably be turned into actual features... for example, I
keep an item at the top of each file whose title is '...'. Its
only purpose is to let me easily add tasks to the top of the
file, by moving the cursor to it and pressing 'o' (like in vim).
It might be better to just have an actual feature which would
either add a task to the top/bottom of the current file. And
another to make it list each task file.
> I performed make todo and ...
Oh, yeah. That's just a reminder for developers. I put comments
in the code with tags of "TODO" or "FIXME" when there's something
I need to remember for later, and then 'make todo' will grep for
those. I should probably export that to a VO file, but haven't
In any case, there is a lot left to do, and I'm finally getting
some of it done again. :)