                              ChangeLog of Tofu
                  Plain (stupid) text-based todo(s) manager




Syntax:

    === <version> (<date>)

        + Normal change

        = Bug fix

=== 3.1 (20100816)

    + Added a new ride=<n> command. It allows to move the selection relatively
      to a task, inserting the selected tasks just before this one.

    + Updated tofu.1 and tofututo.7

=== 3.0 (20100520)

    + Tofu's been heavily rewritten and provides a new interface. The project
      directories have disappeared, all of them being merged in a single todo
      list (file) you can split as you want using tags. The priority level of
      each task is no more defined through built-in categories but merely
      infered from its (of course mutable) position in the todo. From now on,
      it is also possible to execute several selection/action(s) cycles
      on a single command line (glance at the "mono-cycle" option if you prefer
      a more traditional behaviour). Here are the UI changes to consider:

         Removed commands: crown, feed=, fresh, heat, last, move=, top, pr=,
                 print, printf=.
        
         Added commands: all, color=, digest, dump, help, jump=, list, next=,
                 queue=.

         Removed options: auto-repair, check-new-tasks, clean-on-move,
                 edit-on-next, next-on-feed, output-format, safe-feed,
                 source-auto-stamp.

         Added options: mono-cycle, recursive-macro-max, reset-to-main.

    + Added support for macro definitions that allow to create custom commands.

    + Added the new "tofuup" command, which allows to easily update the tofu
      todo layout when needed. Use it to convert your current tofu todo lists
      if you have some.

    + Updated tofu.1, added tofuup.1 and tofututo.7, which replaces the obsolete
      "PLAY" tutorial.

    + Updated and slightly improved the build system. The shebang is now
      automatically set into scripts and a new "--perl=</path/to/perl>"
      configure option's appeared to fix this if needed.

    + Removed the useless and obsolete "playground.pl" script

=== 2.6 (20100107)

    = Don't leave before having stripped the content in delete() when the user
      answers "N".

    + Use the editor defined in the EDITOR environment variable as default
      editor before set it to "vi".
      (thanks to Marcin Mirosław for the suggestion)

    = Fix the non-ASCII represented character in the help digest.
      (thanks to Marcin Mirosław)

=== 2.5 (20090213)

    + New action "printf=<format>", which allows to customize the output of the
      "print" action.

    + New option: "output-format", which redefines the default format used by
      the "print" action.

    + Print the ID of tasks in the separator when several are read.

    = Fixed mispelling of the "-tutorial" option in PLAY.
      (thanks to Corey Beres for the report)

    + Updated tofu.1 and PLAY; from now on the number of each task is documented
      as its ID, not as its name or as the task itself (more intuitive).

=== 2.4 (20081107)

    + New options: "clean-on-move", which deletes all of the markers on the
      moved tasks before transferring them, and "source-auto-stamp", which
      automatically stamps these ones with the name of the source stack (tofu.1
      updated).

    = Create correctly and quietly the "stack" files when they don't exist.

    = Don't increment the value of "next" in the destination stack when deleted
      tasks are moved.

    = Don't empty the selection when max=<n> is the only selector.

    = Don't call the whole stack in the selection when feed is given alone while
      the "next-on-feed" option is set to "true".

=== 2.3 (20080919)

    + New "move=<stack>" action, which allows to move tasks from their stack to
      the targeted one.

    + Removed the support to allow precisely markers in the 'markers' file. As
      a matter of fact, this made the things too much rigid (in particuliar to
      reset markers before moving tasks) for a very little gain. However, this
      support could come back, let me know if you have a real usage of this one
      (ie. other than preventing from typo errors). :)

    + Updated tofu.1 and PLAY.

=== 2.2 (20080823)

    = Made stat() to expect an array to get correctly the total of tasks from
      readstack().

    + Improved and shrunk (a little bit) argon().

    + Merged atomtrap() and banished() in atomtrap().

    + New options: "edit-on-next", which automatically calls "edit" when "next"
      is given, and "next-on-feed", which automatically calls "next" when "feed"
      is given.

=== 2.1 (20080707)

    + New install system, based on a Makefile with the classic "./configure;
      make; make install [DESTDIR=<root>]" sequency (README updated).

    + New "feed=<string>" action, which allows to quickly define the content
      of the next task as <string>.

    + New "top" selector, which refers to the highest task in the stack.
      (NOTE: the warning on line 244 which appears during the *first* change in
       an already existing stack is normal and inconsequential)

    + New "crown" action, which allows to arbitrary define the value of "top".

    + Updated tofu.1, PLAY and playground.pl.

=== 2.0 (20080519)

    + New storage format: all of the *.txt are now concatenated in a single
      "<stack>/content" file after edition. This should save much disk space.

    + New "auto-repair" option, which quietly and automatically reintegrates the
      trailing *.txt in the  "<stack>/content" file.

    + Updated playground.pl to generate the new format and added a "-root=<dir>"
      option to specify an alternative path to "~/.tofu" for stacks.

    + &argon() returns the complete path of the stack.

    + Renamed the configuration settings in a less idle fashion: "color" ->
      "use-color", "globedit" -> "global-edition", "checknew" ->
      "check-new-tasks".

    = Don't attempt to perform any action after &delete() let the selection
      empty.

    + Updated tofu.1.

=== 1.2 (20080502)

    = Removed the idle "-tar=<command>" option from install.sh.

    + Improved install.sh: added a "-root=<path>" and a "-verbose" options.

    + Mention the PLAY tutorial in tofu.1

    + New "clean" action, which wipes out the markers set on the selected tasks.

    + Added the possibility to specify strictly the allowed markers in
      "<stack>/markers" files (tofu.1 updated).

    + Shrunk (a little bit) marksman() and answer().

=== 1.1 (20080418)

    + Improved install.sh: install path is specified directly on the command-
      line, and a support for FHS conformance was added (README updated).

    + New "last" selector, which refer to the last task added to the stack.

    + Shrunk order(), help() and mangle().

    + New actions: "heat"/"fresh" which increase/decrease the priority of the
      selected tasks by one level.

    + Updated tofu.1 and PLAY.

=== 1.0 (20080329)

    + Check if the new tasks are empty before rewriting the stack. Can be
      inhibited with the "checknew" configuration option (tofu.1 updated).

    + Added a default "tutorial" mode to playground.pl, in order to ensure that
      all of the examples in the PLAY tutorial match with the content of the
      stack.

    + Updated PLAY, made minor fixes and clarifications in tofu.1.

=== 0.6 (20080229)

    + Invoke help() directly in argon().

    + New feature, stat() which allows to print a digest of present stacks.

    + playground.pl generates filled tasks.

    + New "max=<int>" selector, which limits the number of tasks printed.

    + New -<n> selectors.

    + Updated tofu.1

=== 0.5 (20080221)

    + Put in a new answer() function the code used to get user choices from
      stdin.

    + Added possibility to read task content before deleting it ("r" answer).
      Documented this in tofu.1.

    = edit(): don't give up when "Y" is given in the interactive mode.

    = Minor fix in PLAY.

=== 0.4 (20080212)

    + First public release.

    = prompt before edit when interactively iterate in edit().

    = match correctly the one-number-named tasks.

    + Updated the README file.

    + Added the playground.pl random stack generator script and the PLAY
      tutorial.

    = stamp(): clean the list of markers between each attribution.

=== 0.3 (20080208)

    = Don't consider a blank string as "true" in confetti().

    = edit() must return "1".

    + Warn (on stderr) when the selection is empty (better for using tofu in
      scripts or .bash_profile).

    + Added the "tofu.1" manual and rewrote the help digest.

=== 0.2 (20080205)

    = Don't edit the first selected task twice.

    = Don't automatically edit the next task.

    + New action: read.

=== 0.1 (20080204)

    + First functionnal and private release.

    + pr, wi, wil, wo, wol, next and individual selectors.

    + edit, print, delete, stamp, erase, [chw]set actions.
    
    + Configuration file ('~/.tofurc') support.


