tasker 0.3

Tasker is a way to organize your scripts. If you have dozens of small scripts and you keep forgetting what they do and what parameters they take, tasker is for you. Instead of scripts you write functions, put them in a file or a couple of them, and then run the main tasker script giving them the task name and the parameters of the task like this:

run_task <task-name> [<task-parameter> ...]

The built-in help task will scan your tasks package and give you the call signatures and docstrings of your tasks (see example below).

To be considered a task, the function name should end with _task. The task name is the function name less the suffix. Suppose we have the following file stucture:

run_task
tasks/
    __init__.py
    module1.py

__init__.py

def taskone_task(p1):
    print('in taskone_task, parameters:', p1)

def tasktwo_task(p1, p2):
    print('in tasktwo_task, parameters:', p1, p2)

and module1.py

def t1_task(p1):
    """This is the task's documentation string.

    Continuation of the docstring
    """
    print('in t1_task, parameters:', p1)

def t2_task(p1, p2=None):
    print('in t2_task, parameters:', p1, p2)

To generate the run_task script in the current directory run:

python -m tasker <fully-qualified-name-of-tasks-package>  # default = "tasks"

A run of the script with the built-in help task produces the following:

./run_task help
Usage: run_task <task-name> [parameter, ...]

All available tasks:
    help      # print help screen

    In module tasks
        taskone :  (p1)
        tasktwo :  (p1, p2)

    In module tasks.module1
        module1.t1 :  (p1)
            This is the task's documentation string.
        module1.t2 :  (p1, p2=None)

We see the call signatures of the tasks. If a task function contains a docstring the first line of the docstring will also be included. To print the complete docstring of a task run:

./run_task help <task-name>

IMPORTANT: All task parameters are strings.

With the above setup we can run the following examples in the current directory (which just print the parameters):

./run_task taskone foo          # prints foo
./run_task tasktwo foo baz      # prints foo, baz
./run_task tasktwo foo          # fails, not enough parameters
./run_task module1.t1 spam      # prints spam
./run_task module1.t2 spam ham  # prints spam, ham
./run_task module1.t2 spam      # works too because of default value of the second parameter

You will want to put something meaningful in your tasks.

You can include tasks with distribution of your project and run them all with a single installed script. Suppose your project looks like this:

myproject/
    __init__.py
    somestuff.py
    tasks/

Then you can include the following snippet in __init__.py:

def run_task():
    import tasker
    tasker.main('tasks')

then include the following in your setuptools-based setup.py:

entry_points={
    'console_scripts': [
        'myproject_run_task = myproject:run_task',
        ],
    }

This setup will create script myproject_run_task, which will run your tasks.

pip install tasker