pavement.py in depth

Paver is meant to be a hybrid declarative/imperative system for getting stuff done. You declare things via the options in pavement.py. And, in fact, many projects can get away with nothing but options in pavement.py. Consider, for example, an early version of Paver’s own pavement file:

from paver.easy import *
import paver.doctools

options(
    setup=dict(
    name='paver',
    version="0.3",
    description='Python build tool',
    author='Kevin Dangoor',
    author_email='dangoor+paver@gmail.com',
    #url='',
    packages=['paver'],
    package_data=setuputils.find_package_data("paver", package="paver",
                                            only_in_packages=False),
    install_requires=[],
    test_suite='nose.collector',
    zip_safe=False,
    entry_points="""
    [console_scripts]
    paver = paver.command:main
    """,
    ),

    sphinx=Bunch(
        builddir="build",
        sourcedir="source"
    )

)

@task
@needs('paver.doctools.html')
def html():
    """Build Paver's documentation and install it into paver/docs"""
    builtdocs = path("docs") / options.sphinx.builddir / "html"
    destdir = path("paver") / "docs"
    destdir.rmtree()
    builtdocs.move(destdir)

This file has both declarative and imperative aspects. The options define enough information for distutils, setuptools and Sphinx to do their respective jobs. This would function just fine without requiring you to define any tasks.

However, for Paver’s ‘paverdoc’ built-in task to work, we need Sphinx’s generated HTML to end up inside of Paver’s package tree. So, we override the html task to move the generated files.

Defining Tasks

Tasks are just simple functions. You designate a function as being a task by using the @task decorator.

You can also specify that a task depends on another task running first with the @needs decorator. A given task will only run once as a dependency for other tasks.

Manually Calling Tasks

Sometimes, you need to do something before running another task, so the @needs decorator doesn’t quite do the job.

Of course, tasks are just Python functions. So, you can just call the other task like a function!

How Task Names Work

Tasks have both a long name and a short name. The short name is just the name of the function. The long name is the fully qualified Python name for the function object.

For example, the Sphinx support includes a task function called “html”. That task’s long name is “paver.doctools.html”.

If you `import paver.doctools` in your pavement.py, you’ll be able to use either name to refer to that task.

Tasks that you define in your pavement are always available by their short names. Tasks defined elsewhere are available by their short names unless there is a conflict where two tasks are trying to use the same short name.

Tasks are always available unambiguously via their long names.

Task Parameters

Tasks don’t have to take any parameters. However, Paver allows you to work in a fairly clean, globals-free manner(*). Generally speaking, the easiest way to work with paver is to just do from paver.easy import *, but if you don’t like having so much in your namespace, you can have any attribute from tasks.environment sent into your function. For example:

@task
def my_task(options, info):
    # this task will get the options and the "info" logging function
    # sent in
    pass

(*): well, there is one global: tasks.environment.

Command Line Arguments

Tasks can specify that they accept command line arguments, via two other decorators. The @consume_args decorator tells Paver that all command line arguments following this task’s name should be passed to the task. You can either look up the arguments in options.args, or just specify args as a parameter to your function. For example, Paver’s “help” task is declared like this:

@task
@consume_args
def help(args, help_function):
    pass

The “args” parameter is just an attribute on tasks.environment (as is help_function), so it is passed in using the same rules described in the previous section.

More generally, you’re not trying to consume all of the remainder of the command line but to just accept certain specific arguments. That’s what the cmdopts decorator is for:

@task
@cmdopts([
    ('username=', 'u', 'Username to use when logging in to the servers')
])
def deploy(options):
    pass

@cmdopts takes a list of tuples, each with long option name, short option name and help text. If there’s an “=” after the long option name, that means that the option takes a parameter. Otherwise, the option is assumed to be boolean. The command line options set in this manner are all added to the options under a namespace matching the name of the task. In the case above, options.deploy.username would be set if the user ran paver deploy -u my-user-name. Note that an equivalent command line would be paver deploy.username=my-user-name deploy

For fine-tuning, you may add optparse.Option instances:

@tasks.task
@tasks.cmdopts([
    make_option("-f", "--foo", help="foo")
])
def foo_instead_of_spam_and_eggs(options):
    pass

You may share @cmdopts between tasks. To do that and to avoid confusion, You have to add share_with argument:

@task
@cmdopts([
    ('username=', 'u', 'Username to use when logging in to the servers')
])
def deploy_to_linux(options):
    pass


@task
@needs(['deploy_to_linux'])
@cmdopts([
    ('username=', 'u', 'Username to use when logging in to the servers')
], share_with=['deploy_to_linux'])
def deploy(options):
    pass

For sharing, following must be fullfilled:

  • Both long and short option names must be same
  • share_with argument must be specified on top-level task

Otherwise, PavementError is raised.

Hiding tasks

Some tasks may only exist as a shared dependency between other tasks and may not make sense to be called directly.

There is no way to provide that, however you can hide them from paver help to reduce noise. Just decorate function with @no_help decorator:

@task
@no_help
def hidden_dependency():
    pass

Of course, this should not be used usually. If task is not to be called at all, why not just make them a g’old function?

More complex dependencies

@needs might not cover all your needs. For example, depending on argument or environment, you might decide to call an appropriate task in the middle of another one.

There are two key options for fixing that:

# ``@might_call`` decorator, which indicates that task might invoke ``call_task`` on one or more of the specified tasks. This allows you to provide command line options to task that might be called (it is inserted in dependency chain)::

@task @cmdopts([

(‘username=’, ‘u’, ‘Whom to greet’)

], share_with=[‘deploy_to_linux’]) def say_hello(options):

if not hasattr(options, “username”):
print ‘SPAM’
else:
print ‘Hello, my dear user %s’ % options.username

@task @might_call(‘say_hello’) def greet_user(options):

setup_environment()

call_task(‘say_hello’)

do_cleanup()

# Providing options and arguments to another tasks directly. Options are provided with final assignet value:

@task
@cmdopts([
    ('long-username=', 'u', 'Whom to greet')
], share_with=['deploy_to_linux'])
def say_hello(options):
    if not hasattr(options, "username"):
        print 'SPAM'
    else:
        print 'Hello, my dear user %s' % options.long_username


@task
def greet_user(options):
    call_task('say_hello', options={
        'long_username' : 'Kitty'
    })

Console arguments (args) should be passed as in console:

@task
@consume_args
def say_hello(args):
    print 'Hello to ALL the users: %s' % ', '.join(args)


@task
def greet_user(options):
    call_task('say_hello', args = [
        'Arthur Pewtey',
        'The Reverend Arthur Belling',
    ])