Files
deb-python-taskflow/taskflow/task.py
Joshua Harlow 23dfff4105 Engine, task, linear_flow unification
In order to move away from the existing flows having their
own implementation of running, start moving the existing
flows to be  patterns that only structure tasks (and impose
constraints about how the group of tasks can run) in useful
ways.

Let the concept of running those patterns be handled by an
engine instead of being handled by the flow itself. This
will allow for varying engines to be able to run flows in
whichever way the engine chooses (as long as the constraints
set up by the flow are observed).

Currently threaded flow and graph flow are broken by this
commit, since they have not been converted to being a
structure of tasks + constraints. The existing engine has
not yet been modified to run those structures either, work
is underway  to remediate this.

Part of: blueprint patterns-and-engines

Followup bugs that must be addressed:
  Bug: 1221448
  Bug: 1221505

Change-Id: I3a8b96179f336d1defe269728ebae0caa3d832d7
2013-09-05 19:26:36 -07:00

175 lines
6.1 KiB
Python

# -*- coding: utf-8 -*-
# vim: tabstop=4 shiftwidth=4 softtabstop=4
# Copyright (C) 2013 Rackspace Hosting Inc. All Rights Reserved.
# Copyright (C) 2013 Yahoo! Inc. All Rights Reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License"); you may
# not use this file except in compliance with the License. You may obtain
# a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS, WITHOUT
# WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the
# License for the specific language governing permissions and limitations
# under the License.
import abc
from taskflow.utils import misc
from taskflow.utils import reflection
def _save_as_to_mapping(save_as):
"""Convert save_as to mapping name => index
Result should follow storage convention for mappings.
"""
if save_as is None:
return {}
if isinstance(save_as, basestring):
return {save_as: None}
elif isinstance(save_as, (tuple, list)):
return dict((key, num) for num, key in enumerate(save_as))
raise TypeError('Task provides parameter '
'should be str or tuple/list, not %r' % save_as)
def _build_rebind_dict(args, rebind_args):
if rebind_args is None:
return {}
elif isinstance(rebind_args, (list, tuple)):
rebind = dict(zip(args, rebind_args))
if len(args) < len(rebind_args):
rebind.update((a, a) for a in rebind_args[len(args):])
return rebind
elif isinstance(rebind_args, dict):
return rebind_args
else:
raise TypeError('Invalid rebind value: %s' % rebind_args)
def _check_args_mapping(task_name, rebind, args, accepts_kwargs):
args = set(args)
rebind = set(rebind.keys())
extra_args = rebind - args
missing_args = args - rebind
if not accepts_kwargs and extra_args:
raise ValueError('Extra arguments given to task %s: %s'
% (task_name, sorted(extra_args)))
if missing_args:
raise ValueError('Missing arguments for task %s: %s'
% (task_name, sorted(missing_args)))
def _build_arg_mapping(task_name, reqs, rebind_args, function, do_infer):
task_args = reflection.get_required_callable_args(function)
accepts_kwargs = reflection.accepts_kwargs(function)
result = {}
if reqs:
result.update((a, a) for a in reqs)
if do_infer:
result.update((a, a) for a in task_args)
result.update(_build_rebind_dict(task_args, rebind_args))
_check_args_mapping(task_name, result, task_args, accepts_kwargs)
return result
class BaseTask(object):
"""An abstraction that defines a potential piece of work that can be
applied and can be reverted to undo the work as a single unit.
"""
__metaclass__ = abc.ABCMeta
def __init__(self, name, provides=None):
self._name = name
# An *immutable* input 'resource' name mapping this task depends
# on existing before this task can be applied.
#
# Format is input_name:arg_name
self.requires = {}
# An *immutable* output 'resource' name dict this task
# produces that other tasks may depend on this task providing.
#
# Format is output index:arg_name
self.provides = _save_as_to_mapping(provides)
# This identifies the version of the task to be ran which
# can be useful in resuming older versions of tasks. Standard
# major, minor version semantics apply.
self.version = (1, 0)
@property
def name(self):
return self._name
def __str__(self):
return "%s==%s" % (self.name, misc.get_task_version(self))
@abc.abstractmethod
def execute(self, *args, **kwargs):
"""Activate a given task which will perform some operation and return.
This method can be used to apply some given context and given set
of args and kwargs to accomplish some goal. Note that the result
that is returned needs to be serializable so that it can be passed
back into this task if reverting is triggered.
"""
def revert(self, *args, **kwargs):
"""Revert this task using the given context, result that the apply
provided as well as any information which may have caused
said reversion.
"""
class Task(BaseTask):
"""Base class for user-defined tasks
Adds following features to Task:
- auto-generates name from type of self
- adds all execute argument names to task requirements
"""
def __init__(self, name=None, provides=None, requires=None,
auto_extract=True, rebind=None):
"""Initialize task instance"""
if name is None:
name = reflection.get_callable_name(self)
super(Task, self).__init__(name,
provides=provides)
self.requires = _build_arg_mapping(self.name, requires, rebind,
self.execute, auto_extract)
class FunctorTask(BaseTask):
"""Adaptor to make task from a callable
Take any callable and make a task from it.
"""
def __init__(self, execute, name=None, provides=None,
requires=None, auto_extract=True, rebind=None, revert=None,
version=None):
"""Initialize FunctorTask instance with given callable and kwargs"""
if name is None:
name = reflection.get_callable_name(execute)
super(FunctorTask, self).__init__(name, provides=provides)
self._execute = execute
self._revert = revert
if version is not None:
self.version = version
self.requires = _build_arg_mapping(self.name, requires, rebind,
execute, auto_extract)
def execute(self, *args, **kwargs):
return self._execute(*args, **kwargs)
def revert(self, *args, **kwargs):
if self._revert:
return self._revert(*args, **kwargs)
else:
return None