You can not select more than 25 topics
Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
151 lines
6.5 KiB
151 lines
6.5 KiB
3 years ago
|
.. Licensed to the Apache Software Foundation (ASF) under one
|
||
|
or more contributor license agreements. See the NOTICE file
|
||
|
distributed with this work for additional information
|
||
|
regarding copyright ownership. The ASF licenses this file
|
||
|
to you 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.
|
||
|
|
||
|
Tutorial
|
||
|
========
|
||
|
|
||
|
This tutorial show you the basic concept of *PyDolphinScheduler* and tell all
|
||
|
things you should know before you submit or run your first workflow. If you
|
||
|
still not install *PyDolphinScheduler* and start Apache DolphinScheduler, you
|
||
|
could go and see :ref:`how to getting start PyDolphinScheduler <start:getting started>`
|
||
|
|
||
|
Overview of Tutorial
|
||
|
--------------------
|
||
|
|
||
|
Here have an overview of our tutorial, and it look a little complex but do not
|
||
|
worry about that because we explain this example below as detailed as possible.
|
||
|
|
||
|
.. literalinclude:: ../../examples/tutorial.py
|
||
|
:start-after: [start tutorial]
|
||
|
:end-before: [end tutorial]
|
||
|
|
||
|
Import Necessary Module
|
||
|
-----------------------
|
||
|
|
||
|
First of all, we should importing necessary module which we would use later just
|
||
|
like other Python package. We just create a minimum demo here, so we just import
|
||
|
:class:`pydolphinscheduler.core.process_definition` and
|
||
|
:class:`pydolphinscheduler.tasks.shell`.
|
||
|
|
||
|
.. literalinclude:: ../../examples/tutorial.py
|
||
|
:start-after: [start package_import]
|
||
|
:end-before: [end package_import]
|
||
|
|
||
|
If you want to use other task type you could click and
|
||
|
:doc:`see all tasks we support <tasks/index>`
|
||
|
|
||
|
Process Definition Declaration
|
||
|
------------------------------
|
||
|
|
||
|
We should instantiate object after we import them from `import necessary module`_.
|
||
|
Here we declare basic arguments for process definition(aka, workflow). We define
|
||
|
the name of process definition, using `Python context manager`_ and it
|
||
|
**the only required argument** for object process definition. Beside that we also
|
||
|
declare three arguments named `schedule`, `start_time` which setting workflow schedule
|
||
|
interval and schedule start_time, and argument `tenant` which changing workflow's
|
||
|
task running user in the worker, :ref:`section tenant <concept:tenant>` in *PyDolphinScheduler*
|
||
|
:doc:`concept` page have more detail information.
|
||
|
|
||
|
.. literalinclude:: ../../examples/tutorial.py
|
||
|
:start-after: [start workflow_declare]
|
||
|
:end-before: [end workflow_declare]
|
||
|
|
||
|
We could find more detail about process definition in
|
||
|
:ref:`concept about process definition <concept:process definition>` if you interested in it.
|
||
|
For all arguments of object process definition, you could find in the
|
||
|
:class:`pydolphinscheduler.core.process_definition` api documentation.
|
||
|
|
||
|
Task Declaration
|
||
|
----------------
|
||
|
|
||
|
Here we declare four tasks, and bot of them are simple task of
|
||
|
:class:`pydolphinscheduler.tasks.shell` which running `echo` command in terminal.
|
||
|
Beside the argument `command`, we also need setting argument `name` for each task *(not
|
||
|
only shell task, `name` is required for each type of task)*.
|
||
|
|
||
|
.. literalinclude:: ../../examples/tutorial.py
|
||
|
:dedent: 0
|
||
|
:start-after: [start task_declare]
|
||
|
:end-before: [end task_declare]
|
||
|
|
||
|
Beside shell task, *PyDolphinScheduler* support multiple tasks and you could
|
||
|
find in :doc:`tasks/index`.
|
||
|
|
||
|
Setting Task Dependence
|
||
|
-----------------------
|
||
|
|
||
|
After we declare both process definition and task, we have one workflow with
|
||
|
four tasks, both all tasks is independent so that they would run in parallel.
|
||
|
We should reorder the sort and the dependence of tasks. It useful when we need
|
||
|
run prepare task before we run actual task or we need tasks running is specific
|
||
|
rule. We both support attribute `set_downstream` and `set_upstream`, or bitwise
|
||
|
operators `>>` and `<<`.
|
||
|
|
||
|
In this example, we set task `task_parent` is the upstream task of task
|
||
|
`task_child_one` and `task_child_two`, and task `task_union` is the downstream
|
||
|
task of both these two task.
|
||
|
|
||
|
.. literalinclude:: ../../examples/tutorial.py
|
||
|
:dedent: 0
|
||
|
:start-after: [start task_relation_declare]
|
||
|
:end-before: [end task_relation_declare]
|
||
|
|
||
|
Please notice that we could grouping some tasks and set dependence if they have
|
||
|
same downstream or upstream. We declare task `task_child_one` and `task_child_two`
|
||
|
as a group here, named as `task_group` and set task `task_parent` as upstream of
|
||
|
both of them. You could see more detail in :ref:`concept:Tasks Dependence` section in concept
|
||
|
documentation.
|
||
|
|
||
|
Submit Or Run Workflow
|
||
|
----------------------
|
||
|
|
||
|
Now we finish our workflow definition, with task and task dependence, but all
|
||
|
these things are in local, we should let Apache DolphinScheduler daemon know what we
|
||
|
define our workflow. So the last thing we have to do here is submit our workflow to
|
||
|
Apache DolphinScheduler daemon.
|
||
|
|
||
|
We here in the example using `ProcessDefinition` attribute `run` to submit workflow
|
||
|
to the daemon, and set the schedule time we just declare in `process definition declaration`_.
|
||
|
|
||
|
Now, we could run the Python code like other Python script, for the basic usage run
|
||
|
:code:`python tutorial.py` to trigger and run it.
|
||
|
|
||
|
.. literalinclude:: ../../examples/tutorial.py
|
||
|
:dedent: 0
|
||
|
:start-after: [start submit_or_run]
|
||
|
:end-before: [end submit_or_run]
|
||
|
|
||
|
If you not start your Apache DolphinScheduler server, you could find the way in
|
||
|
:ref:`start:start Python gateway server` and it would have more detail about related server
|
||
|
start. Beside attribute `run`, we have attribute `submit` for object `ProcessDefinition`
|
||
|
and it just submit workflow to the daemon but not setting the schedule information. For
|
||
|
more detail you could see :ref:`concept:process definition`.
|
||
|
|
||
|
DAG Graph After Tutorial Run
|
||
|
----------------------------
|
||
|
|
||
|
After we run the tutorial code, you could login Apache DolphinScheduler web UI,
|
||
|
go and see the `DolphinScheduler project page`_. they is a new process definition be
|
||
|
created and named "Tutorial". It create by *PyDolphinScheduler* and the DAG graph as below
|
||
|
|
||
|
.. literalinclude:: ../../examples/tutorial.py
|
||
|
:language: text
|
||
|
:lines: 24-28
|
||
|
|
||
|
.. _`DolphinScheduler project page`: https://dolphinscheduler.apache.org/en-us/docs/latest/user_doc/guide/project.html
|
||
|
.. _`Python context manager`: https://docs.python.org/3/library/stdtypes.html#context-manager-types
|