分布式调度框架。
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.
 
 
 
 
 
 

10 KiB

Develop

pydolphinscheduler is python API for Apache DolphinScheduler, it just defines what workflow look like instead of store or execute it. We here use py4j to dynamically access Java Virtual Machine.

Setup Develop Environment

PyDolphinScheduler use GitHub to hold all source code, you should clone the code before you do same change.

git clone git@github.com:apache/dolphinscheduler.git

Now, we should install all dependence to make sure we could run test or check code style locally

cd dolphinscheduler/dolphinscheduler-python/pydolphinscheduler
python -m pip install -e '.[dev]'

Next, we have to open pydolphinscheduler project in you editor. We recommend you use pycharm instead of IntelliJ IDEA to open it. And you could just open directory dolphinscheduler-python/pydolphinscheduler instead of dolphinscheduler-python.

Brief Concept

Apache DolphinScheduler is design to define workflow by UI, and pydolphinscheduler try to define it by code. When define by code, user usually do not care user, tenant, or queue exists or not. All user care about is created a new workflow by the code his/her definition. So we have some side object in pydolphinscheduler/side directory, their only check object exists or not, and create them if not exists.

Process Definition

pydolphinscheduler workflow object name, process definition is also same name as Java object(maybe would be change to other word for more simple).

Tasks

pydolphinscheduler tasks object, we use tasks to define exact job we want DolphinScheduler do for us. For now, we only support shell task to execute shell task. This link list all tasks support in DolphinScheduler and would be implemented in the further.

Test Your Code

Linting and tests is very important for open source project, so we pay more attention to it. We have continuous integration service run by GitHub Action to test whether the patch is good or not, which you could jump to section With GitHub Action see more detail.

And to make more convenience to local tests, we also have the way to run your test automated with tox locally(run all tests except integrate test with need docker environment). It is helpful when your try to find out the detail when continuous integration in GitHub Action failed, or you have a great patch and want to test local first.

Besides automated testing with tox locally, we also have a manual way run tests. And it is scattered commands to reproduce each step of the integration test we told about.

With GitHub Action

GitHub Action test in various environment for pydolphinscheduler, including different python version in 3.6|3.7|3.8|3.9 and operating system linux|macOS|windows. It will trigger and run automatically when you submit pull requests to apache/dolphinscheduler.

Automated Testing With tox

tox is a package aims to automate and standardize testing in Python, both our continuous integration and local test use it to run actual task. To use it, you should install it first

python -m pip install --upgrade tox

After installation, you could run a single command to run all the tests, it is almost like test in GitHub Action but not so much different environment.

tox -e local-ci

It will take a while when you run it the first time, because it has to install dependencies and make some prepare, and the next time you run it will be faster.

If you failed section lint when you run command tox -e local-ci, you could try to run command tox -e auto-lint which we provider fix as many lints as possible. When I finish, you could run command tox -e local-ci to see whether the linter pass or not, you have to fix it by yourself if linter still fail.

Manually

Code Style

We use isort to automatically keep Python imports alphabetically, and use Black for code formatter and Flake8 for pep8 checker. If you use pycharmor IntelliJ IDEA, maybe you could follow Black-integration to configure them in your environment.

Our Python API CI would automatically run code style checker and unittest when you submit pull request in GitHub, you could also run static check locally.

We recommend pre-commit to do the checker mentioned above before you develop locally. You should install pre-commit by running

python -m pip install pre-commit 

in your development environment and then run pre-commit install to set up the git hooks scripts. After finish above steps, each time you run git commit or git push would run pre-commit check to make basic check before you create pull requests in GitHub.

# We recommend you run isort and Black before Flake8, because Black could auto fix some code style issue
# but Flake8 just hint when code style not match pep8

# Run Isort
python -m isort .

# Run Black
python -m black .

# Run Flake8
python -m flake8

Testing

Build Document

We use sphinx to build docs. Dolphinscheduler Python API CI would automatically build docs when you submit pull request in GitHub. You may locally ensure docs could be built successfully in case the failure blocks CI, you can build by tox or manual.

Build Document Automatically with tox

We integrated document build process into tox, you can build the latest document and all document(including history documents) via single command

# Build the latest document in dev branch
tox -e doc-build
# Build all documents, which including the latest and all history documents
tox -e doc-build-multi

Build Document Manually

To build docs locally, install sphinx and related python modules first via:

python -m pip install '.[doc]'

Then go to document directory and execute the build command

cd pydolphinscheduler/docs/
make clean && make html

NOTE: We support build multiple versions of documents with sphinx-multiversion, you can build with command git fetch --tags && make clean && make multiversion

Testing

pydolphinscheduler using pytest to test our codebase. GitHub Action will run our test when you create pull request or commit to dev branch, with python version 3.6|3.7|3.8|3.9 and operating system linux|macOS|windows.

pydolphinscheduler using pytest to run all tests in directory tests. You could run tests by the commands

python -m pytest --cov=pydolphinscheduler --cov-config=.coveragerc tests/

Besides run tests, it will also check the unit test coverage threshold, for now when test cover less than 90% will fail the coverage, as well as our GitHub Action.

The command above will check test coverage automatically, and you could also test the coverage by command.

python -m coverage run && python -m  coverage report

It would not only run unit test but also show each file coverage which cover rate less than 100%, and TOTAL line show you total coverage of you code. If your CI failed with coverage you could go and find some reason by this command output.

Integrate Test

Integrate Test can not run when you execute command tox -e local-ci because it needs external environment including Docker and specific image build by maven. Here we would show you the step to run integrate test in directory dolphinscheduler-python/pydolphinscheduler/tests/integration. There are two ways to run integrate tests.

Method 1: Launch Docker Container Locally

# Go to project root directory and build Docker image
cd ../../

# Build Docker image
./mvnw -B clean install \
    -Dmaven.test.skip \
    -Dmaven.javadoc.skip \
    -Dmaven.checkstyle.skip \
    -Pdocker,release -Ddocker.tag=ci \
    -pl dolphinscheduler-standalone-server -am

# Go to pydolphinscheduler root directory and run integrate tests
tox -e integrate-test

Method 2: Start Standalone Server in IntelliJ IDEA

# Start the standalone server in IDEA

# Go to pydolphinscheduler root directory and run integrate tests
tox -e local-integrate-test

Add LICENSE When New Dependencies Adding

When you add a new package in pydolphinscheduler, you should also add the package's LICENSE to directory dolphinscheduler-dist/release-docs/licenses/python-api-licenses, and also add a short description to dolphinscheduler-dist/release-docs/LICENSE.

Update UPDATING.md when public class, method or interface is be changed

When you change public class, method or interface, you should change the UPDATING.md to notice users who may use it in other way.

Reference