This page describes the mechanics of how to contribute software to ZooKeeper. For ideas about what you might contribute, please see the ProjectSuggestions page.
First of all, you need the ZooKeeper source code. Go to https://github.com/apache/zookeeper (which is a read only mirror of zookeeper git hosted on Apache) and click the fork button to create your own fork. After forking download your fork to local disk. Most development is done on the "master":
$ git clone https://github.com/<userid>/zookeeper.git my-zookeeper |
You may also want to develop against a specific release. You can list the release branches and checkout 3.4 by going into the zookeeper directory and running:
git branch -la git checkout remotes/origin/branch-3.4 |
Before you start, send a message to the ZooKeeper developer mailing list, or file a bug report in Jira. Describe your proposed changes and check that they fit in with what others are doing and have planned for the project. Be patient, it may take folks a while to understand your requirements.
You will want to create a local branch to work under. It's convenient to name it the JIRA number:
git checkout -b ZOOKEEPER-1234 # this is the same as git branch ZOOKEEPER-1234 && git checkout ZOOKEEPER-1234 |
Modify the source code and add some (very) nice features using your favorite IDE.
But take care about the following points
junit.framework.TestCase
and whose class name ends with Test
.test
, and call JUnit's many assert methods to verify conditions; these methods will be executed when you run ant test
./tmp
. Instead, the tests should write to the location specified by the test.build.data
system property.src/java/test
tree.ClientTest.java
is an example of a client-server test.ant test
, or you can run a specific unit test with the command ant -Dtestcase=<class name without package prefix> test
(for example ant -Dtestcase=ClientTest test
)We accept pull requests via github against the apache/zookeeper repository. Submitting to Apache ZooKeeper via github is no different than submitting any other pull request against a github repository. The developer needs to follow the contribution guidelines to produce changes, just like with patch files, and once it is ready, the developer produces a pull request from his/her own branch. The one constraints we have is that the title of the pull request must match the paring JIRA issue. If the JIRA issue is "ZOOKEEPER-222222: My lovely bug", then the pull request issue must also match "ZOOKEEPER-222222: My lovely bug".
We currently perform QA on pull requests, although the process is still being polished. We also have not completed the JIRA workflow when the developer submits a pull request, which means that the JIRA does not transition to Patch Available when a pull request is submitted. We are currently discussing on the list how to adapt our workflow.
there are many ways to setup your git environment but this one example:
go to https://github.com/apache/zookeeper and click the fork button to create your own fork.
on your local machine run
$ git clone https://github.com/<userid>/zookeeper.git my-zookeeper |
$ git remote add apache-github https://github.com/apache/zookeeper.git |
$ git pull apache-github |
$ git checkout remotes/apache-github/master $ git checkout -B ZOOKEEPER-<jiraNumber> |
<jiraNumber> is the JIRA you will be working on.
sorry can't help you here. it's all up to you to fix the bug
do you will want to do some combination of git add
and git commit
to get all your changes committed. when you run git commit, you only need a single line in the commit message it should have the form ZOOKEEPER-jiraNumber: jiraTitle
.
$ git push origin ZOOKEEPER-<jiraNumber> |
go to the github page that lists your branches. you will see a pull request button that you can press. Please make sure your pull request description captures enough information so reviewers can get an idea of what the pull request is targeting without digging into the JIRA or looking into the code. When the pull request is merged, the description will be part of the commit message. Here is an example of a descriptive pull request looks like: https://github.com/apache/zookeeper/pull/211
This is a check list of things we'd like a contributor to do before opening a pull request.
Before submitting your pull request, you are encouraged to run the same tools that the automated Jenkins CI system will run on your patch. This enables you to fix problems with your pull request before you submit it. The test
Ant target will run your patch through the same checks that Jenkins currently does.
To use this target, you must run it from a clean workspace (ie git status
shows no modifications or additions). From your clean workspace, run:
mvn verify spotbugs:check checkstyle:check -Pfull-build -Dsurefire-forkcount=4 |
At the end, you should get a message on your console that indicates success.
Please make sure that all unit tests succeed before submitting pull request and that no new javac compiler warnings are introduced by your pull request
After a while, if you see
BUILD SUCCESSFUL |
all is ok, but if you see
BUILD FAILED |
then please examine error messages in build/test
and fix things before proceeding.
Please also check the javadoc.
> mvn install -DskipTests -pl zookeeper-docs > firefox build/docs/api/index.html |
Examine all public classes you've changed to see that documentation is complete, informative, and properly formatted. Your patch must not generate any javadoc warnings.
The source documentation is available in src/docs directory. It can be regenerated using forrest (as documented here) however the contributor is not responsible for submitting this as part of the pull request. To be clear: the pull request should include changes to src/docs but not the toplevel (of the source tree) docs directory.
Note: This is in case of manual copies, ToDo for me: I will check the commit script to integrate this functionality
There are occasions when there are multiple author for a patch. For example when there is a patch that is abandoned by its original author for any reason, and it can no longer be merged, or it's unfinished, someone might pick it up and finish it. (After asking the original author if he is still working on it or not).
In these occasions, we should also attribute the original author by adding one or more Co-authored-by
trailers to the commit’s message. See the GitHub documentation for "Creating a commit with multiple authors".
As this is a GitHub feature, Co-author will show up on GitHub's statistics.
In short, these are the steps to add Co-authors that will be tracked by GitHub:
Collect the name and email address for each co-author.
Commit the change, but after your commit description, instead of a closing quotation, add two empty lines. (Do not close the commit message with a quotation mark)
On the next line of the commit message, type Co-authored-by: name <name@example.com>
. After the co-author information, add a closing quotation mark.
Here is the example from the GitHub page, using 2 Co-authors:
$ git commit -m "Refactor usability tests. > > Co-authored-by: name <name@example.com> Co-authored-by: another-name <another-name@example.com>" |
Please note that the attachment should be granted license to ASF for inclusion in ASF works (as per the Apache License §5).
Folks should run a full check (mvn verify spotbugs:check checkstyle:check -Pfull-build -Dsurefire-forkcount=4) before submitting pull request. Tests should all pass. Javadoc should report no warnings or errors. Jenkin's tests are meant to double-check things, and not be used as a primary patch tester, which would create too much noise on the mailing list and in Jira. Submitting patches that fail Jenkins testing is frowned on, (unless the failure is not actually due to the patch).
If your PR involves performance optimizations, they should be validated by benchmarks that demonstrate an improvement.
If your PR creates an incompatibility with the latest major release, then you must set the Incompatible change flag on the issue's Jira 'and' fill in the Release Note field with an explanation of the impact of the incompatibility and the necessary steps users must take.
If your PR implements a major feature or improvement, then you must fill in the Release Note field on the issue's Jira with an explanation of the feature that will be comprehensible by the end user.
Once a PR is created, a Jenkins job will be triggered to run a set of pre-commit checks on the pull request, including all Java and C unit tests, find bug test, release audit check, etc. If all checks pass, you should see something like this in your pull request page:
After code review and committers sign off, the pull request will be merged.
If any of the checks fail, you will see this:
In this case, please check the Jenkins output by following the "Details" link, and fix the issue, and try trigger a Jenkins build again. There are these ways to trigger a Jenkins build:
For contributors who owns the pull request: please make sure to get a green build to prepare the patch in a landing state.
For committers: please make sure to get a green build before committing a pull request.
Once a "+1" comment is received from the automated patch testing system and a code reviewer has set the Reviewed flag on the issue's Jira, a committer should then evaluate it within a few days and either: commit it; or reject it with an explanation.
Please be patient. Committers are busy people too. If no one responds to your patch after a few days, please make friendly reminders. Please incorporate other's suggestions into your patch if you think they're reasonable. Finally, remember that even a patch that is not committed is useful to the community.
Should your patch receive a "-1" from the Jenkins testing, select the Resume Progress on the issue's Jira, update pull request with necessary fixes.
Committers: for non-trivial changes, it is best to get another committer to review your pull request before commit. Please also try to frequently review things in the patch queue.
Please comment on issues in Jira, making their concerns known. Please also vote for issues that are a high priority for you.
Please refrain from editing descriptions and comments if possible, as edits spam the mailing list and clutter Jira's "All" display, which is otherwise very useful. Instead, preview descriptions and comments using the preview button (on the right) before posting them. Keep descriptions brief and save more elaborate proposals for comments, since descriptions are included in Jira's automatically sent messages. If you change your mind, note this in a new comment, rather than editing an older comment. The issue should preserve this history of the discussion.
Contributors should join the ZooKeeper mailing lists. In particular, the commit list (to see changes as they are made), the dev list (to join discussions of changes) and the user list (to help others).