Reviewing Zulip code
Code review is a key part of how Zulip does development. It’s an essential aspect of our process to build a high-quality product with a maintainable code base.
Principles of code review
Zulip has an active contributor community, and just a small handful of maintainers who can do the final rounds of code review. As such, we would love for contributors to help each other with making pull requests that are not only correct, but easy to review. Doing so ensures that PRs can be finalized and merged more quickly, and accelerates the pace of progress for the entire project.
If you’re new to open source, this may be the first time you do a code review of anyone’s changes! We have therefore written this step-by-step guide to be accessible to all Zulip contributors.
Reviewing your own code
One of the best ways to improve the quality of your own work as a software engineer is to do a code review of your own work before submitting it to others for review. We thus strongly encourage you to get into the habit of reviewing you own code. You can often find things you missed by taking a step back to look over your work before asking others to do so, and this guide will walk you through the process. Catching mistakes yourself will help your PRs be merged faster, and folks will appreciate the quality and professionalism of your work.
Reviewing other contributors’ code
Doing code reviews is a valuable contribution to the Zulip project. It’s also an important skill to develop for participating in open-source projects and working in the industry in general. If you’re contributing to Zulip and have been working in our code for a little while, we would love for you to start doing code reviews!
Anyone can do a code review – you don’t have to have a ton of experience, and you don’t have to have the power to ultimately merge the PR. The sections below offer accessible, step-by-step guidance for how to go about reviewing Zulip PRs.
For students participating in Google Summer of Code or a similar program, we expect you to spend a chunk of your time each week (after the first couple of weeks as you’re getting going) doing code reviews.
How to review code
Whether you are reviewing your own code or somebody else’s, this section describes how to go about it.
If you are reviewing somebody else’s code, you will likely need to first fetch
it so that you can play around with the new functionality. If you’re working in
the Zulip server codebase, use our Git tool
tools/fetch-rebase-pull-request to check out a pull request locally and rebase
Code review checklist
The following review steps apply to the majority of PRs.
Think about the issue:
Start by rereading the issue the PR is intended to solve. Are you confident that you understand everything the issue is asking for? If not, try exploring the relevant parts of the Zulip app and reading any linked discussions on the development community server to see if the additional context helps. If any part is still confusing, post a GitHub comment or a message on the development community server explaining precisely what points you find confusing.
Now that you’re confident that you understand the issue, does the PR address all the points described in the issue? If not, is it easy to tell without reading the code which points are not addressed and why? Here is a handful of good ways for the author to communicate why the issue as written might not be fully solved by the PR:
The issue explicitly notes that it’s fine for some parts to be completed separately, and the PR clearly indicates which parts are solved.
After discussion of initial prototypes (in GitHub comments or on the development community server), it was decided to change some part of the specification, and the PR notes this.
The author explains why the PR is a better way to solve the issue than what was described.
The solution changed because of changes in the project or application since the issue was written, and the author explains the adjustments that were made.
Think about the code:
Make sure the PR uses clear function, argument, variable, and test names. Every new piece of Zulip code will be read many times by other developers, and future developers will
grepfor relevant terms when researching a problem, so it’s important that variable names communicate clearly the purpose of each piece of the codebase.
Make sure the PR avoids duplicated code. Code duplication is a huge source of bugs in large projects and makes the codebase difficult to understand, so we avoid significant code duplication wherever possible. Sometimes avoiding code duplication involves some refactoring of existing code; if so, that should usually be done as its own series of commits (not squashed into other changes or left as a thing to do later). That series of commits can be in the same pull request as the feature that they support, and we recommend ordering the history of commits so that the refactoring comes before the feature. That way, it’s easy to merge the refactoring (and minimize risk of merge conflicts) if there are still user experience issues under discussion for the feature itself.
Good comments help. It’s often worth thinking about whether explanation in a commit message or pull request discussion should be included in a comment,
/docs, or other documentation. But it’s better yet if verbose explanation isn’t needed. We prefer writing code that is readable without explanation over a heavily commented codebase using lots of clever tricks.
If you can, step back and think about the technical design. There are a lot of considerations here: security, migration paths/backwards compatibility, cost of new dependencies, interactions with features, speed of performance, API changes. Security is especially important and worth thinking about carefully with any changes to security-sensitive code like views.
Think about testing:
The CI build tests need to pass. One can investigate any failures and figure out what to fix by clicking on a red X next to the commit hash or the Detail links on a pull request. (Example: in #17584, click the red X before
49b10a3to see the build jobs for that commit. You can see that there are 7 build jobs in total. All the 7 jobs run in GitHub Actions. You can see what caused the job to fail by clicking on the failed job. This will open up a page in the CI that has more details on why the job failed. For example this is the page of the “Debian 10 Buster (Python 3.7, backend + frontend)” job. See our docs on continuous integration to learn more.
Make sure the code is well-tested; see below for details. The PR should summarize any manual testing that was done to validate that the feature is working as expected.
Think about the commits:
Does the PR follow the principle that “Each commit is a minimal coherent idea”? See the commit discipline guide to learn more about commit structure in Zulip.
Does each commit have a clear commit message? Check for content, format, spelling and grammar. See the Zulip version control documentation for details on what we look for.
You should also go through any of the following checks that are applicable:
Error handling. The code should always check for invalid user input. User-facing error messages should be clear and when possible be actionable (it should be obvious to the user what they need to do in order to correct the problem).
Translation. Make sure that the strings are marked for translation.
Completeness of refactoring. When reviewing a refactor, verify that the changes are complete. Usually, one can check that efficiently using
git grep, and it’s worth it, as we very frequently find issues by doing so.
Documentation updates. If this changes how something works, does it update the documentation in a corresponding way? If it’s a new feature, is it documented, and documented in the right place?
mypy annotations in Python code. New functions should be annotated using mypy and existing annotations should be updated. Use of
ignore, and unparameterized containers should be limited to cases where a more precise type cannot be specified.
The tests should validate that the feature works correctly, and specifically test for common error conditions, bad user input, and potential bugs that are likely for the type of change being made. Tests that exclude whole classes of potential bugs are preferred when possible (e.g., the common test suite
test_markdown.pybetween the Zulip server’s frontend and backend Markdown processors, or the
GetEventsTesttest for buggy race condition handling). See the test writing documentation to learn more.
We are trying to maintain ~100% test coverage on the backend, so backend changes should have negative tests for the various error conditions. See backend testing documentation for details.
If the feature involves frontend changes, there should be frontend tests. See frontend testing documentation for details.
If the PR makes any frontend changes, you should make sure to play with the part of the app being changed to validate that things look and work as expected. While not all of the situations below will apply, here are some ideas for things that should be tested if they are applicable. Use the development environment to test any webapp changes.
This might seem like a long process, but you can go through it quite quickly once you get the hang of it. Trust us, it will save time and review round-trips down the line!
Open up the parts of the UI that were changed, and make sure they look as you were expecting.
Is the new UI consistent with similar UI elements? Think about fonts, colors, sizes, etc. If a new or modified element has multiple states (e.g. “on” and “off”), consider all of them.
Is the new UI aligned correctly with the elements around it, both vertically and horizontally?
If the PR adds or modifies a clickable element, does it have a hover behavior that’s consistent with similar UI elements?
If the PR adds or modifies an element (e.g. a button or checkbox) that is sometimes disabled, is the disabled version of the UI consistent with similar UI elements?
Did the PR accidentally affect any other parts of the UI? E.g., if the PR modifies some CSS, look for other elements that may have been altered unintentionally. Use
git grepto see if the code you modified is being used elsewhere.
Now check all of the above in the other theme (light/dark).
Responsiveness and internationalization:
Check the new UI at different window sizes, including mobile sizes (you can use Chrome DevTools if you like). Does it look good in both wide and narrow windows?
To simulate what will happen when the UI is translated to different languages, try changing any new strings, or ones that are now displayed differently, to make them 1.5x longer, and check if anything breaks. What would happen if the strings were half as long as in English?
Strings (text): If the PR adds or modifies strings, check the following:
Does the wording seem consistent with similar features (similar style, level of detail, etc.)?
If there is a number, are the
N = 1and
N > 1cases both handled properly?
Do elements that require tooltips have them? Check similar elements to see whether a tooltip is needed, and what information it should contain.
Functionality: We’re finally getting to the part where you actually use the new/updated feature. :) Test to see if it works as expected, trying a variety of scenarios. If it works as described in the issue but seems awkward in some way, note this on the PR.
If relevant, be sure to check that:
Live updates are working as expected.
Keyboard navigation, including tabbing to the interactive elements, is working as expected.
Some scenarios to consider:
Try clicking on any interactive elements, multiple times, in a variety of orders.
If the feature affects the message view, try it out in different types of narrows: topic, stream, All messages, PMs.
If the feature affects the compose box in the webapp, try both ways of resizing the compose box. Test both stream messages and PMs.
If the feature might require elevated permissions, check it out as a user who has permissions to use it and one who does not.
Think about how the feature might interact with other features, and try out such scenarios. For example:
If the PR adds a banner, is it possible that it would be shown at the same time as other banners? Does something reasonable happen?
If the feature has to do with topic editing, do you need to think about what happens when a topic is resolved/unresolved?
If it’s a message view feature, would anything go wrong if the message was collapsed or muted? If it was colored like an
@-mention or a PM?
Review process and communication
Asking for a code review
There are a few good ways to ask for a code review:
Are there folks who have been working on similar things, or a loosely related area? If so, they might be a good person to review your PR.
@-mention them with something like “
@person, would you be up for reviewing this?” If you’re not sure whether they are familiar with the code review process, you can also include a link to this guide.
If you would like feedback on user-facing changes, you can
@alyaon your PR. She can also help find someone to review the code once the PR is ready from a product perspective.
Finally, if you are not sure who should review the PR, just indicate clearly that it is ready for review, and the project maintainers will take a look and follow up with next steps.
With any of these approaches, please be patient and mindful of the fact that it isn’t always possible to provide a quick reply. Going though the review process described above for your own PR will make your code easier and faster to review, which makes it much more likely that it will be reviewed quickly and require fewer review cycles.
Reviewing someone else’s code
Fast replies are key
For the author of a PR, getting feedback quickly is really important for making progress quickly and staying productive. That means that if you get @-mentioned on a PR with a request for you to review it, it helps the author a lot if you reply promptly.
A reply doesn’t even have to be a full review; if a PR is big or if you’re pressed for time, then just getting some kind of reply in quickly – initial thoughts, feedback on the general direction, or just saying you’re busy and when you’ll have time to look harder – is still really valuable for the author and for anyone else who might review the PR.
People in the Zulip project live and work in many time zones, and code reviewers also need focused chunks of time to write code and do other things, so an immediate reply isn’t always possible. But a good benchmark is to try to always reply within one workday, at least with a short initial reply, if you’re working regularly on Zulip. And sooner is better.
Any time you leave a code review, be sure to treat the author with respect. Remember that they are generously spending their time on an effort to improve the Zulip project. Thank them for their work, and express your appreciation for anything the author did especially well, whether it’s a nice commit message, a prompt response to earlier feedback, or a well-written test.
Be as clear and direct as you can when describing requested changes. There is no need to apologize when asking for a change; you’re collaborating with the author to make the PR as good as you can together.
Be sure to explain the motivation for the changes you’re requesting if it’s not obvious, so that the author can learn for next time. It may be helpful to point to developer documentation, such as the commit discipline guide.
Fixing up the PR
If a pull request just needs a little fixing to make it mergeable, feel free to do that in a new commit, then push your branch to GitHub and mention the branch in a comment on the pull request. That’ll save the maintainer time and get the PR merged quicker.
Responding to review feedback
Once you’ve received a review and resolved any feedback, it’s critical to update the GitHub thread to reflect that. Best practices are to:
Make sure that CI passes and the PR is rebased onto recent
Post comments on each feedback thread explaining at least how you resolved the feedback, as well as any other useful information (problems encountered, reasoning for why you picked one of several options, a test you added to make sure the bug won’t recur, etc.).
Post a summary comment in the main feed for the PR, explaining that this is ready for another review, and summarizing any changes from the previous version, details on how you tested the changes, new screenshots/etc. More detail is better than less, as long as you take the time to write clearly.
If you resolve the feedback, but the PR has merge conflicts, CI failures, or the most recent comment is the reviewer asking you to fix something, it’s very likely that a potential reviewer skimming your PR will assume it isn’t ready for review and move on to other work.
If you need help or think an open discussion topic requires more feedback or a more complex discussion, move the discussion to a topic in the Zulip development community server. Be sure to provide links from the GitHub PR to the conversation (and vice versa) so that it’s convenient to read both conversations together.
We also recommend the following resources on code reviews.
The Gentle Art of Patch Review article by Sarah Sharp
Zulip & Good Code Review article by Sumana Harihareswara
Code Review - A consolidation of advice and stuff from the internet article by James J. Porter