If you think working with Django is fun, wait until you start working on it. We’re passionate about helping Django users make the jump to contributing members of the community, so there are many ways you can help Django’s development:
That’s all you need to know if you’d like to join the Django development community. The rest of this document describes the details of how our community works and how it handles bugs, mailing lists, and all the other minutiae of Django development.
See also
This document contains specific details for contributing to Django. However, many new contributors find this guide confusing or intimidating at first. For a simpler introduction to becoming a contributor please see the How to contribute to Django guide.
Well-written bug reports are incredibly helpful. However, there’s a certain amount of overhead involved in working with any bug tracking system so your help in keeping our ticket tracker as useful as possible is appreciated. In particular:
Report security issues to security@djangoproject.com. This is a private list only open to long-time, highly trusted Django developers, and its archives are not publicly readable.
In the event of a confirmed vulnerability in Django itself, we will take the following actions:
We’re always grateful for patches to Django’s code. Indeed, bug reports with associated patches will get fixed far more quickly than those without patches.
In an open-source project with hundreds of contributors around the world, it’s important to manage communication efficiently so that work doesn’t get duplicated and contributors can be as effective as possible. Hence, our policy is for contributors to “claim” tickets in order to let other developers know that a particular bug or feature is being worked on.
If you have identified a contribution you want to make and you’re capable of fixing it (as measured by your coding ability, knowledge of Django internals and time availability), claim it by following these steps:
If you have an account but have forgotten your password, you can reset it using the password reset page.
Once you’ve claimed a ticket, you have a responsibility to work on that ticket in a reasonably timely fashion. If you don’t have time to work on it, either unclaim it or don’t claim it in the first place!
If there’s no sign of progress on a particular claimed ticket for a week or two, another developer may ask you to relinquish the ticket claim so that it’s no longer monopolized and somebody else can claim it.
If you’ve claimed a ticket and it’s taking a long time (days or weeks) to code, keep everybody updated by posting comments on the ticket. If you don’t provide regular updates, and you don’t respond to a request for a progress report, your claim on the ticket may be revoked. As always, more communication is better than less communication!
Of course, going through the steps of claiming tickets is overkill in some cases. In the case of small changes, such as typos in the documentation or small bugs that will only take a few minutes to fix, you don’t need to jump through the hoops of claiming tickets. Just submit your patch and be done with it.
Make sure your code matches our coding style.
Submit patches in the format returned by the svn diff command. An exception is for code changes that are described more clearly in plain English than in code. Indentation is the most common example; it’s hard to read patches when the only difference in code is that it’s indented.
Patches in git diff format are also acceptable.
When creating patches, always run svn diff from the top-level trunk directory – i.e., the one that contains django, docs, tests, AUTHORS, etc. This makes it easy for other people to apply your patches.
Attach patches to a ticket in the ticket tracker, using the “attach file” button. Please don’t put the patch in the ticket description or comment unless it’s a single line patch.
Name the patch file with a .diff extension; this will let the ticket tracker apply correct syntax highlighting, which is quite helpful.
Check the “Has patch” box on the ticket details. This will make it obvious that the ticket includes a patch, and it will add the ticket to the list of tickets with patches.
The code required to fix a problem or add a feature is an essential part of a patch, but it is not the only part. A good patch should also include a regression test to validate the behavior that has been fixed (and prevent the problem from arising again).
If the code associated with a patch adds a new feature, or modifies behavior of an existing feature, the patch should also contain documentation.
A “non-trivial” patch is one that is more than a simple bug fix. It’s a patch that introduces Django functionality and makes some sort of design decision.
If you provide a non-trivial patch, include evidence that alternatives have been discussed on django-developers. If you’re not sure whether your patch should be considered non-trivial, just ask.
Unfortunately, not all bug reports in the ticket tracker provide all the required details. A number of tickets have patches, but those patches don’t meet all the requirements of a good patch.
One way to help out is to triage bugs that have been reported by other users. The core team–as well as many community members–work on this regularly, but more help is always appreciated.
Most of the workflow is based around the concept of a ticket’s “triage stage”. This stage describes where in its lifetime a given ticket is at any time. Along with a handful of flags, this field easily tells us what and who each ticket is waiting on.
Since a picture is worth a thousand words, let’s start there:
We’ve got two roles in this diagram:
Second, note the five triage stages:
The second part of this workflow involves a set of flags the describe what the ticket has or needs in order to be “ready for checkin”:
See also
The contributing howto guide has a detailed explanation of each of the triage stages and how the triage process works in Trac.
A ticket can be resolved in a number of ways:
If you believe that the ticket was closed in error – because you’re still having the issue, or it’s popped up somewhere else, or the triagers have made a mistake – please reopen the ticket and provide further information. Please do not reopen tickets that have been marked as “wontfix” by core developers.
See also
For more information on what to do when closing a ticket, please see the contributing howto guide.
Although the core developers make the big decisions in the ticket triage process, there’s a lot that general community members can do to help the triage process. In particular, you can help out by:
However, we do ask the following of all general community members working in the ticket database:
Various parts of Django, such as the admin site and validation error messages, are internationalized. This means they display different text depending on a user’s language setting. For this, Django uses the same internationalization infrastructure available to Django applications described in the i18n documentation.
These translations are contributed by Django users worldwide. If you find an incorrect translation or want to discuss specific translations, go to the translation team page for that language. If you would like to help out with translating or add a language that isn’t yet translated, here’s what to do:
Join the Django i18n mailing list and introduce yourself.
Make sure you read the notes about Specialties of Django translation.
Signup at Transifex and visit the Django project page.
On the “Translation Teams” page, choose the language team you want to work with, or – in case the language team doesn’t exist yet – request a new team by clicking on the “Request a new team” button and select the appropriate language.
Then, click the “Join this Team” button to become a member of this team. Every team has at least one coordinator who is responsible to review your membership request. You can of course also contact the team coordinator to clarify procedural problems and handle the actual translation process.
Once you are a member of a team choose the translation resource you want update on the team page. For example the “core” resource refers to the translation catalogue that contains all non-app translations. Each of the contrib apps also have a resource (prefixed with “contrib-”).
Note
For more information about how to use Transifex, see the Transifex Help
Optionally, review and update the conf/locale/<locale>/formats.py file to describe the date, time and numbers formatting particularities of your locale. These files aren’t covered by the use of Transifex and require a patch against the Django source tree, just as a code change would:
Django’s admin system leverages the jQuery framework to increase the capabilities of the admin interface. In conjunction, there is an emphasis on admin javascript performance and minimizing overall admin media file size. Serving compressed or “minified” versions of javascript files is considered best practice in this regard.
To that end, patches for javascript files should include both the original code for future development (e.g. “foo.js”), and a compressed version for production use (e.g. “foo.min.js”). Any links to the file in the codebase should point to the compressed version.
To simplify the process of providing optimized javascript code, Django includes a handy script which should be used to create a “minified” version. This script is located at /contrib/admin/media/js/compress.py.
Behind the scenes, compress.py is a front-end for Google’s Closure Compiler which is written in Java. However, the Closure Compiler library is not bundled with Django directly, so those wishing to contribute complete javascript patches will need to download and install the library independently.
The Closure Compiler library requires Java version 6 or higher (Java 1.6 or higher on Mac OS X). Note that Mac OS X 10.5 and earlier did not ship with Java 1.6 by default, so it may be necessary to upgrade your Java installation before the tool will be functional. Also note that even after upgrading Java, the default /usr/bin/java command may remain linked to the previous Java binary, so relinking that command may be necessary as well.
Please don’t forget to run compress.py and include the diff of the minified scripts when submitting patches for Django’s javascript.
Various Django-specific code issues are detailed in this section.
Modules should not in general use settings stored in django.conf.settings at the top level (i.e. evaluated when the module is imported). The explanation for this is as follows:
Manual configuration of settings (i.e. not relying on the DJANGO_SETTINGS_MODULE environment variable) is allowed and possible as follows:
from django.conf import settings
settings.configure({}, SOME_SETTING='foo')
However, if any setting is accessed before the settings.configure line, this will not work. (Internally, settings is a LazyObject which configures itself automatically when the settings are accessed if it has not already been configured).
So, if there is a module containing some code as follows:
from django.conf import settings
from django.core.urlresolvers import get_callable
default_foo_view = get_callable(settings.FOO_VIEW)
...then importing this module will cause the settings object to be configured. That means that the ability for third parties to import the module at the top level is incompatible with the ability to configure the settings object manually, or makes it very difficult in some circumstances.
Instead of the above code, a level of laziness or indirection must be used, such as django.utils.functional.LazyObject`, django.utils.functional.lazy or lambda.
Please follow these coding standards when writing code for inclusion in Django:
Unless otherwise specified, follow PEP 8.
You could use a tool like pep8.py to check for some problems in this area, but remember that PEP 8 is only a guide, so respect the style of the surrounding code as a primary goal.
Use four spaces for indentation.
Use underscores, not camelCase, for variable, function and method names (i.e. poll.get_unique_voters(), not poll.getUniqueVoters).
Use InitialCaps for class names (or for factory functions that return classes).
Mark all strings for internationalization; see the i18n documentation for details.
In docstrings, use "action words" such as:
def foo():
"""
Calculates something and returns the result.
"""
pass
Here's an example of what not to do:
def foo():
"""
Calculate something and return the result.
"""
pass
Please don't put your name in the code you contribute. Our policy is to keep contributors' names in the AUTHORS file distributed with Django -- not scattered throughout the codebase itself. Feel free to include a change to the AUTHORS file in your patch if you make more than a single trivial change.
In Django template code, put one (and only one) space between the curly brackets and the tag contents.
Do this:
{{ foo }}
Don't do this:
{{foo}}
In Django views, the first parameter in a view function should be called request.
Do this:
def my_view(request, foo):
# ...
Don't do this:
def my_view(req, foo):
# ...
Field names should be all lowercase, using underscores instead of camelCase.
Do this:
class Person(models.Model):
first_name = models.CharField(max_length=20)
last_name = models.CharField(max_length=40)
Don't do this:
class Person(models.Model):
FirstName = models.CharField(max_length=20)
Last_Name = models.CharField(max_length=40)
The class Meta should appear after the fields are defined, with a single blank line separating the fields and the class definition.
Do this:
class Person(models.Model):
first_name = models.CharField(max_length=20)
last_name = models.CharField(max_length=40)
class Meta:
verbose_name_plural = 'people'
Don't do this:
class Person(models.Model):
first_name = models.CharField(max_length=20)
last_name = models.CharField(max_length=40)
class Meta:
verbose_name_plural = 'people'
Don't do this, either:
class Person(models.Model):
class Meta:
verbose_name_plural = 'people'
first_name = models.CharField(max_length=20)
last_name = models.CharField(max_length=40)
The order of model inner classes and standard methods should be as follows (noting that these are not all required):
If choices is defined for a given model field, define the choices as a tuple of tuples, with an all-uppercase name, either near the top of the model module or just above the model class. Example:
GENDER_CHOICES = (
('M', 'Male'),
('F', 'Female'),
)
We place a high importance on consistency and readability of documentation. (After all, Django was created in a journalism environment!)
We treat our documentation like we treat our code: we aim to improve it as often as possible. This section explains how writers can craft their documentation changes in the most useful and least error-prone ways.
Documentation changes come in two forms:
Our policy is:
All documentation of new features should be written in a way that clearly designates the features are only available in the Django development version. Assume documentation readers are using the latest release, not the development version.
Our preferred way for marking new features is by prefacing the features' documentation with: ".. versionadded:: X.Y", followed by an optional one line comment and a mandatory blank line.
General improvements, or other changes to the APIs that should be emphasized should use the ".. versionchanged:: X.Y" directive (with the same format as the versionadded mentioned above.
There's a full page of information about the Django documentation system that you should read prior to working on the documentation.
These guidelines regulate the format of our reST documentation:
Here are some style guidelines on commonly used terms throughout the documentation:
Please follow these guidelines when committing code to Django's Subversion repository:
For any medium-to-big changes, where "medium-to-big" is according to your judgment, please bring things up on the django-developers mailing list before making the change.
If you bring something up on django-developers and nobody responds, please don't take that to mean your idea is great and should be implemented immediately because nobody contested it. Django's lead developers don't have a lot of time to read mailing-list discussions immediately, so you may have to wait a couple of days before getting a response.
Write detailed commit messages in the past tense, not present tense.
For commits to a branch, prefix the commit message with the branch name. For example: "magic-removal: Added support for mind reading."
Limit commits to the most granular change that makes sense. This means, use frequent small commits rather than infrequent large commits. For example, if implementing feature X requires a small change to library Y, first commit the change to library Y, then commit feature X in a separate commit. This goes a long way in helping all core Django developers follow your changes.
Separate bug fixes from feature changes.
Bug fixes need to be added to the current bugfix branch (e.g. the 1.0.X branch) as well as the current trunk.
If your commit closes a ticket in the Django ticket tracker, begin your commit message with the text "Fixed #abc", where "abc" is the number of the ticket your commit fixes. Example: "Fixed #123 -- Added support for foo". We've rigged Subversion and Trac so that any commit message in that format will automatically close the referenced ticket and post a comment to it with the full commit message.
If your commit closes a ticket and is in a branch, use the branch name first, then the "Fixed #abc." For example: "magic-removal: Fixed #123 -- Added whizbang feature."
For the curious: We're using a Trac post-commit hook for this.
If your commit references a ticket in the Django ticket tracker but does not close the ticket, include the phrase "Refs #abc", where "abc" is the number of the ticket your commit references. We've rigged Subversion and Trac so that any commit message in that format will automatically post a comment to the appropriate ticket.
Nobody's perfect; mistakes will be committed. When a mistaken commit is discovered, please follow these guidelines:
Django comes with a test suite of its own, in the tests directory of the Django tarball. It's our policy to make sure all tests pass at all times.
The tests cover:
We appreciate any and all contributions to the test suite!
The Django tests all use the testing infrastructure that ships with Django for testing applications. See Testing Django applications for an explanation of how to write new tests.
Running the tests requires a Django settings module that defines the databases to use. To make it easy to get started. Django provides a sample settings module that uses the SQLite database. To run the tests with this sample settings module, cd into the Django tests/ directory and run:
./runtests.py --settings=test_sqlite
If you get an ImportError: No module named django.contrib error, you need to add your install of Django to your PYTHONPATH. For more details on how to do this, read Pointing Python at the new Django version below.
The included settings module allows you to run the test suite using SQLite. If you want to test behavior using a different database (and if you're proposing patches for Django, it's a good idea to test across databases), you may need to define your own settings file.
To run the tests with different settings, cd to the tests/ directory and type:
./runtests.py --settings=path.to.django.settings
The DATABASES setting in this test settings module needs to define two databases:
If you're using a backend that isn't SQLite, you will need to provide other details for each database:
You will also need to ensure that your database uses UTF-8 as the default character set. If your database server doesn't use UTF-8 as a default charset, you will need to include a value for TEST_CHARSET in the settings dictionary for the applicable database.
Django's entire test suite takes a while to run, and running every single test could be redundant if, say, you just added a test to Django that you want to run quickly without running everything else. You can run a subset of the unit tests by appending the names of the test modules to runtests.py on the command line.
For example, if you'd like to run tests only for generic relations and internationalization, type:
./runtests.py --settings=path.to.settings generic_relations i18n
How do you find out the names of individual tests? Look in tests/modeltests and tests/regressiontests -- each directory name there is the name of a test.
If you just want to run a particular class of tests, you can specify a list of paths to individual test classes. For example, to run the TranslationTests of the i18n module, type:
./runtests.py --settings=path.to.settings i18n.TranslationTests
Going beyond that, you can specify an individual test method like this:
./runtests.py --settings=path.to.settings i18n.TranslationTests.test_lazy_objects
If you want to run the full suite of tests, you'll need to install a number of dependencies:
If you want to test the memcached cache backend, you'll also need to define a CACHES setting that points at your memcached instance.
Each of these dependencies is optional. If you're missing any of them, the associated tests will be skipped.
Tests for apps in django/contrib/ go in their respective directories under django/contrib/, in a tests.py file. (You can split the tests over multiple modules by using a tests directory in the normal Python way.)
For the tests to be found, a models.py file must exist (it doesn't have to have anything in it). If you have URLs that need to be mapped, put them in tests/urls.py.
To run tests for just one contrib app (e.g. markup), use the same method as above:
./runtests.py --settings=settings markup
We're always trying to make Django better, and your feature requests are a key part of that. Here are some tips on how to make a request most effectively:
As with most open-source projects, code talks. If you are willing to write the code for the feature yourself or if (even better) you've already written it, it's much more likely to be accepted. If it's a large feature that might need multiple developers, we're always happy to give you an experimental branch in our repository; see below.
In general, the trunk must be kept stable. People should be able to run production sites against the trunk at any time. Additionally, commits to trunk ought to be as atomic as possible -- smaller changes are better. Thus, large feature changes -- that is, changes too large to be encapsulated in a single patch, or changes that need multiple eyes on them -- must happen on dedicated branches.
This means that if you want to work on a large feature -- anything that would take more than a single patch, or requires large-scale refactoring -- you need to do it on a feature branch. Our development process recognizes two options for feature branches:
Feature branches using a distributed revision control system like Git, Mercurial, Bazaar, etc.
If you're familiar with one of these tools, this is probably your best option since it doesn't require any support or buy-in from the Django core developers.
However, do keep in mind that Django will continue to use Subversion for the foreseeable future, and this will naturally limit the recognition of your branch. Further, if your branch becomes eligible for merging to trunk you'll need to find a core developer familiar with your DVCS of choice who'll actually perform the merge.
If you do decided to start a distributed branch of Django and choose to make it public, please add the branch to the Django branches wiki page.
Feature branches using SVN have a higher bar. If you want a branch in SVN itself, you'll need a "mentor" among the core committers. This person is responsible for actually creating the branch, monitoring your process (see below), and ultimately merging the branch into trunk.
If you want a feature branch in SVN, you'll need to ask in django-developers for a mentor.
We've got a few rules for branches born out of experience with what makes a successful Django branch.
DVCS branches are obviously not under central control, so we have no way of enforcing these rules. However, if you're using a DVCS, following these rules will give you the best chance of having a successful branch (read: merged back to trunk).
Developers with branches in SVN, however, must follow these rules. The branch mentor will keep on eye on the branch and will delete it if these rules are broken.
Only branch entire copies of the Django tree, even if work is only happening on part of that tree. This makes it painless to switch to a branch.
Merge changes from trunk no less than once a week, and preferably every couple-three days.
In our experience, doing regular trunk merges is often the difference between a successful branch and one that fizzles and dies.
If you're working on an SVN branch, you should be using svnmerge.py to track merges from trunk.
Keep tests passing and documentation up-to-date. As with patches, we'll only merge a branch that comes with tests and documentation.
Once the branch is stable and ready to be merged into the trunk, alert django-developers.
After a branch has been merged, it should be considered "dead"; write access to it will be disabled, and old branches will be periodically "trimmed." To keep our SVN wrangling to a minimum, we won't be merging from a given branch into the trunk more than once.
To use a branch, you'll need to do two things:
To get the latest version of a branch's code, check it out using Subversion:
svn co http://code.djangoproject.com/svn/django/branches/<branch>/
...where <branch> is the branch's name. See the list of branch names.
Alternatively, you can automatically convert an existing directory of the Django source code as long as you've checked it out via Subversion. To do the conversion, execute this command from within your django directory:
svn switch http://code.djangoproject.com/svn/django/branches/<branch>/
The advantage of using svn switch instead of svn co is that the switch command retains any changes you might have made to your local copy of the code. It attempts to merge those changes into the "switched" code. The disadvantage is that it may cause conflicts with your local changes if the "switched" code has altered the same lines of code.
(Note that if you use svn switch, you don't need to point Python at the new version, as explained in the next section.)
Once you've retrieved the branch's code, you'll need to change your Python site-packages directory so that it points to the branch version of the django directory. (The site-packages directory is somewhere such as /usr/lib/python2.4/site-packages or /usr/local/lib/python2.4/site-packages or C:\Python\site-packages.)
The simplest way to do this is by renaming the old django directory to django.OLD and moving the trunk version of the code into the directory and calling it django.
Alternatively, you can use a symlink called django that points to the location of the branch's django package. If you want to switch back, just change the symlink to point to the old code.
A third option is to use a path file (<something>.pth) which should work on all systems (including Windows, which doesn't have symlinks available). First, make sure there are no files, directories or symlinks named django in your site-packages directory. Then create a text file named django.pth and save it to your site-packages directory. That file should contain a path to your copy of Django on a single line and optional comments. Here is an example that points to multiple branches. Just uncomment the line for the branch you want to use ('Trunk' in this example) and make sure all other lines are commented:
# Trunk is a svn checkout of:
# http://code.djangoproject.com/svn/django/trunk/
#
/path/to/trunk
# <branch> is a svn checkout of:
# http://code.djangoproject.com/svn/django/branches/<branch>/
#
#/path/to/<branch>
# On windows a path may look like this:
# C:/path/to/<branch>
If you're using Django 0.95 or earlier and installed it using python setup.py install, you'll have a directory called something like Django-0.95-py2.4.egg instead of django. In this case, edit the file setuptools.pth and remove the line that references the Django .egg file. Then copy the branch's version of the django directory into site-packages.
Whenever possible, we strive for a rough consensus. To that end, we'll often have informal votes on django-developers about a feature. In these votes we follow the voting style invented by Apache and used on Python itself, where votes are given as +1, +0, -0, or -1. Roughly translated, these votes mean:
Although these votes on django-developers are informal, they'll be taken very seriously. After a suitable voting period, if an obvious consensus arises we'll follow the votes.
However, consensus is not always possible. If consensus cannot be reached, or if the discussion towards a consensus fizzles out without a concrete decision, we use a more formal process.
Any core committer (see below) may call for a formal vote using the same voting mechanism above. A proposition will be considered carried by the core team if:
When calling for a vote, the caller should specify a deadline by which votes must be received. One week is generally suggested as the minimum amount of time.
Since this process allows any core committer to veto a proposal, any "-1" votes (or BDFL vetos) should be accompanied by an explanation that explains what it would take to convert that "-1" into at least a "+0".
Whenever possible, these formal votes should be announced and held in public on the django-developers mailing list. However, overly sensitive or contentious issues -- including, most notably, votes on new core committers -- may be held in private.
Django has two types of committers:
These are people who are "domain experts." They have direct check-in access to the subsystems that fall under their jurisdiction, and they're given a formal vote in questions that involve their subsystems. This type of access is likely to be given to someone who contributes a large subframework to Django and wants to continue to maintain it.
Partial commit access is granted by the same process as full committers. However, the bar is set lower; proven expertise in the area in question is likely to be sufficient.
Decisions on new committers will follow the process explained above in how we make decisions.
To request commit access, please contact an existing committer privately. Public requests for commit access are potential flame-war starters, and will be ignored.
Dec 26, 2011