Django's release process
Official releases
Since version 1.0, Django's release numbering works as follows:
- Versions are numbered in the form
A.B
orA.B.C
. A.B
is the feature release version number. Each version will be mostlybackwards compatible with the previous release. Exceptions to this rule willbe listed in the release notes.C
is the patch release version number, which is incremented for bugfixand security releases. These releases will be 100% backwards-compatible withthe previous patch release. The only exception is when a security or dataloss issue can't be fixed without breaking backwards-compatibility. If thishappens, the release notes will provide detailed upgrade instructions.- Before a new feature release, we'll make alpha, beta, and release candidatereleases. These are of the form
A.B alpha/beta/rc N
, which means theNth
alpha/beta/release candidate of versionA.B
.In git, each Django release will have a tag indicating its version number,signed with the Django release key. Additionally, each release series has itsown branch, calledstable/A.B.x
, and bugfix/security releases will beissued from those branches.
For more information about how the Django project issues new releases forsecurity purposes, please see our security policies.
- Feature release
- Feature releases (A.B, A.B+1, etc.) will happen roughly every eight months— see release process for details. These releases will contain newfeatures, improvements to existing features, and such.
- Patch release
- Patch releases (A.B.C, A.B.C+1, etc.) will be issued as needed, to fixbugs and/or security issues.
These releases will be 100% compatible with the associated feature release,unless this is impossible for security reasons or to prevent data loss.So the answer to "should I upgrade to the latest patch release?" will alwaysbe "yes."
- Long-term support release
- Certain feature releases will be designated as long-term support (LTS)releases. These releases will get security and data loss fixes applied fora guaranteed period of time, typically three years.
See the download page for the releases that have been designated forlong-term support.
Release cadence
Starting with Django 2.0, version numbers will use a loose form of semanticversioning such that each version following an LTS willbump to the next "dot zero" version. For example: 2.0, 2.1, 2.2 (LTS), 3.0,3.1, 3.2 (LTS), etc.
SemVer makes it easier to see at a glance how compatible releases are with eachother. It also helps to anticipate when compatibility shims will be removed.It's not a pure form of SemVer as each feature release will continue to have afew documented backwards incompatibilities where a deprecation path isn'tpossible or not worth the cost. Also, deprecations started in an LTS release(X.2) will be dropped in a non-dot-zero release (Y.1) to accommodate our policyof keeping deprecation shims for at least two feature releases. Read on to thenext section for an example.
Deprecation policy
A feature release may deprecate certain features from previous releases. If afeature is deprecated in feature release A.x, it will continue to work in allA.x versions (for all versions of x) but raise warnings. Deprecated featureswill be removed in the B.0 release, or B.1 for features deprecated in the lastA.x feature release to ensure deprecations are done over at least 2 featurereleases.
So, for example, if we decided to start the deprecation of a function inDjango 4.2:
- Django 4.2 will contain a backwards-compatible replica of the function whichwill raise a
RemovedInDjango51Warning
. - Django 5.0 (the version that follows 4.2) will still contain thebackwards-compatible replica.
- Django 5.1 will remove the feature outright.The warnings are silent by default. You can turn on display of these warningswith the
python -Wd
option.
A more generic example:
- X.0
- X.1
- X.2 LTS
- Y.0: Drop deprecation shims added in X.0 and X.1.
- Y.1: Drop deprecation shims added in X.2.
- Y.2 LTS: No deprecation shims dropped (while Y.0 is no longer supported,third-party apps need to maintain compatibility back to X.2 LTS to easeLTS to LTS upgrades).
- Z.0: Drop deprecation shims added in Y.0 and Y.1.
Supported versions
At any moment in time, Django's developer team will support a set of releases tovarying levels. See the supported versions section of the downloadpage for the current state of support for each version.
The current development master will get new features and bug fixesrequiring non-trivial refactoring.
Patches applied to the master branch must also be applied to the last featurerelease branch, to be released in the next patch release of that featureseries, when they fix critical problems:
- Security issues.
- Data loss bugs.
- Crashing bugs.
- Major functionality bugs in new features of the latest stable release.
- Regressions from older versions of Django.The rule of thumb is that fixes will be backported to the last featurerelease for bugs that would have prevented a release in the first place(release blockers).
Security fixes and data loss bugs will be applied to the current master, thelast two feature release branches, and any other supported long-termsupport release branches.
Documentation fixes generally will be more freely backported to the lastrelease branch. That's because it's highly advantageous to have the docs forthe last release be up-to-date and correct, and the risk of introducingregressions is much less of a concern.
As a concrete example, consider a moment in time halfway between the release ofDjango 5.1 and 5.2. At this point in time:
- Features will be added to development master, to be released as Django 5.2.
- Critical bug fixes will be applied to the
stable/5.1.x
branch, andreleased as 5.1.1, 5.1.2, etc. - Security fixes and bug fixes for data loss issues will be applied to
master
and to thestable/5.1.x
,stable/5.0.x
, andstable/4.2.x
(LTS) branches. They will trigger the release of5.1.1
,5.0.5
,4.2.8
, etc. - Documentation fixes will be applied to master, and, if easily backported, tothe latest stable branch,
5.1.x
.
Release process
Django uses a time-based release schedule, with feature releases every eightmonths or so.
After each feature release, the release manager will announce a timeline forthe next feature release.
Release cycle
Each release cycle consists of three parts:
Phase one: feature proposal
The first phase of the release process will include figuring out what majorfeatures to include in the next version. This should include a good deal ofpreliminary work on those features — working code trumps grand design.
Major features for an upcoming release will be added to the wiki roadmap page,e.g. https://code.djangoproject.com/wiki/Version1.11Roadmap.
Phase two: development
The second part of the release schedule is the "heads-down" working period.Using the roadmap produced at the end of phase one, we'll all work very hard toget everything on it done.
At the end of phase two, any unfinished features will be postponed until thenext release.
Phase two will culminate with an alpha release. At this point, thestable/A.B.x
branch will be forked from master
.
Phase three: bugfixes
The last part of a release cycle is spent fixing bugs — no new features willbe accepted during this time. We'll try to release a beta release one monthafter the alpha and a release candidate one month after the beta.
The release candidate marks the string freeze, and it happens at least twoweeks before the final release. After this point, new translatable stringsmust not be added.
During this phase, committers will be more and more conservative withbackports, to avoid introducing regressions. After the release candidate, onlyrelease blockers and documentation fixes should be backported.
In parallel to this phase, master
can receive new features, to be releasedin the A.B+1
cycle.
Bug-fix releases
After a feature release (e.g. A.B), the previous release will go into bugfixmode.
The branch for the previous feature release (e.g. stable/A.B-1.x
) willinclude bugfixes. Critical bugs fixed on master must also be fixed on thebugfix branch; this means that commits need to cleanly separate bug fixes fromfeature additions. The developer who commits a fix to master will beresponsible for also applying the fix to the current bugfix branch.