v0.3.3

Compare Source

v0.3.2

Compare Source

See: https://github.com/astral-sh/ruff/releases/tag/v0.3.2

v0.3.1

Compare Source

See: https://github.com/astral-sh/ruff/releases/tag/v0.3.1

v0.3.0

Compare Source

See: https://github.com/astral-sh/ruff/releases/tag/v0.3.0

v0.2.2

Compare Source

See: https://github.com/astral-sh/ruff/releases/tag/v0.2.2

v3.6

Compare Source

v9.5.13: mkdocs-material-9.5.13

Compare Source

• Updated Slovak translations
• Improved info plugin interop with projects plugin
• Improved info plugin inclusion/exclusion logic
• Fixed info plugin not gathering files recursively
• Fixed #​6750: Ensure info plugin packs up all necessary files

Thanks to @​kamilkrzyskow and @​scepka for their contributions

v9.5.12: mkdocs-material-9.5.12

Compare Source

• Fixed #​6846: Some meta tags removed on instant navigation (9.4.2 regression)
• Fixed #​6823: KaTex not rendering on instant navigation (9.5.5 regression)
• Fixed #​6821: Privacy plugin doesn't handle URLs with encoded characters

v9.5.11: mkdocs-material-9.5.11

Compare Source

• Updated Finnish translation

v9.5.10: mkdocs-material-9.5.10

Compare Source

• Updated Bahasa Malaysia translations
• Fixed #​6783: Hide continue reading link for blog posts without separators
• Fixed #​6779: Incorrect positioning of integrated table of contents

v9.5.9: mkdocs-material-9.5.9

Compare Source

• Fixed navigation pruning with tabs and sections enabled

v0.24.1

Compare Source

Compare with 0.24.0

Code Refactoring

• Support new pymdownx-highlight options (a7a2907 by Timothée Mazzucotelli).
• Backup anchors with id and no href, for compatibility with autorefs' Markdown anchors (b5236b4 by Timothée Mazzucotelli). PR-#​651, Related-to-mkdocs-autorefs#39, Co-authored-by: Oleh Prypin oleh@pryp.in

v1.9.0

Compare Source

Compare with 1.8.0

Dependencies

• Add upper bound on Python-Markdown 3.6 to temporarily prevent breaking changes (cd93ee3 by Timothée Mazzucotelli).

Features

• Add show_labels option to show/hide labels (eaf9b82 by Viicos). [Issue

#​120](https://togithub.com/mkdocstrings/python/issues/120), PR #​130

• Add option to search for stubs packages (0c6aa32 by Romain). [PR

#​128](https://togithub.com/mkdocstrings/python/pull/128), PR griffe#221: : https://github.com/mkdocstrings/griffe/pull/221

Code Refactoring

• Mark all Jinja blocks as scoped (548bdad by Timothée Mazzucotelli).

v0.3.3

Compare Source

Preview features

• [flake8-bandit]: Implement S610 rule (#​10316)
• [pycodestyle] Implement blank-line-at-end-of-file (W391) (#​10243)
• [pycodestyle] Implement redundant-backslash (E502) (#​10292)
• [pylint] - implement redeclared-assigned-name (W0128) (#​9268)

Rule changes

• [flake8_comprehensions] Handled special case for C400 which also matches C416 (#​10419)
• [flake8-bandit] Implement upstream updates for S311, S324 and S605 (#​10313)
• [pyflakes] Remove F401 fix for __init__ imports by default and allow opt-in to unsafe fix (#​10365)
• [pylint] Implement invalid-bool-return-type (E304) (#​10377)
• [pylint] Include builtin warnings in useless-exception-statement (PLW0133) (#​10394)

CLI

• Add message on success to ruff check (#​8631)

Bug fixes

• [PIE970] Allow trailing ellipsis in typing.TYPE_CHECKING (#​10413)
• Avoid TRIO115 if the argument is a variable (#​10376)
• [F811] Avoid removing shadowed imports that point to different symbols (#​10387)
• Fix F821 and F822 false positives in .pyi files (#​10341)
• Fix F821 false negatives in .py files when from __future__ import annotations is active (#​10362)
• Fix case where Indexer fails to identify continuation preceded by newline #​10351 (#​10354)
• Sort hash maps in Settings display (#​10370)
• Track conditional deletions in the semantic model (#​10415)
• [C413] Wrap expressions in parentheses when negating (#​10346)
• [pycodestyle] Do not ignore lines before the first logical line in blank lines rules. (#​10382)
• [pycodestyle] Do not trigger E225 and E275 when the next token is a ')' (#​10315)
• [pylint] Avoid false-positive slot non-assignment for __dict__ (PLE0237) (#​10348)
• Gate f-string struct size test for Rustc < 1.76 (#​10371)

Documentation

• Use ruff.toml format in README (#​10393)
• [RUF008] Make it clearer that a mutable default in a dataclass is only valid if it is typed as a ClassVar (#​10395)
• [pylint] Extend docs and test in invalid-str-return-type (E307) (#​10400)
• Remove . from check and format commands (#​10217)

v0.3.2

Compare Source

Preview features

• Improve single-with item formatting for Python 3.8 or older (#​10276)

Rule changes

• [pyupgrade] Allow fixes for f-string rule regardless of line length (UP032) (#​10263)
• [pycodestyle] Include actual conditions in E712 diagnostics (#​10254)

Bug fixes

• Fix trailing kwargs end of line comment after slash (#​10297)
• Fix unstable with items formatting (#​10274)
• Avoid repeating function calls in f-string conversions (#​10265)
• Fix E203 false positive for slices in format strings (#​10280)
• Fix incorrect Parameter range for *args and **kwargs (#​10283)
• Treat typing.Annotated subscripts as type definitions (#​10285)

v0.3.1

Compare Source

Preview features

• [pycodestyle] Fix E301 not triggering on decorated methods. (#​10117)
• [pycodestyle] Respect isort settings in blank line rules (E3*) (#​10096)
• [pycodestyle] Make blank lines in typing stub files optional (E3*) (#​10098)
• [pylint] Implement singledispatch-method (E1519) (#​10140)
• [pylint] Implement useless-exception-statement (W0133) (#​10176)

Rule changes

• [flake8-debugger] Check for use of debugpy and ptvsd debug modules (#​10177)

(#​10194)

• [pyupgrade] Generate diagnostic for all valid f-string conversions regardless of line length (UP032) (#​10238)
• [pep8_naming] Add fixes for N804 and N805 (#​10215)

CLI

• Colorize the output of ruff format --diff (#​10110)
• Make --config and --isolated global flags (#​10150)
• Correctly expand tildes and environment variables in paths passed to --config (#​10219)

Configuration

• Accept a PEP 440 version specifier for required-version (#​10216)
• Implement isort's default-section setting (#​10149)

Bug fixes

• Remove trailing space from CapWords message (#​10220)
• Respect external codes in file-level exemptions (#​10203)
• [flake8-raise] Avoid false-positives for parens-on-raise with future.exception() (RSE102) (#​10206)
• [pylint] Add fix for unary expressions in PLC2801 (#​9587)
• [ruff] Fix RUF028 not allowing # fmt: skip on match cases (#​10178)

v0.3.0

Compare Source

This release introduces the new Ruff formatter 2024.2 style and adds a new lint rule to detect invalid formatter suppression comments.

Preview features

• [flake8-bandit] Remove suspicious-lxml-import (S410) (#​10154)
• [pycodestyle] Allow os.environ modifications between imports (E402) (#​10066)
• [pycodestyle] Don't warn about a single whitespace character before a comma in a tuple (E203) (#​10094)

Rule changes

• [eradicate] Detect commented out case statements (ERA001) (#​10055)
• [eradicate] Detect single-line code for try:, except:, etc. (ERA001) (#​10057)
• [flake8-boolean-trap] Allow boolean positionals in __post_init__ (#​10027)
• [flake8-copyright] Allow © in copyright notices (#​10065)
• [isort]: Use one blank line after imports in typing stub files (#​9971)
• [pylint] New Rule dict-iter-missing-items (PLE1141) (#​9845)
• [pylint] Ignore sys.version and sys.platform (PLR1714) (#​10054)
• [pyupgrade] Detect literals with unary operators (UP018) (#​10060)
• [ruff] Expand rule for list(iterable).pop(0) idiom (RUF015) (#​10148)

Formatter

This release introduces the Ruff 2024.2 style, stabilizing the following changes:

• Prefer splitting the assignment's value over the target or type annotation (#​8943)
• Remove blank lines before class docstrings (#​9154)
• Wrap multiple context managers in with parentheses when targeting Python 3.9 or newer (#​9222)
• Add a blank line after nested classes with a dummy body (...) in typing stub files (#​9155)
• Reduce vertical spacing for classes and functions with a dummy (...) body (#​7440, #​9240)
• Add a blank line after the module docstring (#​8283)
• Parenthesize long type hints in assignments (#​9210)
• Preserve indent for single multiline-string call-expressions (#​9673)
• Normalize hex escape and unicode escape sequences (#​9280)
• Format module docstrings (#​9725)

CLI

• Explicitly disallow extend as part of a --config flag (#​10135)
• Remove build from the default exclusion list (#​10093)
• Deprecate ruff <path>, ruff --explain, ruff --clean, and ruff --generate-shell-completion in favor of ruff check <path>, ruff rule, ruff clean, and ruff generate-shell-completion (#​10169)
• Remove the deprecated CLI option --format from ruff rule and ruff linter (#​10170)

Bug fixes

• [flake8-bugbear] Avoid adding default initializers to stubs (B006) (#​10152)
• [flake8-type-checking] Respect runtime-required decorators for function signatures (#​10091)
• [pycodestyle] Mark fixes overlapping with a multiline string as unsafe (W293) (#​10049)
• [pydocstyle] Trim whitespace when removing blank lines after section (D413) (#​10162)
• [pylint] Delete entire statement, including semicolons (PLR0203) (#​10074)
• [ruff] Avoid f-string false positives in gettext calls (RUF027) (#​10118)
• Fix ruff crashing on PowerPC systems because of too small page size (#​10080)

Performance

• Add cold attribute to less likely printer queue branches in the formatter (#​10121)
• Skip unnecessary string normalization in the formatter (#​10116)

Documentation

• Remove "Beta" Label from formatter documentation (#​10144)
• line-length option: fix link to pycodestyle.max-line-length (#​10136)

v0.2.2

Compare Source

Highlights include:

• Initial support formatting f-strings (in --preview).
• Support for overriding arbitrary configuration options via the CLI through an expanded --config argument (e.g., --config "lint.isort.combine-as-imports=false").
• Significant performance improvements in Ruff's lexer, parser, and lint rules.

Preview features

• Implement minimal f-string formatting (#​9642)
• [pycodestyle] Add blank line(s) rules (E301, E302, E303, E304, E305, E306) (#​9266)
• [refurb] Implement readlines_in_for (FURB129) (#​9880)

Rule changes

• [ruff] Ensure closing parentheses for multiline sequences are always on their own line (RUF022, RUF023) (#​9793)
• [numpy] Add missing deprecation violations (NPY002) (#​9862)
• [flake8-bandit] Detect mark_safe usages in decorators (#​9887)
• [ruff] Expand asyncio-dangling-task (RUF006) to include new_event_loop (#​9976)
• [flake8-pyi] Ignore 'unused' private type dicts in class scopes (#​9952)

Formatter

• Docstring formatting: Preserve tab indentation when using indent-style=tabs (#​9915)
• Disable top-level docstring formatting for notebooks (#​9957)
• Stabilize quote-style's preserve mode (#​9922)

CLI

• Allow arbitrary configuration options to be overridden via the CLI (#​9599)

Bug fixes

• Make show-settings filters directory-agnostic (#​9866)
• Respect duplicates when rewriting type aliases (#​9905)
• Respect tuple assignments in typing analyzer (#​9969)
• Use atomic write when persisting cache (#​9981)
• Use non-parenthesized range for DebugText (#​9953)
• [flake8-simplify] Avoid false positive with async for loops (SIM113) (#​9996)
• [flake8-trio] Respect async with in timeout-without-await (#​9859)
• [perflint] Catch a wider range of mutations in PERF101 (#​9955)
• [pycodestyle] Fix E30X panics on blank lines with trailing white spaces (#​9907)
• [pydocstyle] Allow using parameters as a subsection header (D405) (#​9894)
• [pydocstyle] Fix blank-line docstring rules for module-level docstrings (#​9878)
• [pylint] Accept 0.0 and 1.0 as common magic values (PLR2004) (#​9964)
• [pylint] Avoid suggesting set rewrites for non-hashable types (#​9956)
• [ruff] Avoid false negatives with string literals inside of method calls (RUF027) (#​9865)
• [ruff] Fix panic on with f-string detection (RUF027) (#​9990)
• [ruff] Ignore builtins when detecting missing f-strings (#​9849)

Performance

• Use memchr for string lexing (#​9888)
• Use memchr for tab-indentation detection (#​9853)
• Reduce Result<Tok, LexicalError> size by using Box<str> instead of String (#​9885)
• Reduce size of Expr from 80 to 64 bytes (#​9900)
• Improve trailing comma rule performance (#​9867)
• Remove unnecessary string cloning from the parser (#​9884)

v7.4.4

Compare Source

• Fix: in some cases, even with [run] relative_files=True, a data file could be created with absolute path names. When combined with other relative data files, it was random whether the absolute file names would be made relative or not. If they weren't, then a file would be listed twice in reports, as detailed in issue 1752_. This is now fixed: absolute file names are always made relative when combining. Thanks to Bruno Rodrigues dos Santos for support.

• Fix: the last case of a match/case statement had an incorrect message if the branch was missed. It said the pattern never matched, when actually the branch is missed if the last case always matched.

• Fix: clicking a line number in the HTML report now positions more accurately.

• Fix: the report:format setting was defined as a boolean, but should be a string. Thanks, Tanaydin Sirin <pull 1754_>_. It is also now documented on the :ref:configuration page <config_report_format>.

.. _issue 1752:https://github.com/nedbat/coveragepy/issues/17522 .. _pull 1754https://github.com/nedbat/coveragepy/pull/175454

.. _changes_7-4-3:

v7.4.3

Compare Source

• Fix: in some cases, coverage could fail with a RuntimeError: "Set changed size during iteration." This is now fixed, closing issue 1733_.

.. _issue 1733:https://github.com/nedbat/coveragepy/issues/17333

.. _changes_7-4-2:

v7.4.2

Compare Source

• Fix: setting COVERAGE_CORE=sysmon no longer errors on 3.11 and lower, thanks Hugo van Kemenade <pull 1747_>_. It now issues a warning that sys.monitoring is not available and falls back to the default core instead.

.. _pull 1747:https://github.com/nedbat/coveragepy/pull/17477

.. _changes_7-4-1:

v8.1.1

Compare Source

pytest 8.1.1 (2024-03-08)

::: {.note} ::: {.title} Note :::

This release is not a usual bug fix release -- it contains features and improvements, being a follow up to 8.1.0, which has been yanked from PyPI. :::

Features

• #​11475: Added the new consider_namespace_packages{.interpreted-text role="confval"} configuration option, defaulting to False.

If set to True, pytest will attempt to identify modules that are part of namespace packages when importing modules.

• #​11653: Added the new verbosity_test_cases{.interpreted-text role="confval"} configuration option for fine-grained control of test execution verbosity. See Fine-grained verbosity <pytest.fine_grained_verbosity>{.interpreted-text role="ref"} for more details.

Improvements

• #​10865: pytest.warns{.interpreted-text role="func"} now validates that warnings.warn{.interpreted-text role="func"} was called with a [str]{.title-ref} or a [Warning]{.title-ref}. Currently in Python it is possible to use other types, however this causes an exception when warnings.filterwarnings{.interpreted-text role="func"} is used to filter those warnings (see [CPython

#​103577](https://togithub.com/python/cpython/issues/103577) for a discussion). While this can be considered a bug in CPython, we decided to put guards in pytest as the error message produced without this check in place is confusing.

• #​11311: When using --override-ini for paths in invocations without a configuration file defined, the current working directory is used as the relative directory.

Previoulsy this would raise an AssertionError{.interpreted-text role="class"}.

• #​11475: --import-mode=importlib <import-mode-importlib>{.interpreted-text role="ref"} now tries to import modules using the standard import mechanism (but still without changing :pysys.path{.interpreted-text role="data"}), falling back to importing modules directly only if that fails.

This means that installed packages will be imported under their canonical name if possible first, for example app.core.models, instead of having the module name always be derived from their path (for example .env310.lib.site_packages.app.core.models).

• #​11801: Added the iter_parents() <_pytest.nodes.Node.iter_parents>{.interpreted-text role="func"} helper method on nodes. It is similar to listchain <_pytest.nodes.Node.listchain>{.interpreted-text role="func"}, but goes from bottom to top, and returns an iterator, not a list.

• #​11850: Added support for sys.last_exc{.interpreted-text role="data"} for post-mortem debugging on Python>=3.12.

• #​11962: In case no other suitable candidates for configuration file are found, a pyproject.toml (even without a [tool.pytest.ini_options] table) will be considered as the configuration file and define the rootdir.

• #​11978: Add --log-file-mode option to the logging plugin, enabling appending to log-files. This option accepts either "w" or "a" and defaults to "w".

Previously, the mode was hard-coded to be "w" which truncates the file before logging.

• #​12047: When multiple finalizers of a fixture raise an exception, now all exceptions are reported as an exception group. Previously, only the first exception was reported.

Bug Fixes

• #​11475: Fixed regression where --importmode=importlib would import non-test modules more than once.

• #​11904: Fixed a regression in pytest 8.0.0 that would cause test collection to fail due to permission errors when using --pyargs.

This change improves the collection tree for tests specified using --pyargs, see 12043{.interpreted-text role="pull"} for a comparison with pytest 8.0 and <8.

• #​12011: Fixed a regression in 8.0.1 whereby setup_module xunit-style fixtures are not executed when --doctest-modules is passed.

• #​12014: Fix the stacklevel used when warning about marks used on fixtures.

• #​12039: Fixed a regression in 8.0.2 where tests created using tmp_path{.interpreted-text role="fixture"} have been collected multiple times in CI under Windows.

Improved Documentation

• #​11790: Documented the retention of temporary directories created using the tmp_path fixture in more detail.

Trivial/Internal Changes

• #​11785: Some changes were made to private functions which may affect plugins which access them:

• FixtureManager._getautousenames() now takes a Node itself instead of the nodeid.

• FixtureManager.getfixturedefs() now takes the Node itself instead of the nodeid.

• The _pytest.nodes.iterparentnodeids() function is removed without replacement. Prefer to traverse the node hierarchy itself instead.

If you really need to, copy the function from the previous pytest release.

• #​12069: Delayed the deprecation of the following features to 9.0.0:

  • node-ctor-fspath-deprecation{.interpreted-text role="ref"}.
  • legacy-path-hooks-deprecated{.interpreted-text role="ref"}.

It was discovered after 8.1.0 was released that the warnings about the impeding removal were not being displayed, so the team decided to revert the removal.

This is the reason for `8.1.0` being yanked.

pytest 8.1.0 (YANKED)

::: {.note} ::: {.title} Note :::

This release has been yanked: it broke some plugins without the proper warning period, due to some warnings not showing up as expected.

See #​12069. :::

v8.1.0

Compare Source

pytest 8.1.0 (YANKED)

[!IMPORTANT]
This release has been yanked: it broke some plugins without the proper warning period, due to some warnings not showing up as expected. See #​12069.

Features

• #​11475: Added the new consider_namespace_packages{.interpreted-text role="confval"} configuration option, defaulting to False.

If set to True, pytest will attempt to identify modules that are part of namespace packages when importing modules.

• #​11653: Added the new verbosity_test_cases{.interpreted-text role="confval"} configuration option for fine-grained control of test execution verbosity. See Fine-grained verbosity <pytest.fine_grained_verbosity>{.interpreted-text role="ref"} for more details.

Improvements

• #​10865: pytest.warns{.interpreted-text role="func"} now validates that warnings.warn{.interpreted-text role="func"} was called with a [str]{.title-ref} or a [Warning]{.title-ref}. Currently in Python it is possible to use other types, however this causes an exception when warnings.filterwarnings{.interpreted-text role="func"} is used to filter those warnings (see [CPython

#​103577](https://togithub.com/python/cpython/issues/103577) for a discussion). While this can be considered a bug in CPython, we decided to put guards in pytest as the error message produced without this check in place is confusing.

• #​11311: When using --override-ini for paths in invocations without a configuration file defined, the current working directory is used as the relative directory.

Previoulsy this would raise an AssertionError{.interpreted-text role="class"}.

• #​11475: --import-mode=importlib <import-mode-importlib>{.interpreted-text role="ref"} now tries to import modules using the standard import mechanism (but still without changing :pysys.path{.interpreted-text role="data"}), falling back to importing modules directly only if that fails.

This means that installed packages will be imported under their canonical name if possible first, for example app.core.models, instead of having the module name always be derived from their path (for example .env310.lib.site_packages.app.core.models).

• #​11801: Added the iter_parents() <_pytest.nodes.Node.iter_parents>{.interpreted-text role="func"} helper method on nodes. It is similar to listchain <_pytest.nodes.Node.listchain>{.interpreted-text role="func"}, but goes from bottom to top, and returns an iterator, not a list.

• #​11850: Added support for sys.last_exc{.interpreted-text role="data"} for post-mortem debugging on Python>=3.12.

• #​11962: In case no other suitable candidates for configuration file are found, a pyproject.toml (even without a [tool.pytest.ini_options] table) will be considered as the configuration file and define the rootdir.

• #​11978: Add --log-file-mode option to the logging plugin, enabling appending to log-files. This option accepts either "w" or "a" and defaults to "w".

Previously, the mode was hard-coded to be "w" which truncates the file before logging.

• #​12047: When multiple finalizers of a fixture raise an exception, now all exceptions are reported as an exception group. Previously, only the first exception was reported.

Bug Fixes

• #​11904: Fixed a regression in pytest 8.0.0 that would cause test collection to fail due to permission errors when using --pyargs.

This change improves the collection tree for tests specified using --pyargs, see 12043{.interpreted-text role="pull"} for a comparison with pytest 8.0 and <8.

• #​12011: Fixed a regression in 8.0.1 whereby setup_module xunit-style fixtures are not executed when --doctest-modules is passed.

• #​12014: Fix the stacklevel used when warning about marks used on fixtures.

• #​12039: Fixed a regression in 8.0.2 where tests created using tmp_path{.interpreted-text role="fixture"} have been collected multiple times in CI under Windows.

Improved Documentation

• #​11790: Documented the retention of temporary directories created using the tmp_path fixture in more detail.

Trivial/Internal Changes

• #​11785: Some changes were made to private functions which may affect plugins which access them:
• FixtureManager._getautousenames() now takes a Node itself instead of the nodeid.
• FixtureManager.getfixturedefs() now takes the Node itself instead of the nodeid.
• The _pytest.nodes.iterparentnodeids() function is removed without replacement. Prefer to traverse the node hierarchy itself instead.

If you really need to, copy the function from the previous pytest release.

v8.0.2

Compare Source

pytest 8.0.2 (2024-02-24)

Bug Fixes

• #​11895: Fix collection on Windows where initial paths contain the short version of a path (for example c:\PROGRA~1\tests).
• #​11953: Fix an IndexError crash raising from getstatementrange_ast.
• #​12021: Reverted a fix to [--maxfail]{.title-ref} handling in pytest 8.0.0 because it caused a regression in pytest-xdist whereby session fixture teardowns may get executed multiple times when the max-fails is reached.

v8.0.1

Compare Source

pytest 8.0.1 (2024-02-16)

Bug Fixes

• #​11875: Correctly handle errors from getpass.getuser{.interpreted-text role="func"} in Python 3.13.
• #​11879: Fix an edge case where ExceptionInfo._stringify_exception could crash pytest.raises{.interpreted-text role="func"}.
• #​11906: Fix regression with pytest.warns{.interpreted-text role="func"} using custom warning subclasses which have more than one parameter in their [__init__]{.title-ref}.
• #​11907: Fix a regression in pytest 8.0.0 whereby calling pytest.skip{.interpreted-text role="func"} and similar control-flow exceptions within a pytest.warns(){.interpreted-text role="func"} block would get suppressed instead of propagating.
• #​11929: Fix a regression in pytest 8.0.0 whereby autouse fixtures defined in a module get ignored by the doctests in the module.
• #​11937: Fix a regression in pytest 8.0.0 whereby items would be collected in reverse order in some circumstances.

v0.23.5.post1: pytest-asyncio 0.23.5.post1

Compare Source

0.23.5 (2024-02-09)

• Declare compatibility with pytest 8 #​737
• Fix typing errors with recent versions of mypy #​769
• Prevent DeprecationWarning about internal use of asyncio.get_event_loop() from affecting test cases #​757

Known issues

As of v0.23, pytest-asyncio attaches an asyncio event loop to each item of the test suite (i.e. session, packages, modules, classes, functions) and allows tests to be run in those loops when marked accordingly. Pytest-asyncio currently assumes that async fixture scope is correlated with the new event loop scope. This prevents fixtures from being evaluated independently from the event loop scope and breaks some existing test suites (see #​706). For example, a test suite may require all fixtures and tests to run in the same event loop, but have async fixtures that are set up and torn down for each module. If you're affected by this issue, please continue using the v0.21 release, until it is resolved.

v0.12.1

Compare Source

• Fixed compatibility with upcoming pytest 8.1.x. (#125_).

.. _#125:https://github.com/pytest-dev/pytest-subtests/issues/1255

v0.12.0

Compare Source

• Python 3.12 is now officially supported (#113_).
• Added typing support (#115_).
• SubTests can be imported from pytest_subtests to type-annotate the subtests fixture.

.. _#113:https://github.com/pytest-dev/pytest-subtests/pull/1133 .. _#115https://github.com/pytest-dev/pytest-subtests/pull/11515

v1.9.0

Compare Source