[dependabot]

Bumps phpunit/phpunit from 9.6.29 to 13.4.5.

Release notes

Sourced from phpunit/phpunit's releases.

PHPUnit 13.4.5

Fixed

• The static analysis cache directory that is automatically created in the operating system's path used for temporary files when no cache directory is configured is not used when the POSIX extension is not loaded

---

Learn how to install or update PHPUnit 13.4 in the documentation.

Keep up to date with PHPUnit:

• You can follow @​phpunit@phpc.social to stay up to date with PHPUnit's development.
• You can subscribe to the PHPUnit Updates newsletter to receive updates about and tips for PHPUnit.

PHPUnit 13.4.4

Fixed

• The TeamCity logger writes testSuiteFinished twice for a test class when one of its before-first-test methods errors or fails
• The TestDox output shows a less important outcome of a test that has more than one outcome, for instance the failure in tearDown() of a test that errored
• When the test run is interrupted by a signal while a test is run in process isolation, the result of the test is discarded and the test is counted as passed, and the TestDox output cannot be printed
• When no static analysis cache directory is configured, the cache directory that is automatically created in the operating system's path used for temporary files is also used when it was created by another user or can be written to by other users
• The directory in the operating system's path used for temporary files in which the temporary files for collecting code coverage for a PHPT test are created has a fixed name, and is also used when it was created by another user, who can then replace the code of the PHPT test before it is executed

---

Learn how to install or update PHPUnit 13.4 in the documentation.

Keep up to date with PHPUnit:

• You can follow @​phpunit@phpc.social to stay up to date with PHPUnit's development.
• You can subscribe to the PHPUnit Updates newsletter to receive updates about and tips for PHPUnit.

PHPUnit 13.4.3

Fixed

• The dependencies of a test that are listed after a dependency on a test that does not return a value, or after a dependency on a larger test, were not checked, so the test was run instead of being skipped when one of these dependencies had not passed
• Tests could not be run in process isolation when the namespace of the test class begins with a name that the code of the child process imports, for instance CodeCoverage or Facade
• A test that is run in process isolation aborts the test run when its data set or a value returned by a test it depends on cannot be serialized, for instance because it is a closure
• When a before-first-test method or an after-last-test method errors or fails, the remaining methods of the same kind are still called, and each of them that does not error or fail itself is reported with the exception of the earlier method
• When a hook method of a test, for instance a #[Before] method, skips the test, the remaining hook methods of the same kind are still called, so the test is reported as errored instead of skipped when one of them errors
• An error or failure in a before-first-test method or an after-last-test method that is called again in the child process of a test that is run in process isolation is reported twice, once for the test class and once for the test
• When a before-first-test method errors, the progress output shows one E for the test class instead of one for each of its tests, only one of its tests is counted as run, and the test that is run next is shown as an error
• When a before-first-test method fails, the progress output shows nothing for the tests of the test class and only one of them is counted as run
• The Open Test Reporting XML logfile has no start for a test that errors before it is prepared, for instance because it depends on a test that does not exist, and no end for the test that is run after a test that is skipped because a test it depends on did not pass or because one of its requirements is not met
• The Open Test Reporting XML logfile has no entry for a test that is marked as incomplete before it is prepared, for instance in setUp(), and reports the test that is run after it as aborted with the message of the incomplete test
• The Open Test Reporting XML logfile has no end for the test that is run after a test whose preparation errors, for instance because setUp() raises an exception or the global state cannot be backed up
• The Open Test Reporting XML logfile reports a test that has more than one outcome, for instance because tearDown() fails after the test errored, with its last outcome instead of its most severe one, and with more than one end when the preparation of the test errored
• The Open Test Reporting XML logfile is not well-formed when a test name, a message, the output of a test, or the string representation of an exception contains ]]>, bytes that are not valid UTF-8, or characters that are not allowed in XML, for instance the escape character of an ANSI escape sequence

... (truncated)

Changelog

Sourced from phpunit/phpunit's changelog.

[13.4.5] - 2026-10-27

Fixed

• The static analysis cache directory that is automatically created in the operating system's path used for temporary files when no cache directory is configured is not used when the POSIX extension is not loaded

[13.4.4] - 2026-10-27

Fixed

• The TeamCity logger writes testSuiteFinished twice for a test class when one of its before-first-test methods errors or fails
• The TestDox output shows a less important outcome of a test that has more than one outcome, for instance the failure in tearDown() of a test that errored
• When the test run is interrupted by a signal while a test is run in process isolation, the result of the test is discarded and the test is counted as passed, and the TestDox output cannot be printed
• When no static analysis cache directory is configured, the cache directory that is automatically created in the operating system's path used for temporary files is also used when it was created by another user or can be written to by other users
• The directory in the operating system's path used for temporary files in which the temporary files for collecting code coverage for a PHPT test are created has a fixed name, and is also used when it was created by another user, who can then replace the code of the PHPT test before it is executed

[13.4.3] - 2026-10-16

Fixed

• The dependencies of a test that are listed after a dependency on a test that does not return a value, or after a dependency on a larger test, were not checked, so the test was run instead of being skipped when one of these dependencies had not passed
• Tests could not be run in process isolation when the namespace of the test class begins with a name that the code of the child process imports, for instance CodeCoverage or Facade
• A test that is run in process isolation aborts the test run when its data set or a value returned by a test it depends on cannot be serialized, for instance because it is a closure
• When a before-first-test method or an after-last-test method errors or fails, the remaining methods of the same kind are still called, and each of them that does not error or fail itself is reported with the exception of the earlier method
• When a hook method of a test, for instance a #[Before] method, skips the test, the remaining hook methods of the same kind are still called, so the test is reported as errored instead of skipped when one of them errors
• An error or failure in a before-first-test method or an after-last-test method that is called again in the child process of a test that is run in process isolation is reported twice, once for the test class and once for the test
• When a before-first-test method errors, the progress output shows one E for the test class instead of one for each of its tests, only one of its tests is counted as run, and the test that is run next is shown as an error
• When a before-first-test method fails, the progress output shows nothing for the tests of the test class and only one of them is counted as run
• The Open Test Reporting XML logfile has no start for a test that errors before it is prepared, for instance because it depends on a test that does not exist, and no end for the test that is run after a test that is skipped because a test it depends on did not pass or because one of its requirements is not met
• The Open Test Reporting XML logfile has no entry for a test that is marked as incomplete before it is prepared, for instance in setUp(), and reports the test that is run after it as aborted with the message of the incomplete test
• The Open Test Reporting XML logfile has no end for the test that is run after a test whose preparation errors, for instance because setUp() raises an exception or the global state cannot be backed up
• The Open Test Reporting XML logfile reports a test that has more than one outcome, for instance because tearDown() fails after the test errored, with its last outcome instead of its most severe one, and with more than one end when the preparation of the test errored
• The Open Test Reporting XML logfile is not well-formed when a test name, a message, the output of a test, or the string representation of an exception contains ]]>, bytes that are not valid UTF-8, or characters that are not allowed in XML, for instance the escape character of an ANSI escape sequence
• The Open Test Reporting XML logfile reports an issue that is triggered before a test is prepared, for instance in setUp(), with an empty ID before the start of the test

[13.4.2] - 2026-10-12

Fixed

• #7033: Compact printer is created although an extension replaces the progress output
• #7035: Sorting tests by duration recomputes the duration of a test suite for every comparison
• Sorting tests by size recomputes the size of a test suite for every comparison
• expectUserDeprecationMessage() and expectUserDeprecationMessageMatches() did not see the deprecations triggered by an attempt of a retried test and saw those of the previously run test instead
• The status of a test whose mock object raises an exception in its destructor is success although the test is reported as errored, so #[Retry] does not retry the test, --repeat does not skip the remaining repetitions, tearDown() sees the wrong status, and code coverage records the test as successful
• Named arguments passed to with() were matched by position instead of by name, so an expectation such as ->with(b: 7) compared the expected value with the first parameter of the method and the parameter named $b was not checked
• String keys of the parameter sets passed to withParameterSetsInOrder(), withParameterSetsInAnyOrder(), and withParameterSetsInPartialOrder() were ignored, so a parameter set such as ['b' => 7] compared the expected value with the first parameter of the method and the parameter named $b was not checked

... (truncated)

Commits

• 941457a Prepare release
• 2080b37 Merge branch '12.5' into 13.4
• 1d6dbe7 Prepare release
• ef98204 Merge branch '12.5' into 13.4
• 15fb431 Do not require the POSIX extension for using the automatically created static...
• 71b681a Prepare release
• c056074 Merge branch '12.5' into 13.4
• 191b2e0 Prepare release
• ae31c78 Create the temporary files for collecting code coverage for a PHPT test in a ...
• c4c4f06 Merge branch '12.5' into 13.4
• Additional commits viewable in compare view

Dependabot will resolve any conflicts with this PR as long as you don't alter it yourself. You can also trigger a rebase manually by commenting @dependabot rebase.

---

Dependabot commands and options

You can trigger Dependabot actions by commenting on this PR:

• @dependabot rebase will rebase this PR
• @dependabot recreate will recreate this PR, overwriting any edits that have been made to it
• @dependabot show <dependency name> ignore conditions will show all of the ignore conditions of the specified dependency
• @dependabot ignore this major version will close this PR and stop Dependabot creating any more for this major version (unless you reopen the PR or upgrade to it yourself)
• @dependabot ignore this minor version will close this PR and stop Dependabot creating any more for this minor version (unless you reopen the PR or upgrade to it yourself)
• @dependabot ignore this dependency will close this PR and stop Dependabot creating any more for this dependency (unless you reopen the PR or upgrade to it yourself)

[dependabot]

Labels

The following labels could not be found: Skip-Changelog, dependencies. Please create them before Dependabot can add them to a pull request.

Please fix the above issues or remove invalid values from dependabot.yml.