Review main Abandoned

93744[BUGFIX] Dump class loading information on extension install

Benjamin Kott CR 0 V +1 Normal +129 −4 updated 04/20/26, 10:41

Commit message

Installing an extension via the Extension Manager in classic
(non-Composer) mode does not rebuild the class loading files in
typo3conf/autoload/. The newly installed extension's classes therefore
remain unknown to the autoloader until another trigger re-dumps them.

DispatchAfterPackageActivationEventOnPackageInitialization previously
obtained its event dispatcher from the container carried by
PackageInitializationEvent. Since no container is attached to that
event, the null guard short-circuits and the follow-up
AfterPackageActivationEvent is never emitted, so
ClassLoadingInformationUpdater never runs.

The listener now receives EventDispatcherInterface via constructor
injection and dispatches AfterPackageActivationEvent directly. As the
listener is resolved from the same container that dispatches the
incoming event, its injected dispatcher is the correct one for
downstream listeners such as ClassLoadingInformationUpdater.

A functional test for InstallUtility covers the scenario by activating
a fixture extension and asserting that autoload_classmap.php and
autoload_psr4.php are written and contain the fixture's class and
PSR-4 namespace after install().

Review

  • Code review CR 0
  • Verified V +1
    • Oliver Klee +1
    • core-ci +1 CI

Issues 1

Files 5

Review log 69

  1. Benjamin Kott uploaded patch set 1 04/20/26, 09:40

  2. Benjamin Kott updated the commit message in patch set 2 04/20/26, 09:41

  3. core-ci voted on patch set 1 V -1 04/20/26, 09:56 Core CI is not happy: https://git.typo3.org/typo3/CI/cms/-/pipelines/100540

  4. Benjamin Kott uploaded patch set 3 04/20/26, 09:59

  5. core-ci voted on patch set 2 V -1 04/20/26, 10:00 Core CI is not happy: https://git.typo3.org/typo3/CI/cms/-/pipelines/100542

  6. core-ci voted on patch set 3 V +1 04/20/26, 10:09 Core CI is happy: https://git.typo3.org/typo3/CI/cms/-/pipelines/100544

  7. Oliver Klee voted on patch set 3 V +1 04/20/26, 10:10

  8. Helmut Hummel commented on patch set 3 04/20/26, 10:39

    @benjamin.kott@outlook.com Thanks for letting me know this change exists and for making me actually write a test for this
  9. Helmut Hummel abandoned the change 04/20/26, 10:41