Fatal error: ForwardLinkBlock assumes a 'node' route parameter is always a node object

Thanks Yahya!

On closer inspection, I think it would be much better to implement the method described in this blog post. That way we lose the hard-coded references to 4 specific entity types, and it's less code :-) I guess the function could look like this (warning, untested code):

public function isAllowed() {
  $allowed = FALSE;

  $parameters = $this->routeMatch->getParameters();
  // Find the first parameter that is an entity.
  foreach ($parameters as $parameter) {
    if ($parameter instanceof EntityInterface) {
      $entity = $parameter;
      $this->entity = $entity;
      $allowed = $this->accessChecker->isAllowed($this->settings, $entity, 'full', $entity->getEntityTypeId(), $entity->bundle());
      break;
    }
  }

  return $allowed;
}

PS. You don't need to do $bundle = $entity->bundle() == $entity->getEntityTypeId() ? "" : $entity->bundle();, because the documentation of EntityInterface::bundle() says: "@return string - The bundle of the entity. Defaults to the entity type ID if the entity type does not make use of different bundles."

I guess I'm still not that confident to change other's code.

Please don't worry too much about that. Other developers may look confident (some are :-)) but often we're just putting ideas out there to collect feedback. It's the fastest way to move the Drupal project and your own learning forward.

Thanks for your contribution! That's a much bigger patch than the first one. While that is not necessarily a problem, it seems that you're trying to solve more issues than just the fatal error this issue started with. Creating a separate issue for each improvement (fix fatal error / create base class / coding standards / etc.) may make it easier for the maintainer to review and commit your patches.

Thanks @YahyaAlHamad, that patch looks very good, although I must add I haven't tested it yet.

You are right that some code is duplicated and abstracting it into a base class would solve that. You don't need to apologize! However in my experience, patches are more likely to be accepted when they only try to solve one problem. That way, if a discussion arises about the newly introduced base class, it will not stop the fatal error from being fixed.