Can't run cim: Argument #2 ($data) must be of type array, bool given

changing like 184 of the Event Sub to this:

if ($ignored === TRUE && !empty($destination_data)) {

fixes this. I'll submit a patch even though it seems that project maintainer may not be interested in fixing this.

why would you assume that I would not be interested in fixing a bug?

Please make sure to include all possible information to reproduce the bug. Ideally with a test case of course, but just detailed instructions for how to reproduce this bug is already helpful in deciding what code to change to fix it.

I think i suggested no interest as there was already an issue posted about this, that i think had been closed.. but maybe I am mixing it up with something else.

And not sure how to reproduce. I suspect it is from config which exists in db that there is no .yml in sync folder for. Which, typically with cim, would simply be removed; but i think with config ignore, it throws an error. With the module disabled i think cim works as expected.

maybe there was a similar issue, I closed all 2.x issues with the same message without reading what the issue was about and checking if it could still be an issue:

This issue is being closed because Config Ignore 2.x is deprecated.

The new 3,x version has been re-written from the ground up and works in a totally different way, yet it provides the same backwards compatible functionality. All the 2.x tests pass in 3.x and the 3.x codebase is easier to maintain.
Users of 2.x are strongly encouraged to upgrade to 3.x, as there is a very simple update path.

So likely this issue is no longer relevant. If you think that this issue still applies please open it again and target the new branch. All issues on the 2.x branch were closed in bulk.

For example #3150307-12: On config export / config page: Exception "Argument 1 passed to Drupal\Component\Utility\NestedArray::setValue() must be of the type array, bool given"" →

But without being able to reproduce the issue, I am not sure we can really do anything and be sure that it is the correct thing.

PS: patches are no longer tested, only merge requests via gitlab.

After Update to Drupal core from 10.2 to 10.3, we are also facing the same issue.

just saying that you experience this without also telling me how is not going to move this forward.

I would like to know how this can happen so that we can test for it and assert that we deal with it correctly.

Hello. I also reproduced the issue. I don't know why you are arguing about how to reproduce, when it is clear....
The error is saying that you can not pass boolean to write method in the second parameter:
$transformation_storage->write($config_name, $destination_data);
One line above is stored the data of $destination_data
$destination_data = $destination_storage->read($config_name);

When you read what can be returned from read's method annotation:
* @return array|false
* The configuration data stored for the configuration object name. If no
* configuration data exists for the given name, FALSE is returned.

So here it is the issue.....

Hello nikolaat.
You see, the thing you say is technically true, except for the conclusion.
line 183 $destination_data = $destination_storage->read($config_name); could return false of course.
But you are ignoring that $config_name comes from line 165 foreach ($destination_storage->listAll() as $config_name)

and that reads:

  /**
   * Gets configuration object names starting with a given prefix.
   *
   * Given the following configuration objects:
   * - node.type.article
   * - node.type.page
   *
   * Passing the prefix 'node.type.' will return an array containing the above
   * names.
   *
   * @param string $prefix
   *   (optional) The prefix to search for. If omitted, all configuration object
   *   names that exist are returned.
   *
   * @return array
   *   An array containing matching configuration object names.
   */
  public function listAll($prefix = '');

So the storage first said that the config with a certain name is in it but then returns false?
How do you do that? That is maybe not a bug in config ignore, therefore, I want people like you who experience this to provide more information and that is how this issue is moving forward.