Fix Notice: Undefined index: import in EmailBuilderManager->getImportInfo()

Thanks for the testing and bug report. My enthusiasm for this project has been boosted by seeing how many sites are using the module already.

I don't see the warning you report. I think this patch shouldn't be needed because there is code in Drupal\symfony_mailer\Annotation\EmailBuilder to initialise $import = '';.

Please could you try clearing caches and see if that fixes it?

I can confirm that I also needed this patch.

I updated a site from Swift mailer to Symphony mailer, following https://www.drupal.org/docs/contributed-modules/symfony-mailer-0/getting....

On the first "import configuration", there was nothing to import.
On the second "import configuration", the error occured.

Drupal 9.3.9

Did you both try clearing caches?

Please could you try and get some debug output with code something like this:

diff --git a/src/Processor/EmailBuilderManager.php b/src/Processor/EmailBuilderManager.php
index 842f521..3b5b5fe 100644
--- a/src/Processor/EmailBuilderManager.php
+++ b/src/Processor/EmailBuilderManager.php
@@ -100,6 +100,7 @@ class EmailBuilderManager extends DefaultPluginManager implements EmailBuilderMa
     $state_all = $this->keyValue->get('import', []);
 
     foreach ($this->getDefinitions() as $id => $definition) {
+      ksm($definition);
       if ($definition['import']) {
         $state = $state_all[$id] ?? self::IMPORT_READY;

Ik did clear the cache multiple times. That normally solves most of Drupal's problems ;-).

I moved away from this module because after testing, i stumbled upon the unability to mail attachments with webform. So unfortunately I cannot provide more info at this moment. Maybe I will revisit this when attachments are supported. I see a lot of activity in that issue.

After updating to the latest version of Symfony mailer 1.0.0-alpha7 and applying the patch. I don't get any error's and I've tested my mailing and works like it should.

Thanks for the updates. The problem I have is that I can't reproduce the error, and I can't see why any change is needed, as explained in #5. In #7 I asked if someone could provide some extra debugging. Any help with this would be welcome.

Ha and now I hit it myself😃.

I added a new patch that I prefer because it goes to the source of the problem by fixing the missing array entry (and any other possible future missing entry). Previous patch just worked around the missing entry.

I would appreciate any testing.