[Paragraphs] Email handler sends "_default" in body when embedding webform in Paragraph

I am having the same problem and the patch fixes it for me. It could also be fixed by comparing to <p>_default</p> or <p>static::DEFAULT_VALUE</p> instead of using strip_tags().

It is also fixed by the work-around described in #5.

The webform does not need to be embedded in a paragraph for this to happen.

Hi everyone,

I stumbled upon this issue because of a problem one of our customers has when translating a webform.

When having an email handler in the source webform, the body of this email handler's email is displayed as a richtext field, which, by default, encapsulates the "_default" string in

tags, which later breaks the sending of the emails:

Also experiencing this issue.

We are not using paragraphs, so nothing to do with paragraphs specifically
We are also not using any custom text format for the elements/emails, everything is default webform.
I checked all the settings for emails, formats and the dedault webform ckeditor profile: everything is exactly as the default webform config provides it after installation.

Mail body texts are translated with a ckeditor, and the "_default" text is wrapped in p-tags. So all translated emails are wrong for us, unless we specifically translate them, or remove the p-tags in the source code of the ckeditor.

Provided patch here, fixes our issue.

I don't like working around the root cause, which is when translating the email body when it is set to default, it should not even display the 'default' value or the HTML editor.

The attached patch goes after the root cause.

The issue is so significant it is very unlikely that no one has allowed the broke configuration to remain, I am not sure we need any update hook to fix a bug in beta software.

The last submitted patch, 13: 3304518-13.patch, failed testing. View results →

This is the patch in #13 plus just enough test changes to make it pass. This is not to commit but to show what tests are failing.

We shouldn't comment out the tests. The selectors need to be updated.

Yes, that patch is not to be committed. That is why I left the issue as "Needs work". The patch is there to demonstrate what tests are not passing and show that the rest of the tests do pass.

@Liam Thanks

If the tests pass this should be RTBC.

Automatically closed - issue fixed for 2 weeks with no activity.