I have given more thought to this PR and I would like two things to be discussed.

1. Calculating the amount of seconds between two dates is apparently easy, but if we are dealing with a timezone with a daylight saving time period, a +/- 1 hour difference will appear if the availableAt DateTime changes between daylight saving period and non-daylight saving period. I am no expert managing date and time functions, but could this be a solution to the problem (basically, comparing dates in UTC timezone)?

            $timezone = new DateTimeZone('UTC');

            $availableAt = new DateTime(
                $availableAtStamp
                    ->getAvailableAt()
                    ->format('Y-m-d H:i:s'),
                $timezone
            );

            $now = new DateTime(
                (new DateTime())
                    ->format('Y-m-d H:i:s'),
                $timezone
            );

            $delay = $availableAt->getTimestamp() - $now->getTimestamp();

2. If an AvailableAtStamp is included, the default Serializer created by the Serializer Transport will now fail when trying to denormalize the DateTime object. The Symfony\Component\Serializer\Normalizer\DateTimeNormalizer should be added in the public static function create(): self method of the Serializer Transport, especially if this stamp is provided as a core function in the Messenger component. Would this be fine?

What's the status here?

It looks like there is an unrelated commit.

Can't we add a DelayStamp::to(\DateTime $datetime): DelayStamp helper instead of a new stamp+middleware?

Can't we add a DelayStamp::to(\DateTime $datetime): DelayStamp helper instead of a new stamp+middleware?

@ogizanagi I agree, it could be a really good idea

I like @ogizanagi idea.

Can't we add a DelayStamp::to(\DateTime $datetime): DelayStamp helper instead of a new stamp+middleware?

I was thinking the same! Maybe DelayStamp::delayUntil()?

We've just moved away from master as the main branch to use 5.x instead. Unfortunately, I cannot reopen the PR and change the target branch to 5.x. Can you open a new PR referencing this one to not loose the discussion? Thank you for your understanding and for your help.

As linked by @maxhelias , I've just proposed the exact same thing in #38459 (sorry for the duplication)...

Please, read my issue as there are some useful inspiration, I think...
Please, read also this comment, because it has some details that maybe of inspiration: #38463 (comment)

I've send a new PR: #38480