Michael Reiche

edit: getTypeToRead() is returning null even though it has enough information to determine the return type. I'm still investigating. ``` 124: private Object execute(ParametersParameterAccessor parameterAccessor) { Class typeToRead = processor.getReturnedType().getTypeToRead();...

Name needs to be a class. This will suffice: ``` class Name(var name: String) { } ``` It seems that some better diagnostics would help.

Yes. An id is(was) always expected for an entity. Also cas. It doesn't matter what it is though. You can have "" as __id, 0 as __cas. There is a...

> I also need to add an __id to my query even if I don't need it. Or you could just use the newer version that doesn't require it.

My mistake - the id is always required when anything other than one simple field is projected. So just project "" as __id.

Your analysis is correct. The cas of the document after the transaction is complete is different than the cas used to update the @Version in the save() call inside the...

> This is the desired behavior? More important than "desired" is "Does it guarantee that the transactional properties (ACID) are always satisfied?" And that's what I've asked the transaction developer....

The short explanation is that replace()/remove() of an entity object inside a transaction that was obtained outside that transaction is susceptible to write-write conflict from an update outside of that...

Following the discussion - the document replaced in a transaction needs to have had in that same transaction an operation that obtained the cas (insert/replace/get). A cas from outside the...

see https://github.com/spring-projects/spring-data-examples/issues/646