Adam Retter

@lschult2 I have added your `test.xlsx` to a Unit Test in - https://github.com/eXist-db/exist/pull/4168 I spent some time looking into your issue, unfortunately it isn't a simple one to trace or...

> I do consider the switch from implicit use-last to use-first a breaking change. @line-o It's a bugfix, but I agree it is also likely a breaking change for people...

Thanks @brihaye I have fixed that now.

@joewiz eXist-db 5.2.0 is still very lax about enforcing return types. It also doesn't cope well with compound types, e.g. `map(xs:string, map(xs:string, xs:int)`, I think I fixed the `document-node(element(x))` compound...

@line-o Once this is rebased, we could merge this. In future when creating a backport, can you do so as a 1-1 mapping between PRs please. i.e. if you want...

I don't think this is the same as #3953 at all as this is a separate issue to CI. 1. `mvn release:perform` should never have updated the `latest` tag -...

Okay good to see that `release` is now correct. I did check Docker Hub after I did the release last night and it was showing as not updated at that...

@duncdrum Okay, but I don't think the way it works now makes sense. If whenever an `mvn release:perform` is done, the Docker `latest` tag is pointed at a fixed release,...

@lcahlander I think it depends... If your drivers are JDBC 4 compliant then I don't think you actually need the `Class.forName` even in a Xar. I might be wrong though......

> Does it hurt anything to do the call? I don't know, but my preference would be not to add it if we don't need it. Can you tell me...