Does Client Control mess up history on Spark 2.8.3?

In the release notes, it says that it, “requires Spark version 2.9.0 and above for proper functionality.” Does that mean that it will cause Spark 2.8.3 to function improperly, or just that users will need 2.9.0 in order to access the new features? It would seem odd to me that an update would be released for the plugin that requires a not-yet-released build of Spark, but the release notes are worded differently than the last two updates were worded (“requires Spark version 2.8.X and above for new functionality”), so I want to be sure I’m not going to break anything before applying the update.

It means that 2.9.0 is needed for the new functionality. It shouldn’t break anything. But it also won’t add anything for Spark 2.8.3. So you can also not update it. It is weird, but that’s just how things go. Spark is lagging behind because of a few bugs that nobody wants/know how to fix, but the changes for the plugin were already submitted, so why not applying them. Otherwise the patch might get lost or become incompatible waiting for Spark to be updated…

That’s good to know. I know that we’ll end up updating to Spark 2.9.0 once it comes out on some PCs but leaving others on 2.8.3 and didn’t want those users complaining about us breaking things that were “working perfectly fine before.”