Jon Pretty

I've been flattered by all the #ScalaThankYou messages this week, but they've revealed to me that there are many more people than I realised quietly using Magnolia for generic derivation! If you're using it and I might not know you are, please let me know so I can compile a list!

Equally, if you're *not* using Magnolia for generic derivation, I'd be interested to know that too, especially if you tried it and it didn't work out for some reason: I'd like to do my best to remove those reasons, if I can!

Happy user in Caliban! There's only 1 thing that could be improved: error reporting. When there are many chained implicits involved (deeply nested case classes), errors get cryptic. Ideally it would tell exactly which deeply nested type is missing an instance.

Have you seen the derivation "stack traces", like the one 2/3 of the way down this page: https://propensive.com/opensource/magnolia … ? These don't appear if you're deriving implicitly, but if you explicitly request a derivation (e.g. something like Type.gen[Type]), you should see it.

Yeah, my advice in case of errors is to add the derivation explicitly but most users wish it would work out without any boilerplate

If I remember correctly, the problem was that it resulted in lots of false positives when Magnolia-derived implicits were considered for types which were later satisfied by different implicits.