Tyler Rick

A workaround for now is to just lock to: ```ruby gem 'tzinfo', '1.2.5' ``` ... but eventually we'll be forced to upgrade tzinfo, so we'll need an actual solution. Has...

Ooh, that would be fun to have. I've wished for something nice like this, too. I know we have `Trailblazer::Developer.render` now (not sure if it existed when you opened this?):...

Note that the problem goes away if you change it to use the `Field` from `'react-final-form'` instead.

As a workaround, you can set `key` to the same thing as you set `required` prop to. But we shouldn't have to resort to this, should we??

Oh dear, so this bug has been around since March? I just ran into this as well and was about to submit a new issue when I saw this existing...

Have only changed it for Rails 5.1 so far. I can update the other patches too if there is interest and if this is likely to get merged.

The reason I wanted this feature, in case it wasn't clear, was because: 1. In general, we should always try to use the official/standard Rails behavior as much as possible,...

Great idea, @bboe. If I'm understand you correctly, you would prefer to configure how the STI class is stored directly within each model, instead of some global configuration done in...

Actually, it looks like the hard-coded use of `base_class` for destroy versions is not a new thing. That behavior has apparently been around since 2011 (https://github.com/paper-trail-gem/paper_trail/commit/f51eb08c13e91a9dbf95b3a9382927afeb48b1a4). So I'm not sure...

Thanks for chiming in, @lorint. I've been watching your PR(s) with great interest to see how it all shakes out. I even tried out your branch for a while. (Ended...