fix: cache in item instead of entity #470
Conversation
this change also makes this the default for a few other funtions, namely sliders. fixes resoai#425 BREAKING CHANGE: The order of attrs and def is changed for a few sliders. This should not normally be a problem, because user would only use one of the possibilities. But... just to give note on that.
This fixes resoai#464 The solution uses the new nullis coalescing operator `??`. It's just like `||`, but it allows `0` as output. Babel traspiles this just fine. I had to update ESLINT config to use ECMA 2020. Not sure, what are the side effects. @rchl
That's fine. It only affects linting. Did you mean to include all those changes in this PR? |
It was kind of the same "issue" for me. As you can see, I did the first version of most of the changes in the very first commit... We could split in three, but it will be some work:
Do you prefer that? |
|
So... I would be fine with multiple semi-related commits but then I'd avoid squashing all commits together on merging so that they are listed separately in changelog and history. But in this case, those separate commits are not really arranged or formatted properly as you've added something in one commit to remove or change it later. And some commits don't have a summary that follows conventional commit convention. Basically, when looking at individual commits from oldest to latest, each should be a self-contained change that can work on its own. So if you think cleaning that up is too much work, then I'd rather squash the changes but I wouldn't want to squash those two separate fixes together so I'd ask you to at least separate it into two commits, one for changing the |
|
OK! I'll re-open new PRs for that. |
fixes #464
fixes #425
@rchl , I have changed
eslintto accept the??operator as of ECMA Script 2020. It transpiles just fine. But I am not aware of the issues this might cause on the rest of the linting.Explanation of that operator:
https://github.com/tc39/proposal-nullish-coalescing
PS: Regarding the breaking change below, we should take a short time and think about which order is more suitable
def ?? attrsorattrs ?? def. I tend to think now thatdefshould have priority, because it is user-defined.BREAKING CHANGE: The order of
attrsanddefis changed for a few sliders. This should not normally be a problem, because the user would only use one of the possibilities. But... just to give note on that.