How/Should perennial be able to use code from phet-core, like IntentionalAny?

[samreid]

How/Should perennial be able to use code from phet-core, like IntentionalAny? Keep in mind we do not want to create phet-core-alias.

[samreid]

I am reluctant to move more code to perennial while we have this problem: #363

[samreid]

@jonathanolson @zepumph and I met today to discuss #363 and agreed that #401 would be good to investigate.

We do not want to move phet-core code such as optionize or IntentionalAny to perennial. Instead, we would prefer that perennial can depend on phet-core. In order to do so without creating phet-core-alias, we will investigate the above approaches.

Until then, perennial should use any instead of IntentionalAny and assignIn or other alternatives to optionize.

[zepumph]

We cannot justify changes to our pattern for perennial-alias over in #363.

Given that, here are solutions we don't want to do:

• phet-core-alias: This won't even work, because you need a different string in the import depending on if it is perennial or perennial-alias. This could work if they were dynamic imports, but seriously. Ew.
• Copy/move all files we want from phet-core over to perennial. Gross!
• Publish an npm module of phet-core so that perennial can use pinned version of it as it changes. No way bronco.

And so, @samreid and I regret to inform you, reader, that it is time to close this issue. We will continue with perennial unable to use phet-core or anything else as a dependency. Closing.

[phet-dev]

Reopening because there is a TODO marked for this issue.

[samreid]

I'm glad this issue reopened, thanks @phet-dev. I was working on #404 and it was feeling unfortunate that perennial cannot call code in chipper. I believe if we can demarcate sha operations in #402 then we can allow our non-sha operation code to depend on chipper and phet-core, etc. So I'd like to take a look.

[samreid]

@zepumph and I agreed with #371 (comment)