Add an updated proposal for DOM Parts.

[rniwa]

No description provided.

[rniwa]

@annevk @justinfagnani @hober @mfreed7

[justinfagnani]

cc @yuzhe-han

[rniwa]

@justinfagnani @yuzhe-han : would you like me to add acknowledgment for your involvement? I wasn't sure if you'd like to put your names on it or not.

[yuzhe-han]

@rniwa : sure. I enjoyed our discussions and look forward to collaborating in the future. Thanks.

[justinfagnani]

That would be great, thanks!

[annevk]

I still need to read more, but this looks interesting and seems like a good iteration given the previous feedback of the approach being too high-level. Once thing that's not entirely clear to me is what the internal model of a part is and why having it is beneficial other than developer ergonomics. Does this enable any performance wins over mutating the tree yourself, for instance?

[rniwa]

I still need to read more, but this looks interesting and seems like a good iteration given the previous feedback of the approach being too high-level. Once thing that's not entirely clear to me is what the internal model of a part is and why having it is beneficial other than developer ergonomics. Does this enable any performance wins over mutating the tree yourself, for instance?

Performance improvement is possible if we introduce some kind of batching.

[rniwa]

@justinfagnani @hober @mfreed7 @yuzhe-han : Added the acknowledgement. Let's merge this so that people can take a look before tomorrow's meeting.

[justinfagnani]

Once thing that's not entirely clear to me is what the internal model of a part is and why having it is beneficial other than developer ergonomics. Does this enable any performance wins over mutating the tree yourself, for instance?

Another big win for template libraries and hand-written DOM code is the cloning behavior. This would be the first way to clone some kind of reference to nodes, along with a subtree.

[rniwa]

Can someone approve this PR?

[leobalter]

I'm interested but I'm trying to fit in time this afternoon to read it.

[annevk]

If people want to read it they can do so at https://github.com/w3c/webcomponents/blob/405d041d8346355d11c1f1d426407488e7c16820/proposals/DOM-Parts.md. It does seem good to address the issues @myakura raised before merging. What's the best way to raise high-level issues? File a new issue with the templates label?

[rniwa]

It does seem good to address the issues @myakura raised before merging.

Fixed. Can we merge it now? It would be really unhelpful for this PR to be continue to be open while more people keep adding more review comments here. It would be far better if this was merged into the repository and we filed issues instead.

[rniwa]

What's the best way to raise high-level issues? File a new issue with the templates label?

File issues with [DOM Parts] probably since this new proposal is no longer anything to do with template elements per se.

[rniwa]

Alright, I added this to wiki page instead since this isn't getting approved by anyone.

[annevk]

You gave me 8 minutes to reply? I'm happy to merge this in a bit.

[rniwa]

You gave me 8 minutes to reply? I'm happy to merge this in a bit.

Oh sorry, I thought you went off somewhere. Thanks for merging it!