Re: [whatwg/dom] Proposal: a DocumentFragment whose nodes do not get removed once inserted (#736)

rniwa left a comment (whatwg/dom#736)

Remainder: I have a [when2meet survey](https://www.when2meet.com/?29803039-IvMvD) for people to pick a date in April.

> I think we are overthinking this idea/proposal (me included) **but** it requires a lot of DOM prototype patching from userland ... which is not too bad, when it comes to polyfills, but imagine two projects land two different versions of that polyfill and you can see how that "_just polyfill it in userland_" solution is not really compelling ... that plus the fact tools need to have guidance around these things so that special comments won't be removed from minifiers or other processing.

Okay, so the argument is interoperability between different libraries & frameworks.

> After a lot of prototyping around this idea I think the ultimate goal would be to have `<!--<>-->` and `<!--</>-->` as special comments, but that comment could also be `<!--<GroupNodes>-->` and `<!--</GroupNodes>-->`.

Formalizing something like that as a way to anchor `NodeGroup` object seems like where we're all converging. Being able to find those grouping natively in the browser was also where @mfreed7 saw the most perf gain from implementing DOM parts so I'd imagine this proposal's version will be perf positive as well.

> Again, happy to provide a concrete example of this idea but it's honestly the easiest I could think about to move this forward, yet the guard around not breaking those start/end comments boundaries via ranges remain the main, if not the only, issue to tackle at the Range level, the rest feels trivial to do.

I guess one feature we could offer in the browser is some ability to adjust the location of start/end comment nodes upon DOM mutations. I suspect we want Range-like behavior to stick around in the tree when DOM is mutated.

-- 
Reply to this email directly or view it on GitHub:
https://github.com/whatwg/dom/issues/736#issuecomment-2762716642
You are receiving this because you are subscribed to this thread.

Message ID: <whatwg/dom/issues/736/2762716642@github.com>