Bernie Reiter

Opening for review. As stated in the PR desc, this is pretty much copy-pasted from @dmsnell's https://github.com/WordPress/gutenberg/pull/46345. Not sure if you'd rather make any changes to the API and/or test...

Closing. Per recent discussions, we're going to unblock the work in the [BHE repo](https://github.com/WordPress/block-hydration-experiments) by implementing missing features in a subclass of `WP_HTML_Tag_Processor` right there ([e.g.](https://github.com/WordPress/block-hydration-experiments/pull/158)). Development of the HTML...

As I've now implemented bookmark invalidation for bookmarks that are "overwritten", I'll tentatively open this for review. Note that since the method itself is `private`, I haven't added any unit...

> The changes look good to me. Let's observe how we end up using this API – same as Dennis I am worried about the possibility of overlapping updates. Thank...

> What does the addition of `lexical_updates` and `attribute_updates` buy us? We can already distinguish these because attribute updates have a string array key while non-attribute updates are appended with...

Rebased, now that the bookmark invalidation PR (https://github.com/WordPress/gutenberg/pull/47559) has been merged.

> with `$this->lexical_updates` already being protected, is this required now? Probably not anymore! > I'm still wondering what separating attribute from lexical updates gain us that we can't already do,...

> > suggests the ability to apply a diff to any part of the markup > > It _shouldn't_ suggest this. It should only suggest that at the point we...

Closing this PR as obsolete, now that we have https://github.com/WordPress/wordpress-develop/pull/4519.

> is this something you consider necessary for 6.1? You mean 6.2 🙂 No, this isn't needed for the pending WP release. > is this necessary at all? I kinda...