| 09:12 | <annevk> | smaug: can I get you to quickly review https://github.com/whatwg/dom/pull/1334? rniwa already looked it over, but since it's fairly significant I thought I should get at least one other person. |
| 10:21 | <smaug> | will do in a moment |
| 10:54 | <sharonne> | Any Cashapp? Chime? Zelle? Btc? Usdt?Skrill? Apple Pay? Pay pal? Venmo? BOA? Wells Fargo? Join my channel https://t.me/+32cFzLuOiacxZmM0 Got any of these Chase? BOA? Wells Fargo? Navy federal? Capital one? Truist? Att &t? PNC? TD bank? Credit union? M&T bank? Santander bank? Northwest bank? Or any other local Bank JOIN THE CHANNEL NOW TO GET PAID💯 ✅ |
| 12:16 | <annevk> | smaug: thanks! So I guess that means that "clone a shadow root" needs to be its independent operation that we call from both "clone a node" and "clone a node's children". And I guess that also goes for running the cloning steps then as they will also append a bunch of nodes in case of the template element. |
| 12:18 | <annevk> | So much unfortunate duplication, but it seems hard to avoid. |
| 12:21 | <smaug> | I guess template element just calls clone for its content. And it doesn't matter much where that happens, since the content lives on its own subtree |
| 12:22 | <smaug> | or does it create document fragment on the new one and clones that as the parent |
| 12:22 | <smaug> | I guess it goes like that |
| 12:22 | <smaug> | but still doesn't matter too much when that happens |
| 12:25 | <annevk> | Oh I guess you're right. It's only shadow-including tree traversal. That doesn't go into all conceptual hosts. |
| 12:52 | <annevk> | smaug: should be good now. |
| 13:06 | <annevk> | smaug: I think I have stared long enough at this now that I can deduplicate "clone a node" and "clone a node's children". Let me try that. |
| 13:09 | smaug | will read the pr after lunch |
| 13:22 | <annevk> | I think this is the approach I had initially in mind, but then forgot how to do it. I still need to push it. |
| 13:29 | <annevk> | I think I could even go back to a single recursive algorithm. The only new parameter it needs is a parent node. But it seems a little clearer with cloning a single node being its own algorithm. And also with most callers not having to pass in a null parent. |
| 17:42 | <smaug> | FWIW, I'm now looking at whether the order of cloning shadowroot vs other nodes can affect to slotchange ordering in case of nested shadow dom |
| 20:42 | <Yagiz Nizipli> | Hey folks, I couldn't find the declaration of URLPattern.compareComponent in the urlpattern spec. Can someone help me to find where is it defined? There is a test in WPT checking for it. |
| 21:41 | <smaug> | And Safari seems to have different behavior there currently vs. Firefox and Chrome. |
| 21:44 | <smaug> | Firefox and Chrome follow the current spec |
| 23:02 | <Meghan Denny> | https://github.com/whatwg/urlpattern/issues/61 looks to be the source of truth atm, which is still open |
| 23:16 | <Yagiz Nizipli> | https://github.com/whatwg/urlpattern/issues/61 looks to be the source of truth atm, which is still open |