| 03:42 | <jojo404> | "I'll help anyone interested on how to earn 100k in just 24hours from the crypto market. But you will have to pay me my commission! when you receive your profit! if interested click on the group link and send me a direct message https://t.me/+FGhsj-vpp7dmZDc0 by asking me HOW |
| 05:33 | <Domenic> | annevk: ping on https://github.com/whatwg/html/pull/10015 , it's ready to merge but ntim was hoping for a check from you it seems. |
| 08:07 | <hsivonen> | Currently, I'm thinking that the right API for IDNA is pub fn process<W: Write + ?Sized>(&self, domain_name: &[u8], mode: Mode, strictness: Strictness, sink: &mut W) -> Result<(), ()> where Mode is ToAscii or ToUnicode and Strictness is Whatwg or Std3. Am I missing something? |
| 08:10 | <hsivonen> | annevk: In UTS 46 section 4 Processing step 4 Convert/Validate substeps 1 & 2, it seems to me that there's a bug in step 1 saying "continue with the next label" but step 2 saying "Attempt to convert the rest of the label". Am I misunderstanding, or is this a spec bug where it first continues to the next label but still wants to do stuff with the non-next label? |
| 08:14 | <hsivonen> | Or maybe Strictness could be something like UserAgent and Validator. |
| 08:14 | <hsivonen> | Also depends on the outcome of https://github.com/whatwg/url/issues/820 and https://github.com/whatwg/url/issues/821 . |
| 08:32 | <hsivonen> | Seems annoying that if an RTL character shows up in a later label, the rules for the earlier labels change. |
| 08:33 | <hsivonen> | Complicates an ASCII fast path. :-( |
| 09:15 | <Noam Rosenthal> | HTML spec local build question: is there a way to make it fail on HTML parsing issues? I often get success on local build and then a failure on PR preview build (e.g. missing end-tags) |
| 16:06 | <annevk> | Is anyone other than Aryeh familiar with WPT dom/common.js? Ms2ger maybe? We need CDATASection coverage, but I've no idea how to go about it. |
| 16:15 | <Ms2ger> | Oh dear, that one |
| 16:21 | <Ms2ger> | annevk: I think you'll need to update all references to Node.TEXT_NODE to also include cdata sections, and then add a couple of cases to testRanges/testNodes and it should just work, maybe? |
| 16:25 | <annevk> | Ms2ger: I will give that a go. Making some changes to WebKit to verify. But debugging for actual correctness seems really hard so maybe I should also write some simple tests on the side... |
| 16:25 | <Ms2ger> | Sounds good |
| 16:27 | <annevk> | So yeah, I found unique test failures in WebKit. I guess that's proof that nobody cares about CDATASection. |
| 16:29 | <annevk> | Domenic: minor [SecureContext] mess: https://github.com/w3c/webref/issues/1142#issuecomment-1930188979. Thoughts? |
| 16:32 | <Noam Rosenthal> | annevk: can I ask for your guidance about https://github.com/whatwg/html/pull/9970#discussion_r1479199162 ? Trying to figure out if the new <link rel=expect> should use URL-parse or encoding-parse (I think the latter to be consistent with existing link/a, but you should be more informed about this than myself) |
| 16:36 | <annevk> | Noam Rosenthal: yeah, for (new) elements using encoding-parse is reasonable. |
| 16:36 | <annevk> | Although maybe if we had a completely new element we'd sidestep it? Not sure. |
| 16:41 | <Noam Rosenthal> | annevk: the guidance in the html spec suggests that new stuff should use "parse", was I reading it wrong? |
| 16:45 | <Noam Rosenthal> | Ah it's for new APIS rather than elements |