| 01:33 | <Nick Johnstone> | Hiya. Possibly a silly question, but I'm implementing bindings for a WebAssembly powered mruby framework, and I was wondering where I can download .idl files for WHATWG specs. I've had a hunt around but struggled to find any downloads. Is the intention to parse the spec itself to create the .idl files? |
| 01:35 | <Nick Johnstone> | Ohh I had just been looking through the WebIDL docs, hadn't realized that the IDL is just at the bottom of the file in each spec. Should be fairly easy to grab. |
| 01:36 | <Andreu Botella (he/they)> | https://github.com/w3c/webref/tree/main/ed/idl has continuously-updated IDL files for every spec, not just WHATWG ones |
| 01:37 | <Nick Johnstone> | Oh perfect, thank you. |
| 01:37 | <Nick Johnstone> | That makes life very easy |
| 06:55 | <sideshowbarker> | annevk: so about the status-code-in-CORS-messages thing, I actually ended up writing a patch — so I’ll raise the bug and then also attach that patch |
| 06:58 | <annevk> | sideshowbarker: okay, I worry a bit about conflicts as my PR is still outstanding and also touches that, but we'll see I guess |
| 06:59 | <sideshowbarker> | oh |
| 06:59 | <sideshowbarker> | which PR? |
| 07:06 | <sideshowbarker> | hsivonen: you know about https://github.com/albfernandez/juniversalchardet already? |
| 07:08 | <hsivonen> | hsivonen: you know about https://github.com/albfernandez/juniversalchardet already? |
| 07:09 | <annevk> | sideshowbarker: https://github.com/whatwg/fetch/pull/1330? |
| 07:09 | <annevk> | Still hoping for more review on that one btw |
| 07:09 | <hsivonen> | Specifically, I thought jchardet was a port of "universalchardet". If not, what was "chardet" and how old is jchardet? |
| 07:11 | <sideshowbarker> | hsivonen: yeah I kind of only vaguely recall any details. But anyway yesterday Yves Lafon pointed ou juniversalchardet to me |
| 07:12 | <sideshowbarker> | annevk: ah that one yeah — agree it would be great to get more review |
| 07:12 | <hsivonen> | I wish I could find the time to try out chardetng on JVM via https://github.com/cretz/asmble |
| 07:12 | <sideshowbarker> | annevk: but about the patch, I meant a Gecko patch, to the networking code |
| 07:13 | <sideshowbarker> | hsivonen: oh wow, had not seen that one yet. I had heard about other WASM to JVM tools, but not that one |
| 07:14 | <hsivonen> | hsivonen: oh wow, had not seen that one yet. I had heard about other WASM to JVM tools, but not that one |
| 07:14 | <annevk> | sideshowbarker: aaah, I was already so confused |
| 07:14 | <annevk> | my bad |
| 07:14 | <annevk> | that's great |
| 07:15 | <sideshowbarker> | hsivonen: TeaVM for one |
| 07:15 | sideshowbarker | looks for others |
| 07:15 | <hsivonen> | TeaVM is the other way round, right? |
| 07:16 | <sideshowbarker> | hsivonen: oh I guess so |
| 07:16 | <sideshowbarker> | ah yeah of course |
| 07:16 | <sideshowbarker> | same for the others I was thinking of |
| 07:16 | <hsivonen> | I see |
| 07:18 | <annevk> | I didn't know Jeff Jaffe played Werewolf; kinda sad I missed that now, although Werewolf was often a bit late for me iirc |
| 07:18 | <annevk> | (watching the TPAC intro video, for context) |
| 07:18 | <sideshowbarker> | Yeah Jeff often joined it at past TPACs |
| 07:20 | <sideshowbarker> | btw, Gecko string-manipulation code is a lot of “fun” |
| 07:22 | <sideshowbarker> | the way I wrote up the patch, I am certain it’s not anywhere close to being the “right” way — but having spent nearly half a day looking through existing code and doing trial-and-error to try to figure out what the right way actually is, I’m now looking forward to somebody enlightening me about how it code the way it should be |
| 07:23 | <sideshowbarker> | even the hacky way I ended up doing it, it I would have never arrived at without some suggestions from Arai |
| 07:30 | <annevk> | sideshowbarker: arai++; there's a necko-reviewers group, FWIW |
| 07:31 | <sideshowbarker> | oh, didn’t know — a mailing list? |
| 07:33 | <annevk> | sideshowbarker: I'm not sure how it works, but basically once you have your patch on Phabricator you put that in the reviewers field |
| 07:45 | <hsivonen> | The group reviews just show up on the review list (on Phab) of every reviewer in the group. |
| 08:12 | <annevk> | Does anyone know if there's a way to get your selected TPAC events into your own calendar? |
| 08:27 | <annevk> | It seems that if you go to an individual session there's another calendar thing that allows for that... |
| 13:10 | <sideshowbarker> | Jake Archibald: for the CORS playground, it would be useful to be able to set a status code for the main response as well (not just for the preflight response) |
| 13:12 | <Jake Archibald> | sideshowbarker: I didn't add it because it has no impact on if something passes a CORS test or not. Or am I missing something? (not against adding it) |
| 13:13 | <sideshowbarker> | Jake Archibald: for demo’ing/emulating the case where a server responds with a 4xx or 5xx error without the Access-Control-Allow-Origin header |
| 13:13 | <Jake Archibald> | ohh yeah |
| 13:14 | <Jake Archibald> | ok, I'll add that |
| 13:14 | <sideshowbarker> | great |
| 13:14 | <sideshowbarker> | yeah, for me right now, I’m trying to test my gecko patch that adds the HTTP status code to some of the CORS messages that get logged to the devtools console |
| 13:18 | <sideshowbarker> | annevk: Jake Archibald: just now raised https://bugzilla.mozilla.org/show_bug.cgi?id=1736026 and will attach the patch there shortly |
| 13:29 | <Jake Archibald> | sideshowbarker: There you go! https://jakearchibald.com/2021/cors/playground/?prefillForm=1&requestMethod=GET&requestUseCORS=1&requestSendCredentials=&preflightStatus=204&preflightAllowOrigin=&preflightAllowCredentials=&preflightAllowMethods=&preflightAllowHeaders=&responseStatus=404&responseAllowOrigin=&responseAllowCredentials=&responseExposeHeaders= |
| 13:30 | <sideshowbarker> | Jake Archibald: beautiful! thanks much |
| 13:31 | <Jake Archibald> | sideshowbarker: are you going to raise the same issue on crbug, or shall I? |
| 13:32 | <sideshowbarker> | Jake Archibald: I will — I’ll also try to write up a Blink patch for it |
| 13:32 | <Jake Archibald> | sideshowbarker: ohhh nice! Much appreciated |
| 13:32 | <sideshowbarker> | gonna raise a WebKit bug too (with patch) |
| 13:32 | <sideshowbarker> | cheers |
| 13:39 | <sideshowbarker> | OK, gecko patch submitted — https://phabricator.services.mozilla.com/D128596 |
| 18:57 | <Andreu Botella (he/they)> | So I just noticed that equality/identity for the purposes of infra sets and maps can't be defined trivially, and might need to be type-specific |
| 18:57 | <Andreu Botella (he/they)> | Because the module map uses (URL record, string) tuples as keys, but the URL record equality must ignore the blob URL entry to remain compatible with the web |
| 18:58 | <Andreu Botella (he/they)> | https://github.com/web-platform-tests/wpt/blob/master/html/semantics/scripting-1/the-script-element/module/dynamic-import/blob-url.any.js#L54 |
| 20:09 | <Domenic> | Because the module map uses (URL record, string) tuples as keys, but the URL record equality must ignore the blob URL entry to remain compatible with the web |