| 05:07 | <canrau> | Hi there, just out of curiosity, why is the colon : the de-facto standard it seems for named groups in URLPattern (and others) instead of delimiting the group name within square brackets [group] or curly braces {group} which both would easily (I think) support pre- & suffixes like @[username] and potentially even things like user-[name]-[surname]? 🤓 surprisingly few (JS) frameworks support this use-case easily out of the box, so far I think I know only of SvelteKit and RakkasJS.With URLPattern I can kind of replicate this with a RegExp of course e.g. new URLPattern({ pathname: "/:username(@.+)" }) though this feels more complex, requires a RegExp and contains the @ in the result.To clarify, this is just curiosity currently to understand what I might be missing, as I'm slightly annoyed by the fact that : is mostly used and makes prefixes (& suffixes) harder 😅 |
| 07:15 | <annevk> | CanRau: I believe it largely followed an existing JS library in terms of syntax. You might be able to find references to it in the early issues/commits. |
| 11:04 | <smaug> | Hmm, Blink allows importScripts("data:text/plain,1"); in workers, Gecko and Webkit don't |
| 11:07 | <smaug> | looks like Blink doesn't have the correct mimetype check |
| 11:14 | <Kaiido> | smaug: https://github.com/whatwg/html/issues/8869 |
| 11:55 | <annevk> | At this point it seems like we should just add a test to WPT that expects MIME type enforcement? |
| 11:58 | <Kaiido> | Isn't it what we did in https://github.com/web-platform-tests/wpt/pull/38430 ? |
| 12:00 | <annevk> | You did indeed, great. I'm going to close the HTML issue then. This is just a bug in Chromium. |
| 16:11 | <CanRau> | hmm, yea path-to-regexp, thank you. maybe I should ask them what the motivation was 🤔 |
| 16:36 | <evilpie> | See also https://github.com/whatwg/urlpattern/issues/241 |
| 17:11 | <annevk> | zcorpan: I think we need CSS to own those concepts |
| 17:13 | <annevk> | zcorpan: see https://github.com/w3c/csswg-drafts/issues/10039 |
| 17:33 | <TabAtkins> | I dropped a comment in the issue for you. |
| 19:34 | <CanRau> | thanks for sharing, must've missed it while looking thru the issues. agreeing with both sides tho I'm curious what "There are also concepts that are possible to express in the current syntax (path/:id) that I don't believe are expressible in the Swagger syntax (path/{id})" might refer to 🤔 maybe I should ask in the issue itself |