profile
viewpoint
Anne van Kesteren annevk @mozilla https://annevankesteren.nl/ Weird, but potentially useful immigrant · engine diversity absolutist · #BlackLivesMatter · he/him

domenic/promises-unwrapping 1010

The ES6 promises spec, as per September 2013 TC39 meeting

heycam/webidl 178

Web IDL

annevk/url 107

URL parser in JavaScript

annevk/html-heading-level-polyfill 7

A polyfill for a proposed HTML Standard "outline algorithm".

annevk/live-url-viewer 6

Live URL Viewer

annevk/xml5 3

Automatically exported from code.google.com/p/xml5

annevk/html-cross-origin-objects 2

Defining Window and Location

darobin/webref 2

References for Web Specifications (http://webref.herokuapp.com/)

fiji-flo/bikeshed-rs 2

[WIP] GSOC2020 Mozilla project

issue commentw3c/webcomponents

2020 Virtual F2F for Accessibility and Web Components

Yeah, I did not realize we would keep the same time. I'm fine with reading the notes and chatting with some folks about it afterwards.

@F1LT3R in principle, yes.

rniwa

comment created time in 5 hours

issue commentwhatwg/html

Opt-in for `<select>` customizability

@gregwhitworth well yes, you would also need pseudo-elements to get at the various bits. It would not give you the ability to add new bits necessarily, but it would address a large swath of the use cases. (And yes, part of disabling the native theming would be disabling the dialog drawing its own window.)

gregwhitworth

comment created time in 6 hours

pull request commentwhatwg/fetch

Partition HTTP Connections

Oh, that sounds great. I'll be out for week but happy to look at that mid-August.

MattMenke2

comment created time in 7 hours

PR closed whatwg/fetch

Add device-memory and network quality client hints to the list needs tests topic: client hints

annevk: please take a look. This is the first PR where the new client hints are added to the spec.

<!-- This comment and the below content is programatically generated. You may add a comma-separated list of anchors you'd like a direct link to below (e.g. #idl-serializers, #idl-sequence):

Don't remove this comment or modify anything below this line.
If you don't want a preview generated for this pull request,
just replace the whole of this comment's content by "no preview"
and remove what's below.

-->


<a href="https://whatpr.org/fetch/725.html" title="Last updated on May 30, 2018, 11:54 PM GMT (43eea2f)">Preview</a> | <a href="https://whatpr.org/fetch/725/3055f13...43eea2f.html" title="Last updated on May 30, 2018, 11:54 PM GMT (43eea2f)">Diff</a>

+8 -3

6 comments

1 changed file

tarunban

pr closed time in 7 hours

pull request commentwhatwg/fetch

Add device-memory and network quality client hints to the list

Closing per 6a644c676558edc7e80548e25c6f85bf0143f395.

tarunban

comment created time in 7 hours

delete branch whatwg/fetch

delete branch : annevk/examples-quoted-string

delete time in 7 hours

push eventwhatwg/fetch

Anne van Kesteren

commit sha 1b1239ad51711aeaed722adbd7857f66fdda2fdb

Add examples for collect an HTTP quoted string And remove a confusing note. Closes #1068.

view details

push time in 7 hours

PR merged whatwg/fetch

Add examples for collect an HTTP quoted string

And remove a confusing note.

Closes #1068.

<!-- This comment and the below content is programatically generated. You may add a comma-separated list of anchors you'd like a direct link to below (e.g. #idl-serializers, #idl-sequence):

Don't remove this comment or modify anything below this line.
If you don't want a preview generated for this pull request,
just replace the whole of this comment's content by "no preview"
and remove what's below.

-->


<a href="https://whatpr.org/fetch/1069.html" title="Last updated on Aug 6, 2020, 2:42 PM UTC (3e0b5da)">Preview</a> | <a href="https://whatpr.org/fetch/1069/44a94a6...3e0b5da.html" title="Last updated on Aug 6, 2020, 2:42 PM UTC (3e0b5da)">Diff</a>

+25 -3

1 comment

1 changed file

annevk

pr closed time in 7 hours

issue closedwhatwg/fetch

"get, decode, and split" seems incorrect

It's wrong for X-Content-Type-Options, but it seems wrong for other callers too. E.g., we don't want to support Content-Type: "text/html".

What needs to happen is that this algorithm preserves the quotes in the output. It needs to be aware of them to ensure they are matched up with another quote, but it shouldn't end up stripping them.

closed time in 7 hours

annevk

issue commentwhatwg/html

Add open state to <select>

(See also #3976 and https://www.w3.org/Bugs/Public/show_bug.cgi?id=25452.)

gregwhitworth

comment created time in 7 hours

issue closedwhatwg/html

System Font Lists / Dialog

Due to the possibilities provided with WebGL v2 and WebAssembly, there are more and more companies pushing into the multimedia space with their Web apps, especially as this gives a unified distribution platform for both the Web and the Desktop.

We are for example providing a full featured painting and image editing application here: www.paintsupreme3d.com.

Now, we draw fonts using the 2D canvas. The biggest drawback right now for web based multimedia applications is that they cannot offer the user a list of the SYSTEM fonts to choose from for drawing text.

For web apps to be able to compete with native desktop apps, this is a major drawback right now.

Would it be possible at one stage to either:

  1. Provide a list of the installed system fonts to the application so that the app can provide its own font dialog to the user.

  2. Provide a browser font dialog where the selected font is passed to the 2D canvas if 1) provides a fingerprinting security issue.

This would be a big step for multimedia web app providers like us.

closed time in 7 hours

markusmoenig

issue commentwhatwg/html

System Font Lists / Dialog

See https://github.com/WICG/local-font-access.

markusmoenig

comment created time in 7 hours

issue closedwhatwg/html

System Font Dialog / Font Data

This is related to #3379 which is about getting a FontFace from a system font dialog to allow the user to select a system font to draw with in the 2D canvas.

Apart from getting a FontFace, it would also be great to get the font data of the font itself (OpenFont / TrueType). Paired with something like https://opentype.js.org one could:

  • Implement custom font rasterization engines which surpass the limited text capabilities of the 2D canvas for sophisticated text rendering applications. This would make Illustrator like applications possible in the browser.

  • Triangulate the font data into 3D meshes which can be processed with WebGL or custom rendering technologies (SDF, Raytracing etc.). This would allow full featured 3D applications in the browser.

closed time in 7 hours

markusmoenig

issue commentwhatwg/html

System Font Dialog / Font Data

There's also https://github.com/WICG/local-font-access now. Closing this.

markusmoenig

comment created time in 7 hours

pull request commentwhatwg/fetch

Partition HTTP Connections

Thanks all! (To be clear, I'm not blocking on tests as we don't have good infrastructure to test connections and I don't really see that changing anytime soon.)

MattMenke2

comment created time in 7 hours

push eventwhatwg/fetch

Matt Menke

commit sha c08e612512a627f9f6b6270e9e529270633457ac

Partition HTTP connections Extend the HTTP cache partitioning logic to partition the connection pool as well and introduce the network partition key as a shared primitive. There is still work remaining for various other network caches that is tracked by #917 for now. See also https://privacycg.github.io/storage-partitioning/.

view details

push time in 7 hours

PR merged whatwg/fetch

Partition HTTP Connections

Extend the HTTP cache partitioning logic to partition connections as well. This doesn't cover everything that will need to be partitioned, since much of it isn't even capturing in the fetch spec yet (browser-managed DNS caches, WebSocket connections over H2/H3 proxies, H2/H3 sessions, TLS/H3 session resumption cache, cache of metadata about H3-capable servers, HTTP auth cache, etc).

This addresses parts of issue #917.

  • [x] At least two implementers are interested (and none opposed):
    • Chrome
    • FireFox
  • [ ] Tests are written and can be reviewed and commented upon at:
    • ...
  • [ ] Implementation bugs are filed:
    • Chrome: https://crbug.com/963480
    • Firefox: https://bugzilla.mozilla.org/show_bug.cgi?id=1572544
    • Safari: …

<!-- This comment and the below content is programatically generated. You may add a comma-separated list of anchors you'd like a direct link to below (e.g. #idl-serializers, #idl-sequence):

Don't remove this comment or modify anything below this line.
If you don't want a preview generated for this pull request,
just replace the whole of this comment's content by "no preview"
and remove what's below.

-->


<a href="https://whatpr.org/fetch/1063.html" title="Last updated on Aug 6, 2020, 2:51 PM UTC (8cbe436)">Preview</a> | <a href="https://whatpr.org/fetch/1063/4ad496d...8cbe436.html" title="Last updated on Aug 6, 2020, 2:51 PM UTC (8cbe436)">Diff</a>

+43 -19

10 comments

1 changed file

MattMenke2

pr closed time in 7 hours

Pull request review commentwhatwg/fetch

Add examples for collect an HTTP quoted string

 and optionally an <var>extract-value flag</var>, run these steps:  inclusive, within <var>input</var>. </ol> -<p class="note no-backref">The <var>extract-value flag</var> argument makes this algorithm suitable-for <a for="header list">getting, decoding, and splitting</a> and <a>parse a MIME type</a>, as well-as other header value parsers that might need this.+<div class=example id=example-http-quoted-string>+ <table>+  <tr>+   <th>Input+   <th>Final position (always starts at 0)

https://whatpr.org/fetch/1069.html#example-http-quoted-string looks reasonable to me. What do you think?

annevk

comment created time in 7 hours

push eventwhatwg/fetch

Anne van Kesteren

commit sha 3e0b5dad406505dea858357c2f035a514e3f14e6

end of day, so play a bit more with this

view details

push time in 7 hours

Pull request review commentwhatwg/fetch

Add examples for collect an HTTP quoted string

 and optionally an <var>extract-value flag</var>, run these steps:  inclusive, within <var>input</var>. </ol> -<p class="note no-backref">The <var>extract-value flag</var> argument makes this algorithm suitable-for <a for="header list">getting, decoding, and splitting</a> and <a>parse a MIME type</a>, as well-as other header value parsers that might need this.+<div class=example id=example-http-quoted-string>+ <table>+  <tr>+   <th>Input+   <th>Final position (always starts at 0)

It's an "outparam", no?

annevk

comment created time in 8 hours

pull request commentwhatwg/html

registerProtocolHandler(): creating a navigable URL cannot fail

Isn't that what asserts are for? And as I mentioned, it's a more complete assert than "not failure".

annevk

comment created time in 8 hours

pull request commentwhatwg/fetch

Add examples for collect an HTTP quoted string

Should I put final position last?

annevk

comment created time in 8 hours

push eventwhatwg/fetch

Anne van Kesteren

commit sha 758979a9165ac904b659bbfa4bdd7f7eb5c97360

oops

view details

push time in 8 hours

PR opened whatwg/fetch

Add examples for collect an HTTP quoted string

And remove a confusing note.

Closes #1068.

+25 -3

0 comment

1 changed file

pr created time in 8 hours

create barnchwhatwg/fetch

branch : annevk/examples-quoted-string

created branch time in 8 hours

PR opened web-platform-tests/wpt

Reviewers
Docs: JavaScript tests

This tries to pull some of the more pertinent content forward.

TODO:

  • create more links into testharness-api * figure out how to weave in testdriver.js references
  • rewrite idlharness to account for all the changes to how we do that today

Feedback appreciated.

+135 -86

0 comment

1 changed file

pr created time in 8 hours

create barnchweb-platform-tests/wpt

branch : annevk/docs-js-tests

created branch time in 8 hours

push eventwhatwg/html

Anne van Kesteren

commit sha e15c649da93e8a894c98e7225ae621ea2e009dba

registerProtocolHandler(): drop scheme from note If you put a % in a scheme, it effectively turns the input from an absolute URL into a relative URL. See https://github.com/web-platform-tests/wpt/pull/24886 for context, again.

view details

push time in 8 hours

delete branch whatwg/html

delete branch : annevk/rph-note-again

delete time in 8 hours

PR merged whatwg/html

registerProtocolHandler(): drop scheme from note topic: custom protocols

If you put a % in a scheme, it effectively turns the input from an absolute URL into a relative URL.

See https://github.com/web-platform-tests/wpt/pull/24886 for context, again.

<!-- This comment and the below content is programatically generated. You may add a comma-separated list of anchors you'd like a direct link to below (e.g. #idl-serializers, #idl-sequence):

Don't remove this comment or modify anything below this line.
If you don't want a preview generated for this pull request,
just replace the whole of this comment's content by "no preview"
and remove what's below.

-->


<a href="https://whatpr.org/html/5794/system-state.html" title="Last updated on Aug 6, 2020, 5:45 AM UTC (ddf33b4)">/system-state.html</a> ( <a href="https://whatpr.org/html/5794/8b2eaf9...ddf33b4/system-state.html" title="Last updated on Aug 6, 2020, 5:45 AM UTC (ddf33b4)">diff</a> )

+1 -1

0 comment

1 changed file

annevk

pr closed time in 8 hours

pull request commentwhatwg/html

registerProtocolHandler(): creating a navigable URL cannot fail

You don't have to know that. All you have to know is that you have a URL that navigate can use. And the URL parser can also return nothing (though only when you have a state override).

annevk

comment created time in 8 hours

push eventmozilla/standards-positions

Anne van Kesteren

commit sha db7d306a2305c8210b9cec7af53fbdba0044f7a1

nit

view details

push time in 8 hours

PR opened mozilla/standards-positions

Meta: make add work for non-spec URLs

This seems like more useful behavior for the various proposals we have to evaluate.

+10 -9

0 comment

1 changed file

pr created time in 9 hours

create barnchmozilla/standards-positions

branch : annevk/activities.py-add

created branch time in 9 hours

issue commentwhatwg/html

Proposal: nested <OPTGROUP>

Semantics are part of a UI. Also, perhaps I wasn't clear, but we cannot change the parser.

oriadam

comment created time in 10 hours

issue commentwhatwg/encoding

Concatenating two ISO-2022-JP outputs from a conforming encoder doesn't result in conforming input

Looks okay. I wonder if @ricea and @cdumez could figure out if Chromium and WebKit would align with any potential resulting Encoding standard changes, but I don't think that should block raising this with Unicode as them giving better security advice is a win either way.

hsivonen

comment created time in 11 hours

pull request commentwhatwg/html

Crbug 1060691 name noopener

You reuse a top-level browsing context, but to be able to reuse it one would first have to be able to create one and I don't see how that would work if COOP is in effect.

ParisMeuleman

comment created time in 11 hours

issue commenttc39/proposal-realms

Examples are not emulating frames

Not sure, but it seems very easy to do the "wrong" thing here.

annevk

comment created time in 11 hours

issue commentwhatwg/meta

Avoid the 100 char wrap

@domenic is there tooling we could use? Something like rustfmt, but for Bikeshed and Wattsi input.

jakearchibald

comment created time in 13 hours

push eventMattMenke2/fetch

Anne van Kesteren

commit sha 8cbe436ea4188bca25bbbae92bfb91e1276c08d0

editorial nits

view details

push time in 13 hours

Pull request review commentweb-platform-tests/wpt

Test some more edge cases for X-Frame-Options

+function xfo_simple_tests({ headerValue, headerValue2, cspValue, sameOriginAllowed, crossOriginAllowed }) {+  const value2QueryString = headerValue2 !== undefined ? `&value2=${headerValue2}` : ``;+  const cspQueryString = cspValue !== undefined ? `&csp_value=${cspValue}` : ``;++  const valueMessageString = headerValue === "" ? "(the empty string)" : headerValue;+  const value2MessageString = headerValue2 !== undefined ? `;${headerValue2}` : ``;+  const cspMessageString = cspValue !== undefined ? ` with CSP ${cspValue}` : ``;++  xfo_test({+    url: `/x-frame-options/support/xfo.py?value=${headerValue}${value2QueryString}${cspQueryString}`,+    check: sameOriginAllowed ? "loaded message" : "no message",+    message: `${valueMessageString}${value2MessageString} ${sameOriginAllowed ? "allows" : "blocks"} same-origin framing${cspMessageString}`+  });++  xfo_test({+    url: `http://{{domains[www]}}:{{ports[http][0]}}/x-frame-options/support/xfo.py?value=${headerValue}${value2QueryString}${cspQueryString}`,+    check: crossOriginAllowed ? "loaded message" : "no message",+    message: `${valueMessageString}${value2MessageString} ${crossOriginAllowed ? "allows" : "blocks"} cross-origin framing${cspMessageString}`+  });+}++function xfo_test({ url, check, message }) {+  async_test(t => {+    const i = document.createElement("iframe");+    i.src = url;++    switch (check) {+      case "loaded message": {+        waitForMessageFrom(i, t).then(t.step_func_done(e => {+          assert_equals(e.data, "Loaded");+        }));+        break;+      }+      case "failed message": {+        waitForMessageFrom(i, t).then(t.step_func_done(e => {+          assert_equals(e.data, "Failed");+        }));+        break;+      }+      case "no message": {+        waitForMessageFrom(i, t).then(t.unreached_func("Frame should not have sent a message."));+        i.onload = t.step_func_done(() => {

I suspect a problem here is that Firefox does not dispatch the load event for (all) network errors. See also https://github.com/whatwg/html/issues/125 (and references to that in other tests). If initially i.contentDocument was available as I suspect it would be given initial about:blank, you could wait for it to not be available using step_wait and friends.

domenic

comment created time in 14 hours

issue commentwhatwg/html

Disable custom protocols in sandboxed iframe

#5289 might obviate the need for this somewhat.

shhnjk

comment created time in 14 hours

issue commentwhatwg/html

Update registerProtocolHandler() and unregisterProtocolHandler()

Yeah, I think that would be ideal.

annevk

comment created time in 14 hours

issue closedwhatwg/html

Make it clear that registerProtocolHandler() only affects navigation

E.g., they cannot change what is being fetched, just what will be navigated towards. Prior history is https://www.w3.org/Bugs/Public/show_bug.cgi?id=24091 and http://krijnhoetmer.nl/irc-logs/whatwg/20150928#l-323.

closed time in 14 hours

annevk

issue commentwhatwg/html

Make it clear that registerProtocolHandler() only affects navigation

Let's fold this into #3043.

annevk

comment created time in 14 hours

issue commentwhatwg/fetch

"get, decode, and split" seems incorrect

Oh wait, I think the current algorithm is fine after all as we don't pass the extract-value flag. Any ideas for how to improve the readability?

annevk

comment created time in 14 hours

issue commentwhatwg/fetch

"get, decode, and split" seems incorrect

Hopefully in the future we'll only have structured fields to (mostly not) deal with. 😊

annevk

comment created time in 14 hours

pull request commentmozilla/standards-positions

Payment Handler API (defer)

I thought that we only used defer if we decided not to look at something. From the description it seems like we looked at this quite a bit and therefore non-harmful seems more appropriate.

marcoscaceres

comment created time in 15 hours

issue commenttc39/proposal-realms

Examples are not emulating frames

console.log(document instanceof Object) would be false.

annevk

comment created time in 15 hours

pull request commentw3c/webappsec-referrer-policy

Make strict-origin-when-cross-origin the default referrer policy

@domfarolino just read your email on blink-dev. Note that since Safari ships a superset of this, I don't think we need to hold off here.

krgovind

comment created time in 15 hours

issue openedtc39/proposal-realms

Examples are not emulating frames

It seems to me that if you do something like realmGlobal.document = new FakeDocument(); you are not emulating frames as FakeDocument will have the "wrong" global associated with it.

created time in 15 hours

PR opened whatwg/html

registerProtocolHandler(): creating a navigable URL cannot fail

Fixes #481.

+2 -0

0 comment

1 changed file

pr created time in 15 hours

create barnchwhatwg/html

branch : annevk/rph-not-failure

created branch time in 15 hours

issue commentwhatwg/html

Meta discussion: Coordination/specing of overlap content between Open UI & WHATWG

(Open UI is https://github.com/WICG/open-ui.)

Is an accurate summary that changes to the web platform are standardized here and Open UI serves as some kind of discussion/documentation hub for the effort? If so, I think that division seems reasonable.

gregwhitworth

comment created time in 16 hours

issue commentwhatwg/html

Opt-in for `<select>` customizability

If the main goal is styling I don't see why we need a new attribute or element. It seems that appearance:none or some such could allow for that.

gregwhitworth

comment created time in 16 hours

issue commentwhatwg/fetch

421 and upload streams

Well, the request cannot be retried and therefore the 421 is forwarded to the application, whereas in other cases retrying is allowed. (I haven't tested what browsers do with it in general though.)

annevk

comment created time in 16 hours

issue commentwhatwg/fetch

A new Request(input, init) constructor step sounds weird.

I see, so yeah, moving step 38 to 39.0 would be the right approach here as we need to know that body/inputBody are non-null.

yoichio

comment created time in 16 hours

PR opened whatwg/html

registerProtocolHandler(): drop scheme from note

If you put a % in a scheme, it effectively turns the input from an absolute URL into a relative URL.

See https://github.com/web-platform-tests/wpt/pull/24886 for context, again.

+1 -1

0 comment

1 changed file

pr created time in 16 hours

create barnchwhatwg/html

branch : annevk/rph-note-again

created branch time in 16 hours

pull request commentweb-platform-tests/wpt

Do not remove %s token when validating (un)registerProtocolHandler's URL

Right, sigh. Basically, if it's in the scheme part its treated as a relative URL as % cannot appear in a scheme.

chromium-wpt-export-bot

comment created time in 16 hours

push eventwhatwg/infra

Anne van Kesteren

commit sha 146d27bf3480d3153ff14bd52f430f94b8de98b3

more exemplary-prose

view details

push time in 17 hours

issue commentwhatwg/fetch

"get, decode, and split" seems incorrect

Well, Content-Type in a way, as text/html;charset="utf-8" ought to work. And if the parser did preserve things, it would also work for X-Content-Type-Options as it would simply not result in a match. (When you do want it to result in a match you can use https://fetch.spec.whatwg.org/#collect-an-http-quoted-string with the extract-value flag as parse a MIME type does.

annevk

comment created time in a day

issue commentWICG/native-file-system

getOriginPrivateDirectory is not a name that's friendly to partitioning

getStorageDirectory() or some such? Or navigator.storage.getDirectory()?

annevk

comment created time in a day

Pull request review commentwhatwg/infra

Specify algorithm declarations

 as well as the single algorithm would. Therefore performance is best left as a f over.  +<h3 id=declaration>Declaration</h3>++<p>Algorithm names are usually verb phrases, but sometimes are given names that emphasize their+standalone existence, so that standards and readers can refer to the algorithm more idiomatically.++<p class=example id=example-algorithm-declaration-names>Some algorithm names in the latter category+include "attribute change steps", "internal module script graph fetching procedure", and "overload+resolution algorithm".++<p>Declare algorithms by stating their name, parameters, and return type, in the following form:++<blockquote>+To <dfn ignore>[algorithm name]</dfn>, given a [type1] <var ignore>[parameter1]</var>, a [type2]+<var ignore>[parameter2]</var>, &hellip;, perform the following steps. They return a [return type].+</blockquote>++<p>(For non-verb phrase algorithm names, use "To perform the+<dfn ignore>[algorithm name]</dfn>&hellip;".)++<p class=example id=example-algorithm-declaration>To <dfn ignore>parse an awesome format</dfn> given+a [=byte sequence=] <var ignore>bytes</var>, perform the following steps. They return a+[=string=] or null.++<p>Algorithms which do not return a value use a shorter form. This same shorter form can be used even+for algorithms that do return a value, since the return type is relatively easy to infer from the+algorithm steps and call sites:++<blockquote>+To <dfn ignore>[algorithm name]</dfn>, given a [type1] <var ignore>[parameter1]</var>, a [type2]+<var ignore>[parameter2]</var>, &hellip;:+</blockquote>++<p class=example id=example-algorithm-declaration-no-return>To+<dfn ignore>parse an awesome format</dfn> given a [=byte sequence=] <var ignore>bytes</var>:++<p>Very short algorithms can be declared and specified using a single sentence:++<p class=example id=example-algorithm-declaration-short>To <dfn ignore>parse an awesome format</dfn>+given a [=byte sequence=] <var ignore>bytes</var>, return the result of+<a lt="ASCII uppercase">ASCII uppercasing</a> the <a lt="isomorphic decode">ismorphic decoding</a>+of <var ignore>bytes</var>.++<p>Types should be included in algorithm declarations, but may be omitted if the parameter name is+clear enough, or if they are otherwise clear from context. (For example, because the algorithm is+a simple wrapper around another one.)++<p class=example id=example-algorithm-declaration-no-types>To+<dfn ignore>load a classic script</dfn> given <var>url</var>, return the result of performing the+internal script-loading algorithm given <var>url</var> and "<code>classic</code>".

I typically use "with" instead of "given" when calling, but no strong opinion.

domenic

comment created time in a day

Pull request review commentwhatwg/infra

Specify algorithm declarations

 leading spaces" or "return false") are to be interpreted with the meaning of the  <p>it would be equivalent to the following:</p>   <blockquote>-  <p>To eat an orange:+  <p>To <dfn ignore>eat an orange</dfn>:

Yeah, I guess it's okay. It's just that other places have a more direct connection with "steps" or "algorithm".

domenic

comment created time in a day

Pull request review commentwhatwg/html

Editorial: modernization of script fetching algorithms

 document.querySelector("button").addEventListener("click", bound);     also finish their work synchronously.</p>    </li> -   <li><p>Let <var>response</var> be <var>response</var>'s <span>unsafe response</span>.</p></li>+   <li><p>Set <var>response</var> to <var>response</var>'s <span>unsafe response</span>.</p></li>     <li>-    <p>If any of the following conditions are met, throw a <span>"<code>NetworkError</code>"</span>-    <code>DOMException</code>:</p>+    <p>If any of the following conditions are met:</p> -    <ul class="compact">+    <ul>      <li><p><var>response</var>'s <span data-x="concept-response-type">type</span> is "<code-     data-x="">error</code>"</p></li>+     data-x="">error</code>"; or</p></li>

I realize, but there's precedent, e.g.:

  • https://url.spec.whatwg.org/#absolute-url-string
  • https://url.spec.whatwg.org/#authority-state (step 2)
domenic

comment created time in a day

issue commentwhatwg/html

Proposal: nested <OPTGROUP>

This isn't compatible with the HTML parser. This might be something https://github.com/WICG/open-ui is willing to investigate.

oriadam

comment created time in a day

issue closedw3c/FileAPI

Store HTTPS state on Blob objects

See https://github.com/whatwg/html/issues/929 for details.

closed time in a day

annevk

issue commentw3c/FileAPI

Store HTTPS state on Blob objects

We achieved an even better solution, removal of HTTPS state: https://github.com/whatwg/fetch/pull/1067.

annevk

comment created time in a day

push eventwhatwg/html

Anne van Kesteren

commit sha 7c544596edc6e4d0a3468d96458f4267faf30cd4

registerProtocolHandler(): correct placement of note %s inside a scheme/host/port causes the URL parser to return failure. As discussed at https://github.com/web-platform-tests/wpt/pull/24886.

view details

push time in a day

delete branch whatwg/html

delete branch : annevk/rph-note

delete time in a day

PR merged whatwg/html

registerProtocolHandler(): correct placement of note clarification topic: custom protocols

%s inside a scheme/host/port causes the URL parser to return failure.

As discussed at https://github.com/web-platform-tests/wpt/pull/24886.

<!-- This comment and the below content is programatically generated. You may add a comma-separated list of anchors you'd like a direct link to below (e.g. #idl-serializers, #idl-sequence):

Don't remove this comment or modify anything below this line.
If you don't want a preview generated for this pull request,
just replace the whole of this comment's content by "no preview"
and remove what's below.

-->


<a href="https://whatpr.org/html/5788/system-state.html" title="Last updated on Aug 5, 2020, 2:59 PM UTC (3f45c01)">/system-state.html</a> ( <a href="https://whatpr.org/html/5788/230af41...3f45c01/system-state.html" title="Last updated on Aug 5, 2020, 2:59 PM UTC (3f45c01)">diff</a> )

+8 -8

0 comment

1 changed file

annevk

pr closed time in a day

issue commentw3c/webappsec-csp

Cross-realm eval() calls and 'unsafe-eval'

I tend to agree that trying to enforce cross-document consistency when documents (can) have different policies is not a good use of time.

cc @domenic

TomiBelan

comment created time in a day

PR opened whatwg/html

registerProtocolHandler(): correct placement of note clarification topic: custom protocols

%s inside a scheme/host/port causes the URL parser to return failure.

As discussed at https://github.com/web-platform-tests/wpt/pull/24886.

+8 -8

0 comment

1 changed file

pr created time in a day

create barnchwhatwg/html

branch : annevk/rph-note

created branch time in a day

pull request commentweb-platform-tests/wpt

Do not remove %s token when validating (un)registerProtocolHandler's URL

I will take care of the note.

chromium-wpt-export-bot

comment created time in a day

issue openedhttpwg/http-extensions

Incrementally Better Cookies vs RFC6265bis

Could someone please explain the history why we have two drafts that are somewhat incompatible with each other? Could the proposals that have support from multiple implementers move into RFC6265bis?

created time in a day

pull request commentwhatwg/html

Crbug 1060691 name noopener

I don't see how your behavior for /b and /c would happen as step 7 of "The rules for choosing a browsing context" would end up setting chosen and step 8 would never run.

ParisMeuleman

comment created time in a day

issue commentwhatwg/html

(un)registerProtocolHandler and localhost URLs

Thank you, it seems like we should be using "Is url potentially trustworthy?" instead here. I wonder if we should rename that to "secure URL" both for brevity and for consistency with secure contexts.

fred-wang

comment created time in a day

pull request commentweb-platform-tests/wpt

Do not remove %s token when validating (un)registerProtocolHandler's URL

  1. I think we should move that note as in fact the URL parser will return failure for such URLs (which I think does align with your tests expecting "SyntaxError", so that part is good):
    • https://jsdom.github.io/whatwg-url/#url=aHR0cCVzOi8vZXhhbXBsZS5jb20v&base=YWJvdXQ6Ymxhbms=
    • https://jsdom.github.io/whatwg-url/#url=aHR0cHM6Ly9leGFtJXNwbGUuY29tLw==&base=YWJvdXQ6Ymxhbms=
    • https://jsdom.github.io/whatwg-url/#url=aHR0cHM6Ly9leGFtcGxlLmNvbTolcy8=&base=YWJvdXQ6Ymxhbms=
  2. If you put %s in front of the scheme it will be parsed as a relative URL so you don't want to do that (or you might want to verify that it does work, but if an implementation did that wrong there'd be bigger problems):
    • https://jsdom.github.io/whatwg-url/#url=JXNodHRwczovL2V4YW1wbGUuY29tLw==&base=aHR0cHM6Ly9leGFtcGxlLmNvbS8=
  3. Firefox has a number of problems regarding exception order, so I suspect that's all there is to it here.
chromium-wpt-export-bot

comment created time in a day

issue commentw3c/input-events

order of events -- where defined

I suspect that for certain algorithms we need to allow for platform variation, but in general that would be the idea, yes.

johanneswilm

comment created time in a day

push eventweb-platform-tests/wpt

Anne van Kesteren

commit sha 7e29b1f803c0b596861731297a3c230cb7ec9229

Update docs/writing-tests/index.md Co-authored-by: jgraham <james@hoppipolla.co.uk>

view details

push time in a day

issue openedwhatwg/fetch

"get, decode, and split" seems incorrect

It's wrong for X-Content-Type-Options, but it seems wrong for other callers too. E.g., we don't want to support Content-Type: "text/html".

What needs to happen is that this algorithm preserves the quotes in the output. It needs to be aware of them to ensure they are matched up with another quote, but it shouldn't end up stripping them.

created time in a day

pull request commentwhatwg/html

Define X-Frame-Options processing

Daniel's comment did make me wonder if X-Content-Type-Options does in fact support "nosniff" as value. And it seems that fetch/nosniff/resources/x-content-type-options.json has no coverage for that.

domenic

comment created time in a day

push eventweb-platform-tests/wpt

Alexey Shvayka

commit sha af4259a9abaeda73de449a4ebcd15155feba8e22

Improve coverage for callback this value Includes CSSPaintCallback, MutationCallback, and IntersectionObserverCallback tests.

view details

push time in a day

PR merged web-platform-tests/wpt

Reviewers
Improve coverage for callback `this` value css-paint-api dom intersection-observer wg-webapps

There are similar tests for PerformanceObserver and ResizeObserver.

+102 -0

1 comment

3 changed files

shvaikalesh

pr closed time in a day

issue commentw3c/input-events

order of events -- where defined

The order of events follows from how the operations invoke each other (as the operations ultimately dispatch events).

johanneswilm

comment created time in a day

Pull request review commentwhatwg/html

Editorial: modernization of script fetching algorithms

 document.querySelector("button").addEventListener("click", bound);     also finish their work synchronously.</p>    </li> -   <li><p>Let <var>response</var> be <var>response</var>'s <span>unsafe response</span>.</p></li>+   <li><p>Set <var>response</var> to <var>response</var>'s <span>unsafe response</span>.</p></li>     <li>-    <p>If any of the following conditions are met, throw a <span>"<code>NetworkError</code>"</span>-    <code>DOMException</code>:</p>+    <p>If any of the following conditions are met:</p> -    <ul class="compact">+    <ul>      <li><p><var>response</var>'s <span data-x="concept-response-type">type</span> is "<code-     data-x="">error</code>"</p></li>+     data-x="">error</code>"; or</p></li>

We typically don't include "or" in these inline lists.

domenic

comment created time in a day

issue commentw3c/input-events

order of events -- where defined

It seems to me that @garykac's model allows for that. The high-level paste operation would at some point realize it needs to do fireInputEventsAndMaybeUpdateDOM and delegates to that.

johanneswilm

comment created time in a day

issue commentw3c/csswg-drafts

[css-ui-3] Can |CSS resize| change the value of the width/height properties which has been set with !Important?

Ah okay, I guess for this the way we could improve clarity then is by having the appropriate (low-level) calls into CSSOM to do just that.

mjf-g

comment created time in a day

Pull request review commentwhatwg/html

Add preservesPitch to HTMLMediaElement

 interface <dfn>MediaError</dfn> {   by script or by being changed directly by the user agent, e.g. in response to user control) the   user agent must <span>queue a media element task</span> given the <span>media element</span> to   <span data-x="concept-event-fire">fire an event</span> named <code-  data-x="event-media-ratechange">ratechange</code> at the <span>media element</span>.</p>+  data-x="event-media-ratechange">ratechange</code> at the <span>media element</span>. The user+  agent must process attribute changes smoothly and must not introduce any perceivable gaps or+  muting of playback in response.</p>++  <p>The <dfn><code data-x="dom-media-preservesPitch">preservesPitch</code></dfn> getter steps are+  to return true if a pitch-preserving algorithm is in effect during playback. The setter steps are+  to correspondingly switch the pitch-preserving algorithm on or off, without any perceivable gaps+  or muting of playback. By default, such a pitch-preserving algorith must be in effect (i.e., the+  getter must initially return true).</p>

This all feels rather implicit, but I guess it works. However, the i.e. bit shouldn't contain a requirement as it was already required. It should be a statement of fact.

domenic

comment created time in a day

Pull request review commentwhatwg/html

Add preservesPitch to HTMLMediaElement

 interface <dfn>MediaError</dfn> {     <p>Can be set, to change the rate of playback.</p>    </dd> +   <dt><var>media</var> . <code subdfn data-x="dom-media-preservesPitch">preservesPitch</code></dt>++   <dd>+

No additional newlines please.

domenic

comment created time in a day

Pull request review commentwhatwg/html

Add preservesPitch to HTMLMediaElement

 interface <dfn>MediaError</dfn> {     <p>Can be set, to change the rate of playback.</p>    </dd> +   <dt><var>media</var> . <code subdfn data-x="dom-media-preservesPitch">preservesPitch</code></dt>++   <dd>++    <p>Returns true if pitch-preserving algorithms are used when the <code+    data-x="dom-media-playbackRate">playbackRate</code> is not 1.0. The default value is true.</p>++    <p>Can be set to false to have the <span>media resource</span>'s audio pitch change up or down+    correspondingly to the <code data-x="dom-media-playbackRate">playbackRate</code>. This is useful
    depending on the <code data-x="dom-media-playbackRate">playbackRate</code>. This is useful

?

domenic

comment created time in a day

pull request commentwhatwg/html

Define navigation parameter: "replacement enabled"

I don't think this change on its own is good as it's really about session history (navigate just passes it on). I think we have to ensure that navigate always passes it to session history (and it's non-optional) if we want to move where it is defined.

jugglinmike

comment created time in a day

issue commentwhatwg/infra

Sketch out prose for algorithm definitions.

@jugglinmike I think in those cases you could also switch to returning a tuple using (value1, value2) and then a caller would use returnValue[0] only or some such.

mikewest

comment created time in 2 days

Pull request review commentwhatwg/infra

Specify algorithm declarations

 as well as the single algorithm would. Therefore performance is best left as a f over.  +<h3 id=declaration>Declaration</h3>++<p>Algorithm names are usually verb phrases, but sometimes are given names that emphasize their+standalone existence, so that standards and readers can refer to the algorithm more idiomatically.++<p class=example id=example-algorithm-declaration-names>Some algorithm names in the latter category+include "attribute change steps", "internal module script graph fetching procedure", and "overload+resolution algorithm".++Declare algorithms by stating their name, parameters, and return type, in the following form:++<blockquote>+To <dfn ignore>[algorithm name]</dfn>, given a [type1] <var ignore>[parameter1]</var>, a [type2]+<var ignore>[parameter2]</var>, &hellip;, perform the following steps. They return a [return type].+</blockquote>++<p>(For non-verb phrase algorithm names, use "To perform the+<dfn ignore>[algorithm name]</dfn>&hellip;".)++<p class=example id=example-algorithm-declaration>To <dfn ignore>parse an awesome format</dfn> given+a [=byte sequence=] <var ignore>bytes</var>, perform the following steps. They return a+[=string=] or null.

Typically we lead with null.

domenic

comment created time in 2 days

Pull request review commentwhatwg/infra

Specify algorithm declarations

 as well as the single algorithm would. Therefore performance is best left as a f over.  +<h3 id=declaration>Declaration</h3>++<p>Algorithm names are usually verb phrases, but sometimes are given names that emphasize their+standalone existence, so that standards and readers can refer to the algorithm more idiomatically.++<p class=example id=example-algorithm-declaration-names>Some algorithm names in the latter category+include "attribute change steps", "internal module script graph fetching procedure", and "overload+resolution algorithm".++Declare algorithms by stating their name, parameters, and return type, in the following form:++<blockquote>+To <dfn ignore>[algorithm name]</dfn>, given a [type1] <var ignore>[parameter1]</var>, a [type2]+<var ignore>[parameter2]</var>, &hellip;, perform the following steps. They return a [return type].

What's the ellipsis for?

domenic

comment created time in 2 days

Pull request review commentwhatwg/infra

Specify algorithm declarations

 as well as the single algorithm would. Therefore performance is best left as a f over.  +<h3 id=declaration>Declaration</h3>++<p>Algorithm names are usually verb phrases, but sometimes are given names that emphasize their+standalone existence, so that standards and readers can refer to the algorithm more idiomatically.++<p class=example id=example-algorithm-declaration-names>Some algorithm names in the latter category+include "attribute change steps", "internal module script graph fetching procedure", and "overload+resolution algorithm".++Declare algorithms by stating their name, parameters, and return type, in the following form:
<p>Declare algorithms by stating their name, parameters, and return type, in the following form:

(It seems this is a more general problem. I'd prefer we stick to explicit <p> rather than a mix.)

domenic

comment created time in 2 days

more