Define preconnect and dns-prefetch with CSP #1620

noamr · 2023-03-20T17:10:31Z

Preconnect & dns-prefetch test a fake prefetch request for CSP, and then obtain a connection/resolve an origin.

At least two implementers are interested (and none opposed):
- …
- …
Tests are written and can be reviewed and commented upon at:
- …
Implementation bugs are filed:
- Chromium: …
- Gecko: …
- WebKit: …
- Deno (not for CORS changes): …
MDN issue is filed: …

(See WHATWG Working Mode: Changes for more details.)

fetch.bs

annevk

It's unfortunate these have to be separate algorithms. I was hoping we could do CSP checks as part of the normal "obtain a connection" routine.

fetch.bs

noamr · 2023-04-24T10:12:34Z

It's unfortunate these have to be separate algorithms. I was hoping we could do CSP checks as part of the normal "obtain a connection" routine.

We can do that but then there would be double CSP checks, one for the request and one for the "fake" request, or we can add a flag to obtain/resolve of whether to perform the CSP check.

annevk · 2023-05-04T08:13:27Z

fetch.bs

-  and <a href="https://en.wikipedia.org/wiki/Proxy_auto-config">proxy auto-config (PAC)</a> come
-  into play. The "<code>DIRECT</code>" value means to not use a proxy for this particular
-  <var>url</var>.
+ <li><p>Let <var>proxies</var> of calling <a>resolve proxies</a>.


This is missing words.

What words are missing? The note was moved to resolve proxies.

annevk · 2023-05-04T08:14:01Z

fetch.bs

+<p>To <dfn>resolve proxies</dfn>, return the result of finding proxies for <var>url</var> in an
+<a>implementation-defined</a> manner. If there are no proxies, return « "<code>DIRECT</code>" ».


Where does url come from here? If it needs to be an argument, let's make it explicit.

annevk · 2023-05-04T08:14:45Z

fetch.bs

@@ -3023,6 +3027,75 @@ details of reused connections are not exposed and time values are coarsened.
 </div>


+<h3 id=preemptive-connections>Preemptive connections</h3>
+
+<div algorithm="preemmptively-obtain-a-connection">


Does algorithm need this value? Normally it works without.

annevk · 2023-05-04T08:15:10Z

fetch.bs

+  <p><a>Check CSP for preemptive operation</a> with <var>environment</var> and <var>url</var>. If
+  that returns <b>allowed</b>, <a>obtain a connection</a> with <var>key</var>, <var>url</var> and
+  <var>credentials</var>.


Oxford comma.

annevk · 2023-05-04T08:16:24Z

fetch.bs

+<div algorithm="check-preemptive-operation-csp">
+<p>To <dfn>check CSP for preemptive operation</dfn>, given an
+<a>environment settings object</a> <var>environment</var>, and a <a for=/>URL</a> <var>url</var>,
+run these steps:


Perhaps this can have a more descriptive name such as "does CSP allow a connection" and we normalize the return value from CSP to true/false while we're here.

Seems this went unaddressed?

fetch.bs

annevk · 2023-05-08T10:50:19Z

fetch.bs

+</ol>
+</div>
+
+<div algorithm="check-preemptive-operation-csp">


This attribute value isn't needed.

annevk · 2023-05-08T10:50:44Z

fetch.bs

+</ol>
+</div>
+
+<div>


This needs an algorithm attribute.

annevk · 2023-05-08T10:51:48Z

fetch.bs

+</div>
+
+<div algorithm="check-preemptive-operation-csp">
+<p>To <dfn>Should connection be allowed by Content Security Policy?</dfn>, given an


This algorithm needs a better name. E.g., "determine if Content Security allows connection". The name shouldn't have a question mark and should read naturally after "To" if you're using to to lead it in.

annevk · 2023-05-08T10:52:00Z

fetch.bs

+<div algorithm="check-preemptive-operation-csp">
+<p>To <dfn>Should connection be allowed by Content Security Policy?</dfn>, given an
+<a>environment settings object</a> <var>environment</var>, and a <a for=/>URL</a> <var>url</var>,
+run these steps:


No need for "run these steps" when you lead it in with "To".

annevk · 2023-05-08T10:52:41Z

fetch.bs

+ <li><p>Return the result of running <a>should request be blocked by Content Security Policy?</a>
+ given <var>request</var>.


We should put the result in a variable first and then convert that to a boolean.

See whatwg/html#9035

Co-authored-by: Valentin Gosu <[email protected]>

Co-authored-by: Anne van Kesteren <[email protected]>

annevk

Apologies for the delay here. Found a couple more issues to address.

annevk · 2023-09-28T11:48:23Z

fetch.bs

+  <p class=note>The user agent can attempt to initiate a preconnect and perform the full
+  connection handshake (DNS+TCP for HTTP, and DNS+TCP+TLS for HTTPS origins) whenever possible, but
+  is allowed to elect to perform a partial handshake (DNS only for HTTP, and DNS or DNS+TCP for
+  HTTPS origins), or skip it entirely, due to resource constraints or other reasons.</p>
+
+  <p class=note>The optimal number of connections per origin is dependent on the negotiated
+  protocol, users current connectivity profile, available device resources, global connection
+  limits, and other context specific variables. As a result, the decision for how many connections
+  should be opened is deferred to the user agent.</p>


These notes feel like normative material. One even contains "should"! In the step above where we obtain the connection perhaps we should have a "with these caveats" ending that then leads into some bullet points?

fetch.bs

Co-authored-by: Anne van Kesteren <[email protected]>

valenting reviewed Mar 20, 2023

View reviewed changes

fetch.bs Outdated Show resolved Hide resolved

noamr mentioned this pull request Mar 21, 2023

Govern dns-prefetch and preconnect with CSP whatwg/html#9035

Open

annevk reviewed Apr 24, 2023

View reviewed changes

noamr force-pushed the preempt branch from e3ae980 to 6cb2721 Compare April 24, 2023 08:15

annevk reviewed May 4, 2023

View reviewed changes

annevk reviewed May 8, 2023

View reviewed changes

noamr and others added 8 commits May 8, 2023 14:06

Define preconnect and dns-prefetch with CSP

ddaaef9

See whatwg/html#9035

Update fetch.bs

daab9c1

Co-authored-by: Valentin Gosu <[email protected]>

Update fetch.bs

2e4bae6

Co-authored-by: Anne van Kesteren <[email protected]>

Update fetch.bs

7e6e995

Co-authored-by: Anne van Kesteren <[email protected]>

PR nits

ef41a25

should

a85fdaa

A few editorial nits

c267d4e

Use boolean for algo

f64322a

noamr force-pushed the preempt branch from c005718 to f64322a Compare May 8, 2023 11:13

annevk reviewed Sep 28, 2023

View reviewed changes

noamr and others added 2 commits September 28, 2023 14:57

Update fetch.bs

b5bc78b

Co-authored-by: Anne van Kesteren <[email protected]>

Update fetch.bs

fa560eb

Co-authored-by: Anne van Kesteren <[email protected]>

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Define preconnect and dns-prefetch with CSP #1620

Define preconnect and dns-prefetch with CSP #1620

noamr commented Mar 20, 2023 •

edited by pr-preview bot

Loading

annevk left a comment

noamr commented Apr 24, 2023

annevk May 4, 2023

noamr May 7, 2023

annevk May 4, 2023

noamr May 7, 2023

annevk May 4, 2023

noamr May 7, 2023

annevk May 4, 2023

noamr May 7, 2023

annevk May 4, 2023

noamr May 7, 2023

annevk May 8, 2023

noamr May 8, 2023

annevk May 8, 2023

noamr May 8, 2023

annevk May 8, 2023

noamr May 8, 2023

annevk May 8, 2023

noamr May 8, 2023

annevk May 8, 2023

noamr May 8, 2023

annevk left a comment

annevk Sep 28, 2023

		<p>To <dfn>resolve proxies</dfn>, return the result of finding proxies for <var>url</var> in an
		<a>implementation-defined</a> manner. If there are no proxies, return « "<code>DIRECT</code>" ».

+              </ol>
+              </div>
+              <div>

		<li><p>Return the result of running <a>should request be blocked by Content Security Policy?</a>
		given <var>request</var>.

Define preconnect and dns-prefetch with CSP #1620

Are you sure you want to change the base?

Define preconnect and dns-prefetch with CSP #1620

Conversation

noamr commented Mar 20, 2023 • edited by pr-preview bot Loading

annevk left a comment

Choose a reason for hiding this comment

noamr commented Apr 24, 2023

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

annevk left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

noamr commented Mar 20, 2023 •

edited by pr-preview bot

Loading