Trusted types attributes #1268

lukewarlow · 2024-03-27T19:22:16Z

This calls the get Trusted Types-compliant attribute value algorithm from Trusted Types (w3c/trusted-types#418) from attribute's change, append, and replace.

Changed the signature of setAttribute and setAttributeNS to accept Trusted Types as values. The underlying Attr node's values continue to be DOMString, so moving nodes across elements or adding standalone attributes to elements can cause TT violations. This matches WPT tests and the Chromium's implementation.

See and #789. Supercedes #809 and #1247

At least two implementers are interested (and none opposed):
- Chromium
- Gecko
Tests are written and can be reviewed and commented upon at:
- https://github.com/web-platform-tests/wpt/tree/master/trusted-types
Implementation bugs are filed:
- Chromium:https://bugs.chromium.org/p/chromium/issues/detail?id=739170
- Gecko: https://bugzilla.mozilla.org/show_bug.cgi?id=1508286
- WebKit: https://bugs.webkit.org/show_bug.cgi?id=266630
MDN issue is filed: …
The top of this comment includes a clear commit message to use.

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

Preview | Diff

lukewarlow · 2024-03-27T19:22:41Z

This is a clone of #1247 where I will finish any outstanding work to get this across the line

lukewarlow · 2024-04-11T12:21:53Z

See also #1258 which is another integration point with the DOM spec that we need for TT.

dom.bs

otherdaniel · 2024-04-15T17:13:57Z

dom.bs

+   <li><p><a>Validate and set attribute value</a> <var>attr</var>'s <a for="Attr">value</a> for
+   <var>attr</var> with <var>element</var>.
+
+   <li><p>If <var>element</var> <a lt="has an attribute">has</a> an <a>attribute</a>


I still find it difficult to wrap my head around this logic. I think conceptually, we want to have the old value -- from before the call, and thus before a default policy might have mucked with it. That's what should go into the change attribute logic. But once we have that, I'm not sure why we'd need to throw an exception here. I'm not sure why we'd have to care whether the default policy does anything funny with the attribute in the mean time.

I've read through this spec path and I believe that, if the default policy removes the existing attribute node, then this will call replace an attribute, which in turn calls replace a list item, which results in a no-op because the old value no longer exists to be replaced.

So I think in this situation you're probably right the spec doesn't need to do anything, will make that change.

Having said that I'm slightly uneasy about stuff like attribute change steps still firing and how that all works spec wise.

Both Chromium and WebKit don't actually follow the spec 100% here and so I think would actually result in a different behaviour to this spec (the index lookup happens after the TT call) so that's also not ideal.

lukewarlow · 2024-04-22T13:18:23Z

I think I've addressed all the comments from #1247.

I do want to point out Chrome and WebKit don't (or at least not in a way obvious to me) 100% follow the flow of the spec and as a result this may result in differences specifically in weird cases with attribute mutation.

So that bit especially it would be good to get feedback on.

It's also worth being aware that like Chromium's implementation this spec means that certain ways to update a nodes value don't work with a trusted type object as a user might expect. (e.g. iframe.getAttributeNode('srcdoc').value = trustedHTMLObj; will throw unless allowed by a default policy). I think in pratice this will be fine, and in some cases is the only real option we have without nodes keeping track of whether they're trusted or not (which would add lots of complexity)

annevk

This change is either incomplete or makes many cosmetic changes that would be best proposed separately as they confuse me quite a bit.

dom.bs

lukewarlow · 2024-04-22T15:47:56Z

This change is either incomplete or makes many cosmetic changes that would be best proposed separately as they confuse me quite a bit.

Apologies I thought I'd reverted all of these changes but I missed a few, it's because I changed stuff and then changed it back and this led to some wonky diffs. Have hopefully reverted all of these unnecessary changes

annevk · 2024-04-22T16:07:37Z

dom.bs

+ <li><p>If <var>attribute</var>'s <a for=Attr>element</a> <a lt="has an attribute">has</a>
+ an <a>attribute</a> <var>attribute</var>, then <a>handle attribute changes</a> for
+ <var>attribute</var> with <var>attribute</var>'s <a for=Attr>element</a>, <var>oldValue</var>, and
+ <var>value</var>.


Should this be value or attribute's value? They can be different, right?

dom.bs

lukewarlow · 2024-05-07T15:54:52Z

@otherdaniel, @annevk , and @smaug---- regarding the case where the default policy changes assumptions about the existence of an attribute mid-way through what would you prefer the spec say to do? Currently I've specced to throw, but Chromium currently re-looks up the index (spec doesn't explicitly work on an index basis but Chromium and WebKit do)

annevk · 2024-05-13T14:58:26Z

Attributes are stored in a list and those do have indices per Infra. What am I missing?

lukewarlow · 2024-05-13T15:03:24Z

Sorry I mean algorithmically the spec and implementations don't follow the same flow. So it's trickier to reason between the spec and implementation. This might just be my lack of familiarity with these APIs too.

annevk · 2024-05-13T15:11:38Z

The path of least resistance is prolly matching Chromium. Introducing new paths that throw is always risky. If you are looking for guidance as to how, I'd need quite a bit more context to provide helpful suggestions.

smaug----

Some questions. I could be missing something in TT spec which somehow covers the problematic cases, but please at least explain and ask review again.

smaug---- · 2024-07-04T13:46:16Z

dom.bs

+<ol>
+ <li><p>Return the result of calling
+ <a abstract-op>get Trusted Types-compliant attribute value</a> for <var>attribute</var>, with
+ <var>element</var>, <var>value</var>. [[!TRUSTED-TYPES]]


So this may throw. Do the callers of this method handle that in some way?

They will end up rethrowing the exception which is what's expected.

smaug---- · 2024-07-04T14:08:56Z

dom.bs

+ <var>attr</var>'s <a for=Attr>value</a> for <var>attr</var>, with <var>element</var>.
+
+ <li><p>Set <var>attr</var>'s <a for=Attr>value</a> to <var>verifiedValue</var>.
+


Hmm, so doing the verification may run scripts. And that means oldAttr might not be anymore in the element it used to be. Could that cause issues? Could the value be validated first for certain kind of element but then used on some other kind of element?

The default policy doesn't provide context about which element an attribute is set on only the name of the attribute. In this case this algorithm is triggered by APIs such as setAttributeNode or setNamedItem.

So I don't think there's anything that can happen here that's too bad. Also any mechanism you use inside of the default policy will itself trigger the default policy so it should be fine?

smaug---- · 2024-07-04T14:22:38Z

dom.bs

+    <li><p>Let <var>verifiedValue</var> be the result of calling <a>verify attribute value</a>
+    <var>value</var> for <var>attribute</var>, with <a>this</a>.
+
+   <li><p><a lt="change an attribute">Change</a> <var>attribute</var> to <var>verifiedValue</var>.


Doesn't this lead to effectively null pointer crashes in algorithms if attribute has been moved out from an element by the tt callback?

I guess potentially if some sub algorithm of change an attribute is assuming that element must exist? Fwiw in practice this doesn't cause issues from what I can see.

Not sure what the best fix here would be?

annevk · 2024-08-13T13:47:23Z

Now that 809bfa2 is merged this should add the Infrastructure changes back in again as this now introduces the dependency on Trusted Types.

See w3c/trusted-types#418 and whatwg#789. Supercedes PR whatwg#809.

…an attribute

… Chrome.

…erved in Chrome." This reverts commit 2f00466.

This reverts commit 6e40964.

https://bugs.webkit.org/show_bug.cgi?id=275352 Reviewed by Darin Adler. The DOM spec PR no longer enforced Trusted Types within toggleAttribute so this removes that from the implementation. See whatwg/dom#1268 * LayoutTests/imported/w3c/web-platform-tests/trusted-types/Element-toggleAttribute-expected.txt: Added. * LayoutTests/imported/w3c/web-platform-tests/trusted-types/Element-toggleAttribute.html: Added. * Source/WebCore/dom/Element.cpp: (WebCore::Element::toggleAttribute): Canonical link: https://commits.webkit.org/279950@main

This is for consistency with how it is done in the HTML specification and in the PR to DOM (whatwg/dom#1268). This is not web-exposed but may matter for implementations. For example, they may generate different C++ structure to represent `(TrustedHTML or DOMString)` and `(DOMString or TrustedHTML)`.

lukewarlow force-pushed the trusted-types-attributes branch 2 times, most recently from 3562fcb to ac5b4aa Compare April 10, 2024 15:52

lukewarlow mentioned this pull request Apr 11, 2024

Draft integration with Trusted Types, take 2. #1247

Closed

5 tasks

lukewarlow force-pushed the trusted-types-attributes branch from 524d8cd to f8877b6 Compare April 11, 2024 12:18

lukewarlow marked this pull request as ready for review April 11, 2024 12:21

lukewarlow marked this pull request as draft April 15, 2024 10:22

lukewarlow force-pushed the trusted-types-attributes branch from f8877b6 to ee9915e Compare April 15, 2024 11:45

otherdaniel reviewed Apr 15, 2024

View reviewed changes

dom.bs Outdated Show resolved Hide resolved

otherdaniel reviewed Apr 15, 2024

View reviewed changes

lukewarlow marked this pull request as ready for review April 22, 2024 12:53

lukewarlow requested review from annevk and smaug---- April 22, 2024 13:13

lukewarlow force-pushed the trusted-types-attributes branch from ba80870 to 36476b5 Compare April 22, 2024 13:57

lukewarlow mentioned this pull request Apr 22, 2024

[Meta] Upstream changes w3c/trusted-types#476

Open

annevk reviewed Apr 22, 2024

View reviewed changes

dom.bs Outdated Show resolved Hide resolved

dom.bs Outdated Show resolved Hide resolved

dom.bs Outdated Show resolved Hide resolved

dom.bs Outdated Show resolved Hide resolved

dom.bs Outdated Show resolved Hide resolved

lukewarlow force-pushed the trusted-types-attributes branch from 370f5c9 to 4421d50 Compare April 22, 2024 15:46

lukewarlow force-pushed the trusted-types-attributes branch 2 times, most recently from 1d42460 to 1eeaf00 Compare April 22, 2024 15:52

annevk reviewed Apr 22, 2024

View reviewed changes

lukewarlow force-pushed the trusted-types-attributes branch 2 times, most recently from 4089eaf to 02db8c7 Compare May 16, 2024 15:41

lukewarlow requested a review from otherdaniel May 16, 2024 15:43

smaug---- reviewed Jul 4, 2024

View reviewed changes

lukewarlow requested a review from smaug---- July 16, 2024 13:58

koto and others added 23 commits August 13, 2024 14:56

Draft integration with Trusted Types, take 2.

184fbb3

See w3c/trusted-types#418 and whatwg#789. Supercedes PR whatwg#809.

Added integration in 'set an attribute value'.

ebae0ea

Removed stringification.

3dc6eba

Fixed formatting.

f88ea29

Fixed copy-paste error in setAttribute() and indentation.

966291f

Fix build

065ee2e

Change code to pass sink values through to TT code.

19e8f54

Move validate and set attribute into append an attribute

964b8cd

Remove sink values from spec

8db6d26

Revert changes to setAttributeNode and setAttributeNodeNS method steps

271a670

Remove throw from set an attribute and move TT check down to replace …

2427ef4

…an attribute

Revert unneeded changes

b5de89b

Address comment

cd0415b

Add missing and

d2adea4

Address some comments

99ae585

Re-add early return

f76755c

Change dfn to use given rather than with

e5be35d

Update how the enforcement is done

982ef71

Add tentative enforcement for toggleAttribute

f584e23

Update enforcement for toggleAttribute to match behaviour observed in…

9b74d93

… Chrome.

Revert "Update enforcement for toggleAttribute to match behaviour obs…

d836a23

…erved in Chrome." This reverts commit 2f00466.

Revert "Add tentative enforcement for toggleAttribute"

b0ecd77

This reverts commit 6e40964.

Reintroduce infrastructure changes

d6b3b5a

lukewarlow force-pushed the trusted-types-attributes branch from 1ec907b to d6b3b5a Compare August 13, 2024 13:58

fred-wang mentioned this pull request Nov 15, 2024

Put trusted type first when declaring union in IDL files w3c/trusted-types#564

Open

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Trusted types attributes #1268

Trusted types attributes #1268

lukewarlow commented Mar 27, 2024 •

edited by pr-preview bot

Loading

lukewarlow commented Mar 27, 2024

lukewarlow commented Apr 11, 2024

otherdaniel Apr 15, 2024

lukewarlow Apr 22, 2024

lukewarlow Apr 22, 2024

lukewarlow commented Apr 22, 2024

annevk left a comment

lukewarlow commented Apr 22, 2024

annevk Apr 22, 2024

lukewarlow commented May 7, 2024 •

edited

Loading

annevk commented May 13, 2024

lukewarlow commented May 13, 2024

annevk commented May 13, 2024

smaug---- left a comment

smaug---- Jul 4, 2024

lukewarlow Jul 16, 2024

smaug---- Jul 4, 2024

lukewarlow Jul 16, 2024

smaug---- Jul 4, 2024

lukewarlow Jul 16, 2024

annevk commented Aug 13, 2024

		<var>attr</var>'s <a for=Attr>value</a> for <var>attr</var>, with <var>element</var>.

		<li><p>Set <var>attr</var>'s <a for=Attr>value</a> to <var>verifiedValue</var>.

Trusted types attributes #1268

Are you sure you want to change the base?

Trusted types attributes #1268

Conversation

lukewarlow commented Mar 27, 2024 • edited by pr-preview bot Loading

lukewarlow commented Mar 27, 2024

lukewarlow commented Apr 11, 2024

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

lukewarlow commented Apr 22, 2024

annevk left a comment

Choose a reason for hiding this comment

lukewarlow commented Apr 22, 2024

Choose a reason for hiding this comment

lukewarlow commented May 7, 2024 • edited Loading

annevk commented May 13, 2024

lukewarlow commented May 13, 2024

annevk commented May 13, 2024

smaug---- left a 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 commented Aug 13, 2024

lukewarlow commented Mar 27, 2024 •

edited by pr-preview bot

Loading

lukewarlow commented May 7, 2024 •

edited

Loading