Navigation: Add optimization to ignore duplicate navigations by llannasatoll · Pull Request #11765 · whatwg/html

llannasatoll · 2025-10-08T06:48:17Z

As proposed in #11743, this prevents an ongoing navigation from being canceled by a new, identical navigation that is initiated within the threshould. This introduces the navigation initiation snapshot params struct, which is used to capture the state of the most recent navigation attempt, allowing the user agent to compare it with subsequent requests at the beginning of the navigation.

At least two implementers are interested (and none opposed):
- Chromium (#11743)
- …
Tests are written and can be reviewed and commented upon at:
- https://chromium-review.googlesource.com/c/chromium/src/+/5856811
Implementation bugs are filed:
- Chromium: https://crbug.com/366060351
- Gecko: …
- WebKit: …
Corresponding HTML AAM & ARIA in HTML issues & PRs:
MDN issue is filed: …
The top of this comment includes a clear commit message to use.

/browsing-the-web.html ( diff )
/document-sequences.html ( diff )
/dom.html ( diff )

llannasatoll · 2026-04-16T02:25:08Z

@rakina Could you take another look? I've added a new struct navigation initiation snapshot params to store information about the most recent navigation so that we can compare the navigations?

rakina

Thanks. I think it mostly looks good. There might be some wording around the cookie checks that might need to be updated later depending on the cookie check discussions.

rakina · 2026-04-22T07:26:18Z

+      <li><p><var>cspNavigationType</var> is "<code data-x="">other</code>" and
+      <var>navigationInitiationSnapshotParams</var>'s <span
+      data-x="nav-initiation-snapshot-params-entry-list">entry list</span> is null;</p></li>


Should this both be checking the entry list? Also should this check documentResource as well?

I've updated the PR to check both, thanks! (I was implicitly checking the entry list via cspNavigationType, but explicitly checking formDataEntryList should be much clearer).

rakina · 2026-06-05T07:52:44Z

+
+      <li><p>If <var>navigationStartTime</var> minus <var>navigationInitiationSnapshotParams</var>'s
+      <span data-x="nav-initiation-snapshot-params-nav-start-time">navigation start time</span> is
+      less than a user-agent provided threshold;</p></li>


There's some suggestion to set this to an actual value, so maybe use the value used in Chromium (3s?): #11743 (comment)

I've updated the threshold to a hardcoded 3 seconds. (also, please let me know if you think we should add a note explaining why 3s.)

rakina · 2026-06-05T07:57:47Z

+   potentially changed</span> to true.</p>
+
+   <p class="note">Other operations that could set cookies, such as the <a
+   href="https://wicg.github.io/cookie-store/">Cookie Store API</a> or receiving a Set-Cookie


We probably need to create a Cookie Store API PR as well. Maybe we need to change this part: https://cookiestore.spec.whatwg.org/#set-cookie-algorithm to get the document's node navigable and update this bit.

I see, sure! I'll create a separate PR in the cookiestore repo, and link it here once it's up.

llannasatoll

Thanks for the review, Rakina! Could you take another look?

llannasatoll · 2026-06-08T02:50:45Z

+
+      <li><p>If <var>navigationStartTime</var> minus <var>navigationInitiationSnapshotParams</var>'s
+      <span data-x="nav-initiation-snapshot-params-nav-start-time">navigation start time</span> is
+      less than a user-agent provided threshold;</p></li>


I've updated the threshold to a hardcoded 3 seconds. (also, please let me know if you think we should add a note explaining why 3s.)

llannasatoll · 2026-06-08T02:51:21Z

+   potentially changed</span> to true.</p>
+
+   <p class="note">Other operations that could set cookies, such as the <a
+   href="https://wicg.github.io/cookie-store/">Cookie Store API</a> or receiving a Set-Cookie


I see, sure! I'll create a separate PR in the cookiestore repo, and link it here once it's up.

rakina

I think this looks good at this point, so let me add reviewers from other browsers who commented in #11743 (feel free to reassign/unassign)

This CL introduces a `cookie_modification_count` in Document and NavigationClient. When a renderer-initiated navigation starts, NavigationClient captures a snapshot of the document's current cookie modification count. Subsequent checks for duplicate navigations now compare this captured count with the current document's count. If the counts differ, the navigation is not considered a duplicate, preventing it from being ignored. New browser tests are added to verify this behavior for both `document.cookie` and `cookieStore.set` modifications. Spec Link: whatwg/html#11765 Bug: 366060351 Change-Id: If5b3e2da6ca233c0eb85df6b08caa6a5959dcfe2 Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/7787903 Reviewed-by: Rakina Zata Amni <rakina@chromium.org> Reviewed-by: Scott Haseley <shaseley@chromium.org> Commit-Queue: Anna Sato <annasato@chromium.org> Reviewed-by: Mingyu Lei <leimy@chromium.org> Cr-Commit-Position: refs/heads/main@{#1648863}

llannasatoll · 2026-07-03T09:00:49Z

Hi @zcorpan @annevk, gentle ping for review when you have time.

annevk

Overall this looks like it does what it set out to do. I found a lot of formatting issues and a couple more substantive issues.

annevk · 2026-07-03T09:36:06Z

+      <span>WebDriver BiDi navigation status</span> whose <span
+      data-x="navigation-status-id">id</span> is <var>navigationId</var>, <span
+      data-x="navigation-status-status">status</span> is "<code
+      data-x="navigation-status-canceled">canceled</code>", and <span


Should we give a more detailed reason?

I checked the definition of WebDriver BiDi navigation status, and it currently doesn't have a reason or message field. Should we expand the struct in this PR to include an optional reason, or leave it as "canceled" for now?

llannasatoll

Thank you for the review! Could you take another look?

llannasatoll · 2026-07-07T11:33:11Z

+      <span>WebDriver BiDi navigation status</span> whose <span
+      data-x="navigation-status-id">id</span> is <var>navigationId</var>, <span
+      data-x="navigation-status-status">status</span> is "<code
+      data-x="navigation-status-canceled">canceled</code>", and <span


I checked the definition of WebDriver BiDi navigation status, and it currently doesn't have a reason or message field. Should we expand the struct in this PR to include an optional reason, or leave it as "canceled" for now?

annevk · 2026-07-17T11:02:38Z

+    potentially changed</span> to true.</p>
+
+    <p class="note">Other operations that could set cookies, such as the <cite>Cookie Store API</cite>,
+    also cause the <span>navigation conditions potentially changed</span> to be set to true.</p>


I'm not seeing a PR for this over at https://github.com/whatwg/cookiestore/pulls

Also, what does "such as" mean?

I was waiting for this PR to land first to avoid cross-reference build errors, but now I've created whatwg/cookiestore#302 with a tentative reference in <pre class="anchors"> (maybe can be removed it once this PR landed?)

Also, what does "such as" mean?

Updated the note to be explicit: "The Cookie Store API also sets navigation conditions potentially changed to true when setting cookies."

rakina mentioned this pull request Oct 8, 2025

Proposal: Add optimization to ignore "duplicate" navigations #11743

Open

rakina reviewed Oct 17, 2025

View reviewed changes

Comment thread source Outdated

llannasatoll force-pushed the ignore-duplicate-navigation branch from 27be0e6 to 4d255eb Compare November 17, 2025 05:05

llannasatoll requested a review from rakina November 17, 2025 05:23

llannasatoll force-pushed the ignore-duplicate-navigation branch 3 times, most recently from cac111e to 1ffc5d1 Compare April 15, 2026 11:44

rakina reviewed Apr 22, 2026

View reviewed changes

llannasatoll force-pushed the ignore-duplicate-navigation branch from f07484b to 7e7c910 Compare June 4, 2026 07:25

rakina reviewed Jun 5, 2026

View reviewed changes

llannasatoll commented Jun 8, 2026

View reviewed changes

rakina requested review from annevk and zcorpan June 8, 2026 05:47

rakina approved these changes Jun 8, 2026

View reviewed changes

llannasatoll mentioned this pull request Jun 22, 2026

WG New Spec: Ignore Duplicate Navigations w3ctag/design-reviews#1240

Open

2 tasks

annevk reviewed Jul 3, 2026

View reviewed changes

llannasatoll added 6 commits July 7, 2026 08:30

feat(navigation): Ignore duplicate navigations

a533b78

Refine duplicate navigation check

7363756

lint and fix typo

d2382ce

add documentResource check

b1404d7

Update duplicate navigation threshold to 3s and drop fetch check

21192c3

Fix Cookie Store API reference

38fe050

llannasatoll force-pushed the ignore-duplicate-navigation branch from 0675e6d to 38fe050 Compare July 7, 2026 08:34

Apply formatting fixes and restructure setter based on review

ff26020

llannasatoll commented Jul 7, 2026

View reviewed changes

llannasatoll requested a review from annevk July 9, 2026 10:20

noamr added the agenda+ To be discussed at a triage meeting label Jul 16, 2026

address a couple issues

2a57588

annevk reviewed Jul 17, 2026

View reviewed changes

llannasatoll mentioned this pull request Jul 21, 2026

Set navigation conditions potentially changed when setting cookie whatwg/cookiestore#302

Open

5 tasks

Clarify Cookie Store API reference note

7aa3db5

jstenback removed the agenda+ To be discussed at a triage meeting label Jul 23, 2026

Uh oh!

Conversation

llannasatoll commented Oct 8, 2025 • edited by pr-preview Bot Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

Uh oh!

llannasatoll commented Apr 16, 2026

Uh oh!

rakina left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

llannasatoll left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

rakina left a comment

Choose a reason for hiding this comment

Uh oh!

llannasatoll commented Jul 3, 2026

Uh oh!

annevk left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

llannasatoll left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Reviewers

Assignees

Labels

Milestone

Development

Uh oh!

5 participants

llannasatoll commented Oct 8, 2025 •

edited by pr-preview Bot

Loading