webdriver: Evaluate script commands via the `WebView` API in servoshell #37663

longvatrong111 · 2025-06-24T05:33:55Z

Let WebDriverCommandMsg::ScriptCommand goes through embedder first.
Give embedder the ability to release webdriver from waiting for a response of ExecuteScript.

Tests: https://github.com/longvatrong111/servo/actions/runs/16071375821
No regression compared to CI run on main branch.

Fixes: #37370

cc: @xiaochengh

wusyong · 2025-06-24T06:46:50Z

What's the motivation here. Is there any related issue?

longvatrong111 · 2025-06-24T06:50:03Z

What's the motivation here. Is there any related issue?

This PR is a follow up of #36714 (comment)

yezhizhen · 2025-06-24T07:18:43Z

I am a bit worried about forward_webdriver_command. Commands that used to directly go to Constellation, now goes to embedder which does nothing but forwards it to Constellation. For example, verify_browsing_context_is_open.

I notice that the test execution speed goes down significantly after #36714. There seems to be an extra round of waiting after each subtest. Previously, it almost has no delay. (For example, many tests which used to took 4 seconds now took 30+ seconds. Probably because verify_browsing_context_is_open now goes through extra forwarding?)
cc @mrobinson @jdm @xiaochengh

longvatrong111 · 2025-06-24T07:24:20Z

I am a bit worried about forward_webdriver_command. Commands that used to directly go to Constellation, now goes to embedder which does nothing but forwards it to Constellation.

I notice that the test execution speed goes down significantly after #36714. There seems to be an extra round of waiting after each subtest. Previously, it almost has no delay. (For example, close_window test which used to took 4 seconds now took 20 seconds. And this is a crucial part for every other test.) cc @mrobinson @jdm @xiaochengh

We need to move all commands to servoshell then evaluate again.
The reason for ScriptCommand was discussed here #36714 (comment).
The extra time is caused by timeout because of no response, which is an issue we can fix later after moving all command. 1 extra forwarding can't take 25s.

yezhizhen · 2025-06-24T07:28:51Z

We need to move all commands to servoshell then evaluate again. The reason for ScriptCommand was discussed here #36714 (comment).

I knew that. But I never think it is good idea to move commands that just does forwarding.

The extra time is caused by timeout because of no response, which is an issue we can fix later after moving all command. 1 extra forwarding can't take 25s.

No it is not TIMEOUT. Please run the test and check for yourself. Maybe checkout before #36714. Please read "There seems to be an extra round of waiting after each subtest."

longvatrong111 · 2025-06-24T07:45:32Z

I knew that. But I never think it is good idea to move commands that just does forwarding.

It serves the purpose of synchronization. If script event has shorter route, it may be done out of order with others. In that case, we need to add a wait as well.

No it is not TIMEOUT. Please run the test and check for yourself. Maybe rollback before #36714. Please read "There seems to be an extra round of waiting after each subtest."

Maybe you want to run with #37555 or #37665
It doesn't take 30s. All webdriver context commands work together, it's too soon to evaluate if we only move a part of them.

I run the CI quite frequently, and with #37555, the time is similar as before.
https://github.com/longvatrong111/servo/actions/runs/15830757765

yezhizhen · 2025-06-24T11:03:13Z

I definitely agree that navigation command should be moved.

It is just that, previously it immediately switches to the next subtest.. My eyes even cannot catch the speed. Right now there is a 1 second gap between each, leading to greatly increased total time. I'll investigate more once recovered.

It serves the purpose of synchronization. If script event has shorter route, it may be done out of order with others. In that case, we need to add a wait as well.

That makes sense. Thanks for clarification❤️ @longvatrong111

longvatrong111 · 2025-06-24T12:17:16Z

I definitely agree that navigation command should be moved.

It is just that, previously it immediately switches to the next subtest.. My eyes even cannot catch the speed. Right now there is a 1 second gap between each, leading to greatly increased total time. I'll investigate more once recovered.

It serves the purpose of synchronization. If script event has shorter route, it may be done out of order with others. In that case, we need to add a wait as well.

That makes sense. Thanks for clarification❤️ @longvatrong111

Let's move this one last. I will also check tests with all pieces of moving commands. Thank you.

xiaochengh · 2025-06-25T08:56:43Z

This one looks pretty straightforward, but please request another review from me when it's ready to merge. Thanks!

ports/servoshell/desktop/app.rs

longvatrong111 · 2025-06-26T10:10:00Z

We have more reason to move ScriptCommand to servoshell: #35734 (comment)

…ndowSize` (#37712) 1. Remove the unnecessary new thread which use GetWindowRect command and blocks for 500ms. Previously this is necessary because constellation forward "resize" to embedder, and WebDriver wait for a constant sufficient time to `GetWindowRect` in the new thread. This caused long delay because there are many subtests and SetWindowRect is called between each. 2. Remove `resize_timeout` 3. Return current dimension instead of 0 from embedder when it fails to resize. 4. Do resizing as long as one of width/height is `Some`, according to spec. Testing: All Conformance test with new passing cases. Fixes: #37663 (comment) --------- Signed-off-by: Euclid Ye <yezhizhenjiakang@gmail.com>

longvatrong111 · 2025-07-08T01:34:48Z

Update expected test result based on: https://github.com/longvatrong111/servo/actions/runs/16120879305

yezhizhen · 2025-07-08T01:48:12Z

Update expected test result based on: https://github.com/longvatrong111/servo/actions/runs/16120879305

This is confusing. Why these test expectations were newly added yesterday in 9afe027, but reverted again today?

longvatrong111 · 2025-07-08T04:18:31Z

Update expected test result based on: https://github.com/longvatrong111/servo/actions/runs/16120879305

This is confusing. Why these test expectations were newly added yesterday in 9afe027, but reverted again today?

I see.
Test result in 9afe027 was updated according to local test. After that I ran the CI:
https://github.com/longvatrong111/servo/actions/runs/16089865593

So /webdriver/tests/classic/perform_actions/key_events.py and some sub-tests in perform_actions were wrongly updated in 9afe027.

Test result in this PR is updated according to https://github.com/longvatrong111/servo/actions/runs/16120879305
This PR actually fixes element_send_key, switch_to_window. For perform_actions it only fixes the wrong test result update from 9afe027

longvatrong111 · 2025-07-08T05:52:16Z

Current test result, some unexpected results remain because of intermittent result or CI give different result with local.
https://github.com/longvatrong111/servo/actions/runs/16133927807

github-actions · 2025-07-08T18:39:36Z

🔨 Triggering try run (#16151610342) for Linux (WPT)

github-actions · 2025-07-08T19:22:43Z

Test results for linux-wpt from try job (#16151610342):

Flaky unexpected result (17)

OK /content-security-policy/frame-ancestors/frame-ancestors-path-ignored.window.html (#36468)
- PASS [expected FAIL] subtest: A 'frame-ancestors' CSP directive with a URL that includes a path should be ignored.
OK /css/css-cascade/layer-font-face-override.html (#35935)
- PASS [expected FAIL] subtest: @font-face override update with appended sheet 1
- PASS [expected FAIL] subtest: @font-face override update with appended sheet 2
OK [expected ERROR] /fetch/fetch-later/quota/same-origin-iframe/multiple-iframes.tentative.https.window.html (#35176)
OK /fetch/metadata/generated/css-font-face.https.sub.tentative.html (#32732)
- PASS [expected FAIL] subtest: sec-fetch-dest
OK /fetch/metadata/generated/css-font-face.sub.tentative.html (#34624)
- PASS [expected FAIL] subtest: sec-fetch-storage-access - Not sent to non-trustworthy cross-site destination
ERROR /fetch/metadata/generated/serviceworker.https.sub.html (#36247)
- PASS [expected FAIL] subtest: sec-fetch-site - Same origin, no options - registration
OK /html/browsers/browsing-the-web/navigating-across-documents/navigation-unload-cross-origin.sub.window.html (#29056)
- PASS [expected FAIL] subtest: Cross-origin navigation started from unload handler must be ignored
OK /html/browsers/browsing-the-web/navigating-across-documents/replace-before-load/a-click.html (#28697)
- PASS [expected FAIL] subtest: aElement.click() before the load event must NOT replace
TIMEOUT /html/browsers/history/the-history-interface/001.html (#12580)
- FAIL [expected PASS] subtest: traversing history must also traverse hash changes
```
assert_equals: (this could cause other failures later on) expected "" but got "test"
```
TIMEOUT [expected OK] /html/infrastructure/urls/base-url/document-base-url-window-initiator-is-not-opener.https.window.html (#30970)
TIMEOUT [expected CRASH] /html/semantics/embedded-content/the-iframe-element/iframe_sandbox_popups_escaping-1.html (#22647)
OK /html/semantics/scripting-1/the-script-element/module/dynamic-import/blob-url.any.worker.html (#33909)
- FAIL [expected PASS] subtest: Revoking a blob URL immediately after calling import will not fail
```
promise_test: Unhandled rejection with value: object "TypeError: Dynamic import failed"
```

OK /html/webappapis/user-prompts/print-during-unload.html (#35944)

FAIL [expected PASS] subtest: print() during unload

assert_array_equals: expected property 1 to be "destination" but got "error: window.print is not a function" (expected array ["start", "destination"] got ["start", "error: window.print is not a function"])

OK /preload/prefetch-document.html (#37210)
- FAIL [expected PASS] subtest: different-site document prefetch with 'as=document' should not be consumed
```
assert_equals: expected 2 but got 1
```
TIMEOUT [expected OK] /service-workers/service-worker/detached-context.https.html
- TIMEOUT [expected FAIL] subtest: accessing navigator.serviceWorker on a detached iframe
```
Test timed out
```
OK [expected TIMEOUT] /webmessaging/with-ports/017.html (#24486)
- PASS [expected TIMEOUT] subtest: origin of the script that invoked the method, about:blank
ERROR [expected OK] /workers/baseurl/alpha/import-in-moduleworker.html (#21315)

Stable unexpected results that are known to be intermittent (24)

FAIL [expected PASS] /_mozilla/css/stacked_layers.html (#15988)
FAIL [expected PASS] /_mozilla/mozilla/sslfail.html (#10760)
TIMEOUT [expected OK] /_mozilla/mozilla/window_resize_event.html (#36741)
- TIMEOUT [expected PASS] subtest: Popup onresize event fires after resizeTo
```
Test timed out
```
OK /css/css-cascade/layer-cssom-order-reverse.html (#36094)
- PASS [expected FAIL] subtest: Delete layer invalidates @font-face
OK /css/css-fonts/variations/at-font-face-font-matching.html (#20684)
- FAIL [expected PASS] subtest: Matching font-weight: '430' should prefer '450 460' over '500'
```
assert_equals: Unexpected font on test element expected 487 but got 532
```
- FAIL [expected PASS] subtest: Matching font-stretch: '90%' should prefer '90% 100%' over '50% 80%'
```
assert_equals: Unexpected font on test element expected 487 but got 532
```
FAIL [expected PASS] /css/css-grid/grid-items/grid-auto-margin-and-replaced-item-001.html (#37162)
TIMEOUT [expected OK] /fetch/api/redirect/redirect-keepalive.https.any.html (#32153)
- TIMEOUT [expected PASS] subtest: [keepalive][iframe][load] mixed content redirect; setting up
```
Test timed out
```
ERROR [expected TIMEOUT] /fetch/fetch-later/quota/same-origin-iframe/max-payload.tentative.https.window.html (#35210)
OK /html/browsers/browsing-the-web/navigating-across-documents/005.html (#27062)
- PASS [expected FAIL] subtest: Link with onclick navigation and href navigation
OK /html/browsers/browsing-the-web/navigating-across-documents/008.html (#24456)
- PASS [expected FAIL] subtest: Link with onclick form submit to javascript url and href navigation
OK /html/browsers/browsing-the-web/navigating-across-documents/empty-iframe-load-event.html (#29066)
- FAIL [expected PASS] subtest: Check execution order from nested timeout
```
assert_equals: Expected nested setTimeout to run second expected true but got false
```
- FAIL [expected PASS] subtest: Check execution order on load handler
```
assert_equals: Expected onload to run first expected false but got true
```
OK /html/browsers/browsing-the-web/navigating-across-documents/initial-empty-document/load-pageshow-events-iframe-contentWindow.html (#28681)
- PASS [expected FAIL] subtest: load & pageshow events do not fire on contentWindow of <iframe> element created with src='about:blank'
OK /html/browsers/history/the-history-interface/traverse_the_history_5.html (#21383)
- FAIL [expected PASS] subtest: Multiple history traversals, last would be aborted
```
assert_array_equals: Pages opened during history navigation expected property 1 to be 5 but got 3 (expected array [6, 5] got [6, 3])
```
FAIL [expected PASS] /html/canvas/element/manual/drawing-text-to-the-canvas/canvas.2d.disconnected-font-size-math.html (#30063)
TIMEOUT /html/interaction/focus/the-autofocus-attribute/supported-elements.html (#24145)
- TIMEOUT [expected FAIL] subtest: Element with tabindex should support autofocus
```
Test timed out
```
OK /html/semantics/embedded-content/the-iframe-element/iframe-loading-lazy-nav-location-assign.html (#32863)
- PASS [expected FAIL] subtest: Navigating iframe loading='lazy' before it is loaded: location.assign
CRASH [expected TIMEOUT] /html/semantics/embedded-content/the-iframe-element/iframe_sandbox_popups_escaping-2.html (#22667)
OK [expected TIMEOUT] /html/semantics/embedded-content/the-iframe-element/iframe_sandbox_popups_escaping-3.html (#24057)
CRASH [expected TIMEOUT] /html/semantics/embedded-content/the-iframe-element/iframe_sandbox_popups_nonescaping-2.html (#22154)
OK [expected TIMEOUT] /html/semantics/embedded-content/the-iframe-element/iframe_sandbox_popups_nonescaping-3.html (#24066)

OK /preload/preload-error.sub.html (#37177)

FAIL [expected PASS] subtest: success (style): main

assert_greater_than: http://web-platform.test:8000/preload/resources/dummy.css?label=style should be loaded expected a number greater than 0 but got 0

PASS [expected FAIL] subtest: 404 (xhr): main

FAIL [expected PASS] subtest: Decode-error (style): main

assert_greater_than: http://web-platform.test:8000/preload/resources/dummy.xml?pipe=header%28Content-Type%2Ctext%2Fcss%29&amp;label=style should be loaded expected a number greater than 0 but got 0

PASS [expected FAIL] subtest: MIME-error (script): main

CRASH [expected TIMEOUT] /trusted-types/trusted-types-navigation.html?06-10 (#37920)
TIMEOUT [expected OK] /webmessaging/with-ports/018.html (#24485)
- TIMEOUT [expected PASS] subtest: origin of the script that invoked the method, javascript:
```
Test timed out
```
OK [expected ERROR] /workers/constructors/Worker/Worker-constructor.html (#22991)

github-actions · 2025-07-08T19:22:51Z

✨ Try run (#16151610342) succeeded.

longvatrong111 · 2025-07-09T02:57:28Z

Hi @mrobinson, I think this one is ready to be merged.

Signed-off-by: batu_hoang <longvatrong111@gmail.com>

mrobinson · 2025-07-09T13:06:26Z

ports/servoshell/desktop/app_state.rs

+    // From <https://w3c.github.io/webdriver/#dfn-execute-a-function-body>:
+    // > The rules to execute a function body are as follows. The algorithm returns
+    // > an ECMAScript completion record.
+    // >
+    // > If at any point during the algorithm a user prompt appears, immediately return
+    // > Completion { Type: normal, Value: null, Target: empty }, but continue to run the
+    // >  other steps of this algorithm in parallel.


Suggested change

// From <https://w3c.github.io/webdriver/#dfn-execute-a-function-body>:

// > The rules to execute a function body are as follows. The algorithm returns

// > an ECMAScript completion record.

// >

// > If at any point during the algorithm a user prompt appears, immediately return

// > Completion { Type: normal, Value: null, Target: empty }, but continue to run the

// > other steps of this algorithm in parallel.

/// From <https://w3c.github.io/webdriver/#dfn-execute-a-function-body>:

/// > The rules to execute a function body are as follows. The algorithm returns

/// > an ECMAScript completion record.

// >

/// > If at any point during the algorithm a user prompt appears, immediately return

/// > Completion { Type: normal, Value: null, Target: empty }, but continue to run the

/// > other steps of this algorithm in parallel.

- Avoid cloning the entire WebDriver message. Just clone the sender. - Adjust the documentation to be rustdoc. Signed-off-by: Martin Robinson <mrobinson@igalia.com>

longvatrong111 requested a review from atbrakhi as a code owner June 24, 2025 05:33

longvatrong111 force-pushed the webdriver-script-command branch from 819e22e to a179956 Compare June 24, 2025 05:36

wusyong approved these changes Jun 24, 2025

View reviewed changes

wusyong added this pull request to the merge queue Jun 24, 2025

wusyong removed this pull request from the merge queue due to a manual request Jun 24, 2025

longvatrong111 force-pushed the webdriver-script-command branch from a179956 to 5ea68e0 Compare June 25, 2025 06:29

xiaochengh approved these changes Jun 25, 2025

View reviewed changes

mrobinson reviewed Jun 25, 2025

View reviewed changes

ports/servoshell/desktop/app.rs Outdated Show resolved Hide resolved

longvatrong111 mentioned this pull request Jun 25, 2025

Move webdriver actions commands to servoshell #37669

Merged

longvatrong111 marked this pull request as draft June 25, 2025 17:23

yezhizhen mentioned this pull request Jun 26, 2025

webdriver: Greatly improve execution speed for all tests using SetWindowSize #37712

Merged

jdm closed this in #37712 Jul 1, 2025

yezhizhen reopened this Jul 1, 2025

longvatrong111 force-pushed the webdriver-script-command branch 5 times, most recently from fdb1500 to 87a1668 Compare July 4, 2025 08:56

longvatrong111 marked this pull request as ready for review July 4, 2025 08:57

longvatrong111 mentioned this pull request Jul 8, 2025

webdriver: Element click waits for navigation complete #37935

Merged

longvatrong111 force-pushed the webdriver-script-command branch 2 times, most recently from e0023dc to 0f5224a Compare July 8, 2025 08:22

longvatrong111 added the T-linux-wpt Do a try run of the WPT label Jul 8, 2025

github-actions bot removed the T-linux-wpt Do a try run of the WPT label Jul 8, 2025

longvatrong111 force-pushed the webdriver-script-command branch from 0f5224a to 1e78893 Compare July 9, 2025 05:56

longvatrong111 requested a review from gterzian as a code owner July 9, 2025 05:56

longvatrong111 added 3 commits July 9, 2025 13:57

Move webdriver script command to servoshell

7c3fec4

Signed-off-by: batu_hoang <longvatrong111@gmail.com>

Fix lint

8fc1302

Signed-off-by: batu_hoang <longvatrong111@gmail.com>

Update wpt expected result

3ec4f85

Signed-off-by: batu_hoang <longvatrong111@gmail.com>

longvatrong111 force-pushed the webdriver-script-command branch from 1e78893 to 3ec4f85 Compare July 9, 2025 05:57

longvatrong111 mentioned this pull request Jul 9, 2025

webdriver: Implement support for simple dialogs #37913

Merged

mrobinson changed the title ~~Move webdriver script command to servoshell~~ webdriver: Evaluate script commands via the WebView API in servoshell Jul 9, 2025

mrobinson force-pushed the webdriver-script-command branch from e7a259d to 03c16e4 Compare July 9, 2025 13:23

mrobinson approved these changes Jul 9, 2025

View reviewed changes

Minor tweaks

b2891e2

- Avoid cloning the entire WebDriver message. Just clone the sender. - Adjust the documentation to be rustdoc. Signed-off-by: Martin Robinson <mrobinson@igalia.com>

mrobinson force-pushed the webdriver-script-command branch from 03c16e4 to b2891e2 Compare July 9, 2025 13:23

mrobinson enabled auto-merge July 9, 2025 13:24

mrobinson added this pull request to the merge queue Jul 9, 2025

Merged via the queue into servo:main with commit 4499fde Jul 9, 2025
21 checks passed

longvatrong111 deleted the webdriver-script-command branch July 9, 2025 15:56

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Uh oh!

Uh oh!

webdriver: Evaluate script commands via the `WebView` API in servoshell #37663

webdriver: Evaluate script commands via the `WebView` API in servoshell #37663

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

webdriver: Evaluate script commands via the WebView API in servoshell #37663

webdriver: Evaluate script commands via the WebView API in servoshell #37663

Uh oh!

Conversation

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

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!

Uh oh!

Uh oh!

webdriver: Evaluate script commands via the `WebView` API in servoshell #37663

webdriver: Evaluate script commands via the `WebView` API in servoshell #37663