Implement "network.responseCompleted" event
Categories
(Remote Protocol :: WebDriver BiDi, task, P1)
Tracking
(firefox110 fixed)
| Tracking | Status | |
|---|---|---|
| firefox110 | --- | fixed |
People
(Reporter: whimboo, Assigned: jdescottes)
References
(Blocks 4 open bugs, )
Details
(Whiteboard: [webdriver:m5][wptsync upstream][webdriver:relnote])
Attachments
(2 files, 1 obsolete file)
|
Assignee | |
Updated•1 year ago
|
|
|
Assignee | |
Comment 1•1 year ago
|
||
Valentin, another question (sorry :) )
We have a PR on the fetch spec to add a trigger for BiDi's "response completed" event: https://whatpr.org/fetch/1540.html#fetch-finale step 3.3.8
If response is a network error, run the WebDriver BiDi fetch error steps with request. Otherwise run the WebDriver BiDi response completed steps with request and response.
Do you what would be a good match to observe this from chrome privileged JS?
At the moment we started our implementation by reusing the DevTools NetworkObserver logic, but to detect the end of a response we wait for the request input stream to be closed (usually seems to trigger from https://searchfox.org/mozilla-central/rev/d62c4c4d5547064487006a1506287da394b64724/devtools/shared/network-observer/NetworkResponseListener.sys.mjs#593)
But for redirects, the redirected request and response are already partly processed when we detect that the original stream closes. So I imagine we should observe something else here. Any idea?
|
|
Assignee | |
Comment 2•1 year ago
|
||
NetworkObserver addResponseContent is the last observable callback for a given request, so let's start by using this as the trigger for
responseCompleted. However, as noted on the bug and on Bug 1809210, this does not meet the expectations from the spec completely so we might have to
change to another observable later.
|
|
Assignee | |
Comment 3•1 year ago
|
||
Depends on D166373
|
||
Updated•1 year ago
|
|
|
||
Updated•1 year ago
|
|
|
Assignee | |
Comment 4•1 year ago
|
||
Depends on D166373
|
|
||
Comment 5•1 year ago
|
||
Comment on attachment 9311845 [details]
Bug 1790372 - [bidi] Do not include the charset in the mimeType
Revision D166599 was moved to bug 1809854. Setting attachment 9311845 [details] to obsolete.
Pushed by jdescottes@mozilla.com: https://hg.mozilla.org/integration/autoland/rev/6956ef91ed6f [bidi] Add basic support for network.responseCompleted event r=webdriver-reviewers,whimboo https://hg.mozilla.org/integration/autoland/rev/89a09894a67b [wdspec] Add wdspec tests for network.responseCompleted r=webdriver-reviewers,whimboo
Created web-platform-tests PR https://github.com/web-platform-tests/wpt/pull/37899 for changes under testing/web-platform/tests
|
||
Comment 8•1 year ago
|
||
| bugherder | ||
https://hg.mozilla.org/mozilla-central/rev/6956ef91ed6f
https://hg.mozilla.org/mozilla-central/rev/89a09894a67b
Upstream PR merged by moz-wptsync-bot
|
Reporter | |
Updated•1 year ago
|
|
||
Updated•1 year ago
|
Description
•