LGTM1 - thanks for the Fetch spec work cleanup.
(Note: I spoke offline with Hubert about
https://github.com/WICG/local-network-access/issues/37 - that will get
landed when someone from Mozilla gets back from PTO, and Hubert will
make sure it happens in a timely fashion).
On 3/6/26 3:25 p.m., Mike Taylor wrote:
Thanks for adding WPTs for WebTransport! I filed
https://github.com/WICG/local-network-access/issues/104, because there
is some more work needed on the spec side of things to meet the
quality bar we require for shipping.
On 3/6/26 11:49 a.m., Hubert Chao wrote:
More WPTs are in, for the 3 types of workers (dedicated, service, and
shared).
/hubert
On Wed, Mar 4, 2026 at 9:26 AM Hubert Chao <[email protected]> wrote:
On Fri, Feb 20, 2026 at 11:35 AM Hubert Chao <[email protected]>
wrote:
On Fri, Feb 20, 2026 at 10:23 AM Mike Taylor
<[email protected]> wrote:
On 2/19/26 2:28 p.m., Chromestatus wrote:
*Contact emails*
[email protected]
*Explainer*
https://github.com/WICG/local-network-access/blob/main/explainer.md
*Specification*
https://wicg.github.io/local-network-access/#integration-with-webtransport
I'd be happier if we spent some time defining the actual
integration with WebTransport - this doesn't really
describe how or when LNA should be applied. i.e., does it
happen in the constructor steps? Or when obtaining a
connection? Somewhere else? Does it ever throw an error? etc
The short answer is that this check happens in the
constructor before a connection is attempted. Re: more spec
work, its something that is on our list of things to do, but
we've chosen to prioritize other LNA landing work to ensure
better user experiences (see the ongoing discussion in
https://groups.google.com/a/chromium.org/g/blink-dev/c/8AK8V4fSZFU)
Looking briefly at the specs, I'm not 100% sure if we'd patch
the fetch spec for "obtaining a connection"
(https://fetch.spec.whatwg.org/#concept-connection-obtain)
which is referenced in webtransport's "obtaining a
connection" section (here
<https://www.w3.org/TR/webtransport/#webtransport-constructor:~:text=Let%20connection%20be%20the%20result%20of%20obtaining%20a%20connection%20with%20networkPartitionKey%2C%20url%2C%20false%2C%20newConnection%2C%20requireUnreliable%20and%20webTransportHashes.>).
If its the former, than we wouldn't need to change the
WebTransport spec at all.
*Is this feature fully tested by web-platform-tests
<https://chromium.googlesource.com/chromium/src/+/main/docs/testing/web_platform_tests.md>?*
No
Why not? It seems like this should be possible
Because we haven't prioritized WPTs for LNA up until
recently; we've been using chromium browser tests and unit
tests. This is on our priority list (again see the discussion
ongoing in
https://groups.google.com/a/chromium.org/g/blink-dev/c/8AK8V4fSZFU)
Update: we have WPT's covering WebSockets
<https://github.com/web-platform-tests/wpt/blob/master/fetch/local-network-access/websocket.tentative.https.html>
and WebTransport
<https://github.com/web-platform-tests/wpt/blob/master/fetch/local-network-access/webtransport.tentative.https.html>.
--
You received this message because you are subscribed to the Google Groups
"blink-dev" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to [email protected].
To view this discussion visit
https://groups.google.com/a/chromium.org/d/msgid/blink-dev/5d61b154-086c-4d1f-a319-8f36d8ab7b48%40chromium.org.
