Remove Yjs room locks

[davidbrochart]

Code changes

Remove the Yjs room locks.

User-facing changes

None

Backwards-incompatible changes

Rename WebSocketProviderWithLocks to WebSocketProvider in @jupyterlab/docprovider.

[jupyterlab-probot]

Thanks for making a pull request to jupyterlab!
To try out this branch on binder, follow this link:

[jtpio]

Thanks @davidbrochart.

Mind providing some more context on what removing the locks will mean in practice?

Or is it a follow-up to #11599, which means they are not needed anymore?

[davidbrochart]

Or is it a follow-up to #11599, which means they are not needed anymore?

Sorry for not providing more context, yes it is. Because it is not backwards-compatible, we decided not to include it in #11599, so that it can be back-ported. But yes, the locks are not needed anymore, now that we save from the back-end.

[jtpio]

OK thanks.

Rename WebSocketProviderWithLocks to WebSocketProvider in @jupyterlab/docprovider.

Let's document this in the extension migration guide? We can also mention the removal of acquireLock and releaseLock from the interface.

https://jupyterlab.readthedocs.io/en/latest/extension/extension_migration.html#api-breaking-changes

[davidbrochart]

It looks like the Linux JS Flaky Tests failures are unrelated, as well as the pre-commit ones.

[afshin]

Is it possible to pass this schema as a local file instead of requiring an outbound HTTP request in the pre-commit workflow?

jupyterlab/.pre-commit-config.yaml

Line 50 in 136fa76

args: ['--schemafile', 'https://json.schemastore.org/github-workflow']

Is that what is breaking here?

[afshin]

cc: @blink1073, RE: pre-commit.ci ⬆️

[blink1073]

Yeah, basically we'd have to completely rethink a bunch of stuff to make this repo compatible with pre-commit.ci, so I removed it. cf jupyterlab/frontends-team-compass#142 (comment)

[jtpio]

Nice cleanup, thanks @davidbrochart!

[github-actions]

Benchmark report

The execution time (in milliseconds) are grouped by test file, test type and browser.
For each case, the following values are computed: min <- [1st quartile - median - 3rd quartile] -> max.

The mean relative comparison is computed with 95% confidence.

Results table

Test file	large_code_notebook	large_md_notebook
open
chromium
actual	3890 <- [4104 - 4190 - 4325] -> 4563	2568 <- [2735 - 2805 - 2863] -> 3220
expected	3923 <- [4150 - 4269 - 4337] -> 4591	2584 <- [2767 - 2830 - 2898] -> 3074
Mean relative change	-0.8% ± 1.0%	-0.9% ± 1.1%
switch-from
chromium
actual	625 <- [691 - 722 - 763] -> 1010	495 <- [550 - 952 - 1022] -> 1177
expected	669 <- [715 - 738 - 804] -> 1031	484 <- [534 - 961 - 1023] -> 1162
Mean relative change	-4.1% ± 3.4%	3.6% ± 8.5%
switch-to
chromium
actual	320 <- [366 - 395 - 414] -> 442	233 <- [283 - 297 - 310] -> 395
expected	312 <- [379 - 407 - 422] -> 496	245 <- [287 - 297 - 310] -> 362
Mean relative change	-2.2% ± 2.1%	-0.5% ± 2.0%
close
chromium
actual	539 <- [958 - 995 - 1026] -> 1173	443 <- [480 - 501 - 516] -> 561
expected	606 <- [988 - 1012 - 1034] -> 1147	443 <- [487 - 512 - 526] -> 570
Mean relative change	-1.7% ± 2.8%	-1.4% ± 1.5%

Changes are computed with expected as reference.