Revert "buffer: move SlowBuffer to EOL"

[panva]

This reverts commit 647175e.

The API in question was runtime-deprecated (#55175) and EOLd (#58008 which i'm reverting here) in the same major release. The intended cycle is that it gets runtime-deprecated as a major, and then EOLd in another later major.

Despite best efforts to detect it in the EOL PR there is breakage in the jsonwebtoken > jws > jwa dependency chain (29-30M weekly downloads on jwa).

This PR reverts to the runtime-deprecation state that was pending release with 24.0.0, EOL shouldn't have landed yet.

https://openjs-foundation.slack.com/archives/C019MGJQ8RH/p1746596637445769

We would re-land the EOL and ship it with a future major release after this revert.

[nodejs-github-bot]

Review requested:

• @nodejs/config
• @nodejs/performance
• @nodejs/tsc

[nodejs-github-bot]

CI: https://ci.nodejs.org/job/node-test-pull-request/66672/

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 90.12%. Comparing base (b21574d) to head (e074ef1).
Report is 9 commits behind head on main.

Additional details and impacted files

@@            Coverage Diff             @@
##             main   #58211      +/-   ##
==========================================
- Coverage   90.13%   90.12%   -0.02%     
==========================================
  Files         629      629              
  Lines      186628   186644      +16     
  Branches    36632    36621      -11     
==========================================
- Hits       168218   168210       -8     
+ Misses      11219    11214       -5     
- Partials     7191     7220      +29     

Files with missing lines	Coverage Δ
lib/buffer.js	100.00% <100.00%> (ø)
src/node_options.cc	85.36% <ø> (ø)

... and 38 files with indirect coverage changes

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.
• 📦 JS Bundle Analysis: Save yourself from yourself by tracking and limiting bundle sizes in JS merges.

[github-actions]

Fast-track has been requested by @panva. Please 👍 to approve.

[panva]

Oy, sorry about the trouble. The EOL was meant to sit in main for 25 and my assumption was that since the semver-major cutoff for 24 had passed it wouldn't have been picked up, but apparently that assumption was incorrect. Let's revert the EOL in 24 but keep it in main. This revert can be treated the same as a back port commit targeting 24.x specifically could it not?

I don't see the point in re-running CI for a different base branch just so that it remains on main. By reverting and re-landing on main as semver-major again (which i plan on doing right after this lands) we will also get a proper v25.x changelog entry. The existing changelog entry also only talked about deprecation and didn't pick up the removal which is why it has slipped my radar. I knew about the deprecation but i wasn't aware of its removal at all, not until i searched the individual semver-major commits.

[jasnell]

I don't see the point in re-running CI for a different base branch just so that it remains on main...

That's fine... but keep in mind it will need to be run on that other branch regardless since this revert will need to be backported.

• Targeting this PR at 24.x now means no additional PRs are needed and CI would need to be run once at best assuming no flaky failures
• Targeting this PR at main now means at least two additional PRs (one to reapply the EOL in main and one to backport the revert) with at least two additional CI runs.

[panva]

I don't get why node-test-commit-osx is failing, it seems to have to do smth with github checks? Its actual test job is passing...

[panva]

If I target 24 directly we'll still have to get the change to 25.x changelog somehow, and I doubt we'll remember to do so in half a year.

[aduh95]

@panva please use Resume build to avoid re-running already-green jobs

[panva]

@panva please use Resume build to avoid re-running already-green jobs

I did two times already before and because node-test-commit-osx failed each time with no indication why (other than pointing to checks API) i suspect that's because there were duplicate jobs started (or resumed) by someone as the ones I resumed were already running.

[aduh95]

node-test-commit-osx failed each time with no indication why

You can see it in the logs:

Build timed out (after 60 minutes). Marking the build as failed.
Build was aborted

Re-running the other jobs is not going to help, I'd be very surprised it has anything to do with duplicated jobs (or that not resuming would even help if duplicate jobs were an issue).

[panva]

node-test-commit-osx failed each time with no indication why

You can see it in the logs:

Build timed out (after 60 minutes). Marking the build as failed.
Build was aborted

Re-running the other jobs is not going to help, I'd be very surprised it has anything to do with duplicated jobs (or that not resuming would even help if duplicate jobs were an issue).

https://ci.nodejs.org/job/node-test-commit-osx/64970/nodes=osx13-x64/console

14:48:49 All tests passed.
...
15:48:53 Build timed out (after 60 minutes). Marking the build as failed.
15:48:53 Build was aborted
...
15:49:03 POST BUILD TASK : SUCCESS
...
15:49:03 Recording test results
15:49:07 [Checks API] No suitable checks publisher found.
15:49:07 Collecting metadata...
15:49:07 Metadata collection done.
15:49:07 [Checks API] No suitable checks publisher found.
15:49:07 Notifying upstream projects of job completion
15:49:07 Finished: FAILURE

[richardlau]

node-test-commit-osx failed each time with no indication why

You can see it in the logs:

Build timed out (after 60 minutes). Marking the build as failed.
Build was aborted

Re-running the other jobs is not going to help, I'd be very surprised it has anything to do with duplicated jobs (or that not resuming would even help if duplicate jobs were an issue).

https://ci.nodejs.org/job/node-test-commit-osx/64970/nodes=osx13-x64/console

14:48:49 All tests passed.
...
15:48:53 Build timed out (after 60 minutes). Marking the build as failed.
15:48:53 Build was aborted
...
15:49:03 POST BUILD TASK : SUCCESS
...
15:49:03 Recording test results
15:49:07 [Checks API] No suitable checks publisher found.
15:49:07 Collecting metadata...
15:49:07 Metadata collection done.
15:49:07 [Checks API] No suitable checks publisher found.
15:49:07 Notifying upstream projects of job completion
15:49:07 Finished: FAILURE

FWIW I think this macOS failure is nodejs/build#4072 where it's taking a very long time to correct file ownership on the ccache files post-build.

[panva]

@richardlau yeah, it's stuck on that step again now.

[nodejs-github-bot]

CI: https://ci.nodejs.org/job/node-test-pull-request/66689/

[nodejs-github-bot]

CI: https://ci.nodejs.org/job/node-test-pull-request/66693/

[nodejs-github-bot]

Landed in da69d13

[jpike88]

any intention on getting a patch release out there for this? it's broken pretty much everything we use, now we need to roll back to 23.

[github-actions]

The notable-change PRs with changes that should be highlighted in changelogs. label has been added by @aduh95.

Please suggest a text for the release notes if you'd like to include a more detailed summary, then proceed to update the PR description with the text or a link to the notable change suggested text comment. Otherwise, the commit will be placed in the Other Notable Changes section.