Plugin: Ensure that Block Hooks work correctly after landing in WP core

[gziolo]

What?

Part of #53987.

Includes necessary changes to ensure that Block Hooks continues working correctly after Porting PHP code to Core.

Why?

At the moment, hooked blocks get inserted twice:

How?

Everything should continue working as in WordPress core. Hooked blocks should be inserted only once:

The challenge is that it has to be tested with WordPress 6.3 and WordPress 6.4 alpha.

Other than that all PHP unit tests should pass.

Testing Instructions

1. Activate Twenty Twenty-Three theme.
2. Download Like Button plugin that contains the hooked block with the Comment Template block.
3. Install and activate the plugin.
4. Open the site and navigate to the single posts page.
5. Ensure that there are comments and the Like Button is rendered once for every comment.

[github-actions]

This pull request has changed or added PHP files. Please confirm whether these changes need to be synced to WordPress Core, and therefore featured in the next release of WordPress.

If so, it is recommended to create a new Trac ticket and submit a pull request to the WordPress Core Github repository soon after this pull request is merged.

If you're unsure, you can always ask for help in the #core-editor channel in WordPress Slack.

Thank you! ❤️

View changed files

❔ lib/compat/wordpress-6.4/block-hooks.php
❔ lib/compat/wordpress-6.4/class-gutenberg-rest-block-patterns-controller.php
❔ phpunit/tests/blocks/renderHookedBlocks.php

[gziolo]

I don't think we need to cover with unit tests the polyfill for Block Hooks anymore so I removed them with b90afcb.

I don't think we run E2E tests with the older major versions of WordPress, so we should just test it manually with WordPress 6.3 before landing the PR.

[github-actions]

Flaky tests detected in b90afcb.
Some tests passed with failed attempts. The failures may not be related to this commit but are still reported for visibility. See the documentation for more information.

🔍 Workflow run URL: https://github.com/WordPress/gutenberg/actions/runs/6259312996
📝 Reported issues:

• [Flaky Test] should partially select with shift + click (@webkit) #53524 in /test/e2e/specs/editor/various/multi-block-selection.spec.js

[gziolo]

I opened a PR against core that documents the latest behavior for edge cases when dealing with Block Hooks at positions firstChild and lastChild: WordPress/wordpress-develop#5279. The behavior differs from what we have in the polyfill developed for the Gutenberg plugin, so we will still have to patch it. It also means that the tests removed with b90afcb could be useful. I will see if we can keep them by moving gutenberg_serialize_blocks outside of the check against WP 6.4 version.

[gziolo]

I brought back unit tests in 7a04903 and updated them to work with the polyfill developed for older versions of WordPress. I applied all necessary updates to handling fistChild and lastChild with 3f56b22 to cover the case when there are no blocks.

[ockham]

• Code looks good ✅
• Tested the plugin zip produced by this PR in a local dev environment running Core trunk. Verified that hooked blocks are correctly inserted both on the frontend and in the editor ✅
• Tested the plugin with a throwaway test site with WP 6.3. Verified that hooked blocks are correctly inserted both on the frontend and in the editor ✅

Thank you!

[gziolo]

@mikachan, it would be nice to backport this fix to the Gutenberg plugin 16.7 release (#54621). We hope to run a call for testing for the Block Hooks feature, and this could interfere in case someone tests Gutenberg 16.7 and WordPress 6.4 Beta 1 together.

[mikachan]

I just cherry-picked this PR to the release/16.7 branch to get it included in the next release: 0790b09

[ockham]

@mikachan Any chance we can also include this in 6.4? Not super urgent, so it's okay if we miss Beta 1, but would be handy to get into e.g. Beta 2. (See WordPress/wordpress-develop#5307)

[mikachan]

Yes! This PR should be included in 6.4 Beta 1, I'll confirm if not and re-add the "Backport to WP Beta/RC" label.

[scruffian]

This PR seems to break template parts: #55202