It is perfectly possible to write a commented regex with layout for readability by using the `x` modifier.
As per the manual:
> x (`PCRE_EXTENDED`)
>
> If this modifier is set, whitespace data characters in the pattern are totally ignored except when escaped or inside a character class, and characters between an unescaped # outside a character class and the next newline character, inclusive, are also ignored. This is equivalent to Perl's /x modifier, and makes it possible to include commentary inside complicated patterns.
>
> Note, however, that this applies only to data characters. Whitespace characters may never appear within special character sequences in a pattern, for example within the sequence (?( which introduces a conditional subpattern.
Reference: [https://www.php.net/manual/en/reference.pcre.pattern.modifiers.php PHP Manual: Pattern Modifiers].
This commit rewrites these two regexes to use the `x` modifier and gets rid of the unnecessary `phpcs:disable` comments.
The tests in the `tests/phpunit/tests/db/dbDelta.php` file cover this change.
Follow-up to [42249].
Props jrf.
See #59650.
Built from https://develop.svn.wordpress.org/trunk@57056
git-svn-id: http://core.svn.wordpress.org/trunk@56567 1a063a9b-81f0-0310-95a4-ce76da25c4cd
This increases the number of times to retry restarting a workflow from 10 to 15, and the `timeout-minutes` value to `30`.
For workflows with complex strategy matrix, the exponential backoff of 10 retries is still not enough to account for the GitHub Actions UI taking a long time to catch up.
Follow up to [56829] and [56830].
See #58867.
Built from https://develop.svn.wordpress.org/trunk@57052
git-svn-id: http://core.svn.wordpress.org/trunk@56563 1a063a9b-81f0-0310-95a4-ce76da25c4cd
The callbacks returned by `make_before_block_visitor` and `make_after_block_visitor`, respectively, (which are passed as arguments to `traverse_and_serialize_block(s)`) currently accept three arguments, all of which are block arrays (i.e. with properties `blockName`, `attrs`, etc.):
- A ''reference'' to the block they're currently visiting, `&$block`;
- the block's `$parent_block`; and
- the `$prev`ious block (for `make_before_block_visitor`), or the `$next` block (for `make_after_block_visitor`), respectively.
Those arguments are passed to the "block visitor" callbacks by `traverse_and_serialize_block(s)` during traversal. The block that the callback is currently visiting is passed ''by reference'' to allow modifying it, which is e.g. used to inject the `theme` attribute into Template Part blocks.
One major limitation of Block Hooks is that they currently only work with templates, parts, and patterns that ''don't have any user modifications'' (i.e. that come straight from the corresponding theme files, rather than from the database). For WordPress 6.5, it is planned to change that to make Block Hooks work for templates, parts, and patterns that ''do'' have user modifications: #59646.
This will be implemented by storing an attribute on the "anchor" block. While working on that feature, it was found that the aforementioned callbacks will need to modify not only the currently visited `$block`, but also the `$parent_block` -- i.e. that the latter argument needs to be passed by reference as well. This is consistent with the requirement of adding an attribute to an anchor block, as it's not only the currently visited block that can serve as an anchor block (in the case of `before` or `after` sibling insertion), but also its parent (for `first_child` and `last_child` insertion).
If the `$parent_block` argument were to be changed to become a reference in a later WordPress version, this could be considered a backwards-compatibility breaking change. For this reason, this change is instead proposed for 6.4 already, which is the cycle during which the relevant functions were first introduced. This should have no impact on existing code, since nothing currently relies on `$parent_block` remaining unmodified by the respective callback, nor is anything currently modifying that argument.
Props hellofromTonya.
Fixes#59776.
Built from https://develop.svn.wordpress.org/trunk@57038
git-svn-id: http://core.svn.wordpress.org/trunk@56549 1a063a9b-81f0-0310-95a4-ce76da25c4cd
This update includes the following bugfixes:
- Fix: Added a new block pattern for the home template that inherits the page query and fits the design of the home page. (more context on the theme repo: https://github.com/WordPress/twentytwentyfour/pull/706)
- Fix: Rely on parent theme data for block style.
- Fix: Categories for some patterns.
- Fix: Minor labeling issues
Follow-up to [56999], [56951], [56813], [56764], [56716].
Props anlino, beafialho, desrosj, devmuhib, didierjm, fabiorubioglio, flixos90, hanneslsm, hellofromTonya, huzaifaalmesbah, ktaron, luminuu, mshowes, onemaggie, phillsav, poena, rajinsharwar, richtabor, shailu25.
Fixes#59770, #59759.
Built from https://develop.svn.wordpress.org/trunk@57036
git-svn-id: http://core.svn.wordpress.org/trunk@56547 1a063a9b-81f0-0310-95a4-ce76da25c4cd
A collection of fixes for `wp_prime_option_caches()`:
* cache arrays and objects in their serialized form for consistency with `get_option()` and `wp_load_alloptions()`
* prevent repeat database queries for falsey and known non-existent options (notoptions)
Additional tests for `wp_prime_option_caches()` to ensure:
* additional database queries are not made repriming options (known, known-unknown and alloptions)
* cache is primed consistently
* `get_option()` returns a consistent value regardless of how it is primed
* database queries do not contain earlier primed options
* `get_option` does not prime the cache when testing the cache has been successfully primed
Fixes a test for `wp_prime_option_caches_by_group()` to ensure `get_option` does not prime the cache when testing the cache has been successfully primed.
Follow up to [56445],[56990],[57013].
Props peterwilsoncc, costdev, flixos90, hellofromTonya, mikeschroder, joemcgill.
Fixes#59738. See #58962.
Built from https://develop.svn.wordpress.org/trunk@57029
git-svn-id: http://core.svn.wordpress.org/trunk@56540 1a063a9b-81f0-0310-95a4-ce76da25c4cd
When setting an RTL language under Settings → General, some RTL stylesheets were not loaded, with LTR stylesheets being loaded instead, meaning that some blocks were not displayed correctly.
This commit ensures that all appropriate RTL stylesheets are loaded when selecting an RTL language.
Follow-up to [56524].
Props mukesh27, maahrokh, hellofromTonya, joemcgill, huzaifaalmesbah, rajinsharwar, devmuhib, swissspidy.
Fixes#59715.
Built from https://develop.svn.wordpress.org/trunk@57028
git-svn-id: http://core.svn.wordpress.org/trunk@56539 1a063a9b-81f0-0310-95a4-ce76da25c4cd
This makes it possible to register a block by passing an array of arguments, without the presence of a `block.json` file.
Follow-up to [48141], [49948].
Props aristath, spacedmonkey, mukesh27, costdev, audrasjb, oglekler, felipeelia, hellofromTonya.
Fixes#56865.
Built from https://develop.svn.wordpress.org/trunk@57026
git-svn-id: http://core.svn.wordpress.org/trunk@56537 1a063a9b-81f0-0310-95a4-ce76da25c4cd
Add a conditional to prevent the `prefers-reduced-motion` `ajaxComplete` listener from observing events not occurring in the plugin installation screen. Improve handling of settings data test.
The listener observing `ajaxComplete` in [56541] was intercepting all `ajaxComplete` events, creating potential for unexpected errors in unrelated functions.
Props bplv, afercia, rudlinkon, hellofromTonya, huzaifaalmesbah, joedolson, jorbin.
Fixes#59689.
Built from https://develop.svn.wordpress.org/trunk@57022
git-svn-id: http://core.svn.wordpress.org/trunk@56533 1a063a9b-81f0-0310-95a4-ce76da25c4cd
This fixes a bug during the database upgrade process where a theme's `functions.php` file may not be loaded, leading to potential exceptions if the theme's pattern files use symbols (classes, functions, constants, etc.) that are declared only when the `functions.php` file is loaded. To do so, a check for `wp_get_active_and_valid_themes()` is added early to `_register_theme_block_patterns()`, which returns early if no active or valid themes are returned.
Props fabiankaegy, rajinsharwar, pbiron, huzaifaalmesbah, hellofromTonya, peterwilsoncc, joemcgill.
Fixes#59723.
Built from https://develop.svn.wordpress.org/trunk@57021
git-svn-id: http://core.svn.wordpress.org/trunk@56532 1a063a9b-81f0-0310-95a4-ce76da25c4cd
As a follow up to [56507], this fixes a bug that could occur for instance when plugins hijack the block template detection process to inject their own block template with entirely custom logic.
Props afragen, hellofromTonya, costdev, mukesh27, huzaifaalmesbah, flixos90.
Fixes#59736.
See #58154.
Built from https://develop.svn.wordpress.org/trunk@57019
git-svn-id: http://core.svn.wordpress.org/trunk@56530 1a063a9b-81f0-0310-95a4-ce76da25c4cd
Tweaks the 6.4 about pages in a couple of ways:
- Decouples the background from the 6.4 logo so the logo can move for RTL.
- Updates a color to improve color contrast.
- Help prevent overlap of long text strings with 6.4 logo.
- Ensure background isn't dark when no background is used on mobile.
Props nudge, jorbin, afercia, sumitsingh, sabernhardt.
See #59289, #59664.
Built from https://develop.svn.wordpress.org/trunk@57018
git-svn-id: http://core.svn.wordpress.org/trunk@56529 1a063a9b-81f0-0310-95a4-ce76da25c4cd
The affected lines already have ignore annotations in the `wp-includes/class-wp-block-parser-block.php` file itself.
Follow-up to [56048], [56738], [56743], [56751], [56752], [56753].
Props jrf, SergeyBiryukov.
See #59161.
Built from https://develop.svn.wordpress.org/trunk@57017
git-svn-id: http://core.svn.wordpress.org/trunk@56528 1a063a9b-81f0-0310-95a4-ce76da25c4cd
Rename the option cache priming functions to more closely follow the naming convention used by other cache priming functions.
* `wp_load_options()` becomes `wp_prime_option_caches()`
* `wp_load_options_by_group()` becomes `wp_prime_option_caches_by_group()`
The unit test files and classes are renamed accordingly.
Unlike the existing cache priming functions, these functions were introduced with the intention of being public so use the `wp_` prefix rather than the `_` prefix used by the functions initially introduced as private functions but since made public.
Follow up to [56445],[56990].
Props flixos90, peterwilsoncc, joemcgill, SergeyBiryukov, desrosj.
Fixes#58962.
Built from https://develop.svn.wordpress.org/trunk@57013
git-svn-id: http://core.svn.wordpress.org/trunk@56524 1a063a9b-81f0-0310-95a4-ce76da25c4cd
[56834] adjusted the order of activity inside the rest server responses. This lead to the `rest_pre_serve_request` filter potentially blocking the sending of the no cache headers. This moves that action back to being after the sending of no cache headers has finished to restore the pre 6.3.2 order of these two actions.
Props perrelet, SergeyBiryukov, peterwilsoncc.
Fixes#59722.
Built from https://develop.svn.wordpress.org/trunk@57012
git-svn-id: http://core.svn.wordpress.org/trunk@56523 1a063a9b-81f0-0310-95a4-ce76da25c4cd
The Customizer live preview broke because of [56635], however the root cause for the bug was a lower-level problem that had been present since WordPress 5.8: The block theme specific functions `_add_default_theme_supports()` and `wp_enable_block_templates()` were being hooked into the `setup_theme` action, which fires too early to initialize theme features. Because of that, theme functionality would be initialized before the current theme setup being completed. In the case of the Customizer, that includes overriding which theme is the current theme entirely, thus leading to an inconsistent experience.
This changeset fixes the bug by moving those two callbacks to the `after_setup_theme` action, which is the appropriate action to initialize theme features.
Props karl94, hellofromTonya, joemcgill, flixos90.
Fixes#59732.
See #18298, #53397, #54597.
Built from https://develop.svn.wordpress.org/trunk@57009
git-svn-id: http://core.svn.wordpress.org/trunk@56520 1a063a9b-81f0-0310-95a4-ce76da25c4cd
This update includes updates to patterns to correct color issues and some code quality fixes.
Follow-up to [56951], [56813], [56764], [56716].
Props luminuu, richtabor, onemaggie, kafleg, swissspidy, huzaifaalmesbah, neilorangepeel, shailu25, lada7042, mukesh27, nilovelez, jorbin.
Fixes#59711.
Built from https://develop.svn.wordpress.org/trunk@56999
git-svn-id: http://core.svn.wordpress.org/trunk@56510 1a063a9b-81f0-0310-95a4-ce76da25c4cd
The method attempted to check if there is already a file with the same name, however the conditional used an undefined variable.
This commit prevents directory creation if a file or directory with the same name already exists, bringing consistency with the PHP `mkdir()` implementation.
Includes adding missing documentation for the method.
Reference: [https://www.php.net/manual/en/streamwrapper.mkdir.php PHP Manual: streamWrapper::mkdir()].
Follow-up to [49230].
Props david.binda, sadizaman, rajinsharwar, SergeyBiryukov.
Fixes#59406.
Built from https://develop.svn.wordpress.org/trunk@56998
git-svn-id: http://core.svn.wordpress.org/trunk@56509 1a063a9b-81f0-0310-95a4-ce76da25c4cd
Fixes static property handling for `WP_Duotone::$block_css_declarations` in the `Tests_Block_Supports_Duotone::test_css_declarations_are_generated_even_with_empty_block_content()`:
* Fixes `ReflectionProperty::setValue()` to use an instance of `WP_Duotone`.
* Adds an inline comment to explain why a static class (i.e. a class that is not intended to be an object by design as it only contains static properties and methods) needs an instance, i.e. needed for PHP 8.3 and higher.
* Resets the static property's value to its original value, i.e. before the test started.
Follow-up to [56991].
Props costdev.
See #59694.
Built from https://develop.svn.wordpress.org/trunk@56996
git-svn-id: http://core.svn.wordpress.org/trunk@56507 1a063a9b-81f0-0310-95a4-ce76da25c4cd
Make layout support compatible with enhanced pagination by ensuring that generated class names are stable across pagination, even when the number of rendered posts is different.
With the previous implementation of enhanced pagination, the CSS corresponding to each block was not detected. Therefore, for enhanced pagination to work correctly, the CSS of the blocks present in the Post Template must be stable on all pages.
The number of posts rendered by the Query block is always the same, except in the last page, where it can be only a fraction. If any of the blocks rendered by the Post Template used the `wp_unique_id` function, the ID (which is incremental) would have been different than in the previous pages and the class names would have varied.
This is remediated by this changeset by replacing the usage of `wp_unique_id` in the layout support (which is used by the Query block) with an implementation that uses IDs that are incremental only for that block. That way, the generated class names are never affected by the number of times `wp_unique_id` runs.
Props luisherranz, andrewserong, isabel_brison, costdev, mukesh27, cbravobernal, hellofromTonya, jorbin.
Fixes#59681.
Built from https://develop.svn.wordpress.org/trunk@56994
git-svn-id: http://core.svn.wordpress.org/trunk@56505 1a063a9b-81f0-0310-95a4-ce76da25c4cd
Some blocks do not have content. For duotone support, blocks without content still need to run through the `render_duotone_support()` to render their duotone CSS.
This fix makes the duotone compatible with the enhanced pagination (introduced in 6.4.0) by making sure that the CSS is always on the page, even when the posts have no featured image. It also prevents the duotone from interfering with other blocks using `wp_unique_id()`.
References:
* [https://github.com/WordPress/gutenberg/pull/55415 Gutenberg PR 55415]
Follow-up to [56226].
Props cbravobernal, luisherranz, hellofromTonya, isabel_brison, jorbin.
Fixes#59694.
Built from https://develop.svn.wordpress.org/trunk@56991
git-svn-id: http://core.svn.wordpress.org/trunk@56502 1a063a9b-81f0-0310-95a4-ce76da25c4cd
This clearly separates these functions which are intended to be used by external developers from the existing `_prime_*_caches()` functions which are primarily intended for internal usage. The term "load" is additionally more accessible than "prime".
This changeset renames the above function, as well as the wrapper function `prime_options_by_group()` to `wp_load_options_by_group()`.
Props peterwilsoncc, joemcgill, hellofromTonya, poran766, flixos90.
Fixes#58962.
Built from https://develop.svn.wordpress.org/trunk@56990
git-svn-id: http://core.svn.wordpress.org/trunk@56501 1a063a9b-81f0-0310-95a4-ce76da25c4cd
When the XML sitemaps feature was originally introduced, the `lastmod` field was omitted because guidance at the time indicated it was less important for search engines, plus for some entities it was computationally expensive to add. Now that the guidance has slightly changed, we are revisiting this and adding `lastmod` where easily possible.
- Adds `lastmod` to all individual post objects (of any post type) in the sitemap
- Adds `lastmod` to the homepage sitemap entry if the homepage is set to display the latest posts.
No `lastmod` is added for the individual sitemap pages in the sitemap index, nor for term archives or user archives. Those enhancements require additional changes, such as storing the modified date for a taxonomy term when something is added to that term. They can be revisited in separate follow-up tickets.
Props swissspidy, joemcgill.
Fixes#52099
Built from https://develop.svn.wordpress.org/trunk@56985
git-svn-id: http://core.svn.wordpress.org/trunk@56496 1a063a9b-81f0-0310-95a4-ce76da25c4cd
In [56765], theme block pattern files were cached to a transient as a performance enhancement. However, transients are not easily clearable when caches are flushed on environments not using a persistent cache, which can lead to errors if the theme files are renamed, edited, or moved.
This changes the caching mechanism to use `wp_cache_set()` instead, and caches these values to the global group so they are still persistent on environments using an object cache, and will be cleared by a cache flush.
In addition, the helper `_wp_get_block_patterns` has been moved `WP_Theme::get_block_patterns` for consistency with other block related theme methods and cache helpers for these values, `WP_Theme::get_pattern_cache` and `WP_Theme::set_pattern_cache`, have been made private.
Relevant unit tests updated.
Props: afercia, flixos90, mukesh27, joemcgill.
Fixes#59633. See #59591, #59490.
Built from https://develop.svn.wordpress.org/trunk@56978
git-svn-id: http://core.svn.wordpress.org/trunk@56489 1a063a9b-81f0-0310-95a4-ce76da25c4cd
A very common contributor setup is having a copy of the `gutenberg` development repository within a checkout of the `wordpress-develop` repository. On occasion, there are some strange incompatibilities that come up when using this setup. A few examples can be seen in #58671 and #59634.
This changeset helps ensure that these edge cases are not introduced by testing the Gutenberg plugin’s build process within WordPress configured to run from both the `src` and `build` directories.
This also renames the “Test npm” workflow to a more general “Test Build Processes”, which more accurately describes what is actually being tested within it and allows these new test jobs to be grouped in.
And finally, the logic within the workflow has been split out into two callable workflows. This helps avoid code duplication within the workflow, and allows for better grouping on the workflow run screen.
Props swissspidy, aferica, SergeyBiryukov, antonvlasenko, desrosj.
Fixes#59632. See #58671, #59634.
Built from https://develop.svn.wordpress.org/trunk@56976
git-svn-id: http://core.svn.wordpress.org/trunk@56487 1a063a9b-81f0-0310-95a4-ce76da25c4cd
The globals introduced in [56635] to cache the current theme directories in memory were not considering switching sites in a multisite network. This changeset addresses the bug including test coverage.
Props codex-m, jeremyfelt.
Fixes#59677.
See #18298.
Built from https://develop.svn.wordpress.org/trunk@56974
git-svn-id: http://core.svn.wordpress.org/trunk@56485 1a063a9b-81f0-0310-95a4-ce76da25c4cd
When a branch is created, there is no previous commit to reference in the `github.event.before` context, which causes the performance workflow to fail because there is no previous commit to perform a comparison with.
This adds a condition to check that `github.event.before` is not set to `0000000000000000000000000000000000000000`, which is the default value when there are no previous commits.
Props swissspidy.
See #588867.
Built from https://develop.svn.wordpress.org/trunk@56972
git-svn-id: http://core.svn.wordpress.org/trunk@56483 1a063a9b-81f0-0310-95a4-ce76da25c4cd
Peter Wilson
Sergey Biryukov
Aaron Jorbin
desrosj
hellofromTonya
John Blackbourn
Bernhard Reiter
Tammie Lister
joedolson
Joe McGill
Felix Arntz
Pascal Birchler
gziolo