Started pipeline but the mariadbd
name used only in 10.6 image, so fixed 10.6 and 10.3.22 images
Filed new issue as it's the blocker for new infra #3449463: Update DB images to prevent server restart →
andypost → created an issue.
Testing new approach https://git.drupalcode.org/project/drupalci_environments/-/commit/de898f...
Follow-up could be useful, at least a test to iterate not in production
I bet even start using local runner's cache can speed-up pipelines
Simplest approach is just add fallback cache https://docs.gitlab.com/ee/ci/caching/#per-cache-fallback-keys
Then tune-up push/pull policy https://docs.gitlab.com/ee/ci/caching/#use-a-variable-to-control-a-jobs-...
Alternatively, cspell for example (7.5MB) can use artifacts via https://docs.gitlab.com/ee/ci/jobs/job_artifacts.html#with-a-cicd-job-token
Is there a way to test it to prevent this kind of errors? for example assert()
in parser...
There's a lot of auto instrumentation packages available
- https://github.com/open-telemetry/opentelemetry-php-instrumentation
- https://packagist.org/packages/open-telemetry/opentelemetry-auto-pdo
- https://packagist.org/packages/open-telemetry/opentelemetry-auto-symfony
There's only 2 child issues left, both are features so the meta could be considered fixed
Added and enabled extension to 8.3-dev image and started CI using it to check if any effect
$ docker run --rm drupalci/php-8.3-apache:dev php --ri opentelemetry
opentelemetry
opentelemetry hooks => enabled
extension version => 1.0.3
Directive => Local Value => Master Value
opentelemetry.conflicts => no value => no value
opentelemetry.validate_hook_functions => On => On
opentelemetry.allow_stack_extension => Off => Off
All issues are resolved and extension since 1.0.2 looks stable https://github.com/open-telemetry/opentelemetry-php-instrumentation/rele...
I'd better keep to throw exception for security reasons, assertions are usually disabled
I find it ready now
btw deprecations should be updated to 11.0.0 for removal in 12.0.0
Thank you! Now only automated fxes left 📌 [PHP 8.4] Fix implicitly nullable type declarations Active
I'm using following config ATM for Nginx Unit
{
"access_log": "/dev/stdout",
"listeners": {
"*:80": {
"pass": "routes/main"
}
},
"routes": {
"main": [
{
"match": {
"uri": [
"!*/.well-known/*",
"/vendor/*",
"/core/profiles/demo_umami/modules/demo_umami_content/default_content/*",
"*.engine",
"*.inc",
"*.install",
"*.make",
"*.module",
"*.po",
"*.profile",
"*.sh",
"*.theme",
"*.tpl",
"*.twig",
"*.xtmpl",
"*.yml",
"*/.*",
"*/Entries*",
"*/Repository",
"*/Root",
"*/Tag",
"*/Template",
"*/composer.json",
"*/composer.lock",
"*/web.config",
"*sql",
"*.bak",
"*.orig",
"*.save",
"*.swo",
"*.swp",
"*~"
]
},
"action": {
"return": 404
}
},
{
"match": {
"uri": [
"/core/authorize.php",
"/core/install.php",
"/core/modules/statistics/statistics.php",
"~^/core/modules/system/tests/https?\\.php",
"/core/rebuild.php",
"/update.php",
"/update.php/*"
]
},
"action": {
"pass": "applications/drupal/direct"
}
},
{
"match": {
"uri": [
"!/index.php*",
"*.php"
]
},
"action": {
"return": 404
}
},
{
"match": {
"uri": [
"~^.*css_[a-zA-Z0-9-_]+\\.css(?:\\?.*)?$",
"~^.*js_[a-zA-Z0-9-_]+\\.js(?:\\?.*)?$"
],
"headers": [
{
"Accept-Encoding": "*gzip*"
}
]
},
"action": {
"pass": "routes/assets_gz"
}
},
{
"action": {
"share": "/var/www/html/web$uri",
"fallback": {
"pass": "applications/drupal/index"
}
}
}
],
"assets_gz": [
{
"action": {
"share": "/var/www/html/web${uri}.gz",
"response_headers": {
"Content-Encoding": "gzip"
},
"fallback": {
"pass": "routes/assets"
}
}
}
],
"assets": [
{
"action": {
"share": "/var/www/html/web${uri}",
"fallback": {
"pass": "applications/drupal/index"
}
}
}
]
},
"applications": {
"drupal": {
"type": "php",
"stdout": "/dev/stdout",
"stderr": "/dev/stderr",
"processes": {
"max": 4,
"spare": 2,
"idle_timeout": 120
},
"limits": {
"timeout": 300,
"requests": 1500
},
"options": {
"admin": {
"apc.serializer": "igbinary",
"memory_limit": "1G",
"opcache.jit_buffer_size": "20M"
}
},
"targets": {
"direct": {
"root": "/var/www/html/web/"
},
"index": {
"root": "/var/www/html/web/",
"script": "index.php"
}
}
}
}
}
Re-rolled and re-based https://git.drupalcode.org/project/drupal/-/merge_requests/8004 as reproducible set of commits
PS: Draft MR !7976 is just a attempt to run current state on PHP 8.4
andypost → changed the visibility of the branch 3427999-php-8.4-fix to hidden.
is there a measurements of effect on CPU?
beta updates (via commit 2ad2c6b1c4) fixed https://github.com/php-fig/http-factory/commit/2e735aa9fca24c41454faae25...
As I get https://github.com/Roave/BetterReflection/pull/1387 already fixed and BetterReflection should work with v5 too
Just a question why new conflict added to composer?
"nikic/php-parser": ">=5"
ATM there's 2 main forks
needs re-run failed test as it known as random-failure
No, for reason of tests
There was 1 error:
1) Drupal\Tests\contact_storage\Functional\ContactStorageTest::testContactStorage
Behat\Mink\Exception\ElementNotFoundException: Form field with id|name|label|value "field_storage[subform][settings][allowed_values]" not found.
/builds/issue/contact_storage-3158051/vendor/behat/mink/src/WebAssert.php:731
/builds/issue/contact_storage-3158051/web/core/tests/Drupal/Tests/UiHelperTrait.php:85
/builds/issue/contact_storage-3158051/web/core/modules/field_ui/tests/src/Traits/FieldUiTestTrait.php:105
/builds/issue/contact_storage-3158051/tests/src/Functional/ContactStorageTest.php:218
/builds/issue/contact_storage-3158051/vendor/phpunit/phpunit/src/Framework/TestResult.php:729
ERRORS!
Tests: 8, Assertions: 229, Errors: 1.
Left review on MR - we can add proper types with deprecation message as 11.x going beta
API is ready api.drupal.org
Merged current state to create minor release but it needs work to address my comments (maybe not all)
+1 but does it need change record?
I bet there's a few issues left mostly about upload progress (apcu, vs native)
Added branch with deprecation for 10.3 and filed CR https://www.drupal.org/node/3446709 →
Also fixed remaining CS, except phpcs:disable Drupal.Commenting.DocComment.ShortSingleLine
as DiffFormatter.php
is deprecated
There's also rector rule https://github.com/rectorphp/rector-src/commit/ff32c0c08a89f27ea34187d00... since 1.0.4
If we gonna inject something via environment it will need policy/guards to prevent sensible information leaking, a-la contrib project/dotenv →
Personally I prefer to have kind of wrapper ✨ Use symfony/runtime for less bespoke bootstrap/compatibility with varied runtime environments Active
Thanks! That's another reason to have a base build of php image to produce from it shared extensions for sodium and others
alphawebgroup → credited andypost → .
+1 to deprecate (not sure there's BC except magic get/set) for removal in 11.x
Discussed in slack with @catch and @smushgrave and decided to not split automated fixes
The issue is postponed follow-up
📌
[8.4] Fix implicitly nullable type declarations in docblocks
Postponed
That's because docblocks are not yet covered with sniffer
#2572645: [Meta] Fix 'Drupal.Commenting.FunctionComment' coding standard →
Repurposed it as there's already started work on interface fixes
postponed on parent issue
Discussed to not split fixes https://drupal.slack.com/archives/C04CHUX484T/p1715262829267749
Discussed to not split fixes https://drupal.slack.com/archives/C04CHUX484T/p1715262829267749
Discussed to not split fixes https://drupal.slack.com/archives/C04CHUX484T/p1715262829267749
The module can just decorate FeaturesManager
There's SlevomatCodingStandard.TypeHints.NullableTypeForNullDefaultValue
sniffer to catch such issues
updated IS with remaining issues
It looks ready to go except renaming of public properties probably needs change record
There's one more deprecation leftover with own issue 🐛 Leftover D10 deprecations in ajax.js Needs work
Filed 📌 Remove core/scripts/drupal.sh Needs review
remaining deprecation in test-suites according to #3443732-20: Remove deprecated PHPUnit test suites →
This one will be eventually fixed in 📌 Upgrade PHPUnit to 10, drop Symfony PHPUnit-bridge dependency Needs work
andypost → created an issue.
I see no reason to allow weight
to be nullable
Thank you! Curious if both should be mentioned as dependencies from config affects a install-from-config order
+1 to provide submodule to export as recipes
thank for elaboration
Only 11.x fail but 11.0.x does not https://git.drupalcode.org/project/drupal/-/pipelines?page=1&scope=all&r...
is how it looks
something is broken when rendering login and logout confirmation forms
Does it mean it needs to update dump or we can just add extra fixtures to existing dumps?
Created 2 MRs - both enables testing and fixes install of the module
Re-formatted according to current standards
- 1.0.x - missing property
- 1.1.x - does not get controller resolver
andypost → created an issue.
All reported places are fixed, hope tests are green as changes makes sense
updated steps to reproduce
Thank you, makes sense to add some test and setup CI
Thank you! Looks the remaining manual changes also backportable #3444232-13: [8.4] Fix implicitly nullable type declarations for phpstan job (manual ones) →
Re-rolled after 📌 [PHP 8.4] Fix phpstan job for ContentModerationConfigureEntityTypesForm Needs review and added 10.3.x MR to make sure it's backportable
Make title clear
Variation cache API added in 10.2 https://www.drupal.org/node/3365546 →
so it needs changes to ^10.2
https://git.drupalcode.org/project/cache_review/-/commit/c6374867771ce83...
Yes, idea is to fix places which require manual intervention in 📌 [8.4] Fix implicitly nullable type declarations for phpstan job Needs review
The rest should be done using fixer to get reproducible results
I split out 📌 [PHP 8.4] Fix phpstan job for ContentModerationConfigureEntityTypesForm Needs review but looks we can just make argument required
andypost → created an issue.
Thank you!
still needs work to split removal of status-code checks, so probably 12.x material
2 child issues left
both needs work, but 📌 Finish deprecating status code and header assertions in JS tests Active can just update for removal in 12.x
book is done
Composer facade fixed
It means modules should be fixed to require EventDispatcherInterface
instead of specific class
one test failed and I left a question for constructor's BC