In the past overviews, the date order is reversed. There it says: “No incidents reported between 2025-01-13 and 2025-01-07”, but the general way of speaking is to take the day further in the past first. It should therefore read “No incidents reported between 2025-01-17 and 2025-01-13”. The date selection fields are also reversed accordingly.
Found 65 issues
First time here? 👋
Welcome to Find a PR.
Find a PR is an open-source site that is built to help developers find projects so that they can submit their very first pull request.
If you're a contributor looking to find a project to contribute to, feel free to browse through the list below.
If you're a maintainer looking to list your project on Find a PR, you can read how to do this in the documentation.
Many of them are out of date, some by a few major versions!
It would be great to get these into good order and start keeping them up to date
simonhamp
24th Oct 2024 @ 11:53
Figure our a way to show a grayed out Laravel logo as a default avatar instead of the current default GitHub one.
driesvints
29th Sep 2021 @ 10:48
It should be possible to configure Cachet via a cachet:install
Artisan command.
v2.x had an interactive command that would ask you questions and store the configuration.
For v3.x, this won't work exactly the same as we now have a mix of .env
and database settings, but the idea is the same.
jbrooksuk
9th Oct 2024 @ 20:21
Describe the bug
I have a table which has an order
column. When I reorder the elements I update each model with the new order.
Because the table has some large JSON columns I select only id
and order
columns for efficiency. I set the new order and save the model.
foreach (Faq::query()->get(['id', 'order']) as $faq) {
$faq->order = $newOrder[$faq->id];
$faq->save();
}
The diffs created in the activity_log
table look like this:
{
"old": {
"order": 1,
"answer": null,
"question": null
},
"attributes": {
"order": 2,
"answer": "long JSON content",
"question": "long JSON content"
}
}
From what I see laravel-activitylog fetches the model with all columns from the database before creating a log entry which kind of defeats the purpose of my "select list optimization" and produces an incorrect diff.
Generally I want these large JSON columns to be logged in case a user changes them but in this case I only change the order
column so I would like to see only the order
column in the diff.
I have tried running these updates straight off the Eloquent builder:
Faq::query()->where('id', $id)->update(['order' => $newOrder[$id]]);
but in this case nothing is logged.
I have the following activitylog configuration for my models:
public function getActivitylogOptions(): LogOptions
{
return LogOptions::defaults()
->logAll()
->logOnlyDirty()
->logExcept([
'id',
'created_at',
'updated_at',
]);
}
To Reproduce Select a subset of columns from the database, change these columns, save the model and see that other (not initially selected) columns are present in the diff.
Expected behavior Only columns that've actually been changed should be present in the diff.
Versions
- PHP: 8.1
- Database: MySQL 8
- Laravel: 9.31.0
- Package: 4.6.0
KKSzymanowski
28th Sep 2022 @ 21:53
Describe the bug
logUnguarded()
works provided that $guarded
is set explicitly in the model. If it is not set explicitly but globally set, say in AppServiceProvider
using Model::unguard();
then nothing is logged.
To Reproduce
- remove
$fillable
and$guarded
from a model that use usesLogsActivity
- call
Model::unguard();
inAppServiceProvider
in theboot
method - set
public function getActivitylogOptions() : LogOptions
{
return LogOptions::defaults()->logUnguarded();
}
- Make some changes to a model and save them
Expected behavior Some changed properties should be logged but none are
Versions (please complete the following information)
- PHP: 8.1
- Database: MySql
- Laravel: 9.38
- Package: larvel-activitylog
colinmackinlay
20th Nov 2022 @ 12:31
When encountering nested tokens, they are parsed only during formatting, not when compiling the pattern. And the type of tokens are validated only during the rendering as well. But the native implementation detects such error during the instantiation: https://3v4l.org/Sf69D This means that the error handling required for the polyfill is not the same than for the native implementation.
stof
23rd Oct 2018 @ 15:58
On the record update for an active incident, there needs to be a tie to the component to be update the current component status.
glipschitz
19th Jan 2025 @ 03:05
Hi! If I use this package with my project I has a problem. My package.json has a "git dependency" like this:
{
"peerDependencies": {
"package": "git+ssh://[email protected]/team/project.git"
}
}
And npm-install-peers
breaks.
JWo1F
18th Dec 2017 @ 08:52
As an extension of https://github.com/cachethq/core/issues/93 it'd be cool to have an option that shows completed schedules in the incident timeline.
jbrooksuk
24th Nov 2024 @ 15:42
Bug description
Currently, the plugin calculates width/height incorrectly when swapping aspect ratios, leading to incorrect HTML output. When changing the aspect ratio of an image, the longest side should remain fixed, and the other side should be calculated accordingly.
How to reproduce
For an image of 3000x1688px as a source:
{{ responsive:image ratio="16/9" md:ratio="9/16" }}
it returns:
<picture>
<source [...] >
<source [...] >
<source [...] >
<img src="/img/asset/hash/image.jpg?w=3000&h=1688&s=9358bb54be5d388a9a2529a4196dcfc6" alt="" width="3000" height="1688">
</picture>
Now for the same image as source but with:
{{ responsive:image ratio="9/16" md:ratio="16/9" }}
it returns:
<picture>
<source [...] >
<source [...] >
<source [...] >
<img src="/img/asset/hash/image.jpg?w=3000&h=5333&s=6c5c0bb8447dbd7ff61a692f98b5f46d" alt="" width="3000" height="5333">
</picture>
The calculated aspect ratio is correct, but the dimensions are not. The generated values exceed the original image size.
In this last example, the correct height should be 3000px, and the width should be 1688px to ensure the image stays within its original dimensions.
Environment
Environment
Application Name: local
Laravel Version: 11.41.3
PHP Version: 8.3.14
Composer Version: 2.8.5
Environment: development
Debug Mode: ENABLED
URL: local.test
Maintenance Mode: OFF
Timezone: Europe/Paris
Locale: en
Cache
Config: NOT CACHED
Events: NOT CACHED
Routes: NOT CACHED
Views: CACHED
Drivers
Broadcasting: null
Cache: file
Database: sqlite
Logs: stack / single
Mail: smtp
Queue: redis
Session: file
Statamic
Addons: 3
Sites: 1
Stache Watcher: Disabled (auto)
Static Caching: Disabled
Version: 5.46.0 PRO
Statamic Addons
rias/statamic-redirect: 3.9.3
spatie/statamic-responsive-images: 5.2.1
statamic/seo-pro: 6.5.0
Installation
Fresh statamic/statamic site via CLI
Antlers Parser
regex (default)
jimblue
3rd Feb 2025 @ 23:33
Reporting this as a bug, even if it may be I use the ux-icons bundle not in the indended fashion. Feel free to change issue type if this is the case.
I installed the UX icons package and use it to just download the icons with like
console ux:icons:import flowbite:user-solid
and reference them as background images with css.
The icons download just fine but I was surprised the browsers could not render them. It appears the xmlns attribute is missing
xmlns="http://www.w3.org/2000/svg"
so the browsers do not render the icons as icons but as xml.
This happens independent of what library I use. I tried akar-icons, devicons and fa6-brands.
While in my case i can add the xmlns manually, the point of using ux-icons for me is not to manually wrangle files but to just download and enjoy.
As stated, maybe I am using it wrong :)
eigentor
10th Feb 2025 @ 09:34
Bug description
Chrome, Firefox, and Safari now support setting the width
and height
attributes on <source>
elements within <picture>
elements. You can read more about this in the web.dev article on optimizing CLS. This enhancement allows browsers to better handle layout shifts and varying aspect ratios for responsive images.
How to reproduce
Current Behavior:
Currently, Statamic Responsive Images does not take advantage of this browser capability. As a result, when multiple sources are provided, the appropriate width
and height
attributes are not applied to <source>
elements, which may lead to layout shifts due to varying aspect ratios.
Proposed Enhancement:
Set appropriate width
and height
attributes on <source>
elements based on their aspect ratios.
Benefits:
- Improved Visual Stability: Reduce Cumulative Layout Shift (CLS) by providing browsers with the correct dimensions upfront.
- Better Performance: Modern browsers can allocate the correct space for images during page load.
- Enhanced Responsiveness: Accurately reflect different image aspect ratios for a more consistent user experience.
Environment
Environment
Application Name: local
Laravel Version: 11.41.3
PHP Version: 8.3.16
Composer Version: 2.8.5
Environment: development
Debug Mode: ENABLED
URL: local.test
Maintenance Mode: OFF
Timezone: Europe/Paris
Locale: en
Cache
Config: NOT CACHED
Events: NOT CACHED
Routes: NOT CACHED
Views: NOT CACHED
Drivers
Broadcasting: null
Cache: file
Database: sqlite
Logs: stack / single
Mail: smtp
Queue: redis
Session: file
Statamic
Addons: 3
Sites: 1
Stache Watcher: Disabled (auto)
Static Caching: Disabled
Version: 5.46.1 PRO
Statamic Addons
rias/statamic-redirect: 3.9.3
spatie/statamic-responsive-images: 5.2.1
statamic/seo-pro: 6.5.0
Installation
Fresh statamic/statamic site via CLI
Antlers Parser
regex (default)
Additional details
Leveraging native width and height support on <source>
elements can lead to significant improvements in layout stability and overall performance on modern browsers. This enhancement would allow the package to better handle responsive images by using up-to-date browser capabilities.
Thanks for considering this enhancement!
jimblue
7th Feb 2025 @ 22:03
Steps to reproduce this issue:
- having a migration file in the package.
- the migration file is already prefixed by a date and time.
- the package is configured with
->runsMigrations()
in it's ServiceProvider. - publish the migrations of the package. then the published migration name would prefixed by date again!
I reproduced this issue not only in my private package, but also in filament-attachmate package, which has a migration named as
database/migrations/2024_02_24_192352_create_attachments_table.php
and the package's service provider is allowed to run migrations:
class FilamentAttachmateServiceProvider extends PackageServiceProvider
{
public function configurePackage(Package $package): void
{
$package
->name('filament-attachmate')
->hasMigrations([
'2024_02_24_192352_create_attachments_table'
])
->runsMigrations();
}
}
everything is ok and the migration is successfully loaded. but after publishing the migration, the published filename would be prefixed again with a new date time string:
Copying file [vendor/zeeshantariq/filament-attachmate/database/migrations/2024_02_24_192352_create_attachments_table.php] to [database/migrations/2024_10_07_135947_2024_02_24_192352_create_attachments_table.php] DONE
the published file is prefixed by 2024_10_07_135947_2024_02_24_192352_
which is the date in vendor's filename prefixed by published date!
Some changes in the generateMigrationName method might fix this issue.
imami
7th Oct 2024 @ 14:42
I think a "Manage Localization" settings page would be more useful for language settings or something else like display timezone...
jbrooksuk
30th Oct 2024 @ 16:37
In raising this issue, I confirm the following (please check boxes):
- I have read and understood the contributors guide.
- I have checked the pull requests tab for existing solutions/implementations to my issue/suggestion.
- I have checked that the bug-fix I am reporting can be replicated.
Description of the problem
One of the important reasons for using hyperscript is that it handles escaping of characters which are reserved in html. Which prevents XSS.
spatie/html-element does not do this. Mainly because it uses strings as intermediate format instead of an object representation of the DOM.
Erikvv
1st Feb 2018 @ 02:19
- standard tags
- input fields
- components
- patterns
Add code snippets for each block
willemvb
25th Jul 2016 @ 12:49
Laravel Version
11.x
PHP Version
any
Database Driver & Version
irrelevant
Description
This is similar to #53692 but a separate issue.
It seems creating a DB connection too early (in a service provider, in a specific way) and using it later causes some weird "desynchronization" when using parallel tests.
I haven't had the time to do a deep dive yet, but the issue is likely related to some reconnecting logic similar to the linked issue.
The reproduction steps may seem very specific, i.e. why am I instantiating a cache store in a service provider. In practice this happens if you just use RateLimiter::for('foo', static fn () => null)
, but that under the hood creates the cache store which seems to be the underlying cause. So the reproduction steps are as low level as I managed to trace this issue.
In a real app, e.g. one based on Fortify which rate limits logins, any call to cache()
while using the database
cache driver, will cause weird behavior. There is a test in Jetstream that directly reproduces this (it makes a request to the login route) but the side effects don't have a perceptible effect. If there were some more DB-related assertions after the request, especially comparing DB state before the request and after, you'd see it.
Steps To Reproduce
-
laravel new
, no starter kit, select Pest for instance - Make the following changes in
phpunit.xml
:<env name="CACHE_STORE" value="database"/> <env name="DB_CONNECTION" value="sqlite"/> <env name="DB_DATABASE" value="database/tests.sqlite"/>
-
touch database/tests.sqlite
- Force cache store creation in a service provider:
// AppServiceProvider::boot() cache()->getStore();
- Add a test:
test('foobar', function () { expect(DB::select('select * from users'))->toHaveCount(0); User::create(['name' => 'John Doe', 'email' => '[email protected]', 'password' => 'foobar']); expect(DB::select('select * from users'))->toHaveCount(1); cache()->put('foo', 'bar'); expect(cache('foo'))->toBe('bar'); // users table is empty! expect(DB::select('select * from users'))->toHaveCount(1); });
-
artisan test -p
should fail on the last assertion - If you remove the service provider call, the test passes
stancl
19th Feb 2025 @ 20:07
AST, or "abstract syntax tree", creates the expectation of a nested tree structure. In the case of a query string, the AST will only ever be one level deep; as multi nodes are combined into one.
If anyone wants to suggest a better, more clear name: you're welcome.
brendt
20th Dec 2018 @ 13:41
What happened?
I just installed package and complement for AI powered solution. When I create an issue to test it I get :
I had to change the links function like this to make it work for me:
public function links(): array
{
$rawText = Str::finish($this->rawText, 'ENDLINKS');
$textLinks = $this->between('LINKS', 'ENDLINKS', $rawText);
$textLinks = explode(PHP_EOL, $textLinks);
$links = [];
foreach ($textLinks as $textLink) {
$textLink = str_replace('\\', '\\\\', $textLink);
$textLink = str_replace('\\\\\\', '\\\\', $textLink);
$decodedLink = json_decode($textLink, true);
if ($decodedLink !== null) {
$links[$decodedLink['title']] = $decodedLink['url'];
}
}
return $links;
}
How to reproduce the bug
When I install the package in my project and use it with AI powered
Package Version
1.0
PHP Version
8.2.14
Laravel Version
11.14
Which operating systems does with happen with?
Windows
Notes
No response
bestmomo
21st Jul 2024 @ 10:57
This is a big one, but would resolve https://github.com/cachethq/cachet/issues/4425
- Opt components into being monitored.
- Needs some kind of configuration (what if a Header is required to access the URL? Frequency?)
- Add a cron job that runs every minute and checks the actively components.
- Track component availability.
- Dashboard to manage component monitors, track availability etc.
- API
jbrooksuk
4th Jan 2025 @ 18:12
Laravel Version
11.41.3
PHP Version
8.4.3
Database Driver & Version
Mysql 8.0.40 - Debian 12
Description
When pushing a job to the database queue using the displayName()
method to set a custom display name in the queue worker, and with the $deleteWhenMissingModels
property set to true
, a ModelNotFoundException
is still logged and the job is going through the fail()
lifecycle if the model is deleted in the meantime.
This happens because the handleModelNotFound
() method in CallQueueHandler
resolves the job name using $class = $job->resolveName();
, which calls JobName::resolve($this->getName(), $this->payload());
This method returns the displayName, which is not a valid class name because the displayName is changed.
As a result, ReflectionClass fails to reflect the class, causing the system to incorrectly determine that shouldDelete is false, leading to running the fail() lifecycle of the job in stead of quietly discard the job.
Steps To Reproduce
- Create a database job with the
$deleteWhenMissingModels
property set totrue
- Add a model to the constructor of the job
- Add an custom displayName() method to the job that returns an non existing class
- Create a test that creates the model, dispatches the job and delete the model before running the (DB) queue worker.
- Make sure to run the test expecting no errors to be logged.
JochemLettink
7th Feb 2025 @ 10:42
When calling mb_convert_encoding()
with $fromEncoding === 'HTML-ENTITIES'
, the polyfill does not return functionally equivalent strings to the native function. This is because mb_convert_encoding()
uses html_entity_decode()
when $fromEncoding === 'HTML-ENTITIES'
and that function does not return characters for many numeric entities 0-31 and 127-159. For example:
<?php
require "vendor/symfony/polyfill-mbstring/Mbstring.php";
use Symfony\Polyfill\Mbstring as p;
for($i = 0; $i < 1024; $i++) {
$string = "&#" . $i . ";";
$mbstring = mb_convert_encoding($string, 'UTF-8', 'HTML-ENTITIES');
$polyfill = p\Mbstring::mb_convert_encoding($string, 'UTF-8', 'HTML-ENTITIES');
if($mbstring != $polyfill) {
echo "Mismatch: $string - mbstring: $mbstring; polyfill: $polyfill\n";
}
}
outputs:
Mismatch: � - mbstring: ; polyfill: �
Mismatch:  - mbstring: ; polyfill: 
Mismatch:  - mbstring: ; polyfill: 
Mismatch:  - mbstring: ; polyfill: 
Mismatch:  - mbstring: ; polyfill: 
Mismatch:  - mbstring: ; polyfill: 
Mismatch:  - mbstring: ; polyfill: 
Mismatch:  - mbstring: ; polyfill: 
Mismatch:  - mbstring:; polyfill: 
Mismatch:  - mbstring:
; polyfill: 
Mismatch:  - mbstring:
; polyfill: 
Mismatch:  - mbstring: ; polyfill: 
Mismatch:  - mbstring: ; polyfill: 
Mismatch:  - mbstring: ; polyfill: 
Mismatch:  - mbstring: ; polyfill: 
Mismatch:  - mbstring: ; polyfill: 
Mismatch:  - mbstring: ; polyfill: 
Mismatch:  - mbstring: ; polyfill: 
Mismatch:  - mbstring: ; polyfill: 
Mismatch:  - mbstring: ; polyfill: 
Mismatch:  - mbstring: ; polyfill: 
Mismatch:  - mbstring: ; polyfill: 
Mismatch:  - mbstring: ; polyfill: 
Mismatch:  - mbstring: ; polyfill: 
Mismatch:  - mbstring: polyfill: 
Mismatch:  - mbstring: ; polyfill: 
Mismatch:  - mbstring: ; polyfill: 
Mismatch:  - mbstring: ; polyfill: 
Mismatch:  - mbstring: ; polyfill: 
Mismatch: ' - mbstring: '; polyfill: '
Mismatch:  - mbstring: ; polyfill: 
Mismatch: € - mbstring: ; polyfill: €
Mismatch:  - mbstring: ; polyfill: 
Mismatch: ‚ - mbstring: ; polyfill: ‚
Mismatch: ƒ - mbstring: ; polyfill: ƒ
Mismatch: „ - mbstring: ; polyfill: „
Mismatch: … - mbstring:
; polyfill: …
Mismatch: † - mbstring: ; polyfill: †
Mismatch: ‡ - mbstring: ; polyfill: ‡
Mismatch: ˆ - mbstring: ; polyfill: ˆ
Mismatch: ‰ - mbstring: ; polyfill: ‰
Mismatch: Š - mbstring: ; polyfill: Š
Mismatch: ‹ - mbstring: ; polyfill: ‹
Mismatch: Œ - mbstring: ; polyfill: Œ
Mismatch:  - mbstring: ; polyfill: 
Mismatch: Ž - mbstring: ; polyfill: Ž
Mismatch:  - mbstring: ; polyfill: 
Mismatch:  - mbstring: ; polyfill: 
Mismatch: ‘ - mbstring: ; polyfill: ‘
Mismatch: ’ - mbstring: ; polyfill: ’
Mismatch: “ - mbstring: ; polyfill: “
Mismatch: ” - mbstring: ; polyfill: ”
Mismatch: • - mbstring: ; polyfill: •
Mismatch: – - mbstring: ; polyfill: –
Mismatch: — - mbstring: ; polyfill: —
Mismatch: ˜ - mbstring: ; polyfill: ˜
Mismatch: ™ - mbstring: ; polyfill: ™
Mismatch: š - mbstring: ; polyfill: š
Mismatch: › - mbstring: ; polyfill: ›
Mismatch: œ - mbstring: ; polyfill: œ
Mismatch:  - mbstring: ; polyfill: 
Mismatch: ž - mbstring: ; polyfill: ž
Mismatch: Ÿ - mbstring: ; polyfill: Ÿ
While many of these are control characters (and the native function does return them), the single quote (dec 39) is particularly problematic.
cpeel
18th Mar 2021 @ 04:11
Laravel Version
11.42.0
PHP Version
8.4.3
Database Driver & Version
No response
Description
Define a custom attribute with a convoluted name using the Attribute
syntax:
Example:
protected function foo1Bar(): Attribute
{
return Attribute::make(
get: fn () => 'yay',
);
}
Accessing a custom attribute is internally done by a snake-to-camel case conversion. So this means both foo1_bar
and foo_1_bar
will yield the expected value.
However, some features in Eloquent do the inverse by calling upon the attribute cache ($getAttributeMutatorCache
): a camel-to-snake case conversion. This means that foo1Bar
now only translates to foo1_bar
, and not anymore to foo_1_bar
even though its value would get returned when retrieving it.
echo $model->foo1_bar; // "yay"
$model->append('foo1_bar');
$model->toArray(); // Works fine.
echo $model->foo_1_bar; // "yay"
$model->append('foo_1_bar');
$model->toArray(); // Call to undefined method Model::getFoo1BarAttribute()
On the contrary, when resorting back to the good old getFoo1BarAttribute()
everything works as expected for both foo1_bar
and foo_1_bar
.
protected function getFoo1BarAttribute()
{
return 'yay';
}
echo $model->foo1_bar; // "yay"
$model->append('foo1_bar');
$model->toArray(); // Works fine.
echo $model->foo_1_bar; // "yay"
$model->append('foo_1_bar');
$model->toArray(); // Works fine.
So, because of internal two-way case conversions to support the newer attribute syntax some unexpected and disfunctional ambiguity gets introduced.
I think both syntaxes (Attribute
vs getXXXAttribute()
) should behave equally. But I'm not quite sure how to proceed with this. Should foo_1_bar
get blocked for access to prevent this kind of expectations further down the line? Or should it also resolve properly when doing the camel-to-snake conversion? Or...?
Steps To Reproduce
protected function foo1Bar(): Attribute
{
return Attribute::make(
get: fn () => 'yay',
);
}
$model->append('foo_1_bar');
$model->toArray();
Propaganistas
11th Feb 2025 @ 20:29
When attempting to filter tests by test or file name pressing the "Enter" key has no effect.
Pressing "Enter" at the default screen correctly triggers a "test run" where all tests are run.
ntwb
25th Feb 2018 @ 10:49
Describe the bug
When you attempt to dump a binary string or anything containing one (like an array or an object), Ray does not dump anything. It just fails silently.
Versions
Ray version (you can see this in "About Ray"): 2.8.1
You can use composer show
to get the version numbers of:
- spatie/ray package version: 1.41.2
- spatie/laravel-ray package version (if applicable): 1.37.1
PHP version: 8.2.17 Laravel version (if applicable): 11.18.1
To Reproduce
Add this in your code:
ray(hex2bin('ae0f3d'));
Expected behavior
Ray should display this (like the console when dump
is used):
b"®\x0F="
Desktop (please complete the following information):
- OS: MacOs
- Version: 14.6.1
sebj54
27th Aug 2024 @ 14:57
Discussed in https://github.com/spatie/schema-org/discussions/202
Originally posted by indyjonesnl January 10, 2024
$lodgingBusiness = Schema::lodgingBusiness()
->openingHours(
Schema::openingHoursSpecification()
->dayOfWeek([Schema::dayOfWeek()::Monday])
->opens(new DateTime('09:00:00'))
->closes(new DateTime('17:00:00'))
)
->checkinTime(new DateTime('14:00:00'))
->checkoutTime(new DateTime('11:00:00'))
This results in the current date + time being included in the output "opens":"2024-01-10T09:00:00+00:00","closes":"2024-01-10T17:00:00+00:00"
and ..."checkinTime":"2024-01-10T14:00:00+00:00","checkoutTime":"2024-01-10T11:00:00+00:00"...
While the Schema should contain only the time, formatted as 14:30:00+08:00
.
Can someone point me in the right direction?
Gummibeer
11th Jan 2024 @ 13:12
Calling for help from Docker experts. We need to create the best possible docker-compose.yml
file for this project. The application requirements are well defined (we use env vars, Webpack Encore, PHP 7.1, Symfony 4.1, SQLite database, etc.) so it should be possible to create that file.
javiereguiluz
21st May 2018 @ 09:37
Laravel Version
11.41.3
PHP Version
8.2.27
Database Driver & Version
No response
Description
Laravel's Concurrency::run() does not properly handle exceptions that require multiple constructor arguments. When an exception is thrown inside a concurrent task, Laravel only serializes the message property, and when it reinstantiates the exception in the parent process, it calls the constructor with only $message instead of all expected arguments.
This causes an ArgumentCountError when the exception constructor requires multiple parameters.
I used debugger to investigate what happens
In my code example below I was able to find that throw new ApiRequestException() happens once, but parent::construct called twice, first time using proper passed args, but the second time it goes only with one argument which is "API request to {$uri} failed with status $statusCode $reason"
in my case. I have tried to put args to parent::__construct like ("message", $code, ...) but it wouldn't work for me (I assume that it is because it calls construct of Exception class). ChatGPT explained that issue like that (I'm not saying that it is correct thing and also it was unable to give me any workaround of that):
This problem occurs because Laravel’s Concurrency::run() spawns subprocesses for concurrent tasks. When an exception is thrown inside a subprocess, Laravel serializes it to pass it back to the main process. However, Laravel only retains the message property when serializing, and when it attempts to reinstantiate the exception, it calls the constructor with only $message, instead of all required arguments, causing an ArgumentCountError.
Steps To Reproduce
- Create a Custom Exception with Multiple Arguments
<?php
namespace App\Exceptions;
use Exception;
class ApiRequestException extends Exception
{
public function __construct(
public string $uri,
public int $statusCode,
public string $reason,
public string|array $responseBody = '',
) {
parent::__construct("API request to {$uri} failed with status $statusCode $reason");
}
}
- Throw the Exception Inside a Concurrent Task
<?php
use Illuminate\Support\Facades\Concurrency;
use App\Exceptions\ApiRequestException;
[$result1, $result2] = Concurrency::run([
fn () => throw new ApiRequestException('https://api.example.com', 400, 'Bad Request', 'Invalid payload'),
fn () => 'Task 2 completed',
]);
max-shteklia
7th Feb 2025 @ 02:18
Provide updates to new functionalities in the v2 of the package.
kwakuOfosuAgyeman
5th Jan 2025 @ 05:45
Hi,
With the polyfill, var_dump(mb_strlen(chr(254)))
return 0
.
With the php8.0-mbstring extension, var_dump(mb_strlen(chr(254))) return 1
;
versions : * v1.26.0
Thanks, Alex
alexchuin
18th Oct 2022 @ 12:51
For some reason, our test suite is super slow. We should look into this and try to decrease memory usage and speed it up.
Parallel run:
Tests: 316 passed (785 assertions)
Duration: 13.64s
Parallel: 10 processes
Regular run:
Tests: 316 passed (785 assertions)
Duration: 96.30s
driesvints
29th Dec 2023 @ 10:43
We are logging activity of several models, some using SoftDeletes trait and others not.
When we set config option 'subject_returns_soft_deleted_models' => true
then calling subject()
on any model not using SoftDeletes trait throws an exception: Call to undefined method Illuminate\Database\Eloquent\Builder::withTrashed()
.
Can the subject()
method below be made to check first whether the Model uses the SoftDeletes trait? Or perhaps check whether withTrashed()
is a defined method on the Model? I tried a few things but couldn't figure out how to get the class of the Model..
bluec
7th Nov 2018 @ 13:44
Add in PHP 7.3 as Normalizer::normalize() argument for NFKC_Casefold normalization.
nicolas-grekas
7th Feb 2019 @ 10:11
Octane Version
2.6.1
Laravel Version
11.41.3
PHP Version
8.3
What server type are you using?
Roadrunner
Server Version
2024.3.2
Database Driver & Version
No response
Description
ok this is in the borderline to be a bug but i write it anyway.
i am deploying using deployer , i suppose will happen for any Zero Downtime Deployments.
/releases/99/
/releases/100/
/releases/101/
and then a symbolic link from the latest release to /current
, all pretty standard.
the problem come when running artisan octane:start
vendor/laravel/octane/src/Commands/StartRoadRunnerCommand.php
will set roadrunner server with
'-o', 'server.command='.(new PhpExecutableFinder)->find().','.base_path(config('octane.roadrunner.command', 'vendor/bin/roadrunner-worker')),
where base_path(...) will resolve to /releases/101/
not to /current
so in next deployment e.g. (/releases/102/) , when we run artisan octane:reload
it will still use old /releases/101/
source code
(i checked and yes if i go back to /releases/101/
and make a change it will be present after artisan octane:reload
).
the solution is very simple, in config/octane.php
you just need to add:
'roadrunner'=>[
'command' => env('OCTANE_ROADRUNNER_WORKER_PATH', base_path('vendor/bin/roadrunner-worker')),
]
OCTANE_ROADRUNNER_WORKER_PATH=../../current
would be nice is this added to config/octane.php
, and maybe don't use base_path
or even better a option to provide the path in octane:start
Steps To Reproduce
deploy a octane laravel app using deployer , octane:reload will not work as expected
inikoo
13th Feb 2025 @ 14:32
Laravel Version
12
PHP Version
any
Database Driver & Version
irrelevant
Description
Given:
- A route like
Route::get('/{foo:slug}/example/{bar}', ...)->name('example');
-
URL::defaults(['foo' => 'some_value']);
route('example', $bar)
will produce a "Missing required parameter" exception.
On the other hand, route('example', ['bar' => $bar])
will work fine.
The issue comes from UrlGenerator::toRoute()
:
public function toRoute($route, $parameters, $absolute)
{
// $parameters = [$bar]
$parameters = Collection::wrap($parameters)->map(function ($value, $key) use ($route) {
// $value = $bar, $key = 0
// true since it's a model && true because field '0' (foo:slug) has a binding field (slug)
return $value instanceof UrlRoutable && $route->bindingFieldFor($key)
? $value->{$route->bindingFieldFor($key)}
: $value;
})->all();
// $parameters = [null because we've incorrectly done $bar->slug]
array_walk_recursive($parameters, function (&$item) {
if ($item instanceof BackedEnum) {
$item = $item->value;
}
});
return $this->routeUrl()->to(
$route, $this->formatParameters($parameters), $absolute
);
}
The default parameter only gets used in $this->routeUrl()
(RouteUrlGenerator
), but by that point the UrlGenerator
has broken the provided parameters, passing a [null]
array.
I think the solution would be something along the lines of: Actually probably a bit more complex than this because defaults can work for latter parameters too, not just the earliest ones. I'll try to send a PR with some reasonable implementation.
- If
$key
is numeric, aka we've passedroute('example', $bar)
notroute('example', ['bar' => $bar])
- And
array_key_exists($route->parameterNames()[$key], $this->getDefaultParameters())
- Skip the
[$key]
parameter and move on to the next one - (Repeat as many times as needed — in my example there's one parameter with a default value but there can be several)
So in this case $bar
should be matched against the second parameter, not the first one.
Steps To Reproduce
See above
stancl
25th Feb 2025 @ 23:46
Octane Version
2.6.2
Laravel Version
11.16.0
PHP Version
8.3.6
What server type are you using?
Swoole
Server Version
5.1.2
Database Driver & Version
No response
Description
Before i start, this is not a duplicate to https://github.com/laravel/octane/issues/903, this is about Swoole. Also large stream responses are fixed with https://github.com/laravel/octane/issues/636, but not this case.
While using Storage::download or Storage::response, laravel is creating a stream response that uses php fpassthru($stream). Somehow the buffering between fpassthru and ob_start does not work, instead it tries to load the full file into the memory. This leads to out of memory errors while downloading very big files.
If i replace the fpassthru with a fread-while-loop it is working as expected.
I can basically make a pull request, but need help with the solution... Is this more a problem in the laravel framework or PHP itself?
Steps To Reproduce
- Write a laravel controller that uses Storage::download() to download file with e.g. 1 GB
- Call the controller and see the out of memory error
NiroDeveloper
24th Jul 2024 @ 11:18
Dashboard showing Operational
Front end showing no issues, but the status seems to still be linked to the incident even if it's closed as fixed
glipschitz
25th Jan 2025 @ 11:01
We need a method of migrating users from 2.4 to 3.x.
Things to consider:
- Migrate settings
- Migrate resources
- Migrate subscriptions / subscribers
- Migrate API Keys
AOB (please add to this list)
As a comment on the latter, once #183 is merged, we'll need to programmatically migrate users 2.4 API Keys to 3.x using something like the following
$v2ApiKey = $v2User->api_key;
$v3User->tokens()->create([
'name' => 'v2.4 API Key',
'token' => hash('sha256', $v2ApiKey),
'abilities' => ['*'],
'expires_at' => now()->addDays(90),
]);
@jbrooksuk We need to decide on the abilities
and expires_at
. IMO it makes sense to throw something into the upgrade documentation that old 2.4 API Keys will continue to work for up to 90 days. But if they want to use the API after that, then they'll need to generate a new API Token using the 3.x dashboard.
joelbutcher
15th Jan 2025 @ 18:19
Let's try out Filament and move our own admin panels to it. This should provide us with a much more powerful admin backend.
driesvints
22nd Dec 2023 @ 09:06
Octane Version
v2.3.10
Laravel Version
v11.6.0
PHP Version
v8.3.6
What server type are you using?
FrankenPHP
Server Version
v1.1.4 PHP 8.3.6 Caddy v2.7.6
Database Driver & Version
Postgress
Description
Last week we updated our app previously using php-fpm
running on Forge to use Laravel Octane with FrankenPHP. Our site is mostly an API that handles analytics events (Like google analytics). It uses the default Laravel api
throttling.
In staging our app worked fine (30 req/sec same IP), but when deploying to production (1400 req/sec, different IPs) it started to fail, giving a lot of 429 Too Many Requests.
I quickly rolled back to php-fpm
and after a few hours tried again with the same problem. Rolled back and the next day I switched to Swoole and it worked perfectly without changing a single line of code nor having to redeploy anything. So I can confidently say that is NOT a bug in my code, but rather a bug with FrankenPHP or the Octane integration with FrankenPHP.
My theory is that the RateLimiter is not reseting between requests so it's shared between different users. So multiple different users trigger the rate limiter:
This is my Rate limiter configuration:
// AppServiceProvider
RateLimiter::for('api', function (Request $request) {
return Limit::perMinute(60)->by($request->user()?->id ?: $request->ip());
});
our production CACHE_STORE
is redis
. Throttling worked perfectly fine without octane and with octane but using Swoole. It failed with hundred of 429 Too Many Requests after installing FrankenPHP.
This is our bootstrap/app.php
:
<?php
use Illuminate\Foundation\Application;
use Illuminate\Foundation\Configuration\Exceptions;
use Illuminate\Foundation\Configuration\Middleware;
use Illuminate\Support\Facades\App;
return Application::configure(basePath: dirname(__DIR__))
->withRouting(
commands: __DIR__.'/../routes/console.php',
health: '/up',
then: function () {
Route::middleware('api')
->prefix('api')
->as('api.')
->domain(config('app.domain'))
->group(base_path('routes/api.php'));
Route::middleware('web')
->domain(config('app.domain'))
->group(base_path('routes/web.php'));
Route::middleware('web')
->domain(config('playsaurus.ads.domain'))
->group(base_path('routes/ads.php'));
}
)
->withMiddleware(function (Middleware $middleware) {
$middleware->throttleApi();
$middleware->redirectTo(
guests: '/login',
users: '/',
);
$middleware->web(append: [
\App\Http\Middleware\HandleInertiaRequests::class,
\Illuminate\Http\Middleware\AddLinkHeadersForPreloadedAssets::class,
]);
$middleware->api(append: [
\App\Http\Middleware\ConfigureLocale::class,
]);
$middleware->alias([
'localize' => \App\Http\Middleware\ConfigureLocale::class,
'embed' => \App\Http\Middleware\AllowsEmbeding::class,
]);
})
->withExceptions(function (Exceptions $exceptions) {
$exceptions->dontReport([
\App\Services\Announcements\InvalidVariantKey::class,
\App\Exceptions\CouponRedeemException::class,
]);
})->create();
Steps To Reproduce
It's difficult to reproduce. Because I can't test it in production because that would mean a lot of downtime for our users.
My theory is that it would be possible to reproduce from multiple different IPs. But since I don't have the means to test it, I don't know.
jhm-ciberman
14th May 2024 @ 00:15
Cachet would benefit from upgrading to Tailwind v4. This could lead to a really powerful theming system beyond the accents feature we currently have.
jbrooksuk
31st Jan 2025 @ 16:44
Octane Version
2.8.1
Laravel Version
10.48.28
PHP Version
8.3.16
What server type are you using?
Swoole
Server Version
6.0.1
Database Driver & Version
No response
Description
I'm using Envoyer to deploy a Laravel Octane application, taking advantage of its zero-downtime deployment features.
However, Octane currently does not support zero-downtime deployment because it does not follow symlink directories. It always remains in the actual directory (instead of the symlinked one) where the Octane server was initially started. When the outdated release directory is deleted, Octane continues running in that location, causing errors on every request and resulting in 500 response codes on the live application.
Error thrown:
Warning: require(/var/www/domain.com/releases/202502010200023/vendor/laravel/octane/bin/bootstrap.php): Failed to open stream: No such file or directory in /var/www/domain.com/releases/202502010200023/vendor/laravel/octane/bin/swoole-server on line 18
Fatal error: Uncaught Error: Failed opening required '/var/www/domain.com/releases/202502010200023/vendor/laravel/octane/bin/bootstrap.php' (include_path='.:/usr/bin/[email protected]/8.3.16/share/[email protected]/pear') in /var/www/domain.com/releases/202502010200023/vendor/laravel/octane/bin/swoole-server:18
Stack trace:
#0 /var/www/domain.com/releases/202502010200023/vendor/laravel/octane/bin/swoole-server(95): {closure}(Array)
#1 [internal function]: {closure}(Object(Swoole\Http\Server), 0)
#2 /var/www/domain.com/releases/202502010200023/vendor/laravel/octane/bin/swoole-server(170): Swoole\Server->start()
#3 {main}
thrown in /var/www/domain.com/releases/202502010200023/vendor/laravel/octane/bin/swoole-server on line 18
#1 [internal function]: {closure}(Object(Swoole\Http\Server), 1)
#1 [internal function]: {closure}(Object(Swoole\Http\Server), 3)
#1 [internal function]: {closure}(Object(Swoole\Http\Server), 2)
#1 [internal function]: {closure}(Object(Swoole\Http\Server), 4)
#1 [internal function]: {closure}(Object(Swoole\Http\Server), 5)
Steps To Reproduce
Use any 0-downtime deployment or test it manually using the following instructions:
- Use the Swoole driver as an example.
-
cd
one directory up from the project's base path. - Create a
current
symlink directory for your project using the command:ln -nsf ./octane-project-test ./current
- Start the Octane server:
php ./current/artisan octane:start
- Copy your project to another directory:
cp -R ./octane-project-test ./octane-project-test-new
- Activate the new release:
ln -nsf ./octane-project-test-new ./current
- Reload the Octane server:
php ./current/artisan octane:reload
- Remove the original project directory:
rm -rf ./octane-project-test
mikkpokk
2nd Mar 2025 @ 01:35
using prompt engineering techniques provide enhancement to the Gemini service class in the Gemini prompt class, creating functions commonly used in Laravel applications.
kwakuOfosuAgyeman
5th Jan 2025 @ 05:43
It should be possible to report an incident and attach components (with statuses) to incidents in a single API request.
jbrooksuk
24th Dec 2024 @ 12:16
Description: I am using Flasher 2.0.1 in my Laravel 11 project and would like to know how to properly use the Flasher library in JavaScript. Additionally, I want to ensure that the JavaScript configuration is consistent with the settings defined in the flasher.php configuration file.
Details: Flasher Version: 2.0.1 Laravel Version: 11 Problem: Need guidance on using the Flasher library with JavaScript in a way that mirrors the configuration set in flasher.php.
raseldev99
16th Sep 2024 @ 04:50
The goal is to make abstract code based on the package notorm to build a new orm system
ambroisehdn
15th Jun 2022 @ 12:34
The solutions widget pulls the last 5 people with solutions but if there's only three people given solutions then we should only show those three. Filter out the ones without solutions.
driesvints
24th Feb 2025 @ 10:04
Currently, for plural rules, the MessageFormatter polyfills uses the English rules for all locales (it ignores the locale).
To be consistent with what we do in symfony/intl
(used by symfony/polyfill-intl-icu
to implement NumberFormatter and DateFormatter), we should rather fail explicitly here (using the English rules for a different locale would not give the right result anyway)
stof
23rd Oct 2018 @ 15:34
When using the addToMediaFromURL method, I'm receiving a Could Not Load Image error on the server, but not locally (I'm using Laravel Vapor) and not with other image formats.
The code below is where the error is occuring and that code is receiving an absolute file path to the image in s3, which is available.
public function addToMediaLibrary($file_path) { $media = $this->addMediaFromUrl($file_path) ->toMediaCollection("", 's3'); return $media; }
As an example the file path when logged out is https://s3.sitesforsaas.com/tenants/d221c93a-3757-466c-bf83-b74a9ee64c04/sites/9cfc24c7-cb2a-4c5a-8c0f-73ecdc79c166/2024/09/4e9854b0-4f2d-4931-a726-01f557bd7fc6.webp
, but the error is looking for the image in local 'tmp' storage for some reason.
Could not load image at path
/tmp/storage/tenantd221c93a-3757-466c-bf83-b74a9ee64c04/media-library/temp/j7pNMerUdReNx9MfAG5rKjcmGHWu4i99/LM2UYMTcYU0BO8sLsTfeOBpS9eXximejlarge.webp`",
josiahmann
11th Sep 2024 @ 21:32
Found the \Illuminate\Database\Eloquent\Concerns\HasAttributes::originalIsEquivalent()
method during code diving and it should be usable for our trait! 🎉
This PR would be mainly reasearch when that method was added, if it's compatible with our version support rules and switching our own change detection logic with the core one.
Gummibeer
20th Oct 2021 @ 10:03
v2.4 allows users to add Google 2FA to their account for an added layer of security. 3.x should also implement this in some way.
Things to consider
- How to generate the code
- How do verify via 2FA when this is enabled on a users account.
- Can admins reset 2FA for a user if they've lost their 2FA device?
- How should backup codes work?
joelbutcher
15th Jan 2025 @ 18:25
Laravel Version
11.41.3
PHP Version
8.4.3
Database Driver & Version
No response
Description
For some reason it seems that calling /link?url=https://www.foo.com/?utm_campaign=some%20campaign
in my application will fail to validate the url
parameter there. Calling Str::isUrl('
https://www.foo.com/?utm_campaign=some%20campaign')directly in tinker returns
truethough. However, dumping
$request->input('url')in the controller reveals that the
%20character there has done MIA. That also happens if I use
+` as a space encoder.
Steps To Reproduce
routes/web.php
:
use Illuminate\Http\Request;
use Illuminate\Support\Facades\Route;
Route::get('/link', function(Request $request) {
$request->validate(['url' => 'required|url']);
});
tests/Feature/UrlValidatorTest.php
:
<?php
use function Pest\Laravel\getJson;
test('passes 1', function() {
getJson('/link?url=https://www.foo.com/?utm_campaign=some%2520campaign')->assertSuccessful();
});
test('fails 1', function() {
getJson('/link?url=https://www.foo.com/?utm_campaign=some%20campaign')->assertSuccessful();
});
test('passes 2', function() {
getJson('/link?url=https%3A%2F%2Fwww.foo.com%2F%3Futm_campaign%3Dsome%2520campaign')->assertSuccessful();
});
test('fails 2', function() {
getJson('/link?url=https%3A%2F%2Fwww.foo.com%2F%3Futm_campaign%3Dsome%20campaign')->assertSuccessful();
});
Test 2 and 4 will then fail here.
Is this supposed to fail like this? My understanding is that https://www.foo.com/?utm_campaign=some%20campaign
is a completely valid URL though (and directly calling Str::isUrl('https://www.foo.com/?utm_campaign=some%20campaign')
confirms that). Or should spaces in query parameters always be double encoded when passing them to Laravel for validation?
One workaround/hack is to add these lines before the validation is done:
$request->merge([
'url' => Str::replace(' ', '+', $request->input('url')),
]);
Reproduction repo: https://github.com/carestad/laravel-url-validation-bug
carestad
10th Feb 2025 @ 13:48
What happened?
After installation, I noticed that in the browser console it outputs an error of wireEl is undefined.
How to reproduce the bug
Just install this package in a livewire project, preferable with Filament's admin package. And see the browser console.
composer create-project laravel/laravel test;
composer require filamentphp/filament;
composer require spatie/laravel-blade-comments;
Then go through the browser console and see the error
Package Version
1.0.1
PHP Version
8.2.0
Laravel Version
10.13.0
Which operating systems does with happen with?
macOS
Notes
Filament ......................................................
Packages ...... filament, forms, notifications, support, tables
Version .............................................. v2.17.44
Views ..................................... PUBLISHED: filament
sawirricardo
2nd Jun 2023 @ 07:15
Hello,
I noticed that there is an incompatibility with the mbstring polyfill and PHP 8.1 / Alpine Linux, which breaks a lot of my projects as soon as the php81-mbstring is not installed, but php81-iconv is installed:
Example:
Warning: iconv(): Wrong encoding, conversion from "ASCII" to "UTF-8//IGNORE" is not allowed in phar:///var/www/localhost/htdocs/phpstan.phar/vendor/symfony/polyfill-mbstring/Mbstring.php on line 736
It looks like //IGNORE
is not accepted since echo iconv('UTF-8', 'UTF-8', 'test');
works, while echo iconv('UTF-8', 'UTF-8//IGNORE', 'test');
doesn't
danielmarschall
7th Jan 2022 @ 00:12
I would be good to support normalizer_get_raw_decomposition function that appeared in PHP 7.3. I've tried to implement it but seems that existing decomposition data is already optimized to get the final decomposition. Example test case is available here.
Pinging @nicolas-grekas as he is the author of Normalizer
polyfill.
IonBazan
12th May 2018 @ 23:26
using prompt engineering techniques provide enhancement to the Claude service class in the Claude prompt class, creating functions commonly used in Laravel applications. For example, brand builder, seo product optimizer, video and or image captioner, automated chatbots #goodfirstissue
kwakuOfosuAgyeman
5th Jan 2025 @ 05:38
Octane Version
2.5.9
Laravel Version
11
PHP Version
8.3
What server type are you using?
Swoole
Server Version
8.3
Database Driver & Version
No response
Description
Wildcard paths like config/**/*.php
in the watch
array of config/octane.php
aren't working properly.
Steps To Reproduce
- Install chokidar
- Run
octane:start --watch
(I'm using Swoole) - Modify a file in the
config
directory. Octane won't restart - Change
config/**/*.php
toconfig
in thewatch
array ofconfig/octane.php
- Restart Octane server
- Modify a file in the
config
directory. Octane will now correctly restart
Same problem with anything else using wildcards eg. database
(try modifying a seeder to test).
binaryfire
14th Nov 2024 @ 11:57
The docs aren't finished. This issue exists to track the missing pieces:
- App icons
- Context menus
- Dock
- Printing
- ProgressBar
- Queue
- Screens
- Settings
- Shell
- System
- Touch ID
- Testing
- Build signing (#361)
- macOS
- Windows
- How to customise Electron / PR to NativePHP
- Security disclosures process
- Highlight current section
- Show ToC persistently on the right-hand side
- Sponsorship page
- Contributors page
- Show sponsors in sidebar
- Add next/previous section buttons to the bottom of pages
- Add Edit on GitHub link to pages
simonhamp
8th Sep 2024 @ 15:01