Trait not found by Laravel base class using Composer 2 autoloader? - laravel

I'm on a Laravel project using new-ish versions of PHP, Laravel and Composer 2, as of this writing. I added a new app/Traits/MyTrait.php file beside several existing trait files but unfortunately Composer absolutely will not detect the new file. I'm getting this error:
Trait 'App\Traits\MyTrait' not found
Similar to:
Laravel Custom Trait Not Found
Here is the general layout of the code:
# app/Traits/MyTrait.php:
<?php
namespace App\Traits;
trait MyTrait {
// ...
}
# app/Notifications/MyBaseClass.php:
<?php
namespace App\Notifications;
use App\Traits\MyTrait;
class MyBaseClass
{
use MyTrait;
// ...
}
# app/Notifications/MyChildClass.php
<?php
namespace App\Notifications;
class MyChildClass extends MyBaseClass
{
// ...
}
The weird thing is that this code runs fine in my local dev, but no matter what I try, it won't work when deployed to the server while running in a Docker container. I've tried everything I can think of like saving "optimize-autoloader": true in composer.json and running composer dump-autoload -o during deployment, but nothing fixes it:
https://getcomposer.org/doc/articles/autoloader-optimization.md
I'm concerned that this inheritance permutation may not have been tested properly by Composer or Laravel, so this may be a bug in the tools. If worse comes to worse, I'll try these (potentially destructive) workarounds:
Calling composer dump-autoload -o (greatly slows deployment, as this is a large project, and so far doesn't seem to fix it anyway)
Deleting via rm vendor/composer/autoload_classmap.php, rm vendor/composer/autoload_psr4.php and/or rm vendor/composer/autoload_namespaces.php (or similar) in the vendor folder before each deployment to force Composer to rebuild.
Deleting via rm -rf vendor
The sinister part about this is that we must have full confidence in our deploy process. We can't hack this in our server dev environments by manually deleting stuff like vendor and then have it fail in the production deploy because Composer tripped over stale data in its vendor folder. My gut feeling is that this is exactly what's happening, perhaps due to an upgrade from Composer 1 to Composer 2 or version change or stale cache files from work in recent months.
Even a verification like "this minimal sample project deployed to Docker works for us" would help to narrow this down thanks.
Edit: this is a useful resource on how the Composer autoloader works: https://jinoantony.com/blog/how-composer-autoloads-php-files

The problem turned out to be caused by the container/filesystem on AWS being case-sensitive, but my local dev environment on macOS being case-insensitive.
My original trait (kept secret) ended with URL in its name, but I was including its path as, and using it in the base class as, Url.
So this issue had nothing to do with traits, base classes or Composer. It also didn't require any modification of composer.json or the way we call it during deployment. But I think it's still best practice to have this in composer.json, I use it this way in local dev too currently (good/bad?):
"config": {
"optimize-autoloader": true
},
The real problems here (industry-wide) are:
Vague error messages
Lack of effort by code to drill down and find actual causes (by attempting to load as case-insensitive and returning a warning when found, for example)
Lack of action items for the user (have you checked the case? checked that the file exists? checked file permissions? etc etc, written into the error message itself, with perhaps a link to a support page/forum)
It wasn't convenient to ssh into the server (by design). So to troubleshoot, I temporarily committed this onto my branch:
# app/Http/Controllers/TestController.php
class TestController extends Controller
{
public function test()
{
return response('<pre>' .
'# /var/www/html/vendor/composer/autoload_classmap.php' . "\n" . file_get_contents('/var/www/html/vendor/composer/autoload_classmap.php') . "\n" .
'# /var/www/html/vendor/composer/autoload_files.php' . "\n" . file_get_contents('/var/www/html/vendor/composer/autoload_files.php') . "\n" .
'# /var/www/html/vendor/composer/autoload_namespaces.php' . "\n" . file_get_contents('/var/www/html/vendor/composer/autoload_namespaces.php') . "\n" .
'# /var/www/html/vendor/composer/autoload_psr4.php' . "\n" . file_get_contents('/var/www/html/vendor/composer/autoload_psr4.php') . "\n" .
'# /var/www/html/vendor/composer/autoload_real.php' . "\n" . file_get_contents('/var/www/html/vendor/composer/autoload_real.php') . "\n" .
'# /var/www/html/vendor/composer/autoload_static.php' . "\n" . file_get_contents('/var/www/html/vendor/composer/autoload_static.php') . "\n"
);
}
}
# routes/api.php
Route::get('/test', 'TestController#test');
Then deployed without merging in GitLab and compared the response to the error in AWS Cloudwatch, which is when the typo jumped out.
Then I removed the temporary commit with:
git reset --soft HEAD^
And force-pushed my branch with:
git push --force-with-lease
So was able to solve this without affecting our CI/CD setup or committing code permanently to the develop or master branches.
I've been doing this for a lot of years, and even suspected a case-sensitivity issue here, but sometimes we're just too close to the problem. If you're knee-deep in code and about to have an anxiety attack, it helps to have another set of eyes review your thought process with you from first principles.
I also need to figure out how to run my local Docker containers as case-sensitive as well, to match the server (since that's the whole point of using Docker containers in the first place).

I had the same problem and it was related to my file name. I had put it in lowercase at the beginning, that is: apiResponser.php. I added some changes and renamed my file to ApiResponser.php and sent it to production, but ... oh, oh!
I had the same problem.
The only way it worked for me was, do the git name replacement:
📦 git mv app/Traits/apiResponser.php app/Traits/ApiResponser.php
This way I was able to solve. I understand that you have solved it in
another way, however this may help another developer. 🙂

Related

Joomla "Fatal error: Cannot redeclare jblogerror() in on line 0"

Here is a weird problem i'm facing; after updating Joomla to the latest version, website failed to up load but, as i've made a backup manually from these folders:
administrator
bin
cli
components
includes
layouts
libraries
modules
plugins
yt-assets
and all the root files, after restoring the backup still the website is not loading and just says
"Fatal error: Cannot redeclare jblogerror() in on line 0" !
Any suggestion? Thank you.
Apparently, the version of jBlog you are using is loading its attempting to load its own classes twice.
Let's assume the jBlog developers already fixed it and you carelessly ignored the warning to ensure all your extensions are compatible before updating.
Are you able to access administrator? Simply go there and upgrade jBlog including any modules.
Else, the issue lies within a plugin. In order to access the administrator and perform the update, you will need to manually disable the offending plugin.
A simple way is to rename its folder: start from plugins/system then plugins/content hopefully you'll have spotted it by then. Look into the subfolders of each and spot any that may be relevant to jBlog (or grep through the folder to locate the specific string)
If you have console access, simply run
# grep -rl jblogerror plugins/system
and you should see all the files that include such string, just rename their main plugin folder. But you could also do all this through ftp and guessing.

SFTP Namespace with phpseclib installed manually

I am trying to connect using SFTP (phpseclib) with a manual installation. That's the very first time I'm using namespaces so I don't know if what I'm doing is the way I have to.
I did download phpseclib from GitHub as ZIP and put it in /home/libs_web/php/class/phpseclib3
After that I used this code (just as example) :
require('/home/libs_web/php/class/phpseclib3/Net/SFTP.php');
use phpseclib3\Net\SFTP;
$sftp = new SFTP('localhost');
$sftp->login('username', 'password');
As mentioned here but with an include on top : https://phpseclib.com/docs/sftp
Here is my error : Fatal error: Uncaught Error: Class 'phpseclib3\Net\SSH2' not found in /home/libs_web/php/class/phpseclib3/Net/SFTP.php:52
It seems like my Namespace doesn't work correctly. I do not use autoloader, composer, and tried to set the working directory to /home/libs_web/php/class/phpseclib3/ & /home/libs_web/php/class/.
Don't know what to do more. If I include the file /Net/SSH2.php I'll have another error about another file. I think this isn't the proper way to work.
Could you please provide some help ?
phpseclib3 is best installed with Composer. eg. on the CLI do composer init; composer require phpseclib/phpseclib:~3.0. You'd also put require __DIR__ . '/vendor/autoload.php'; at the top of your file
If you wanted to do something like make your source code available for people to download and make it work on shared hosts were CLI access might not be available I guess you could do Composer and then upload the vendor/ directory by itself or include it with your zip file or whatever.

CakePHP 3.5: Cannot access Plugin Classes after installing it via Composer

So I have a CakePHP 3 project and want to load FluentDOM, a PHP plugin not specifically written for CakePHP.
According to both software documentations, Composer is the way to go. In my understanding, all I would have to do is the following:
run composer require fluentdom/fluentdom in powershell
run composer require fluentdom/selectors-phpcss in powershell
OR
add the following to composer.json in the project's root directory:
"require": {
"fluentdom/fluentdom": "^7.0",
"fluentdom/selectors-phpcss": "^1.1"
}
run composer update in powershell
Both ways will install the desired plugins to vendor/fluentdom/{pluginname}/ as expected, but /vendor/cakephp-plugins.php won't include them, as implied by CakePHP's plugin installation manual.
The following attempt to load either plugin in a controller by writing
use Cake\Core\Plugin;
Plugin::load('fluentdom/fluentdom');
Plugin::load('fluentdom/selectors-phpcss');
would cause an exception that the desired plugins were not found in plugins/ :
Make sure your plugin fluentdom/fluentdom is in the {absolute project path}\plugins\ directory and was loaded
-- Which is already odd, because Composer wouldn't install anything there to begin with.
I found that I might get around this issue by manually extending vendor/cakephp-plugins.php to include the correct paths:
'fluentdom/fluentdom' => $baseDir . '/vendor/fluentdom/fluentdom/',
'fluentdom/selectors-phpcss' => $baseDir . '/vendor/fluentdom/selectors-phpcss/'
(However, that doesn't seem the way to go, because this file is auto-generated and overwritten by Composer after every update.)
And even then, the final issue still persists: although the plugins seem to be loaded successfully (confirmed by running Plugin::loaded()), I'd finally get the following exception when trying to access FluentDOM's classes as described in their wiki:
$document = new FluentDOM\DOM\Document();
Class 'App\Controller\FluentDOM\DOM\Document' not found
Does the plugin miss out on having its' autoload executed?
Even extending the line in my controller to Plugin::load('fluentdom/fluentdom', ['autoload' => true]);, but doesn't seem to help either; according to CakePHP's doc, that shouldn't be necessary anyway.
So what am I missing?
Found it! First of all, I had the false presumption that Plugins and Vendor Packages are more or less the same: they are not; thanks to Greg Schmidt for pointing this out in the question's comments.
The issue was in the line of how I tried to access FluentDOM's class. While
$document = new FluentDOM\DOM\Document();
worked in a standalone php file, it didn't within the Cake project; I was missing a backslash:
$document = new \FluentDOM\DOM\Document();
So, the entire path of actions to load a Vendor Package is merely:
run composer require fluentdom/fluentdom in powershell
run composer require fluentdom/selectors-phpcss in powershell
Use the new classes right away with $document = new \FluentDOM\DOM\Document();
No further steps required. Side note: Composer seems to refresh autoload config after installing a vendor file with composer require {vendor}/{package}, but in case it doesn't, or autoload config is messed up from earlier experiments, composer dumpautoload should fix it.

Composer [UnexpectedValueException] error will trying to use composer to install a github project

I am trying to install a github project using composer and get the following error
Composer [UnexpectedValueException]
Your Github oauth token for github.com contains invalid characters: ""
Can anyone explain what I need to do to correct this error?
I am using the following command
composer create-project --prefer-dist --stability=dev vova07/yii2-start yii2-start
Thank you
I started getting a similar error and the reason was that Github recently changed the format of their auth tokens:
https://github.blog/changelog/2021-03-31-authentication-token-format-updates-are-generally-available/
To resolve the error:
Find the composer/auth.json file (if you're running the project in a container, you'll have to bash into it and find the file in there)
Remove its github.com entry. Your file will probably look like the following after removing the entry: {"github-oauth": {}}
Run composer self-update. The issue got resolved in version 2.0.12. See the first item in the changelog for that version here: https://getcomposer.org/changelog/2.0.12
After that, you can restore your composer/auth.json file to its initial state as the newer version of composer will recognize the new key format.
You can try Basic Auth instead:
Change this (oauth):
"github-oauth": {
"github.com": "ghp_[YOUR-PERSONAL-TOKEN]"
}
To this (basic auth):
"http-basic": {
"github.com": {
"username": "[YOUR-GITHUB-USERNAME]",
"password": "ghp_[YOUR-PERSONAL-TOKEN]"
}
}
You can find instructions on how to create a Personal Access Token
Inspired from github docs. Apparently, you can use Basic Authentication with a Personal Access token instead of oauth in some cases (e.g. like mine: installing a private git repo with composer).
I fixed it.
Goto C:\Users\XXXXX\AppData\Roaming\Composer
Open the auth.json
delete the github.com entry under "github-oauth": {}
That's it.
Update answer for Masiorama and Ruchir Mehta:
If you looking for file auth.json but don't know how, use this command:
locate auth.json
And here's the result:
You can see that auth.json will look like this:
/home/{your user name}/.config/composer/auth.json
Then you could use this command to edit the file:
sudo gedit /home/dev/.config/composer/auth.json
And remove content inside github-oauth.
If you're on MacOS, the auth.json file is at ~/.composer/auth.json. Then from there, you can remove the value for github-oauth. I tried fully deleting the file but I got a parse error, Expected one of: 'STRING', 'NUMBER', 'NULL', 'TRUE', 'FALSE', '{', '['. Your auth.json file should look like this:
{
"github-oauth": {}
}
This is similar to other answers posted but I wasn't able to use the locate command on MacOS so this might be helpful to other Mac users
This error recently popped up from nowhere.
Simply deleting the whole auth file worked for me..! Not sure why / when it appeared in the first place.
~/.composer/auth.json
As far as I know (I'm a beginner with composer too), the problem is with your authentication, so you have to fix your credentials in auth.json inside path-to-composer/.composer/
Inside you will find a json which will probably looks like:
{
"github-oauth": {
"github.com": null
}
}
Fix that and you should be ok ;)
The solution is just to upgrade your Composer version
using command composer self-update.
Go to C:\Users\UserName\AppData\Roaming\Composer
Open the auth.json file.
Clear everything and paste the below code
{
"bitbucket-oauth": {},
"github-oauth": {},
"gitlab-oauth": {},
"gitlab-token": {},
"http-basic": {},
"bearer": {}
}
I hope it will be solved
I run in the same problem after upgrading githup api token to the new format.
The answer is you need to upgrade composer version 1.10.21 or higher that fixes this problem.
Same solution as the answer of Paulina Khew but with command lines on MacOS :
cd ~/.composer/
nano auth.json
Delete what is inside th bracket :
{
"github-oauth": {}
}
When you're ready to save the file, hold down the Ctrl key and press the letter O
Press the Enter key on your keyboard to save.
When finished, press Ctrl + X to close nano and return to your shell.
Edit the composer authentication configuration file ~/.composer/auth.json
Then replace the following.
"http-basic": {
"github.com": {
"username": "[YOUR-GITHUB-USERNAME]",
"password": "ghp_[YOUR-PERSONAL-TOKEN]"
}
}
Now run the command composer install
That's a bug.
If you have Debian or Ubuntu, try this patch. Otherwise read the last line.
Quick copy-paste patch
If you have Debian 10 buster or Ubuntu 20.LTS or similar distributions, try this copy-paste command:
wget https://gist.githubusercontent.com/valerio-bozzolan/84364c28a3bba13751c504214016adcf/raw/c1356d529c89c10de4c959058e2e86ffe58fa407/fix-composer.patch -O /tmp/fix-composer.patch
sudo patch /usr/share/php/Composer/IO/BaseIO.php /tmp/fix-composer.patch
If it does not work, write it in the comments.
Step-by-step explaination
Your Composer version has a bug: you are able to save a valid GitHub token, but then it's not able to read that token again because Composer thinks that your GitHub token cannot contain underscores or stuff like that. Moreover, it's strange that Composer checks its syntax only the second time. Why? that's another story.
The fix is simple. You can temporary disable that wrong validation in your Composer version. Also because GitHub is a proprietary service and their specifications can change over time (as you demonstrated today). So it makes sense not to validate the syntax of GitHub tokens. The only person who should hard-validate GitHub tokens is GitHub itself, not Composer.
If you installed Composer via apt install composer, probably you will not have any update available and surely you cannot use self-update because Composer is read-only for security reasons (and for a similar reason, you should not execute Composer from root). Instead, you can create a safe hot-patch to fix that specific issue.
To create a patch, create a file called /tmp/fix-composer.patch with this exact content:
103,105c103,105
< if (!preg_match('{^[.a-z0-9]+$}', $token)) {
< throw new \UnexpectedValueException('Your github oauth token for '.$domain.' contains invalid characters: "'.$token.'"');
< }
---
> // if (!preg_match('{^[.a-z0-9]+$}', $token)) {
> // throw new \UnexpectedValueException('Your github oauth token for '.$domain.' contains invalid characters: "'.$token.'"');
> //
That content can also be seen from here:
https://gist.github.com/valerio-bozzolan/84364c28a3bba13751c504214016adcf
Then run this command to apply that patch:
sudo patch /usr/share/php/Composer/IO/BaseIO.php /tmp/fix-composer.patch
If it does not work, probably you have not installed composer via apt.
In short, whatever operating system, and whatever installation method, locate the file BaseIO.php in your Composer and comment out the validation check.

Changing cache_dir and log_dir with Symfony2

Because of deployment constraints, I would like to have the log and cache directories used by my Symfony2 application somewhere under /var/... in my file system. For this reason, I am looking for a way to configure Symfony and to override the default location for these two directories.
I have seen the kernel.cache_dir and kernel.log_dir and read the class Kernel.php. From what I have seen, I don't think that it is possible to change the dir locations by configuration and I would have to patch the Kernel.php class.
Is that true, or is there a way to achieve what I want without modifying the framework code?
Add the following methods to app/AppKernel.php (AppKernel extends Kernel) making them return your preferred paths:
public function getCacheDir()
{
return $this->rootDir . '/my_cache/' . $this->environment;
}
public function getLogDir()
{
return $this->rootDir . '/my_logs';
}
I was happy to find your post, but I was a little bit confused of the unhelping answers.
I got the same problem and found out that the logs are depending on the config parameter
kernel.logs_dir.
So I just added it to my config.yml parameters:
kernel.logs_dir: /var/log/symfonyLogs
I hope it will helpfull for you even, if its a late answer.
i think the easiest way is to link the folder to another place. We have made this on the prod server but when you develop local perhaps on windows its a bit complicated to set the symlinks.
ln -s /var/cache/ /var/www/project/app/cache
something like this.
I would like to offer an alternative and that is to set environment variables to change these directories. This way it's easier to set depending on the stage. (testing, production or development)
export SYMFONY__KERNEL__CACHE_DIR "/your/directory/cache"
export SYMFONY__KERNEL__LOGS_DIR "/your/directory/logs"
Environment variables can also be set in the virtual host with SetEnv.
When reading kernel parameters symfony will look for all the $_SERVER variables that start with SYMFONY__, strip the first part and convert all the double underscores into a .
Source code
See line 568 to 608
In symfony you can override the cache (and logs) directory by extending the method in AppKernel.
// app/appKernel.php
class AppKernel extends Kernel
{
// ...
public function getCacheDir()
{
return $this->rootDir.'/'.$this->environment.'/cache';
}
}
Check out http://symfony.com/doc/current/cookbook/configuration/override_dir_structure.html#override-cache-dir
I used the configuration solution from Dragnic but I put the paths in the parameters.yml file because this file is ignored by git. in other words, it's not synchronized from my PC to the git repository so there is no impact in the prod environment.
# app/config/parameters.yml
parameters:
database_driver: pdo_mysql
[...]
kernel.cache_dir: "T:/project/cache"
kernel.logs_dir: "T:/project/logs"
Configuration: Windows7, WAMP 2.4 and Symfony 2.3.20.
But you have to know that:
Overwriting the kernel.cache_dir parameter from your config file is a very bad idea, and not a supported way to change the cache folder in Symfony.
It breaks things because you would now have different cache folders for the kernel Kernel::getCacheDir() and for the parameter.
Source: https://github.com/symfony/AsseticBundle/issues/370
So you should use it only in dev environment and if you don't want to change the content of the app/AppKernel.php file, otherwise see the other answers.
No accepted answer, and a really old question, but I found it with google, so I post here a more recent way to change the cache directory, and the logs directory, (source here)
remember, short syntax for arrays require php 5.4
you can select the env to modify, and manage different cache and logs directories if you want
public function getCacheDir()
{
if (in_array($this->environment, ['prod', 'test'])) {
return '/tmp/cache/' . $this->environment;
}
return parent::getCacheDir();
}
public function getLogDir()
{
if (in_array($this->environment, ['prod', 'test'])) {
return '/var/log/symfony/logs';
}
return parent::getLogDir();
}

Resources