I use php less compiler to compile bootstrap.less. When I upload on the server bootstrap.less via smartFTP everything works fine.
But as soon as I open and save bootstrap.less on the server (without changing it, just save it using remote editor linke PSPad or Remote System Explorer in Eclipse) I got the error:
Failed to assign arg #list: ../_css/less/bootstrap.less on line 51
So saving the file does something strange, but why and what?
As mentioned by #max-k it's a bug. All above version 3.0.0 don't work.
Lessphp seems to break on mixins like .col(#index) when (#index = 1) { // initial.
Solutions: switch back to TB3.0.0, wait for the bug fix, use a alternative library like less.php
Related
Hello i have installed Parse server and parse sdk for php on my Linux Centos Server.
Parse Dashboard: 1.3.3,
Server version: 4.3.0
I'm managing those db's using php code. This parser was installed by someone else. The thing is I'm not sure how to install LiveQuery for my server. I don't know if my version of server is good for this kind of operations. I found some solutions how to enable this feature, but there was something like "index.js" - in my file system it's app.js. Screen of file system
In app.js i added lines:
liveQuery: {
classNames: ["Test"] //List of classes to support for query subscritions
}
I've created class in my db named "Test", then i found solution i need to add
this line
liveQueryUrl: keyLiveQueryUrl, // Required if using LiveQuery
I don't really know how to get this key.
How can i check if LiveQuery works? It will be used by flutter code, but i'd like to check it first in PHP.
My problem has been solved in my other extended topic here.
Few weeks ago I set up a Message system with Symfony Messenger and it worked great.
Today I wanted to create new object through message, so I went to my server and type the command to consume message
First I had this result
$ bin/console messenger:consume-messages amqp_notifications
/usr/bin/env: ‘php\r’: No such file or directory
It never happened before with my files, and I never changed the line ending or encoding of my file sin PHPstorm.
I tried to use $ php bin/console messenger:consume-messages amqp_notifications
but then I had this error.
Attempted to load class "AMQPConnection" from the global namespace.
Did you forget a "use" statement?
Pretty weird, because I have have the php-amqp ext installed as you can see on the screenshot of my phpinfo
I didn't change anything in my Message class or Handler.
Also, I tried to call new AMQPConnection() on a random action, just to try, and I didn't get the error.
I'm completely lost with this error this time, as everything is installed.
I use PHP 7.3.1 and symfony Messenger 4.2.2
It seems your second issue was already solved by ccKep on his comment.
The first one is that the specific shebang line #!/usr/bin/env php executes the first php found in the $PATH. So if you already have uninstalled it, which seems the case, or it has a symbolic link to another php version, you can get a wrong result.
Tries to check what is inside the $PATH and replace the PHP path for the correct one. You might get the place running which php.
I was trying to run phpdox on windows server 2012 - but I am getting the error:
PHP Version: 7.0.5 (WINNT)
PHPDox Version: 0.8.1.1
Exception: TheSeer\phpDox\Generator\TokenFileException (Code: 1)
Location: phar://D:/htdocs/AscPro/bin/phpdox-0.8.1.1.phar/phpdox/generator/pro
ject/TokenFile.php (Line 19)
File 'file:/D:/htdocs/AscPro/build/phpdox/tokens/AppBundle/AppBundle.php.xml'
not found
I've checked the location, the file is not missing.
I'm running into this problem during continuous integration process with jenkins. It is very strange, because the same phpdox version did work for me on ubuntu. Maybe this is related to the fact, that all programs lay on "C:", including Jenkins - but the jenkins workspace lays on "D:"?
I think that the problem lies within incorrect using WINDOWS path. Please see this question:
Help with windows path - PHP
As a quick test, try to change file_exists check to harcoded argument:
file:///C:\htdocs\AscPro\build\phpdox\tokens\AppBundle\AppBundle.php.xml
Im afraid it`s a bug of phpdox.
Phpdox convert all paths to unix style before check to exists.
You can see this code here.
Maybe you can hack function exists
replace:
return file_exists($this->getPathname());
to:
return file_exists(parent::getPathname());
I have used PHPExcel for my codeigniter app and it is working perfectly in localhost, but when I host this to server, I am getting following error :
Fatal error: Class 'PHPExcel_Shared_String' not found in \xx\xx\xx
third_party\PHPExcel\Autoloader.php on line 36
There was a change introduced to the autoloader in the latest version of PHPExcel that appears to have broken backward compatibility with versions of PHP < 5.3.0
If you edit the Classes/PHPExcel/Autoloader.php file and change line 58, which should read
return spl_autoload_register(array('PHPExcel_Autoloader', 'Load'), true, true);
to
return spl_autoload_register(array('PHPExcel_Autoloader', 'Load'));
I've already made a change to the develop branch on github to test for the PHP version and execute the appropriate line
While this was not deliberate, please note that we really are trying to get users to upgrade to at least version 5.3.0 of PHP, because we can't address any of the memory/performance issues that users working with large spreadsheets complain about until we can use some of the new features available in more recent versions of PHP. Version 5.2 of PHP is no longer supported, and even version 5.3 is end-of-life and will be unsupported before the end of this year
struggled with this issue for a long time under Linux and PHP 5.4x. In the end, in addition to the fix above, I resorted to changing the code on line 73 of the Autoloader file which sets the $pClassFilePath variable from relative (using PHPEXCEL_ROOT) to absolute following the machines file tree. This might only be a hack, but it saved my sanity after several days of trying. Hope this helps someone. Cheers
I had this issue too and i solved it by changing permissions on "Shared" directory to 655.
I Hope it helps
If your server is on Linux, it can be permission problem... Just add all permissions for PHPExcel Folder in you Vendor (on server side) and all subfolders for it. I have same problem and i have solved it by this way...
What worked for me was changing PHPExcel/Autoloader.php line 81 from
if ((file_exists($pClassFilePath) === FALSE) || (is_readable($pClassFilePath) === FALSE)) {
to
if ((stream_resolve_include_path($pClassFilePath) === FALSE)) {
I prefer this approach because it didn't require me to modify file permissions and it should work in PHP 5.3.2 and later.
I'm trying to use the template from http://jenkins-php.org in my installation of Jenkins. As soon as I go in to configure the project I get a grey screen appearing and the words 'Loading'
If I look in the Browser error console I get TypeError: 'undefined' is not an object (evaluating 'registry.get')
Has anyone else experienced this problem? How can this be fixed?
We had this problem when upgraded to 1.528.
Digging further, we found out that the problem is originated from one of the client-side script files called "hudson-behavior.js" and is caused by the "Publish Over SSH" plugin. There have been other plugins that have caused this in the past, as well. We couldn't disable the plugin anyway. So we searched further.
Long story short, we ended up changing the script for the time being. If you, like us, install Jenkins using the war file, then the script file is probably extracted and cached in "/var/cache/jenkins/war/scripts/hudson-behavior.js".
The following change we made was in a function called "registerValidator":
if (depends==null) { // legacy behaviour where checkUrl is a JavaScript
try {
return eval(url); // need access to 'this', so no 'geval'
} catch (e) {
// set depends to an empty array.
depends = [];
}
}
var q = qs(this).addThis();
The URL may not always be valid. Thus, a try-catch is in order. Hope this helps!
I've just had the same issue, I'm using version 1.523 of Jenkins on an Ubuntu server.
The solution I found was to manually edit the project config.xml file.
sudo vim /var/lib/jenkins/jobs/yourProjectName/config.xml
I had images in the description and I removed those leaving the description tag empty.
I also had an empty plot section which I removed completely from the publishers section.
After saving the config and going back to 'manage Jenkins' and then 'Reload Configuration from Disk' I then reopend my project and the config opened normally again.
I wasn't getting quite the same error message as you in the browser console to start with, but I hope that helps.
I too downgraded to 1.523 and then it worked... So it's obviously some bug introduced after 1.524.
Upgrading from 1.530 to 1.531 fixed it for me.