I have a typo3 plugin, that handles the /code/xyz.
When a code is called the first time it doesn't work and results in a 404. When it's called the second time it's running into the controller (I've set breakpoints there).
What could be the reason for the difference of behavior between the two calls. With nothing changed by me two calls right after each other exhibit the same behavior.
Thanks in advance.
Ok, this one was a hard one to crack.
Even though we thought is was Realurl the problem was a different one.
The integrator doing the upgrade from 7 to 8 missed one breaking change:
https://docs.typo3.org/c/typo3/cms-core/master/en-us/Changelog/8.5/Breaking-78002-EnforceCHashArgumentForExtbaseActions.html
Related
Well, I will explain the case, this is the third time that it happens to me that the reports that I make in Jasperserver work perfectly for me, and after 3 days that I use the program again, without any explanation one of the input parameters (That has a query) stops working, the report is loading and after that it doesn't happen, the only solution I've found is to remove and reinstall jasperserver (community version) but it becomes tedious, some explanation about it or does it only happen to me?
Edit: Now I am placing the wrong query with data that is not in the database and the loading continues to appear... it does not even mark an error, it is a simple selection query.
I have the weirdest bug. If I edit anything in the 'Script Execution Order' list in Project Settings, and click 'Apply', it just reverts back to how it was before immediately. This includes any type of change: removal of entries, adding entries, or changing the number of the entry. Everything just plops right back to the previous state on clicking Apply - no error messages whatsoever.
Visually, it's exactly the same result as clicking the 'Revert' button instead of 'Apply' - except that the editor actually starts doing its refresh thing afterwards (I thought for a moment maybe the Apply button was broken, but this kinda rules that out).
After hours of trying stuff I figured out that it works if I remove the 'Unity UI' package, so I guess the problem may have to do something with that. However, another project that includes this package does not have the same problem.. so it's still project-related, somehow. I tried uninstalling and reinstalling all packages 10 times, no luck..
I had submitted this as a bug to Unity support, who have just responded.
After some investigation, I've concluded that this is currently by
design. You're trying to make changes to scripts from Packages,
currently, you can't do that. If you would add your own script and try
to change the execution order, you'd notice that everything works
correctly. (Matas, Customer QA Team)
He is correct, and indeed, changing/adding entries from your own code DOES work. In hindsight, this makes sense, even if it's a bit rigid. Problem 'solved', as in, I can live with packages' entries being readonly.
I have had this problem as seen in the title in the past and I always resolved this issue by adding the psr-4 autoloader in the emconf and simply reactivate my extensions.
Now I'm facing the same problem:
Could not analyse class: maybe not loaded or no autoloader?
but I have set the autoloader correctly as always. This also happens in more than one Extension right now.. After deleting the php cache in the install tool und dump autoload and reactivating my extension, the error was gone for some time, a couple hours later its back again.. Therefore I think it must have something to do with temp files, but I can't figure out what it is exactly..
Does anyone have a solution? I have seen plenty of topics about this issue on stackoverflow, I used them in the past, but unfortunately none is working for me right now.
Important fact: This error is happening on my new server now. On my old server (with the same code in the extensions) this didn't occur and they worked fine.
Thanks in advance.
Edit: Vendornames etc. are set correctly and there are no errors in the syntax whatsoever. As I said, the extensions worked fine.
Edit2: I just found this changelog of Typo3:
https://docs.typo3.org/typo3cms/extensions/core/Changelog/8.4/Breaking-78222-ExtensionAutoloadInformationIsNowInTypo3confautoload.html
But there is no solution for the impact for none composer installations. Can someone provide one for me?
When you use composer installation and you use extensions which are not installed via composer, you need to add the autoload information in the root composer.json of your project and then run composer dump-autoload. (ext_emconf.php dont works in composer mode?)
{
...
"autoload": {
"psr-4": {
"Vendor\\ExtensionNameA\\": "public/typo3conf/ext/extension_name_a/Classes",
"Vendor\\ExtensionNameB\\": "public/typo3conf/ext/extension_name_b/Classes"
}
}
}
A possible explanation for strange timing thing "works and later not". Maybe it has something to do with the red clear cache button in the TYPO3 backend (Clear all caches). Maybe it start not working when u hit this button and cache files get cleared. Then you need to reinstall extensions to get the autoloader "temporarly" working, till the point you hit the clear all caches button again. With the solution i mentioned above, it works permanet.
Have you left any configuration in your ext_tables.php?
As the TCA configuration, which is cached, is expected in Configuration/TCA/[Override/] any code in ext_tables.php might get lost.
If you want some configuration to be executed for each run you need to put it in ext_localconf.php
Thanks for all the help, I found the solution myself now.
Actually it was not really caused by any autoloader configuration, but and old version of fpdf which apparently caused two extensions to not load their classes properly. The exception thrown was simply misleading. I have upgraded the version of fpdf and now it works properly. It is not clear to me why the same code worked a week ago and now it failed, but atleast I have found the solution to my problem.
First, I am new to Octave, but fairly familiar with Matlab. After using Octave for a short time I've noticed that the cursor in the command window will regularly add a new, empty line. Sometime this can be in the middle of typing a command. If I let the window sit with no input, the cursor will eventually advance all the way down the page and keep going.
I have never seen anything like this, and I do not know where to even begin to fix it.
Thanks!
Command Window Screenshot
My alternative answer is that that a bug report should be filed so that this issue gets fixed. I have the exact same problem. Every 30 seconds, when system time hits an even minute and 30 seconds past the minute, the command window cursor skips to the next line. One can still type to the command line and hit enter in the end. The command line now split at random points to several lines will be successfully computed. The problem is that you can not edit the line back past the inserted line break and the whole thing looks really messy. I tried to un-install and re-install but that did not help in my case. I use octave-4.2.1-w64-installer.exe
Boring solution (and hopefully permanent): I uninstalled and reinstalled the same version of Octave and this bug seems to have gone away. Not very convenient, but worked for me.
This might be a problem caused by a 3rd party application. I've seen this issue on a computer with Druva inSync. Disabling the Druva inSync service (not the client) stops this. inSync injects itself in applications using the Windows file dialog. Apparently this code contains a bug and accidentally writes a line feed (LF) to a file handle belonging to the application. The problem might also go away with a different Octave version. In particular the 4.2.x and 4.4.x seem to behave differently.
I am using the Flex Builder 3 debugger almost every day and it's starting to be a real pain that the 'Variables' tab in the 'Flex Debugging' view doesn't show local variables, only 'this' is displayed.
Also I can not add Watch Expressions for local variables.
Am I forgetting something here or is the debugger just very limited?
Thanks
I have the same problem that motto described. Local variables are not displayed when debugging, "hover" method doesn't work either. You can always use trace() or Loging API.
The problem occurred to me yesterday and the only thing I noticed is that the project stopped to compile using ant script due to OutOfMemoryError (I usually build project using FB, but sometimes FlexBuilder doesn't show errors/warnings and compiling via ant script is the only way I know to find them).
To see if Flex Builder is even tracking your local variables properly, set a breakpoint somewhere in a function. Once you hit that breakpoint, you can also "hover" over a variable in your source code, and a tooltip should show the current value of that variable.
Sometimes, FB has problems if the current function is further down in a larger file. Moving the function code up solves the problem (but exposes the same for another funciton further doen then). It seems that FB can only gather a certain amount of metadata for the currently open file.