-
Notifications
You must be signed in to change notification settings - Fork 1
1.0
1.0
java.lang.NullPointerException at java.io.StringReader.(StringReader.java:50) at org.tangram.components.editor.EditingHandler.contentImport(EditingHandler.java:546)
1.0
When the Authorization Service needs a login, it does not check, whether or not it needs to ask the user, which login provider he or she wants to use. It always redirects to the selection page even if therer will be just one entry on that page.
1.0
1.0
Google and other services tend to remove trailing /. The system treats this
as a 404 HTTP code instead of the expected redirect to e.g. the home page.
1.0
1.0
Since the instanciation has changed this is more than just updating the numbers in the plugin's version number list.
1.0
For easier configuration and migration, it is necessary to let a custom attribute from the set of attributes of an external user be taken as the id of that user, like e.g. the email for google users or the screen_name for twitter.
1.0
1.0
Avoid the local copy of the code in the repository and use references like with JAR libraries to integrate it into the system.
1.0
Avoid the local copy of the code in the repository and use references like with JAR libraries to integrate it into the system.
1.0
(Optionally) avoid the local copy of the CKEditor in the JAR and use the CDN provided by AWS to load the editor efficiently.
1.0
This should facilitate caching or add basic CDN support.
1.0
It must be possible to add users and their passwords at runtime and not only in config files, like it is already possible for the FTP service. Of course this has to work with the hashed passwords requested in another ticket.
1.0
The Sping Framework integration got broken somewhere along the way. Fix it again.
1.0
And get rid of the plain text storing of password for the SimpleUserPasswordMapping
and other options.
402658 Only add Users to the List of currently logged in Users when they are not already part of it (Closed)
1.0
Only add Users to the List of currently logged in Users when they are not already part of it.
1.0
We need to discover of two object refer to the same user.
1.0
Use : instead of @.
1.0
Some parts of the Google Guice integration are not stock Google Guice or other
external modules. We use the term Guicy for this part. This has to be reflected
in folder names and the like.
1.0
Since any of the authentication providers will use sessions to store the login, we simply would like to invalidate this session for logout.
1.0
Current releases of the JSP specification and the corresponding containers don't allow this as EL valid expressions.
1.0
The .name. and .description. templates display the technical derived ORM
class. With OpeJPA when using subclassing this e.g. means org$tangram$example$Topic$pcsubclass.
Whenever possible show the class created by the developer and not derived
classes. They are known in other parts of the editor and should be passed to
the templates for display to the user.
1.0
This placeholder configurer was introduced to deal with a very special
configuration chalange introduced by the database configuration for
run@cloudbees.
There is a pattern based solution which doesn't need this special property
splitting anymore. Furtermore run@cloudbees is not available anymore since
they closed this part of their business.
1.0
It should be possible to use Google Guice as a Dependency Injection framework. This also includes, that we need an extension supporting JSR250 and means to generically inject the optional values.
1.0
Alongside stop using a mobile and a desktop view. We developed these back when support for limited mobile devices and editing through EDGE networks was needed.
1.0
Due to some integration errors, references to entities when using EBean can be used transiently but don't get persisted.
1.0
Up to now we don't have any possibility to register new users on the website easily when not using Google App Engine's APIs.
1.0
During the class path scan wrong file names get generated. As a result no codes can be found.
1.0
Don't use the most times working but in fact incorrect text/xml and
text/javascript anymore. Perhaps cope with them as input.
1.0
The source code must not contain other dependency injection related annotations except from JSR330. Don't use Spring Framework custom annotation even in the spring related modules for easier maintenance and migration into common modules.
1.0
Just a change in the FTP service.
1.0
The JSPs views of the Edior rely on this method but right now they are only available in all storage backends by accident. This needs to be changed.
1.0
Some storage backends need to generate classes from the original classes created by the developer to support their respective storage solution. Thus the names of the classes in use by the system are not meaningful for the developer and the user.
The system should derive the original names from the classes in use and only show thoses original classes' names in most places.
1.0
Update to the latest version available - e.g. 3.19. Additionally make the version status of these components visible in the soure code.
1.0
Since there are a number of command line paramters to pass to the server for correct operation with the now common JDO 3 setup, it would be far easier to add the call of the app server to the Script and use it there. Additionally this would be similar to the non Google App Engine scenarios.
1.0
Groovy based classes in the repository register custom views by directly modifying the custom view collection in the DefaultController. Thus they cannot be unregistered on unloading or modification of the classes. We need a complete rewrite of that mimik, which facilitates this correctly.
1.0
Google App Engine now suppurts newer versions of JDO and DataNucleus. Migrate to those as soon as possible.