EJBBuilder is a plugin for the eclipse platform. It will provide the ability to build and deploy EJBs within the eclipse platform. The target is to have an oss end-to-end solution including build, o/r mapping, (re-)deploy, debug.
EJBBuilder consists of different (sometimes remotely) related components to reach the final goal at some point in time.
This wizard will allow creation of EJB's, with support for Entity, Session, and Message-Driven Beans. All source files necessary for each type will be created. If an ejb-jar.xml file already exists, the bean descriptor information will be added to it, otherwise a new one will be created first. Eventually, support for application-server specific deployment descriptors will be added. First candidate as target environment is likely JBoss to be.
Status: There is a pre-alpha, but working, version of the wizard in the cvs repository, though its feature reachness holds some potential ;-) Loads of stuff is not implemented, but it shows how it should basically work.
Before pushing it to a release it is likely to rework the code generation. We are considering velocity, a template engine, as the future core of our code generation.Any thoughts on that?
Status: planning.
This will provide the ability to launch an application server inside eclipse to allow integrated running and debugging of the EJB's. The goal is to initially provide a JBoss (http://www.jboss.org) launcher, with others added in a future version.
Status: pre-planning (varporware ;-).
This will provide the ability to package the EJB project into an ejb jar file suitable for deployment into application servers. The packaging will consist of creating a jar file containing the EJB class files, any required deployment desriptor files, and any other supporting files identified by the developer.
Status: pre-planning.
A deploy/undeploy mechanism will be made available to easily deploy and undeploy ejb jars into an application server. Initially support will be included for JBoss, but others will be added in the future.
Status: pre-planning.
This will be a view where JNDI servers can be configured and browsed. Probably this will be implemented via JMX.
Status: planning.
The main motivation for the SQL Browser is to give developers the ability to browse an existing data model and execute db queries inside eclipse. It is intended to be used by developers rather than database administrators.
So reagarding the datamodel it is read-only. It is not in scope yet to empower the user build a data model, but to support developers using a database.
I just found out (8th of June) that there is another plugin in development with a much broader scope ecldbtool. Maybe you should check it out too.
And another one... (5th of July) eclipsedata. I am not sure if this project is still active though?
Status: pre-alpha.
It is likely that an alpha of SQL Browser will be released until mid of June '02.
To Dos for 0.0.1 (alpha) (mid June)The listed features are ideas and the order represents the current way of thinking. Nobody knows when or if at all these features will see the light of the day.
Have a look at sourceforge.
You can access the source code via anonymous cvs. You can also browse the code online.
We would appreciate input on these questions a lot. Please send feedback to the developers list.
You are also invited to join the party. Contribute bug reports (be the first ;-)[pending: link], feature requests [pending: link], documentation or code.
The commmunication is done using two mailing lists:
| Developer List | Intended for discussions about the development of the plugin | un-/subscribe | archive |
| User List | Intended for discussions about the usage of the plugin (be the first ;-) | un-/subscribe | archive |
Currently only the postgresql and Oracle JDBC drivers are contained in the plugin (to have Oracle in there is hopefully no licence breach). How to tell eclipse to use jar files not delivered with the plugin? Is changing the plugin.xml file and adding the libs by hand a/the only solution?
Currently the ui locks up on long lasting calls, e.g. accessing a slow database. Regarding the article "How to use the JTree Viewer, FAQ Section" [pending: link] the proper solution to that is to use multi-threading.
This would likely make it neccessary to add connection pooling. A candidate component for that is DBCP from jakarta-commons [pending: link]
How is it to be accomplished that cvs does send a message to a mailinglist after a commit happened?
When uploading files to sourceforge via scp the created files only allow rw access for the owner. Is there anyway to change that behavior? Does scp/ssh execute some script when logging in?
$Revision: 1.1 $