We found a nice article about a simple CRUD application, created with JSP/Spring/Hibernate - read more....
This example is great for a comparison with JVx, because JVx is perfect for CRUD applications.
We used our existing app JVxFirstApp as reference because it is comparable to the Spring app. Our app has some extra features like authentication and a specific security manager, but that does not change anything.
Both applications provide CRUD operations on a table called CONTACTS. A big difference is the used technology: Web (JSP) vs. Desktop (Swing), but it is no extra coding effort to start our app as Web application - with our WebUI.
Why the comparison?
A big advantage of JVx is to simplify application development specifically of database applications. We want to show the difference in complexity and compare some statistics.
Let's start with the statistics:
| |
|
JSP/Spring/ Hibernate |
|
JVx |
| Files |
|
17 |
|
9 |
| LoC |
|
265 |
|
208 |
| Config lines |
|
74 |
|
15 |
| JSP lines |
|
230 |
|
- |
| Libraries |
|
37 |
|
2 |
We used the same code formatter for both projects, therefore the Lines of code should be meaningful. The config files do not contain a lot of empty lines and the jsp files are standard html. We counted every line.
As you can see there is a big difference. Less lines of code do not automatically mean less complexity, but in this concrete example is it so. If your application has few lines of code, it is easier to understand how things work together and it is easier to maintain your application. And if you are a beginner it is better to get fast results without losing the control. That is the case if your project contains many files.
And if you are not a beginner it is important that you have the best flexibility with your framework. Both technologies allow the unrestricted use of other frameworks. But more dependencies means more problems if you plan to update to newer versions. The testing effort should not be underestimated. Fewer dependencies are an advantage!
What is your opinion?
We proudly present JVx 0.9!
This release of JVx contains about 190 changes
Originally we wanted to move some features to 1.0 because they needed a considerable amount of time, but we kept them for 0.9 because each feature is great and saves a lot of work... And now, some months after JVx 0.8 we have a very powerful JVx with tons of new features, performance improvements and still very simple and small. We don't know a framework with approximately the same features as JVx which is about 900Kb. If someone knows a framework, please let us know!
And before we continue with some new features, many thanks to all the database vendors
We had a lot of work with different Standard interpretations (quoting, case sensitive, OS differences, imcomplete JDBC implementations, ... )
What's new? ...too much to describe everything in one post, but let's start:
- Database
- Manual quoting via DBStorage, if you need specific database features and don't change the database
- Automatic quoting via DBStorage, if you plan to change your database or if you use a storage for different databases
- Change your database(s) without changing you client code read more...
- Use existing java.sql.Connection instances to create DBAccess instances
- PK, FK, UK detection with or without JDBC driver, when needed (DBAccess dependent)
- Server
- Introduced DBCredentials and DataSourceHandler to change DB connections without changing source code read more...
- Default storage implementation for automatic caching (AbstractCachedStorage)
- Default storage implementation for memory data (AbstractMemStorage)
- Server side triggers and POJO support for storages read more...
- Use custom SessionManager and ObjectProvider implementations
- Global meta-data caching read more...
- Better Exception handling for debugging
- User Interface
- Simple charting
- Default layouts for Container
- Save UIImages
- Changed automatic (master/detail) link editor visibility
- Introduced TreePathFinder
- login/logout event handler for applications
- Modern Message layout
- Border on/off option for TextFields
- Translation for Cell Editors (e.g. Date/Time Picker)
- Utilities
- Added FileUtil.replace, FileUtil.zip, FileUtil.unzip, FileUtil.copy (directories and files), FileUtil.move, FileUtil.delete (directories and files)
- Added StringUtil.getCharacterType, StringUtil.getCaseSensitiveType, StringUtil.replace (faster than String.replace)
- Added <XmlNode>.toString(boolean), <XmlNode>.getXmlValue, <XmlNode>.indexOf (deep search)
- Added CommonUtil.getFirstCause, CommonUtil.equals (supports null comparison)
- Added ResourceUtil.getResource and ResourceUtil now supports different ClassLoaders
- Added simple namespace support for XmlWorker
- XmlWorker has a new option for writing line-feeds between tags
- Added <FileSearch>.searchFirstFile and support for file/directory pattern search
The above list contains only some cool changes. You find the complete list in the Changelog.
But that was not all.... We added some new features to our WebUI and released the second beta, which is fully compatible with JVx 0.9.
Our Showcase application is up-to-date and uses JVx 0.9 and WebUI 0.8 beta-2.
And last but not least, we look forward to your feedback!
Some days ago, we updated our showcase application and especially the WebUI implementation.
But what is new?
Our WebUI has some great new features, in addition to our Plat_Forms release.
Check out the Help menu - the help is now embedded in the application. Another great feature is the download option for grids - press the small export button beside the grids. Or try uploading an image in the contacts screen - works great
Smaller changes are grid specific, e.g. the sort operation is now a remote operation, same for inserting new lines.
We welcome your feedback!
Because of the Plat_Forms competition, we released special versions of JVx, WebUI and OnlineHelp. All releases are available from our project page.
We implemented some amazing new features for JVx and WebUI.
Check out the changelogs for details.
Some tags:
- Simple charting
- Default layouts
- Global metadata caching (server-side)
- Modal frames in WebUI
- Table handling in WebUI
It is available from the project page.
What’s New?
The complete list of changes can be found here.
- Default Database (CoC)It is now possible to configure multiple db datasources via config.xml
<?xml version="1.0" encoding="UTF-8"?>
<application>
<securitymanager>
<class>com.sibvisions.rad.server.security.DBSecurityManager</class>
</securitymanager>
<datasource>
<db name="default">
<url>jdbc:oracle:thin:@localhost:1521:mydb</url>
<username>user</username>
<password>password</password>
</db>
</datasource>
</application>
If the DBSecurityManager does not define a specific datasource, the datasource with the name "default" is used.
- Quoting in DBAccessDBAccess supports case sensitive database access. To enable the feature, call
on your DBAccess instance.
- Custom SessionManager and ObjectProviderIt was possible to use a custom ObjectProvider but it was not possible to extend the DefaultObjectProvider. Now this limit is gone. Additional it is now possible to use a custom SessionManager and extend the DefaultSessionManager.
Configure the objects in your server config.xml like this:
<server>
<sessionmanager>com.sibvisions.AnotherSessionManager</sessionmanager>
<objectprovider>com.sibvisions.AnotherObjectProvider</objectprovider>
<server>
- Server Side TriggersOur DataBooks now supports specific events like before insert or after delete. The server side storage did not have this events, so you had to overwrite the data manipulation methods to call your business logic or to change the default behaviour. That was not very cool.
We have implemented a very powerful feature, with which you have full control of your data sent to the data tier. Do you know database triggers?
Now you have the same power in your storages.
A short example:
public void getUsers
()
{
DBStorage dbsUsers
= new DBStorage
();
dbsUsers.
setDBAccess(getDBAccess
());
dbsUsers.
setFromClause("V_USER_USERS");
dbsUsers.
setWritebackTable("USERS");
dbsUsers.
open();
dbsUsers.eventBeforeInsert().addListener(this, "doEncryptPwd");
dbsUsers.eventBeforeUpdate().addListener(this, "doEncryptPwd");
}
public void doEncryptPwd(StorageEvent pEvent)
{
IBean bn = pEvent.getNew();
bn.put("PASSWORD", AbstractSecurityManager.getEncryptedPassword(
SessionContext.getCurrentSessionConfig(),
(String)bn.get("PASSWORD")));
pEvent.setNew(bn);
}
Use the event handling as usual.
- Flat POJO support for StoragesInstead of using the dynamic property access via IBean, it is possible to work with POJOs:
public void doEncryptPwd
(StorageEvent pEvent
)
{
User user
= pEvent.
getNew(User.
class);
user.setPassword(AbstractSecurityManager.getEncryptedPassword(
SessionContext.getCurrentSessionConfig(),
user.getPassword()));
pEvent.setNew(user);
}
You can work with every useful POJO because the storage mapps only available properties to the POJO.
- Useful bugfixes
We look forward to your reviews
Check out our new JVx version. It is available from the project page.
What's in the current beta?
The complete list of changes can be found here.
Some features, however, at this point:
- Introduced DBCredentials and DataSourceHandlerIt is now possible to configure multiple db datasources via config.xml
<?xml version="1.0" encoding="UTF-8"?>
<application>
<securitymanager>
<class>com.sibvisions.rad.server.security.DBSecurityManager</class>
<database datasource="mydb" />
</securitymanager>
<datasource>
<db name="mydb">
<url>jdbc:oracle:thin:@localhost:1521:mydb</url>
<username>user</username>
<password>password</password>
</db>
<db name="masterdb">
<url>jdbc:derby://localhost:1527/masterdb</url>
<username>master</username>
<password>master</password>
</db>
</datasource>
</application>
and use the configured security manager connection in the Session:
IConfiguration config
= SessionContext.
getCurrentSessionConfig();
dba = DBAccess.getDBAccess(DBSecurityManager.getCredentials(config));
dba.open();
or any connection, by name:
IConfiguration config
= SessionContext.
getCurrentSessionConfig();
DBCredentials cred = DataSourceHandler.createDBCredentials(config, "masterdb");
dba = DBAccess.getDBAccess(cred);
dba.open();
It is still possible to use the old configuration format:
<?xml version="1.0" encoding="UTF-8"?>
<application>
<securitymanager>
<class>com.sibvisions.rad.server.security.DBSecurityManager</class>
<database>
<driver>org.hsqldb.jdbcDriver</driver>
<url>jdbc:hsqldb:hsql://localhost/demodb</url>
<username>sa</username>
<password></password>
</database>
</securitymanager>
You can create DBCredentials or read the config programatically, as it was with JVx 0.8.
- Introduced AbstractCachedStorageThe AbstractCachedStorage is the new base class for all IStorage or ICachedStorage implementations. It implements the metadata caching and offers simple export functionality. With this class it is easy to develop new storages.
- Introduced AbstractMemStorageThe AbstractMemStorage extends the AbstractCachedStorage and offers a MemDataBook on the server-side. With this class it is very simple to create storages with data kept in memory. You don't need a special backend, e.g. you can use a List to fill the storage. Our TwitterStorage uses this storage.
- DBAccess supports pre-configured java.sql.ConnectionCreate a DBAccess instance with a pre-configured connection, e.g.
Connection con
= getJNDIConnection
();
DBAccess.getDBAccess(con)
- XmlNode performance tuningOur XmlNode allows XML access with following syntax:
XmlWorker xmw = new XmlWorker();
xmnRead = xmw.read(new FileInputStream("example.xml"));
xmnRead.setNode("/archive/element(1)/user", "xml");
xmnRead.setNode("/archive/new/element", "car");
The class offers powerful and simple XML handling - check it out!
- Bugfixes and Test cases
We look forward to your reviews
We described all DataBook events and row states with some articles in our Forum. The events and states are very important if you need special data/row handling on client side e.g. ask the user before delete record(s).
The articles includes flow charts and practical examples. If anyone has questions, just post to the forum.
Direct links to the events and states.
We have a lot of useful features in JVx. It is very easy to develop multi tier applications, two tier applications or simple desktop applications - started as RIA, Web or Standalone application.
Every tier is technology independent and does not require third party frameworks. We can use every useful framework to create business logic.
But we don't have a lot of connectors out-of-the box (at the moment). There are a lot of technologies out there, e.g. NOSQL, Cloud services, Facebook, Twitter, YouTube, REST, ...
Not all technologies are useful for business apllications, but it is a lot of fun to work with new technologies
We spent some hours to develop a simple Twitter client (rather a connector) to read/delete/insert Tweets. The result can be seen here:

Twitter client with JVx
What we did?
- Created a new Storage, called TwitterStorage
- extended AbstractMemStorage
- Implemented loadData, insert, delete and getRowDefinition
- used twitter4j
After 290 lines of code we were finished. The standard RemoteDataBook works perfect with the new Storage (full support for filtering, sorting, master-detail relations).
JVx 0.8 ist out now!
We put much time and work into the new release. It contains almost 100 improvements!
But before we go into the details, it should be mentioned that JVx is now hosted on SourceForge. This step was particularly important to us, in order to be able to develop together with our Community.
Our Forum contains detailed instructions for working together.
We appreciate any help.
Download JVx 0.8 here
What's new?
- API changes
- We removed set/getTableColumnNames, set/getTreeColumnNames, set/getEditorColumnNames and introduced ColumnView. Now it is possible to set a column view per UI control e.g.: ITable, ITree, IEditor, null (ALL)
Old:
rdbInvoice.getRowDefinition().
setTableColumnNames(new String [] { "POS", "PRICE"});
New:
rdbInvoice.getRowDefinition().
setColumnView(null, new ColumnView("POS", "PRICE"));
- DBAccess has now a static method, setAutomaticLinkColumnNameTranslation, to translate the automatic link column names into a user-defined name.Default:
tmpAutoLinkColumnNames.put("*_id", "*");
tmpAutoLinkColumnNames.put("*_ID", "*");
- Changed, automatic, visible column detection from case insensitive to case sensitive, for compatibility reasons!Old:
setDefaultIgnoredColumnNames("*_ID");
New:
setDefaultIgnoredColumnNames("*_ID", "*_id");
Default:
"*ID", "*id", "*_ID", "*_id", "*_INTERN", "*_intern"
- Renamed countRows to getEstimatedRowCount in IStorage
- MetaData handlingReduced client/server requests,
Reduced creation loops on server side,
Caching for sub storages,
Fixed Nullable column detection,
Client Cache roles,
Label detection,
Split Client and Server MetaData usage
- SerializerUniversalSerializer is now the standard serializer,
Support for async serialization with new TypeCache,
Bean packages unified,
New init method for ISerializer,
The serializer name is now included in the connection properties
- JNDI supportServer is now available via JNDI or Singleton
- DB supportHSQLDB driver is now registered,
HSQLDB locking mechanism fixed,
Derby locking support,
close open statements,
Check constraint detection with 'OR', 'AND', 'IN',
Automatic Link Column Name detection works now DB independent,
Fetch optimization (reduced method calls),
BigDecimal DataType contains always a BigDecimal - DB independent
- SecurityDifferend Password algorithms (SecureHash),
Automatic Security detection of lifecycle objects,
XmlSecurityManager now supports more than 1 User
- Smaller changes and BugfixesAll details are documented in our Support System or the ChangeLog
The source code of JVx is now hosted on SourceForge!
As a little bonus, we gave following packages in the repository:
- JVx Core
- JVx Android AddOns
- JVx Swing UI
- JVx QTJambi UI
- JVx Web UI
- JVx Flex client and server (proof of concept)
And finally, we decided to release the complete source code of our JVx .NET AddOns. The source code was not yet available and is still in development.
We look forward for many helping hands!