|
FAQ
1. I am trying to run the samples. I see "Processing....." and nothing
happens. What is wrong?
You might be missing one or more required jar files. Please make sure
that you have the following jar files:
1. Latest SwingSet binary jar file (e.g. swingset-bin_0.8.0_beta.jar)
2. Latest PostgreSQL JDBC drive (e.g. pg74.214.jdbc3.jar) from:
http://jdbc.postgresql.org/download.html
3. For JDK/JRE prior to 1.5, early RowSet implementation (e.g.
rowset.jar) from:
http://java.sun.com/developer/earlyAccess/jdbc/jdbc-rowset.html
If you still have problems:
1. Unzip the demo jar file and extract the files to a folder.
2. From a command prompt go to that folder and type:
java MainClass
3. You will now see any remaining errors on the command prompt screen.
2. I have set the headers for a SSDataGrid, so why I don't see them on the
screen?
You have to set the headers before you set the rowset. If you reverse the
order, you will not see the headers. If you want to specify headers, use
the empty constructor.
3. Why am I getting an exception, while setting the renderers.
You have to set the renderer only after you have set the rowset. Until
you set the rowset the JTable does not have any columns, so when you try
to set a renderer SSDataGrid tries to access the specified column from
JTable, which causes an exception.
4. What happens to my default values and any other custom settings when I
change my underlying rowset?
All the defaults and other information like headers, renderer will remain
the same even after changing the rowset for a SSDataGrid. If you want to
change any settings or remove the defaults and/or headers, you have to
call the setXXXXX() functions to change the behavior.
Example: You have a rowset for a SSDataGrid and have set the default value
for the third column to be "XYZ". If you then change the query
for the rowset and want to remove the defaults for the SSDataGrid,
you have to call the setDefaultValues() method like this:
dataGrid.setDefaultValues(null,null);
This will remove all the default values used.
5. I setup a SSDataGrid based on a rowset querying columns "X," "Y," and "Z."
I made a small change to the underlying query but return the same set of
columns. Why are my renderers and/or hidden columns not working as
expected?
When modifying a query for a SSDataGrid's rowset, you must insure that the
columns are returned in the same order! When you set a renderer, or
specify a column as hidden you specify either a column number or column
name (in which case SwingSet internally converts the name to a column
number). If you change the column ordering for the underlying query and
apply the modified rowset to your SSDataGrid, SwingSet continues to use
the "old" column numbers for any renderers or hidden columns. You will
have to supply new renderers/hidden columns for the reordered columns.
6. I'm having trouble getting SwingSet to work with MySQL. Are there any
special instructions?
Yes. The mysql-connector-java MySQL driver has a unique ResultSet
implementation. The default ResultSet is not updatable so an
UpdatableResultSet is needed for creating updatable GUI's with SwingSet.
Rather than creating a RowSet using a connection string and query,
the developer should instantiate their RowSet with an UpdatableResultSet.
In addition, since there is no connection string or query, the
SSDataNavigator needs to be explicitly told not to call the execute()
method of RowSet and to assume that the RowSet already contains the query
results. This is done with the setCallExecute() method.
Example:
Let rs be your UpdatableResultSet object.
JdbcRowSetImpl rowset = new JdbcRowSetImpl(rs);
SSDataNavigator dataNavigator = new SSDataNavigator();
dataNavigator.setCallExecute(false);
dataNavigator.setRowSet(rowset);
If you want to change the query, change the rs object as needed,
re-instantiate your RowSet, and re-call the setRowSet method of your
SSDataNavigator.
(P.S.: This is a work around provided for MySQL DB users based on the
feedback from lopes. See
http://sourceforge.net/forum/forum.php?thread_id=1005899&forum_id=313186)
7. I'm currently using a SwingSet release prior to 0.9.0. What is this new
SSConnection and SSRowSet? How do I fix my programs?
For version 0.9.0 a datasource abstraction layer was added to SwingSet. All
of the SwingSet components are now based on a new SSRowSet interface rather
than Sun's existing RowSet interface. The SSRowSet differs from RowSet in
two important ways:
1. SSRowSet extends serializable which greatly facilitates serialization/
deserialization in the rest of the SwingSet components
2. SSRowSet only contains methods necessary to support the data types in
used by SSTextDocument which will make writing SSRowSet implementations
for non-updatable RowSets and other non-database datasources (e.g. a
HashMap) much easier
A SSJdbcRowSetImpl implementation of SSRowSet is provided to replace the
JdbcRowSetImpl used in most existing SwingSet applications. It is basically
a serialized wrapper of JdbcRowSetImpl. It can be used in conjunction with
the new SSConnection, a serialized wrapper of the Connection interface, which
handles serialization/deserialization of database connection info (path,
username, password, etc.). Finally, an SSRowSetAdapter is provided with
empty method implementations of everything in SSRowSet. This adapter can be
easily extended with non-empty method implementations for non-database
datasources.
To accommodate non-updatable RowSets, SSJdbcRowSetImpl can be extended with
custom updateXYZ() methods to handle database updates via INSERT/UPDATE
queries. SSJdbcRowSetImpl can also serve as a template for writing SSRowSet
wrappers for other RowSets (e.g. CachedRowSet, WebRowSet, etc.).
Unfortunately, the introduction of the SSRowSet requires modification of
existing SwingSet applications, but with the SSJdbcRowSetImpl and
SSConnection, these changes should be minimal. Below is an example of changes
required to transition to 0.9.0 and later versions of SwingSet:
***********************
OLD - Connection/RowSet
***********************
import java.sql.*;
import com.sun.rowset.JdbcRowSetImpl;
Connection conn = null;
JdbcRowSetImpl rowset = null;
Class.forName("org.postgresql.Driver");
conn = DriverManager.getConnection
("jdbc:postgresql://pgserver.greatmindsworking.com/suppliers_and_parts",
"swingset","test");
rowset = new JdbcRowSetImpl(conn);
***************************
NEW - SSConnection/SSRowSet
***************************
import java.sql.*; // still needed
import com.nqadmin.swingSet.datasources.SSJdbcRowSetImpl;
import com.nqadmin.swingSet.datasources.SSConnection;
SSConnection ssConnection = null;
SSJdbcRowSetImpl rowset = null;
ssConnection = new SSConnection
("jdbc:postgresql://pgserver.greatmindsworking.com/suppliers_and_parts",
"swingset", "test");
ssConnection.setDriverName("org.postgresql.Driver");
ssConnection.createConnection();
rowset = new SSJdbcRowSetImpl(ssConnection);
8. I'm building some screens graphically in NetBeans using the SwingSet
JavaBeans. I've setup my SSConnection and SSRowSet properly, but when I
launch the screen, I'm getting null pointer exceptions on the bind() methods
of the individual SwingSet components? How do I correct this?
When building an application in NetBeans you have to call the execute()
method explicitly on SSJdbcRowSetImpl. This has to be done in the post
initialization code for the last rowset/connection property you set.
Assuming that you are setting the four connection-related properties for
SSConnection followed by the two connection-related properties for
SSJdbcRowSetImpl, you would put the following code in the post initialization
code for the last property set for SSJdbcRowSetImpl:
// ASSUMING THAT THE ROWSET VARIABLE NAME IS sSJdbcRowSetImpl1
try {
sSJdbcRowSetImpl1.execute();
} catch(SQLException se) {
se.printStackTrace();
}
Note that in NetBeans you should add your SSConnection and SSJdbcRowSetImpl,
set their properties, and add the above call to execute() before adding and
binding individual SwingSet components to SSRowSet column names.
|
