Develop Reference

Creating new H2 server programmatically

In H2 there are two ways to create a new in-memory database. In the first, you explicitly create the database with a CREATE DATABASE.. SQL statement. In the other, if you attempt to connect to a non-existent database, H2 will simply create it. I've elected the first way because if I don't get some kind of error back how will I know to create the single table (with only two columns).
The problem is that H2 doesn't like he SQL I'm using and flags an error. This SQL statement:
String sql = "CREATE DATABASE Tickets, " + USER + ", " + PASS;
throws this exception:
org.h2.jdbc.JdbcSQLException: Syntax error in SQL statement "CREATE DATABASE[*] TICKETS, USERNAME, PASSWORD "; expected "OR, FORCE, VIEW, ALIAS, SEQUENCE, USER, TRIGGER, ROLE, SCHEMA, CONSTANT, DOMAIN, TYPE, DATATYPE, AGGREGATE, LINKED, MEMORY, CACHED, LOCAL, GLOBAL, TEMP, TEMPORARY, TABLE, PRIMARY, UNIQUE, HASH, SPATIAL, INDEX"; SQL statement:
Any idea about what going on in the above? Or, can you tell me how I can tell that the DB was auto-created so that I can proceed to create the table?

I don't believe that you're correct when you suggest that you can create a H2 database via SQL - I think that's your basic issue...
Just connect to your DB (and it's the jdbc URL that defines the database involved) and if you don't get an exception, carry on and use it. (Create your table, etc.)

Java DB Database, check for all table names

I've got an embedded Derby Database in my java application, and have multiple table's (that are created and deleted, so nothing is set in stone). I wanted to be able to return a list of names of all the tables currently in the database as I have to display the list in the application as well as get all the information from them.
Easiest way to do so? I don't need code just a method or methods. I'm a terrible google-fu user.
Currently my code works by grabbing a ResultSet from a specific table name entered, but it's only for testing purposes and I need to be able to display the full list of tables I have.
EDIT: My current workaround is actually different than posted. I simply have another table that holds all the table names created and updates when one is created/deleted. Obviously not the best approach but it works for me right now.

DatabaseMetaData metaData = connection.getMetaData();
ResultSet resultSet = metaData.getTables(null, "schenaName", "%" ,new String[] {"TABLE"} );
while (resultSet.next()) {
System.out.println(resultSet.getString(3));
}
Adding new answer:
Connection connection = getDBConnection();
DatabaseMetaData dbMetaData = connection.getMetaData();
//getting catalogs for mysql DB, if it is not working for your DB, try dbMetaData.getSchemas();
ResultSet catalogs = dbMetaData.getCatalogs();
while(catalogs.next()){
String catalogName = catalogs.getString(1);
//excluding table names from "mysql" schema from mysql DB.
if(!"mysql".equalsIgnoreCase(catalogName)){
ResultSet tables = dbMetaData.getTables(catalogName, null, null, null);
while(tables.next()){
System.out.println(catalogName + "::"+tables.getString(3));
}
}
}

Using metadata is the (somewhat) more portable solution. Note that you don't need the catalog stuff with Derby, as there are no catalogs. You can issue dmd.getTables(...) directly with null for the catalog. If all the tables you track are in a single schema, (and there aren't any other tables in that schema), getTables(null, "schemaName", null, null) should do the trick.
If need more fancy querying and you're not concerned about portability, you can check out
the dataBaseMetaData tool which gives you access to metadata as tables so that you can perform joins and other sophisticated queries on them.

Try this:
select tableName from sys.systables
You should get all the tables your system.

How to create multiple schema connections at runtime?

I have a a project name drop-down list, and based on the selected value a schema drop-down list is generated.
After the schema list is generated, there is a 'choose file' option where a script name is chosen which is to be run on multiple schemas.
Both the project list values and the schema list values are stored in the database. There is another table in the database where the schema credentials are stored.
I want to know that how create connections on those multiple schemas, and how to run a script in multiple schemas selected from that schema drop-down list.
I have executed the code in one schema using iBatis framework using scriptrunner method.
Table 1: ProjectName
PROJECT_PK,
PROJECT_CODE,
PROJECT_NAME
Table2: ComponentName
COMPONENT_PK,
COMPONENT_CODE,
COMPONENT_NAME,
PROJECT_PK
Table 3: SchemaName (This table contains the credentials of Other Schemas)
SCHEMA_PK,
SCHEMA_NAME,
PASSWORD,
ORACLE_SID,
HOST_NAME,
PORT_ID
Table 4: PROJECT_DETAIL
PROJECT_DETAIL_PK,
COMPONENT_PK,
SCHEMA_PK
Table5: COMPONENT_DETAILS
COMPONENT_DETAILS_PK,
PROJECT_PK,
SCHEMA_PK
I am attaching the scenario image.

I propose you create a 'super-schema' with the appropiate grants on every other schema, and create your JDBC connection for this 'super-schema'. This will make you need some tampering with the sql script - you will need to write it with some marker to easily substitute the schema, something like
ALTER TABLE ${SCHEMA_MARKER}.INVOICES ADD CLIENT_ADRRESS Varchar2(35);
And your java code would substitute it for the schema you are targeting. With some imagination you can extend this idea to execute bulk DDL scripts on all your schemas.
By the way, I understand by your question that you have many schemas with the same structure. I was forced once to work with such a legacy structure, and so I know there is a lot of momentum in corporate structures; nonetheless I must recommend you to redesign such system. Think for example in creating materialized views on your super-schema, replicating your schema structure but with a new field in every table primary key (this new field would fill the gap that made whomever did it separe the data in many schemas for starters).
This will surely make your query cache suffer a lot less, and will make easier the development of any new software that needs to work with that 'distributed' data.

In Oracle, a schema is basically a user. So to create a table FOO in the schema SX, you just login as user SX and execute create table FOO (...) (without specifying a schema). The user == schema convention in Oracle will make sure this works.
To login as user SX, get hold of your DataSource and use getConnection(String username, String password) instead of the default getConnection()
Alternative ways are to add placeholders in your script that you process to generate valid SQL first. See the answer by Jorge_B for an example.
Lastly, you can change the default schema using ALTER SESSION (see Default Schema in Oracle Connection URL). The problem with this approach is that you must restore the schema before closing the connection when using a web container with a DataSource - in this scenario, connections are never really closed, so the next piece of code asking for a connection will get one with an unexpected default schema -> hard to find errors.

Here is the DAO class to create multiple schema connection on the same database at run time.
We need to enter the schema names at run time through and call that in some servlet.
The schema details i.e username, password,host Id,port Id and SID are to be stored in the database tables though which the connections will be created.
The created connections are stored here in the list which can be used at a later point of time.
The credential DTO here is the object which maps to the database tables in the database and then
Connection conn=DBUtil.getConnection(constants.DB_DS_NAME);
Statement stmt=null;
stmt=conn.createStatement();
ResultSet rs= null;
Connection [] con=new Connection[schemaname.length];
int i,j;
String [] url=new String[schemaname.length];
String [] username=new String[schemaname.length];
String [] password=new String[schemaname.length];
List<CredentialDTO> creDTOlist=new ArrayList<CredentialDTO>();
String query1="insert into dba_project_master VALUES(9,'abc','abc','abc','abc',40)";
String query2="CREATE TABLE EMP(EMPNO NUMBER(4,0) NOT NULL ENABLE,ENAME VARCHAR2(10 BYTE),JOB VARCHAR2(9 BYTE), SAL NUMBER(7,2),DEPTNO NUMBER(2,0))";
try
{
for(i=0;i<schemaname.length;i++){
String query=" select * from dba_schema_details where schema_name="+DBUtil.enquoteString(schemaname[i]);
rs=stmt.executeQuery(query);
while(rs.next()){
CredentialDTO creDTO=new CredentialDTO();
creDTO.setSCHEMA_PK(rs.getString("SCHEMA_PK"));
creDTO.setSCHEMA_NAME(rs.getString("SCHEMA_NAME"));
creDTO.setPASSWORD(rs.getString("PASSWORD"));
creDTO.setORACLE_SID(rs.getString("ORACLE_SID"));
creDTO.setHOST_NAME(rs.getString("HOST_NAME"));
creDTO.setPORT_ID(rs.getString("PORT_ID"));
creDTOlist.add(creDTO);
}
}
System.out.println("creDTOlist size:"+creDTOlist.size());
//create URL for the schema name
int m=creDTOlist.size();
Iterator<CredentialDTO> LItr= creDTOlist.iterator();
String [] username1=new String[m];
String [] password1=new String[m];
i=0;
while(LItr.hasNext()){
System.out.println("iteration "+i);
CredentialDTO temp = LItr.next();
String URL="jdbc:oracle:thin:#"+temp.getHOST_NAME()+":"+temp.getPORT_ID()+":"+temp.getORACLE_SID();
System.out.println("URL:"+URL);
username1[i]=temp.getSCHEMA_NAME();
System.out.println("iteartion "+i+" username="+username1[i]);
password1[i]=temp.getPASSWORD();
System.out.println("iteartion "+i+" password="+password1[i]);
Class.forName("oracle.jdbc.OracleDriver");
con[i]=DriverManager.getConnection(URL, username1[i], password1[i]);
System.out.println("Connection Name:" +con[i]);
Statement st1=con[i].createStatement();
con[i].setAutoCommit(false);
st1.addBatch(query1);
st1.addBatch(query2);
int [] update=st1.executeBatch();
i++;
}
}
catch(Exception ex){
ex.printStackTrace();
}finally
{
if (conn != null) try{conn.close();} catch(SQLException ignore) {}
if (stmt!= null) try{stmt.close();} catch(SQLException ignore) {}
if (rs != null) try{rs.close();} catch(SQLException ignore) {}
}
return creDTOlist;
}
}

Return Timestamp With Prepared Statement

I have an auto generated timestamp that is created each time a record is inserted or updated in a mysql table. Is there a way to return this timestamp in a way similar to how I would use a keyholder to return a newly created id?
KeyHolder keyHolder = new GeneratedKeyHolder();
JdbcTemplate jdbcTemplate = new JdbcTemplate(dataSource);
//Insert Contact
jdbcTemplate.update(new PreparedStatementCreator() {
#Override
public PreparedStatement createPreparedStatement(Connection connection) throws SQLException {
PreparedStatement preparedStatement = connection.prepareStatement(SQL_ADD, Statement.RETURN_GENERATED_KEYS);
preparedStatement.setString(1, contact.getFirstName());
preparedStatement.setString(2, contact.getLastName());
preparedStatement.setInt(3, contact.getOrganizationId());
preparedStatement.setString(4, contact.getType());
preparedStatement.setInt(5, contact.getUserId());
return preparedStatement;
}
}, keyHolder);
//Use keyholder to obtain newly created id
contact.setId(keyHolder.getKey().intValue());
Is there some way to also return the new timestamp without having to requery the table? I have been looking for ways to return it along with the id as a key in the keyholder, but it doesn't seem to be returned as a key?

Not very satisfying, but I think "no" is the answer to your question. I don't know any of the Spring stuff, but I think this is due to the basic JDBC that it's wrapping. See http://docs.oracle.com/javase/6/docs/api/java/sql/Statement.html#getGeneratedKeys%28%29
You only option would be to create a stored procedure on MySQL that has an out parameter and call that. See http://dev.mysql.com/doc/refman/5.0/en/call.html.

There are few options for solving this issue on the MySQL database server side. You could start with creating a TRIGGER on the table. As TRIGGER has a restriction and cannot return data, you can set the TIMESTAMP value to a variable:
DEMILITER //
CREATE TRIGGER ai_tbl_name AFTER INSERT ON tbl_name
FOR EACH ROW
BEGIN
SET #TimeStamp = NEW.timestamp_column;
END;//
DELIMITER ;
To retrieve this timestamp value, run the following command:
SELECT #TimeStamp;
Since the variables are stored in the memory, there will be no need to open any tables again.
You go even further. You could create a STORED PROCEDURE in MySQL to automate all the above (sample code, as I do not know your table's details):
DELIMITER //
DROP PROCEDURE IF EXISTS sp_procedure_name //
CREATE PROCEDURE sp_procedure_name (IN col1_val VARCHAR(25),
IN col2_val VARCHAR(25),
IN col3_val INT)
BEGIN
INSERT INTO tbl_name (col1, col2, col3)
VALUES (col1_val, col2_val, col3_val);
SELECT #TimeStamp;
END; //
DELIMITER ;
You can run this procedure with the following code:
CALL sp_procedure_name(col1_val, col2_val, col3_val);
As I'm not familiar with the Java, you'll need to finish it up with your side of code.

It seems that the variable contact is an instance for the newly inserted record. As it contains the newly generated id (primary key) field value, you can execute a new query to return the required timestamp field value for this new id.
The query may look like this:
select timestamp_field from my_table where id=?
Use PreparedStatement to input new id value and execute it to fetch required timestamp field value.

GeneratedKeyHolder also has two methods: getKeyList() that returns Map<String,Object> of generated fields; and getKeyList() that produces a list of generated keys for all affected rows.
See java doc of GeneratedKeyHolder and Spring tutorial of auto generated keys
In addition Spring's SimpleJdbcInsert has methods for generated key retrieval. See also method SimpleJdbcInsert#usingGeneratedKeyColumns

There are 2 methods in java.sql.Connection class causing PreparedStatement execution to return selected key columns :
PreparedStatement prepareStatement(String sql,
int[] columnIndexes)
throws SQLException
PreparedStatement prepareStatement(String sql,
String[] columnNames)
throws SQLException
You don't need to use Spring KeyHolder & JDBCTemplate to do this.
The give hope you could number/name your timestamp column. But the javadoc doesn't require or suggest that any JDBC implementation can return non-key columns, so your out of luck with this approach:
Creates a default PreparedStatement object capable of returning the auto-generated keys
designated by the given array. This array contains the names of the columns in the target
table that contain the auto-generated keys that should be returned.
As suggested in another answer, can switch to a stored procedure that does exactly what you want (CallableStatement is actually a PreparedStatement that executes storedprocedures - i.e. a subclass).
Can populate the timestamp column within the prepared statement via new Timestamp(new Date()) - but you should have in place a mechanism to sync times across your various servers (which is often used in windows and *nix environments). Your trigger could set the timestamp only if a value wasn't already provided.
As part of your app & DB design, you need to commit to a philosophy of where certain operations occur. If the DB derives needed data, the app needs to refresh data - you must pay the price of separate query executions or a combined stored proc that inserts & retrieves.

Error while trying to test if a specific table exists in a MySQL database using JDBC

I want to check wether a table named basestation exists in a MySQL table from a Java program using JDBC. My code is:
conn = DriverManager.getConnection(url, username, password);
DatabaseMetaData dbm = conn.getMetaData();
tables = dbm.getTables(null, null, "basestations", null);
if (tables.next())
{
//Table exists in database
//...
}
else
{
//Table does not exist
//...
}
but although table basestations exists in my database it seems that my code never enters the first clause and always decides that table does not exist, causing naturally a table already exists in database SQL error when trying to create the table from the beginning.

Not the best way to do it below.. But you could do the following:
Use JDBC to send the following statement:
select count(*) from information_schema.tables where information_schema.table_schema='your_schema' and information_schema.table_name='basestations';
Parse the jdbc result set.
That should return either 1 or 0. If the resulting count is 1, then the table exists, otherwise, it does not.

I suspect your getTables() not returning anything here. As per getTables documentation Only table descriptions matching the catalog, schema, table name and type criteria are returned. In your case catalog and schema are null, which may not be correct (my guess), populate them with proper values.