SQLite JDBC driver version 3.21.0 (the latest).
Summary
A prepared statement is opened for multiple insert operations on a table, but is unable to survive primary key violations.
If one "bad" insert fails due to a Primary Key violation, the prepared statement is unable to process subsequent "good" inserts. An exception "statement is not executing" is raised when calling pstmt.setString().
Tracing the error into org.Sqlite.core.CorePreparedStatement the call is failing in checkOpen() pointer==0.
Example is below.
Does anyone know why this is happening? I see a similar bug report was raised but supposedly fixed.
Connection conn = DriverManager.getConnection("jdbc:sqlite:./test.db");
String createtable = "CREATE TABLE dummy(ID text, VAL text, PRIMARY KEY(ID) )";
String psSQL = "INSERT INTO dummy (ID, VAL) VALUES (?,?)";
String id = "123456789";
String val = "FooBar";
String id2 = "123456789";
String val2 = "FooBar2";
String id3 = "345678901";
String val3 = "FooBar3";
try {
Statement st= conn.createStatement();
st.executeUpdate(createtable);
PreparedStatement pst = conn.prepareStatement(psSQL);
// 1 insert good entry
pst.setString(1, id);
pst.setString(2, val);
pst.executeUpdate();
System.out.println("1st insert OK for " + id);
// 2. try to insert bad duplicate entry with pkey violation
try {
pst.setString(1, id);
pst.setString(2, val);
pst.executeUpdate();
System.out.println("2nd insert OK for " + id);
} catch (SQLException e) {
System.out.println("2nd insert Failed for " + id);
e.printStackTrace();
}
// 3. try to insert 3rd good value
try {
pst.setString(1, id3); // exception raised here
pst.setString(2, val3);
pst.executeUpdate();
System.out.println("3 insert OK for " + id3);
} catch (SQLException e) {
System.out.println("3 insert Failed for " + id3);
e.printStackTrace();
}
pst.close();
st.executeUpdate(droptable);
} catch (SQLException e) {
e.printStackTrace();
}
This bug has been fixed in sqlite-jdbc version 3.25.2, see https://github.com/xerial/sqlite-jdbc#news.
Related
I would like to delete a row in tableview but also in the
underlying SQLite Database which populate the tableview
Here I get the selectedRow
public void deleteDBRow() {
if (tableV.getSelectionModel().getSelectedItem() != null) {
Bew selBew = tableV.getSelectionModel().getSelectedItem();
System.out.println(selBew.getName());
}
}
and can delete it with casual code
DELETE FROM Table WHERE name = ""+selBew.getName()");
But I would like to delete the entry in the sqlite database also
From time to time I have rows with the same text in every column - so this way
was critical - can I use rowID to delete the selected row in sqlite?
Here one try from me to get rowid in tableview
ObservableList bewList = FXCollections.observableArrayList();
try {
String sql = "SELECT rowID, Name, Date, Action, Info FROM tab1";
Connection conn = DriverManager.getConnection("jdbc:sqlite:TestB1.db");
Statement stmt = conn.createStatement();
ResultSet rs = stmt.executeQuery(sql);
while (rs.next()) {
bewList.add(new Bew(rs.getInt("rowID"), rs.getString("name"), rs.getString("date"),
rs.getString("action"), rs.getString("Info")));
System.out.println("rs.next : " + rs.getInt("rowID") +" - " + rs.getString("date") +" " +
rs.getString("action") +" " + rs.getString("Info"));
}
}catch (Exception e) {
System.out.println("SQLiteDB.getData ---> Error RS");
System.err.println("*E"+e.getClass().getName() + ": " + e.getMessage());
}
return bewList;
String sql = "CREATE TABLE IF NOT EXISTS tab1" +
"(rowID INT PRIMARY KEY," +
"NAME CHAR(50) NOT NULL ,"+
"DATE CHAR(15) ,"+
"ACTION CHAR(50) ,"+
"INFO CHAR(3));";
conn.createStatement().executeUpdate(sql);
rowID ist always 0 - don't know why ???
edit:
Controller
ObservableList bewList = DB.getData();
tableV.setItems(bewList);
Edit:
Maybe error in here - add data
i add 4 values - rowid missing??
public void add(String Name, String Date, String Action, String Info) {
try {
Connection conn = DriverManager.getConnection("jdbc:sqlite:TestB2.db");
stmt = conn.createStatement();
String ValStr = "\'"+Name+"\' ,\'"+Date+"\',\'"+Action+"\' ,\'"+Info+"\'";
String sql = "INSERT INTO tab1 (rowID, NAME,DATE,ACTION, INFO) VALUES ("+ValStr+")";
// System.out.println("Button Click add"+conn.createStatement().toString());
stmt.executeUpdate(sql); //geändert
ObservableList<Bew> list = getData();
conn.close();
System.out.println("DB ROW add");
} catch (SQLException e) {
System.out.println(e.getMessage());
}
}
I perform a DELETE in my table from my servlet by calling a method action.deleteBox(box); which executes the delete function
deleteBox method
Connection c = null;
PreparedStatement stm = null;
sq = "DELETE FROM BOXES WHERE ...";
try {
Class.forName(typeDB);
c = DriverManager.getConnection(path);
stm = c.prepareStatement(sq);
//...
stm.executeUpdate();
} catch (SQLException e) {
System.out.println(e.getMessage());
} finally {
System.out.println("my stm is closed : " + stm.isClosed());
if (stm != null)
stm.close();
if (c != null)
c.close();
}
the delete is executed fine. Then I want to check how many records were deleted from the previous delete: So I call this method:
public int countDeletesRecords()
throws SQLException, ClassNotFoundException {
Connection c = null;
PreparedStatement stm = null;
sq = "SELECT changes() as \"deletedRows\" ";
int deletedRows=-1;
try {
Class.forName(typeDB);
c = DriverManager.getConnection(path);
stm = c.prepareStatement(sq);
ResultSet rs = stm.executeQuery();
if (rs.next()) {
deletedRows = rs.getInt("deletedRows");
}
} catch (SQLException e) {
System.out.println(e.getMessage());
} finally {
System.out.println("my stm is closed : " + stm.isClosed());
if (stm != null)
stm.close();
if (c != null)
c.close();
}
return deletedRows; //I get 0..
}
and I get 0, while 1 records where deleted.
While this does not directly answer the question an alternative, simpler, approach would capture the return value of executeUpdate(), that returns the number of affected (in this case, deleted) rows:
final int deletedRowCount = stm.executeUpdate();
The documentation says:
This function returns the number of rows modified, inserted or deleted by the most recently completed INSERT, UPDATE or DELETE statement on the database connection
You are creating a new database connection in which no DELETE statement was ever executed.
i am trying to perform insert but getting this error:
[Exception: One or more errors occurred during processing of command.
ORA-00936: missing expression]
it works for select query.
table structure is as follow
database-oracle 10g
table name-investor_info
investor_id-number
first_name-varchar
lastname-varchar
age-number
location-varchar
contact_number-varchar
email-varchar
checked-number-number
public void insert_details(string fname,string lname, int age, string location, string contactnumber, string email)
{
int id = get_id()+1;
int check=0;
string query = "INSERT INTO INVESTOR_INFO (INVESTOR_ID,FIRST_NAME,LAST_NAME,AGE,LOCATION,CONTACT_NUMBER,EMAIL,CHECKED) VALUES (#val1,#val2,#val3,#val4,#val5,#val6,#val7,#val8);";
try
{
conn.Open();
OleDbCommand command1 = new OleDbCommand(query,conn);
command1.Parameters.AddWithValue("#val1", id);
command1.Parameters.AddWithValue("#val2", fname);
command1.Parameters.AddWithValue("#val3", lname);
command1.Parameters.AddWithValue("#val4", age);
command1.Parameters.AddWithValue("#val5", location);
command1.Parameters.AddWithValue("#val6", contactnumber);
command1.Parameters.AddWithValue("#val7", email);
command1.Parameters.AddWithValue("#val8", check);
command1.CommandType = CommandType.Text;
command1.ExecuteNonQuery();
// conn.Close();
}
catch (Exception e)
{
throw new Exception(e.Message);
}
finally
{
conn.Close();
}
}//end of insert
Remove the semicolon in string QUERY
it should be
string query = "INSERT INTO INVESTOR_INFO (INVESTOR_ID,FIRST_NAME,LAST_NAME,AGE,LOCATION,CONTACT_NUMBER,EMAIL,CHECKED) VALUES (#val1,#val2,#val3,#val4,#val5,#val6,#val7,#val8)";
Remove semicolon from the string query and try like below,
string query = "INSERT INTO INVESTOR_INFO (INVESTOR_ID,FIRST_NAME,LAST_NAME,AGE,LOCATION,CONTACT_NUMBER,EMAIL,CHECKED) VALUES (#val1,#val2,#val3,#val4,#val5,#val6,#val7,#val8)";
Always specific the schema name with table name.(schemeName.tableName)
I have not worked on Oracle for a while, but have you tried replacing the # with a : for the parameters. Refer to the following artice:
http://msdn.microsoft.com/en-us/library/system.data.oracleclient.oraclecommand.parameters.aspx
I am working on asp.net(c#) project with SQL SERVER 2008. I want to update three tables using one query. Please suggest me how to do that. thnaks
You can't. Update statement works for a single table. You have to write three different queries for three tables.
You can use transaction to make sure that your update statements are atomic.
BEGIN TRANSACTION
UPDATE Table1
Set Field1 = '1';
Where Field = 'value';
UPDATE Table2
Set Field1= '2'
Where Field = 'value';
UPDATE Table3
Set Field1= '3'
Where Field = 'value';
COMMIT
For C# you can use SqlTransaction. An example from the same link (bit modified)
private static void ExecuteSqlTransaction(string connectionString)
{
using (SqlConnection connection = new SqlConnection(connectionString))
{
connection.Open();
SqlCommand command = connection.CreateCommand();
SqlTransaction transaction;
// Start a local transaction.
transaction = connection.BeginTransaction("SampleTransaction");
// Must assign both transaction object and connection
// to Command object for a pending local transaction
command.Connection = connection;
command.Transaction = transaction;
try
{
command.CommandText =
"UPDATE Table1 Set Field1 = '1' Where Field = 'value';";
command.ExecuteNonQuery();
command.CommandText =
"UPDATE Table2 Set Field1= '2' Where Field = 'value'";
command.ExecuteNonQuery();
command.CommandText =
"UPDATE Table3 Set Field1= '3' Where Field = 'value'";
command.ExecuteNonQuery();
// Attempt to commit the transaction.
transaction.Commit();
Console.WriteLine("Both records are written to database.");
}
catch (Exception ex)
{
Console.WriteLine("Commit Exception Type: {0}", ex.GetType());
Console.WriteLine(" Message: {0}", ex.Message);
// Attempt to roll back the transaction.
try
{
transaction.Rollback();
}
catch (Exception ex2)
{
// This catch block will handle any errors that may have occurred
// on the server that would cause the rollback to fail, such as
// a closed connection.
Console.WriteLine("Rollback Exception Type: {0}", ex2.GetType());
Console.WriteLine(" Message: {0}", ex2.Message);
}
}
}
}
Write three queries and put them all into a single transaction, either in a stored procedure or using TransactionScope from your c# code.
using System.Transactions;
using( var ts = new TransactionScope() ){
// execute your queries
ts.Complete();
}
Full example here: http://msdn.microsoft.com/en-us/library/system.transactions.transactionscope.aspx
One of the way will be create a procedure and inside that procedure update all the table. You can also use transaction if you want to insert all the three or non of them. Here is an example
You can also maintain transaction from C#. here is an example
Transaction Stored Procedure C#
Example
using (var connection = new SqlConnection("your connectionstring"))
{
connection.Open();
using (var transaction = connection.BeginTransaction())
{
try
{
using (var command1 = new SqlCommand("SP1Name", connection, transaction))
{
command1.ExecuteNonQuery();
}
transaction.Commit();
}
catch
{
transaction.Rollback();
throw;
}
}
}
I identified a bug in my code and I'm baffled as to how it could have occurred in the first place.
My question is, I found a skip in the database ID fields (it is an incremented identity field), indicating some records weren't inserted - which means my SQL sprocs (probably) threw an error. Thinking backwards from there, that means that my business object should have caught that error and thrown it, exiting immediately and returning back to the page's codebehind (instead it continued right on to the second stored procedure). I don't understand why or how this is possible. What's going on in respect to the code execution skipping my try catches?
Page code behind:
protected void submitbutton_click(object sender, EventArgs e){
try{
mybusinessobject.savetodatabase()
} catch( Exception ex) {
Response.Redirect("Error.aspx");
}
}
business object code:
public static void savetodatabase(){
int ID1=-1;
int ID2=-1;
//store the billing contact
SqlCommand cmd1 = new SqlCommand("SaveInfo1", con);
cmd1.CommandType = CommandType.StoredProcedure;
//...
cmd1.Parameters.Add("#Ret", SqlDbType.Int);
cmd1.Parameters["#Ret"].Direction = ParameterDirection.ReturnValue;
try
{
con.Open();
cmd1 .ExecuteNonQuery();
ID1 = Convert.ToInt32(cmd1.Parameters["#Ret"].Value);
}
catch (Exception ex) { throw ex; }
finally { con.Close(); }
if (ID1 > 0)
{
SqlCommand cmd = new SqlCommand("SaveInfo2", con);
cmd.CommandType = CommandType.StoredProcedure;
//...
try
{
con.Open();
cmd.ExecuteNonQuery();
ID2= Convert.ToInt32(cmd.Parameters["#Ret"].Value);
}
catch (Exception ex) { throw ex; }
finally { con.Close(); }
}
}
SQL Code:
PROCEDURE [dbo].[SaveInfo1]
(
-- ... parameters ...
)
AS
INSERT INTO Table1 ( ... ) Values ( ... )
RETURN SCOPE_IDENTITY
PROCEDURE [dbo].[SaveInfo2]
(
-- ... parameters ...
)
AS
DECLARE #SpecialID INT
INSERT INTO Table2 ( ... ) Values ( ... )
SET #SpecialID = SCOPE_IDENTITY()
INSERT INTO Table3 ( [ID], ... ) Values ( #SpecialID, ... )
RETURN SCOPE_IDENTITY()
Your exception handling is horrible. Never do this:
catch (Exception ex) { throw ex; }
All that accomplishes is to screw up the stack trace in the exception. It makes it look like the exception originated at the point of the throw.
Never do this:
try{
mybusinessobject.savetodatabase()
} catch( Exception ex) {
Response.Redirect("Error.aspx");
}
You don't know what exception happened. You have no idea whether or not it's safe to redirect, and on top of it all, you lose all information about what the exception was!
You should also get into the habit of implementing using blocks:
public static void savetodatabase()
{
using (SqlConnection con = new SqlConnection("Connectionstring"))
{
int ID1;
//store the billing contact
using (SqlCommand cmd1 = new SqlCommand("SaveInfo1", con))
{
cmd1.CommandType = CommandType.StoredProcedure;
//...
cmd1.Parameters.Add("#Ret", SqlDbType.Int);
cmd1.Parameters["#Ret"].Direction = ParameterDirection.ReturnValue;
con.Open();
cmd1.ExecuteNonQuery();
ID1 = Convert.ToInt32(cmd1.Parameters["#Ret"].Value);
}
if (ID1 <= 0)
{
return;
}
int ID2 = -1;
using (SqlCommand cmd = new SqlCommand("SaveInfo2", con))
{
cmd.CommandType = CommandType.StoredProcedure;
//...
con.Open();
cmd.ExecuteNonQuery();
ID2 = Convert.ToInt32(cmd.Parameters["#Ret"].Value);
}
}
}
A using block will ensure that the resource will have its Dispose method called, whether or not an exception is thrown.
Isn't the more likely scenario that someone just deleted some records from the table?
If records are deleted, their unique identifiers will not be recycled, even when new records are later inserted. You can use RESEED in SQL to reset the identity seed to 0 if you desire, but I suggest against that unless you wipe the table. Otherwise you could end up with primary key violations.
Also, make sure your column's identity seed is set to increment 1 at a time.
Your code doesn't matter, just go to Web.config and play with appropriate node:
<customErrors mode="On|Off" />
P.S.
Use the using clause to auto-close a connection, instead of manual in the finally clause
you can test the catch. just change the procedure:
PROCEDURE [dbo].[SaveInfo1]
(
-- ... parameters ...
)
AS
INSERT INTO Table1 ( ... ) Values ( ..., some_out_of_range_value_here, ....)
RETURN SCOPE_IDENTITY()
to have some hard coded out of range value (so the insert fails), and then run your application...